The Lucifer Effect: Understanding How Good People Turn Evil by Philip G. Zimbardo
Reviewed by Matha Nussbaum
Times Literary Supplement
In August 1971, the Stanford University psychologist Philip Zimbardo and his team of investigators selected twenty-four young men to participate in their study of the psychology of imprisonment. The men, only a few of whom were students, had answered an ad placed in both the student newspaper and the local town daily that offered subjects fifteen dollars per day for two weeks to participate in a study of "prison life". The successful applicants were randomly assigned to the roles of prisoner and guard, fifty-fifty. Prisoners were to stay in the prison for the entire two weeks; guards served in eight-hour shifts, three groups per day. Thus began the now famous Stanford Prison Experiment.
The prison was built in University facilities, after local police refused to allow the use of the real town jail. They did, however, agree to "arrest" the future prisoners, coming unannounced to their homes in a way that enhanced the verisimilitude of the situation. Because Zimbardo, who had been teaching a course on "Psychology of Imprisonment", initially conceived the study as an investigation of the isolation and loss of individuality that occur during imprisonment, he gave the prisoners no detailed instructions, although he initially told them that they, like the guards, were free to leave the experiment at any time (forfeiting all the cash). He also assured them that there would be no physical abuse. (This assurance proved false, since guards were permitted from the beginning to deprive prisoners of sleep, a very damaging form of physical abuse.) Guards, by contrast, initially seen as "ensemble players" whose role was to help Zimbardo study the prisoners, were given a detailed "orientation". Zimbardo told them that in order for the study of prisoner psychology to be successful, they had to play their roles with vigour. He urged them to create an experience that included frustration, fear and loss of control. "In general, what all this should create in them is a sense of powerlessness. We have total power in the situation. They have none. The research question is, What will they do to try to gain power, to regain some degree of individuality, to gain some freedom, to gain some privacy." He told the guards that initially the prisoners would think of the situation as just a game, but "it was up to all of us as prison staff" (Zimbardo doubled as head experimenter and prison superintendent) "to produce the required psychological state in the prisoners for as long as the study lasted". From the beginning, the prisoners were rendered anonymous, made to wear baggy uniforms and stocking caps that concealed their hair; they had to be referred to by number rather than by name. Guards were required to wear reflecting sunglasses, which inhibited any human connection with the prisoners.
In very short order, the situation began to go bad. The prisoners (deprived of sleep) began to exhibit symptoms of depression and dislocation. Guards, meanwhile, engaged in acts of humiliation, which escalated as several aggressive guards took the lead and the more sympathetic guards failed to protest. As time went on, one prisoner, either cracking up or pretending to, succeeded in dropping out of the experiment; two others engaged in individual forms of resistance against the guards, but, much to Zimbardo's disappointment, nobody organized any group protest. On the fifth day of the expected two weeks, Zimbardo called a halt to the experiment. His future wife, the psychologist Christina Maslach, shocked by the abuse she witnessed, persuaded him that it was unethical to allow the experiment to continue.
In Zimbardo's new book, The Lucifer Effect, the shocking events of the SPE (later documented in the film Quiet Rage) provide the lead-in to a detailed examination of psychological research showing the power of situations to overcome people's better judgement. Zimbardo usefully describes a large body of research: Solomon Asch's research on perceptual judgement, which documents the power of peer pressure to lead people to make statements about lines and shapes that they can easily see are untrue; Stanley Milgram's experiments on authority, often replicated in many countries, which showed that about three-quarters of subjects would administer a shock labelled as seriously harmful to a person who was supposed to be a subject in an experiment on learning, if ordered to do so by the researcher; and a host of less famous but equally convincing experiments, all showing disturbing and even cruel behaviour by ordinary people. One particularly chilling example involves schoolchildren whose teacher informs them that children with blue eyes are superior to children with dark eyes. Hierarchical and vindictive behaviour ensues. The teacher then informs the children that a mistake has been made: it is actually the brown-eyed children who are superior, the blue-eyed inferior. The behaviour simply reverses itself: the brown-eyed children seem to have learned nothing from the pain of discrimination.
Zimbardo concludes that situational features, far more than underlying dispositional features of people's characters, explain why people behave cruelly and abusively to others. He then connects these insights to a detailed account of the abuses by United States soldiers at Abu Ghraib prison, where, he argues, the humiliations and torments suffered by the prisoners were produced not by evil character traits but by an evil system that, like the prison system established in the SPE, virtually ensures that people will behave badly. Situations are held in place by systems, he argues, and it is ultimately the system that we must challenge, not the frequently average actors. He then sets himself to analyse the features that make systems and situations bad, and to suggest ways in which they might be remedied.
The Lucifer Effect focuses on the SPE and its aftermath. The story Zimbardo tells, however, is not well served by his own experiment, which has not been replicated and which is profoundly flawed. Asch and Milgram (like most psychologists doing this sort of work) do not inform subjects about the real object of their study. Subjects think that they are engaged in a study of perception or of learning. In this way, they are prevented from playing into a preconceived role or gratifying the experimenter's wishes. Zimbardo's subjects, by contrast, not only knew what he was studying, but were even encouraged to act their roles to the hilt to make the study work. The "orientation" given the guards is particularly problematic, in the light of their subsequent behaviour.
Much though we may suspect that some of the behaviour that emerged in the SPE was behaviour that a non-fictional situation of similar type would also have elicited from those people, solid scientific results cannot be built on such self-conscious role-playing. (When the guards wanted to take off the reflecting glasses that dehumanized their relationship with the prisoners, they were reminded that they were required to wear them as a part of the experiment.) There are other problems. The drama of arrest and imprisonment, for most prisoners, is centrally a drama of guilt and innocence. If a prisoner is innocent, he is preoccupied with proving his innocence; if he is guilty, he is probably trying to assess the weight of the evidence against him and to decide whether a guilty plea is his best strategy.
Finally, there is the drama of Zimbardo's own dual role as investigator and prison superintendent. He is much too emotionally involved in the outcome, and too present on the scene, steering his actors around, for the resulting behaviour to be scientifically reliable. He himself criticizes this aspect of the SPE, but he doesn't fully see how thoroughly his own personal drama permeates the entire set of events -- in particular the fact that the experiment was brought to an end because his own future wife (to whom he is still married, more than twenty-five years later, and whom he describes as "the serene heroine of my life") got very upset with him.
All in all, then, the reader should regard the SPE as a side issue, a curiosity. Its conclusions are very likely true, or at least somewhat true, but it did little to establish them. It is not surprising that Zimbardo -- still, thirty-five years later, with much investment in the significance of the SPE -- devotes about half of this book to it, but the real work of establishing his conclusions is done in other research that he describes. He is at his best, then, when analysing the current state of our knowledge about the role of situations in eliciting bad behaviour. Research has amply confirmed that people of many different kinds will behave badly under certain types of situational pressure. Through the influence of authority and peer pressure, they do things that they are later amazed at having done, things that most people think in advance they would never themselves do.
Zimbardo's first plea, appropriately, is for humility: we have no reason to say that atrocities are the work of a few "bad apples", nor have we reason to think that they are done only by people remote from us in time and place. We should understand that we are all vulnerable, and we should judge individuals, accordingly, in a merciful way, knowing that we don't really know what we would have done, had we faced similar pressures. His second appropriate plea is that we learn to "blame the system": namely, to look at how situations are designed, and to criticize people who design them in ways that confront vulnerable individuals with pressures that human beings cope with badly. Zimbardo served as an expert witness on behalf of one of the officers accused of presiding over torture at Abu Ghraib, and his point was that we must not think that this is the work of a few unusual "defectives". We must understand that good people can do bad things under pressure, and we must learn how to structure situations so that they do not put such pressures on individuals. In short, he calls for collective responsibility -- not as a total replacement for personal responsibility, but as its necessary concomitant, if people are not to be faced, again and again, with demands to which they are very unlikely to respond well.
Zimbardo sometimes speaks as if situations are all that is important, and the insides of people explain nothing at all. That is clearly a wild over-extrapolation from his data, which support a much more qualified view. Even if people in these experiments had all behaved in precisely the same way, we could not conclude that their inner psychology supplies nothing at all: for they might be motivated by common human emotions and tendencies, one or many, and it would then be extremely important to know what these forces are. Fear and insecurity have often been linked with aggressive behaviour -- at least as long ago as the days of Jean-Jacques Rousseau, who concluded that education should aim to give people strategies to cope with life's uncertainties, and, importantly, to make them understand how compassion and cooperation can support our intrinsic weakness. Recent psychological research on disgust, by Paul Rozin and others, suggests a connection between aggression and an inability to tolerate one's bodily nature, with the ominous forecasting of death and decay that its effluvia suggest. Again, thinking about that source of aggression has consequences for education, not far removed from those Rousseau emphasized: we would encourage people not to repudiate their animality, but to understand that we are all mortal together and must learn strategies of mutual aid. Yet other research suggests that deformed images of masculinity are to blame for at least some of the readiness for aggression that we see around us: people are brought up to think that a "real man" should not need others, should not ever be weak, should not be human, really. So they react with violence against a reminder of their weakness. This research, again, suggests a Rousseauesque conclusion: we should bring up young men (and women) to understand that mutual interdependence and caring are not shameful or "unmanly". These intersecting areas of psychological research are of vital importance to Zimbardo's overall project, but he ignores them, suggesting that situations are all there are. Situations, however, don't elicit bad behaviour from a stone, or even an elephant. So what is it about many humans that makes them vulnerable? Zimbardo should press this question.
People, moreover, are not all alike. The research described by Zimbardo shows a surprising level of bad behaviour in the experimental situations, but nothing like uniformly bad behaviour. First, there are active perpetrators and fearful but humane collaborators. Both of these are morally defective, but in different ways. Finally, there are whistle-blowers who do have the strength to challenge the system, and Zimbardo devotes his final chapter to the characteristics of such people. So he himself knows that the individual does matter, and he is actually very interested in asking not only how situations can be better designed but also how people can be brought up to be good actors in bad situations. (Unfortunately, the sort of self-report questionnaire used by psychologists before such experiments can tell us little about subtle differences in upbringing and education that contribute to these individual differences.) What would a good society focus on, according to Zimbardo, if it wanted people to be capable of behaving well under situations of moral stress?
First of all, he calls for a society-wide emphasis on critical thinking, beginning in childhood. From their earliest days, children should be encouraged to think about the traditions and norms that govern their lives, and to ask uncomfortable searching questions. We need a culture of timely whistle-blowers, and we will only get this, he rightly argues, if we encourage Socratic questioning of authority both in the family and in the classroom. Beyond this, Zimbardo emphasizes the importance of fostering a culture of personal accountability. As anyone who has ever violated the speed limit knows, when people are anonymous or think they will not be seen, they are more willing to do bad things: so, making people aware of themselves as individuals, with personal responsibility, makes them less likely to act on aggressive impulses that they might have. Training individual responsibility is both a developmental and a situational matter. Zimbardo emphasizes the latter (foster structures that promote individual accountability), but we, with John Stuart Mill, should also pursue the former, thinking of ways to bring up children so that they are ready and eager to see themselves not as the anonymous heirs of a tradition, but as individuals. We should promote the perception of others as individuals: when people are led to see others as anonymous members of a group, they are more willing to do bad things to them than when they see them as individuals with names and distinctive histories.
Philip Zimbardo does not focus on emotional development, but it is surely a key part of the future of any society that is going to refuse to go down the road of the SPE and Abu Ghraib. What the guards in the experiment crucially lacked, when they lacked the ability to see the other as human, was empathy and its close relative, compassion. Compassion, as Daniel Batson's wonderful research has shown, is closely linked to the ability to follow the story of another's plight with vivid imagination. Situations can certainly encourage this ability, as Batson's experimental situation did. Nonetheless, the imagination is a muscle that gets weak from routinized thinking and strong from vigorous challenges, and this suggests a vital role for the arts and humanities in any curriculum for good citizenship.
Let us hope that The Lucifer Effect, which confronts us with the worst in ourselves, stimulates a critical conversation that will lead to more sensible and less arrogant strategies for coping with our shared human weaknesses.
Martha Nussbaum is Professor of Law and Ethics at the University of Chicago and a visiting professor at Harvard University. Her publications include The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy, 1986, Hiding From Humanity: Disgust, Shame, and the Law, 2004, and Frontiers of Justice: Disability, Nationality, Species Membership, 2006.