Obedience in perspective: Psychology and the Holocaust

Paradigm shifting

Stanley Milgram once stated that:

on the basis of having observed a thousand people in the experiment and having my own intuition shaped and informed by these experiments, that if a system of death camps were set up in the United States of the sort we had seen in Nazi Germany, one would find sufficient personnel for those camps in any medium-sized American town. (Milgram, 1979, 2nd quote)

Phillip Zimbardo’s Stanford Prison Study (Haney, Banks, & Zimbardo, 1973), seemingly confirmed in another way the idea that ordinary people could easily be transformed into evildoers by situations. The rhetorical power of the Milgram and Zimbardo studies lies in part in the apparent ease and rapidity with which ordinary people can be transformed into brutes and even killers. The entire experience of the Milgram experiment for individual research participants began and ended in a few hours. The Zimbardo study lasted only six days. The seeming ordinariness of Milgram’s participants, the random assignment of Zimbardo’s participants as prisoners or guards, coupled with the almost instantaneous transformation of these ordinary folk effected by the experiment, appeared at a stroke to moot discussion of ideology, of long-term historical, political, economic, or cultural factors in producing such behavior.

Just as B. F. Skinner had claimed to eliminate the necessity for tiresome discussions of internal motivations and rationales for explaining human behavior by appealing to nothing more than the explanatory power of reinforcement histories, so too the situationists sought to eliminate the necessity for discussions of ideology, morality, and belief with the explanatory power of the situation. On the situationist account, it simply does not take years of exposure to pernicious propaganda or authoritarian child-rearing and educational practices or ugly beliefs about other people to get ordinary citizens to abuse or even kill other citizens: it just takes a few minutes and the right (wrong) situation.

In fact, Milgram’s claim that a system of Nazi-like death camps could be readily staffed by contemporary American citizens raises an important question: If it is so easy to harness the power of destructive obedience latent in all of us, why haven’t we seen more examples outside social psychology laboratories? There appears to be no shortage of people with evil intent and access to introductory psychology texts in contemporary society. Even recent replications of the Milgram study produce compliance rates very similar to those found by Milgram (Burger, 2009), so it is not the case that general awareness of these mechanisms has reduced their potential power. And yet it is only in rare and highly specialized circumstances, such as the atrocities at My Lai or the abuse of detainees at Abu Ghraib, that the putative power of situational forces to lead to destructive behavior is invoked. Even in these rare instances the obedience and/or conformity explanation has not fared well as a legal strategy. Given that defense attorneys are practical people, one must conclude that obedience and conformity do not play as great a role in everyday life as the situationists might have us think, or that the obedience defense is not especially plausible to judges and juries.

Phillip Zimbardo’s energetic and persistent identification of the abuses visited upon Iraqi detainees in the fall of 2003 by a small group of American soldiers as real-life instances of the kind of behavior occurring in his Stanford Prison Study is a case in point (Zimbardo, 2007). Zimbardo claims that the soldiers who committed the abuses were exemplary, outstanding soldiers who were transformed by a corrosive situation and induced to treat detainees cruelly. He also claimed that the abuses were the result of “migration” of enhanced interrogation techniques, specifically from Guantanamo Bay, that were applied to the detainees in the now-famous photographs that received worldwide publicity in the wake of the abuses.

Just as Milgram’s focus on obedience does not square with the historical truth about the Holocaust, neither does Zimbardo’s focus on role-specific behavior square with the reality of what happened at Abu Ghraib. The soldiers who committed the abuses at Abu Ghraib were not randomly assigned to their roles as guards, as were Zimbardo’s research participants, and there is ample evidence that the most serious abuses were committed by individuals with significant histories of deviant and aggressive behavior. Moreover, the specific abuses that were prosecuted did not take place in the context of interrogation. Virtually none of the Iraqi detainees depicted in the infamous photographs were ever interrogated, as they were not suspected terrorists, but instead were common criminals or innocent Iraqis who had been caught up in massive sweeps by American troops (Mastroianni, 2013).

Phillip Zimbardo himself offered expert testimony at the sentencing hearing of one of the Abu Ghraib defendants (Graveline & Clemens, 2010), and Stephan Mestrovic (2007) offered similar situationist-based testimony at the trial of another of the defendants. In neither case did the situationist defense appear to be very effective on behalf of the defendants. This is in marked contrast to the reception Zimbardo’s interpretation has received among psychologists and the general public, where it has been widely accepted: it is to be found virtually unquestioned in most psychology textbooks. Why do judges and juries see things so differently from psychology textbook authors and the lay public? Perhaps because judges and juries are exposed to the facts of these cases more thoroughly and completely than textbook authors and journalists explain them.

The fact pattern associated with Abu Ghraib simply does not support a narrowly situationist explanation of these events. Like it or not, there were dispositional variables that played a role: some of the perpetrators had a history of deviant and aggressive behavior, and had these people not been present, events almost certainly would have unfolded very differently. Some of the individuals present behaved admirably and attempted to report the abuses as they happened, so clearly the situation was not so powerful as to be anything close to universally compelling.

This is not to say that contingent situational factors played no role in what happened. The failure of unit leaders at every level to establish and sustain a proper leadership climate conducive to good behavior among all members of the unit was identified as contributory by every investigation conducted on Abu Ghraib. One of the lessons to be drawn from Abu Ghraib, and one that I emphasize in my own teaching on leadership, is that good leadership establishes conditions that make it easier for soldiers to do the right thing, rather than the wrong thing. At Abu Ghraib, it was all too easy to do the wrong thing on the night shift in the hard site: poor leadership opened wide a door through which some soldiers chose to walk. Good leadership leaves fewer doors ajar, and also promotes and enables better outcomes when soldiers are confronted with difficult choices.

A lack of detailed knowledge about events such as Abu Ghraib or the Holocaust may explain why so many people readily accept the situationist explanation of these events. Many people, including many psychologists, have only the most superficial acquaintance with the details of what actually happened during the Holocaust. This can be explained by three things. First, the public understanding of the Holocaust was shaped by an incomplete and sometimes inaccurate account emerging from the Nuremberg Tribunals following the defeat of Germany in World War II and the trial of Adolf Eichmann in Jerusalem in 1961 (Arendt, 1963). Second, historical scholarship has continued to expand and refine our understanding of the Holocaust until the present day, and many people have simply not kept up with these new findings. Third, Stanley Milgram told psychologists that historical context should be ignored. In his reply to Diana Baumrind’s (1964) critical analysis of the obedience studies, Milgram argued that:

Baumrind mistakes the background metaphor for the precise subject of the investigation. The German event was cited to point up a serious problem in the human situation: the potentially destructive effect of obedience. But the best way to tackle the problem of obedience, from a scientific standpoint, is in no way restricted by “what happened exactly” in Germany. What happened exactly can never be duplicated in the laboratory or anywhere else. The real task is to learn more about the general problem of destructive obedience using a workable approach. Hopefully, such inquiry will stimulate insights and yield general propositions that can be applied to a wide variety of situations. (1964, p. 851)

Of course, it does matter “what happened exactly” in Germany, because if what happened exactly in Germany does not resemble what happened exactly in Milgram’s laboratory, then the insights emerging from the latter may or may not be relevant to the former, any more than they may be relevant to any other historical event. In fact, it is precisely our new knowledge of what happened exactly in Germany and occupied Europe and Russia that has eroded support for Milgram’s identification of destructive obedience as a primary explanatory tool for the behavior of Holocaust perpetrators. Newman and Erber [following Blass] aver that:

Milgram might have shed light on an interesting aspect of human behavior, but the phenomenon he studied might have little to do with what happened to the victims of the Holocaust or with the behavioral dynamics involved in any episode of genocide. Indeed, the idea that all, most or even many of the acts of cruelty perpetrated during the Holocaust were carried out by people who were grimly following orders is remarkably easy to disprove. (2002, p. 335)

None of the foregoing is meant to suggest that laboratory experiments must exactly replicate the natural phenomena we wish to study in order for them to be valid or interesting. It is important, however, to be careful about generalizing the findings from simplified situations constructed in a laboratory to more complicated real-world interactions. Stanley Milgram ran two dozen variations of his obedience experiment. In one variation, all or nearly all of the participants complied with the experimenter’s instructions. In others, none or almost none did. That range of compliance perhaps more faithfully reflects the historical record than the obsessive focus on the 62.5% who complied in the voice-feedback condition most commonly cited in discussions of the Holocaust (Perry, 2013).

However slow textbooks may be to recognize the fact, the paradigm is shifting when it comes to psychological thinking about the Holocaust, and genocide. While obedience has not been abandoned as an explanatory mechanism for the behavior of Holocaust perpetrators, a process is underway which promises to displace it from center stage and assign it a more modest supporting role.

Toward a more complete psychology of the Holocaust

Psychological theorizing about the Holocaust has built on Milgram’s situationist approach but remains mainly located in social psychology. James Waller (2002), Roy Baumeister (1999), and Ervin Staub (2011) have offered theoretical accounts of the psychology of genocide. These three theories usefully expand the psychology of the Holocaust somewhat beyond the narrow situationist approaches of Milgram and Zimbardo.

James Waller’s 2002 book, Becoming Evil: How Ordinary People Commit Genocide and Mass Killing, offers a psychological model intended to explain “how ordinary people commit extraordinary evil.” Waller further limits the scope of his inquiry to the rank-and-file killers:

I am not interested in the higher echelons of leadership who structured the ideology, policy, and initiatives behind a particular genocide or mass killing. Nor am I interested in the middle-echelon perpetrators, the faceless bureaucrats who made implementation of those initiatives possible. (p. 14)

Limiting the explanatory target in this way narrows the kinds of explanations one is likely to find. Waller’s focus on low-level perpetrators seemingly favors the identification of mechanisms of coercion and social influence as important while downplaying the roles of ideology and belief likely to animate the actions of the instigators, leaders, and bureaucrats who played such an important role in the Holocaust. The explanation of extraordinary human evil outlined by Waller includes four primary elements: (a) our ancestral shadow, (b) the identities of the perpetrators, (c) a culture of cruelty, and (d) social death of the victims. While Waller does allow for cultural and educational histories that might predispose some to genocidal behavior more than others, his theory relies on biological and social mechanisms that tend to universalize the potential for genocidal behavior and downplay contextual historical factors. Waller argues that situational forces are so powerful that “any deed that perpetrators of extraordinary evil have ever done, however atrocious, is possible for any of us to do – under particular situational pressures” (2002, p. 228).

Missing from this account is an understanding of why particular situational pressures might not have the same result in different historical, cultural, economic, or political contexts. The universalizing impulse central to the kind of model-making undertaken by Waller glosses over the fact that situational pressures simply are not as powerful as situationists would have us believe: it is not the case that it is possible for any of us to perform any atrocious deed ever done by perpetrators of extraordinary evil. Most Italian Jews survived German occupation, while most Dutch Jews perished under German occupation. Were the “situations” created by Nazi occupation that different in the two countries? Or were the cultural, historical, and ideological differences between these two societies more important than the situational similarities?

Roy Baumeister’s psychological explanation of genocide is based on four “roots of evil.” These are idealism, threatened egotism, instrumentalism, and sadism. Of these, Baumeister sees idealism and threatened egotism as the primary factors relevant to the explanation of the Holocaust. The architects of the Holocaust did not see the enterprise in which they were engaged as primarily destructive or anti-social. Rather, they viewed their mission in world-historic terms. The Nazi reordering of Europe, with its attendant dislocation, resettlement, and eventual extermination of millions of human beings, was undertaken to secure a brighter future: an idealistic goal. Baumeister argues that the (to the Nazis) noble end of creating a brighter future for Germans and Germany, and indeed the world as a whole was understood by them as justifying the horrific means that were eventually employed to achieve that goal (Baumeister, 1999).

Ervin Staub has written extensively about genocide and mass violence, including the Holocaust, but he has also worked tirelessly in both prevention and reconciliation efforts around the world. His theorizing is thus informed by his experience on the ground in Rwanda and Congo, as well as by the historical record of other genocides. Staub’s theory begins with difficult life conditions. Societies confronting economic or political upheaval are more prone to the development of mass violence. Difficult life conditions frustrate basic human needs: needs for security, self-esteem, and control, for example. Attempts by groups or individuals to explain and address these difficult conditions can, given the nature of human intergroup relations and individual psychology, operate to promote intergroup hostility and violence (Staub, 2011).

Many countries with histories of inter-group competition or conflict confront difficult life conditions without descending into genocide. What characteristics of societies make them more prone to such a reaction? Staub identifies cultural devaluation, authority orientation, cultural factors, an aggressive past, and the lingering effects of past victimization as risk factors.

Staub (2011) points out that dispositional explanations implicating the authority orientation or obedient character of individuals fall short in explaining the rise and initial acceptance of the Nazi movement, but that such explanations may be relevant to understanding those most immediately responsible for the physical destruction of Jews. Staub cites evidence that members of the SS may have had a stronger authority orientation than most people. In addition to personality factors that may have played a role, ideological commitment was clearly important in many of those most responsible for the killing. Finally, the vital role played by particular leaders, such as Hitler, in enabling and organizing violence cannot be ignored. Staub’s approach usefully combines sensitivity to the historical and political factors at a societal level that may enhance the risk of genocidal behavior in a country with an awareness of the individual and social psychological mechanisms that may operate to elevate or diminish the risk of such behavior in a particular case.

Do we need a psychological theory of genocide?

Psychological theorizing about genocide is complicated by a fundamental problem: the role of free will in a deterministic explanatory framework. The seemingly exculpatory nature of scientific explanations of perpetrator behavior has vexed many psychologists who do not wish to be interpreted as excusing genocidal behavior. Psychology has challenged our common-sense understanding of why we do what we do, and offered alternate causal frameworks rooted in a changing set of explanatory constructs that have in common only that they are not our conscious thoughts and beliefs (Baumeister, Massicampo, & Vohs, 2011). For Freud our behavior was driven by unconscious motivations; for Skinner, by our reinforcement histories; for Milgram, the situation. While many would agree that our internal thoughts and motivations are far from the whole story when it comes to human agency, the issue of free will matters because only a person who is free to choose an act can be blamed for that act. Surely there are constraints and limitations on our freedom to act in specific contexts, but we must admit some level of freedom if our actions are to be judged morally.

Invalidation of individual volition has two unhappy consequences. First, it renders moral judgments of behavior problematic. Psychologists often assert that their deterministic explanations of genocidal behavior should not be construed as exculpatory (Miller, Buddie, & Kretschmar, 2002), but this is wishful thinking: insofar as factors outside an individual are thought to cause particular behaviors, those behaviors cannot be subjected to the same moral calculus as voluntary actions. Second, it deflects attention away from the very beliefs and ideas that seem to play such an important role in genocide. Psychologists have become so attentive to and focused on instances of behavior in which our conscious ideas, beliefs, and motivations have somehow been disengaged or overridden that it has seemingly become difficult for many to accept that some, maybe much, human behavior actually is a result of our thoughts, ideas, beliefs, and feelings.

William James grappled with the question of free will in an 1884 lecture later published as an essay entitled, “The Dilemma of Determinism” (James, 1886/1992). James was troubled by the morally exculpatory consequences of so-called hard determinism: determinism that sees every action as the inevitable and accidental consequence of all the accumulated billiard-ball like actions that have previously inevitably and accidentally occurred. James argued that if one accepts human actions as following this same rigid model of causality, it is impossible to praise or blame the actor for them.

James was unwilling to accept the personal and social consequences of a world without praise or blame, and adopted the pragmatic solution of simply choosing to believe in choice: in free will. He did so with eyes wide open as to his inability to articulate a coherent philosophical or scientific basis for such a belief. Many modern psychologists, including some social psychologists who address themselves to matters such as genocide, find themselves in the same quandary. Few, however, are as willing as was James to admit and embrace a contradiction that lies (mostly) quietly beneath much of what we do as psychologists committed to a scientific epistemology.

Aside from the thorny issue of the moral assessment of the behavior of perpetrators, theories of genocide confront other obstacles. The data available to develop and test theories of genocide are naturally very limited. There are problems of definition, at the outset, and then serious limitations in obtaining and interpreting empirical evidence of what transpires during chaotic and violent upheavals. Memories may fade or be unknowingly or deliberately distorted.

One way to approach the explanation of genocide from a psychological viewpoint is to acknowledge in advance that psychology is ill-positioned to address the issue of why genocide occurs. We may want or think we ought to have a “theory” of the Holocaust, or of genocide more broadly, but do we really need one? Genocides apparently occur because of a combination of contingent factors, many of which lie in the domains of other disciplines, such as political science, economics, and history. Barbara Harff has built an empirical model (2003) based on political, economic, and historical variables that correctly detects many, though not all, instances of genocide since 1955. This approach is used in the context of prediction and prevention of genocide. Particular historical, political, and economic conditions at the societal level may be correlated with psychological profiles that can help to further refine our understanding of when genocide is most likely to occur.

Perhaps genocidal behavior is not psychologically extraordinary in any meaningful sense, but is readily explicable in terms already familiar to psychologists because such behavior arises from the very same mechanisms we use to explain other behavior. We don’t have a special theory to explain slavery or Jim Crow: these lamentable examples of human behavior are consistent with what we already know about prejudice, stereotyping, and so on. What psychologists are less able to explain is why slavery was eventually abolished in the United States and Jim Crow has gradually given way to sometimes halting but generally positive progress in achieving racial equality. The answer to why this happened is to be found not in psychology but in history, political science, and economics, among other disciplines. Similarly, we don’t look first to psychologists to explain why particular wars or conflicts occur, though psychologists have contributed a great deal to our understanding of human behavior in wars.

Psychologists are similarly not in the best position to answer the “why?” question about the Holocaust, but we are superbly equipped to answer the “how?” question. That is, we psychologists know a great deal about the determinants of human behavior—biological, social, psychological—and once we accept that our burden is limited to explaining how humans placed in circumstances we might label “genicidogenic” come to commit the behaviors we label genocide, our task is much easier. Why they were in such conditions in the first place is not necessarily squarely within the psychological portfolio. Moreover, “genocide” is a term fraught with definitional ambiguities and subject to political influences: agreement is far from universal on what events deserve that label, and there are sometimes political pressures to apply or withhold that label in light of the international legal consequences of invoking the term.

While it might be difficult to abandon the field of psychology to another field, think of it this way: the answer, “obedience,” is demonstrably unsatisfactory and incomplete as a response to the question, “Why did the Holocaust happen?” It is hard to take seriously Milgram’s claim that the Holocaust could easily be reproduced in any medium-sized American city, simply because the (putative) mechanism of destructive obedience exists. But in response to the question “How did the Holocaust happen?,” “obedience” is one very sensible element of a psychological response that would include other psychological mechanisms as well.

Toward a more complete psychology of the Holocaust

It is time to expand the terms of psychological discussion of the Holocaust. Rather than focusing on explanations based on finding abnormal people (characteristic of the early, dispositional era of explanation), or on finding mechanisms that make normal people act abnormally (the situationist era), or quibbling about how much each of these approaches contributes, we should instead seek to understand much of the behavior that enabled and constituted the Holocaust as normal people acting normally, or perhaps as ordinary people acting ordinarily.

This is not to suggest that psychopathology or evil intent or malign social influence play no role in genocide and mass violence: they surely do. But much behavior of perpetrators, bystanders, victims, and instigators can be understood as the consequence of normal mechanisms of perception, learning, socialization, and development operating just as we might expect. Even though behavior may occur in the context of a very special event, the behavior itself can still be quite ordinary. It is not uncommon to detect in postwar testimony of perpetrators a kind of puzzlement as to what all the excitement is about, suggesting that at the time the events seemed to be unremarkable (Browning, 1992,p. 72). Much of the Holocaust, staggering in its scale and barbarity when seen as a unitary event, was unexceptional to perpetrators in its pieces and parts. The mechanisms that produced these behaviors may also be unexceptional. Why do we need a theory of unexceptional behavior?

The approach suggested here is intended to continue the current trajectory of explanatory expansion and draw broadly on other areas of modern psychology to better understand behavior in the Holocaust. If we could go back to a time before psychology’s position on the Holocaust had evolved as it has, assemble a group of psychologists from a wide swath of the discipline and ask them what psychology can contribute to understanding the behavior of Holocaust perpetrators, what range of answers might we have on offer? What might experimental psychologists, cognitive psychologists, developmental psychologists, biopsychologists, experts on memory and perception, political psychologists, and the many more psychologists represented by the dozens of Divisions of the American Psychological Association, as well as the Association for Psychological Science have to say on the topic? Of course the Holocaust had a social dimension, but it was enacted by individuals, and surely there is something of value to be learned by treating them as individuals with psychologically relevant personal histories, and not solely as subjects of others’ influence or orders, or as creatures shaped by evolution to devalue outgroups.

The extant psychology of the Holocaust is focused at the social level: intergroup conflict, interpersonal influence. Individual psychological factors have received only fragmentary and incomplete treatment. Social categorization is often implicated as a fundamental mechanism underpinning the kind of stereotyping, prejudice, and discrimination that can lead to mass violence. But social categorization and its unfortunate sequelae are in part, at least, an expression of more basic adaptations that economize cognitive effort at the expense of complete accuracy in our understanding of our world. Fiske and Taylor (1991) used the term “cognitive miser” to describe our tendency to use heuristics and other shortcuts to achieve a good-enough solution to most problems we confront. Heuristics and cognitive biases, such as the confirmation bias and representativeness heuristic, are natural economies of cognition that can be shown to fail spectacularly in certain circumstances, often (not always) circumstances created especially for the purpose by psychologists in a laboratory. But the fact that these mechanisms exist must mean that they were once, and perhaps still are, adaptive: there must also be occasions on which cognitive shortcuts produce the right answer, or a good-enough answer.

A poignant postwar insight into the origins of one kind of thinking that contributed to the Holocaust can be found in the musical South Pacific. When Joe Cable sings “You’ve Got to Be Carefully Taught” he touches on something profoundly important: prejudice is not an inevitable aspect of human life. The implication in the song is that racial prejudice is something that is contingent, something that arises when purposely nurtured in a particular cultural environment. Nellie Forbush, otherwise the very archetype of wholesome American Midwestern purity and goodness, is afflicted with and deeply conflicted by the prejudice she has learned at her parents’ knees. It is worth remembering that concern over the perceived approval of race-mixing implicit in “You’ve Got to Be Carefully Taught” threatened the viability of South Pacific in the same America that had helped defeat Nazi Germany only four years before (South Pacific, n.d.).

Milgram and Zimbardo are correct when they suggest that the potential for evil behavior is not confined to a small, deviant sub-group of humanity. They are wrong when they narrowly locate that potential in mechanisms like obedience and role conformity. These mechanisms are an adequate explanation for some small fraction of Holocaust perpetrators, but what about the economists who earnestly planned the starvation of millions of people, or those who stood by as their neighbors were rounded up and then looted their belongings? What stands between an ordinary American (or contemporary German) from a medium-sized town and a concentration camp guard is a great deal more than a few hours or days in a social psychology laboratory. Perhaps if genocide were mainly produced by mechanisms like those engaged at Yale or Stanford, it would be simpler to prevent. But the sad truth is what makes genocide possible often seems to be the product of generations of human experience, of elaborate rational, emotional, and logical justifications that cannot be created or overcome in a few hours or days. A more complete psychology of the Holocaust can help clarify that more complex reality.

In the immediate aftermath of the Holocaust many sought to locate the source of genocidal evil in a deviant sub-group (a “them”) that could be neatly separated from the rest of us. After Hannah Arendt’s (1963) interpretation of Eichmann and the Milgram experiments, though, it seemed clear that absolutely anyone could be a genocidal perpetrator under the right situational pressures. These two species of explanation can be thought of as representing the opposite ends of a continuum: at one end, very few humans have the potential for genocidal violence; at the other end, all of us have that potential. Neither of these extreme positions is tenable any longer. Research has consistently shown that perpetrators were not, and are not, afflicted with mental disorders, nor do they constitute a readily recognizable sub-group identified by particular personality traits, so the “them” end of the continuum cannot explain very much.

On the “us” end, Milgram’s obedience studies had great dramatic and theatrical appeal, and have thus secured a place not only in academic discussions of the psychology of the Holocaust but in the popular imagination as well. The Milgram studies and Hannah Arendt’s term “banality of evil” seemingly democratized and universalized the potential for genocidal evil. After Milgram and Arendt, perpetrators could no longer be comfortably thought of as “them”: instead perpetrators could as easily be “us.” Two developments have eroded the power of that simple idea: our increasingly sophisticated understanding of the history of the Holocaust, and our increasingly sophisticated understanding of the Milgram experiments. The former development has exposed the willing and even enthusiastic participation of many more institutions and individuals in German society and the occupied territories in the Holocaust than previously thought. There is now a much wider range of behavior to be explained, and obedience of the reluctant sort demonstrated at Yale can now be seen to have played at most a small role in the Holocaust. The latter development requires us to modify our assessment of the Milgram studies themselves. New questions have been raised about the conduct of the studies and the rigor of the experimental procedures. Other papers in this special issue have addressed these concerns directly. Just as our growing historical knowledge challenges the external validity of the Milgram studies as having explanatory power for the Holocaust, recent scholarship about the studies themselves also raises questions about their internal validity.

James Waller, Roy Baumeister, and especially Ervin Staub have, to varying degrees, expanded psychological thinking about genocidal behavior beyond the relatively narrow situationist interpretation of Milgram. Much more remains to be done, however. As we learn more about the varieties of response to and participation in the perpetration of atrocities during the Holocaust, openings for psychological explanation outside the realm of obedience and social influence begin to appear. Were young people more susceptible to the Nazi message than older people? If so, why? What psychological mechanisms produced this state of affairs? Were there gender differences in support for Nazism? Did self-interest play a role in the attempt to exterminate Jews?

The idea that any of us could be transformed into genocidaires in a few hours in a social psychology laboratory is wrong. But it is the case that growing up a certain way, in a particular culture, steeped in destructive ideologies can produce people who will commit terrible acts of destruction. Understanding that a more complicated, long-term process is the task ahead of psychologists interested in explaining the Holocaust, and this task will necessarily draw on psychological knowledge quite broadly.