Our topic today is an ugly one, one that we all hate to think about, to read about, to listen to. It is Evil. It is not pleasant to think about the nature of evil, of collective violence, of genocide, ethnic cleansing, mass rape, brutal tortures, and bestial acts of human against human that challenge our basic conception of human nature.
From early Christian history and the personification of Lucifer as the embodiment of evil, the dispositional analysis of evil has focused analytical attention on identifying those individuals who are evil by nature, and indeed there are some people who have directed collective violence, such as Hitler, Stalin, Idi Amin, and Pol Pot. But those evil men are dead, yet evil continues.
It is more profitable, I believe, for us to focus our analytical energies on understanding their followers—why they killed for their leaders, or why, once they started the killing, their leaders became irrelevant, and that once the machinery of mass murder had been installed and lubricated, it required only persistent dedication to one's job and the knowledge that it is being executed effectively. We don't need evil demons for those deeds, only compliant workers or willing soldiers.
What is that line, that cosmic boundary, that one crosses to go from being a good person, a dutiful citizen, to a mass murderer with no conscience for evil deeds and no remorse for destroying human lives? And how is this line maintained? What would it take for you to slide across it? We want to believe it is impermeable, with us here forever and them over there permanently, when in fact, that line is permeable—we good ones could become those bad ones.
My modest task is to outline some of the psychological processes that I believe are involved in human evil, more specifically, to share with you some ideas I have about the social psychological strategies and tactics that can facilitate the transformation of good, or at least ordinary, people into monsters, who are perpetrators of evil. I try to counteract the fundamental attribution error, the human tendency to overemphasize the dispositional while simultaneously underplaying the situational. I do so by demonstrating that seemingly trivial aspects of social situations can influence our behavior in profound ways, more so than we can imagine or give credit for—in the extreme—that can make us do the unimaginable.
Imagine a traitor is sentenced to death by firing squad and the government wants to recruit his peers, civilians, to shoot him. Few volunteer. If, however, they add a condition that only one of six guns will have a real bullet in the chamber, thus each gun would have only a small likelihood of being the lethal weapon, typically more volunteer. Why? Employing the tactic of diffusion of responsibility greases that line, and some good people are ready to slide across the boundary, become killers for the state.
One Principle for Line Bending, More?
Indeed there are many situational variables that subtly change key elements of the behavioral space and shift the behavioral dynamics away from standard operating procedures toward novel relationships and contingencies for which the actor does not have a prepared script to guide behavior down familiar paths—and so becomes more vulnerable to the demands of the immediately present behavioral context. Let's see what this means in four experiments.
Obedience to Authority
Stanley Milgram's Jewish heritage contributed to his intellectual and personal concern for finding an answer to the question: "If Hitler asked you, would you execute a stranger?" Despite cultural differences, historical-setting differences, and the absence of the charismatic power of Adolf Hitler, could it be demonstrated that thousands of us could be led down the same path as them, to inflict extreme harm on another human being? How could an experimental research paradigm provide answers to such vital questions?
Milgram recruited more than 1,000 participants from all walks of life to be a part of his studies. They would arrive individually in the lab and be told they were helping psychological science to find new ways to improve memory through punishment and thereby help in the education processes (Milgram, 1974).
Teacher, the role assigned to the participant, helps the Experimenter, who is wearing the white lab coat, symbolic of his status, to connect the Learner, a lovable middle-aged man, to the electrical shock apparatus; the victim is in an adjacent room. On the first trials, learning is going well, the word associates are being recalled, and Teacher says, "Good, fine."
But then the Learner starts making errors and punishment begins, first small, then ever escalating. As it does, the Learner begins complaining, then yelling and screaming. The Teacher is upset, having never imagined it would come to this. Turning to the Experimenter, Teacher dissents, indicating he or she does not want to continue, which is cast aside as the Teacher is reminded of the contract agreed to previously. More shock, more yelling, complaining of a heart condition, and insisting he wants to quit. "Who will be responsible if something bad happens in there to the Learner, Sir?" asks the Teacher. "I will; please continue, Teacher." At 375 volts the Learner screams, there is a loud thud, and then only silence from the shock chamber thereafter. Teacher is now really distressed (the women often cry, the men wince), says the experiment should be terminated because the Learner has stopped responding.
Not so easy. "Remember the rules," reminds the Experimenter, "Failure to respond is an error, and all errors must be punished immediately with the appropriate level of reaction, Teacher." And there are five more higher levels possible to the extreme of 450 volts.
Before starting his research, Milgram invited 40 psychiatrists to predict the percentage and type of person who would indeed go all the way in this study that he described to them in detail. In their collective wisdom based on their medical training in dispositional, individualistic analysis, they concluded that fewer than 1% of the Teacher-Participants would go all the way, and they would be the sadists.
So the psychiatrists were all wrong; everyone's predictions were all wrong. Not 1% compliance, but 65% compliance—two thirds of the subjects went all the way up to the final level. That quantification of evil went as high as 90% or down to 10% compliance across 18 studies in Milgram's research program—each study varying one aspect of the social situation.
So what is the answer, that my friend and New York High School classmate, Stanley Milgram, found to his initial question? Yes, sadly, by exploiting our deeply ingrained learned behavioral patterns of obedience, ordinary people could be seduced, initiated into behaving in ways that might lead to killing innocent victims.
Let me outline the lessons I think we should take to heart from this experiment, which help us to understand some fundamental process in making that line between good and evil more permeable:
Start with an ideology (justifying beliefs for actions).
Use authority to legitimate that ideology.
Give people desirable roles to play with meaningful status.
Have rules that channel behavioral options.
Employ semantic distortion to disguise truth (help = hurt).
Arrange for contractual agreement with the game rules before the game begins.
Make situation give permission to engage in usually taboo acts.
Make initial harmful act minimal, minor, trivial.
Enable subsequent acts to escalate only gradually, minimally, but their cumulative impact can be deadly.
Displace responsibility for consequences on authority or others.
Get actors involved in action, in technology, in details, without time to think through the meaning of their actions.
Don't allow usual forms of dissent to work; undercut them so dissent does not lead to disobedience.
Put actors in novel setting, without familiar referents.
Have authority transform gradually from just to unjust.
Give no training in how to challenge unjust authority.
Do not provide apparent means for exiting the situation.
In much evil in the world there is no strong leader always present insisting that you must go on to do evil; that would be inefficient. Instead, the leader creates conditions that facilitate evil in his absence, conditions of deindividuation:
Take away people's sense of uniqueness and individuality, because that encourages spontaneity, rebelliousness, and independence.
Do so by submerging them in groups.
Put them in uniforms.
Disguise them with hoods or masks.
Can we demonstrate experimentally that making people feel anonymous will facilitate their crossing that line over to the evil side?
To go from that imagined reality to behavioral reality, I did a series of experiments on the concept of deindividuation (Zimbardo, 1970), in which college students' identities were concealed in a small-group setting, their names were replaced by numbers, their clothing covered with baggy lab coats, and their faces covered by hoods, or masks. Their task was to shock other participants who were allegedly in a related experiment on the effects of stress on creativity; these other participants tried to be creative while being stressed by these random shocks that the members of the observing group were administering to them. In the first study,
I stacked the cards against an easy outcome by having groups of women be the agents of pain for other women who were their victims. Later studies used males and military personnel, with comparable results. Those in a deindividuated state delivered twice as much shock as controls who were individuated.
My colleague, Albert Bandura, and his students continued this line of research by extending the basic paradigm here to study the minimal conditions necessary to create dehumanization (Bandura, Underwood, & Fromson, 1975). What they manipulated was only the actors' perception of their victims—no authority pressures, no induced anonymity. A group of college students expected to help train another group of students from a nearby college by collectively shocking them when they erred on the task.
Just as the study was about to begin, the participants overhead the assistant tell the experimenter one of three phrases—Neutral: "The subjects from the other school are here." Humanized: "The subjects from the other school are here; they seem 'nice.'" Dehumanized: "The subjects from the other school are here, they seem like 'animals.'" Mind you, they never saw those other students, or heard anything directly from them, it is only this label that they had to go on in imaging what they were like.
On trial one, the manipulation failed to have a differential effect on their aggression, and had the researchers ended the study there, we would conclude that dehumanizing labels have no behavioral impact, but as the study wore on, it had a major impact. The boys, who imagined their victims as "animals," progressively elevated their shock levels over each trial after the first, significantly more than the neutral control. Humanizing labels helped to reduce the aggression significantly below the level of the neutral control.
When the participants were interviewed subsequently about why they behaved as they did, what the researchers found was that the experimental condition enabled them to become morally disengaged, to activate a set of psychological mechanisms that minimized the evil of their deeds, while justifying it in a variety of ways. So a one-word label can create a stereotype of the victim, of the enemy, that also lowers the height of that line between good and evil and enables more good people to cross over and become perpetrators.
The external validity of this construct has been demonstrated by the lynching and burning alive of untold numbers of Black men in the U.S., usually based on fears of their sexual conquest of White women, facilitated by the dehumanizing label of "nigger."
Stanford Prison Experiment
What happens when we aggregate many of these processes that contribute to the making of perpetrators: dehumanize victims, deindividuate potential perpetrators, put them in a new strange, anonymous environment, and give the perpetrators total power and render the victims powerless? The answer is: the Stanford Prison Experiment, which contains further lessons in the transformation of good people into evil perpetrators (Haney, Banks, & Zimbardo, 1973; Zimbardo, Haney, Banks, & Jaffe, 1973). In this study, a group of ordinary college students were divided arbitrarily into "prisoners" and "guards." Hidden cameras captured what happened: the "guards" became more sadistic, devising cruel mental tortures, while the "prisoners" either broke down or succumbed in cowed and mindless obedience. I had to end this 2-week study after only 6 days because it had become too real, too volatile. [Editor's note: The website www.prisonexp.org (Zimbardo, 1999), on which the photograph above appears, offers an Internet slide show about this experiment.]
Violence of War
We have seen how social psychologists have isolated variables that can contribute to the creation of perpetrators, but how do leaders of nations do it? How do national policies and agencies do it to transform idealistic, innocent children into those who hate other people enough to want them all to cease to exist, to exterminate them, to eliminate them and their very memory from the face of the earth forever? Briefly, it involves agents and agencies of socialization, propaganda, and education.
Socialization. Socialization means that parents and adult caretakers shape the values and ideas of the next generation. It is the basic means of civilizing children, of transmitting the lessons of the past to the current generation, of modeling what the culture says is the right way to behave in order to be a good person. When parents become agents of the state whose agenda includes learning to hate select others, and in doing so provide compelling reasons and examples for their children to believe in and emulate, that is one powerful way for ruling elites to spread their ideals and political agendas across generations to create a youth that will fight and die for their cause.
Propaganda. Central to genocide is the psychological and sociological construction of the concept of the enemy, an abstraction into which the propagandist can embed all the fears and loathing of the citizenry, all their primal anxieties about survival and well-being. Every nation that goes to war must first construct the face of the enemy for its soldiers to want to kill and its citizens to want to work and sacrifice to prevent its takeover of their land, home, way of life, women and children, and even their god. Recall Bandura's research on the power of a dehumanizing label.
Education. All nations educate children to learn information that the state believes is vital for them to know. What happens when that educational process becomes distorted so that biases and self-serving values prevail and knowledge of the truth is suppressed? It happens to some extent in most countries when they fail to include the history of their atrocities or defeats in history texts. Beside these omissions, education is compromised when teachers and text writers are subjugated to powerful national forces that instruct them to teach hate and lies and falsehoods to children as if they were merely facts about the enemies of the state.
Where is That Elusive Line?
So what have we learned so far about that line between good and evil? Alexander Solzhenitsyn offers an insight into where it can be found: "The line between good and evil lies in the center of every human heart"--not in some abstract moral, celestial space, but right here in each of our individual and collective beings.
We have seen that education can be corrupted and turn out mindless nazis and perpetrators of evil, when the state controls what students and teachers must think. It is imperative that we each make a personal commitment never again to allow education to be perverted into a tool for prejudice, an instrument for demeaning human nature, or an intellectual weapon for justifying the evils of inhumane treatment of our brothers and sisters of any race, religion, ethnicity, or political persuasion. Education must be our salvation, not our damnation.
Evil is not simply in the past and in far-off lands. There are still forces in our own country and throughout the world that promote evil. We must not make them our enemies in the abstract, but monitor their ideas and actions in the concrete, and do all in our power to oppose their distorted values by promoting human understanding, compassion, and commitments that foster peace at home and abroad. You must sustain a sense of moral and social intelligence, to always think critically, to stand up for your beliefs regardless of the immediate consequences, to oppose unjust authorities and reveal the true nature of fools. I urge you to make a life goal to be heroic, to learn how to resist undesirable social pressures and be willing to stand alone. Otherwise, we join ranks with the brute beast and get in line as passive witnesses to the next generation of dictators and perpetrators of evil.
Recall the UNESCO Charter, which declares: "Since wars begin in the minds of men, it is in the minds of men that we must erect the ramparts of peace."
Thus, the first step in preventing genocide begins by promoting peace, love, and understanding in your minds and in mine. It is our first line of defense against evil, and it is the source of strength we all need to resist the ever-present, pervasive, powerful forces in the world that would lure us across that seductive line to descend into the realm of the next generation of evil perpetrators. Resist those pressures, and help others to also dissent, disobey, rebel.
So go in peace, Shalom.
Bandura, A., Underwood, B., & Fromson, M. E. (1975). Disinhibition of aggression through diffusion of responsibility and dehumanization of victims. Journal of Personality and Social Psychology, 9, 253-269.
Milgram, S. (1974). Obedience to authority: An experimental view. New York: Harper & Row.
Haney, C., Banks, W. C., & Zimbardo, P. G. (1973). Interpersonal dynamics in a simulated prison. International Journal of Criminology & Penology, 1, 69-97.
Zimbardo, P. G. (1970). The human choice: Individuation, reason, and order versus deindividuation, impulse, and chaos. In W. J. Arnold & D. Levine (Eds.), Nebraska Symposium on Motivation: Vol. 17 (pp. 237-307). Lincoln: University of Nebraska Press.
Zimbardo, P. G. (1999). Stanford Prison Experiment Slide Show [On-line]. Available: http://www.prisonexp.org
Zimbardo, P. G., Haney, C., Banks, W. C., & Jaffe, D. (1973, April 8). The mind is a formidable jailer: A Pirandellian prison. The New York Times Magazine, 122, 38-60.
This article is condensed and adapted from Dr. Zimbardo's Psi Chi Distinguished Lecture, "The Psychology of Evil: Seducing Good People Into Evil Deeds," delivered on March 31, 2000, at the annual meeting of the Southeastern Psychological Association in New Orleans, Louisiana.