When is 'obedience' obedience? Conceptual and historical commentary. Neil Lutsky.

This article reassesses the role of obedience to authority in the Milgram experiment and in the Holocaust. I argue that the term "obedience" can be used to both describe and explain the behavior of subjects in Milgram's experiment, and that a failure to distinguish the two uses conceptually has led to an inflated sense of the extent to which Milgram's experiment demonstrates underlying obedience to authority. The article also reviews empirical evidence and alternative analyses to show that when authority influences behavior in the experiment, it may do so for reasons other than the subjects' felt obligation to obey. Finally, I suggest that contemporary history presents a more complex and problematic view of the Holocaust than that implied by social psychology's application of obedience to authority.

These questions, posed by Abraham Lewin in his diary of life and death in the Warsaw Ghetto (posthumously published in 1988), haunt a social psychology of our times. To its enduring credit, the work of Stanley Milgram on what we commonly call obedience to authority stands out as one of social psychology's signal attempts to address Lewin's pleas, as recent scholarship on the Holocaust demonstrates (see, e.g., Bauman, 1989; Blass, 1993; Browning, 1992; Katz, 1993).

Social scientists often take Milgram's research to suggest how strongly individuals adhere to a felt duty to obey, so much so that normal people can commit monstrously inhumane acts at the direction of authority. For example, Holocaust historian Christopher Browning (1992) cites Milgram's conception of obedience as "deference" in his attempt to explain the murders of Jews in Polish villages by the "ordinary men" of a German Order Police unit. And sociologist Zygmunt Bauman argues Milgram demonstrated that "cruelty . . . correlates very strongly indeed with the relationship of authority and subordination" (1989, p. 153). Milgram himself attributed "the extreme willingness of adults to go to almost any lengths on the command of an authority" (1974, p. 5) to their "sense of obligation" (1974, p. 6).

In what follows, I want to suggest that this emphasis on obligatory obedience in the Milgram experiment has led us astray in understanding behavior in the experiment, obedience outside the laboratory, and destructive behavior in the Holocaust. Specifically, I will advance the following four related claims:

1. "Obedience" as a description must be carefully distinguished from obedience as an explanation. Although the behavior of subjects in the Milgram experiment can be described as obedient in the specific sense that subjects ultimately tended to act in a manner consistent with the experimenter's instructions, that does not mean that obedience to authority explains why subjects acted as they did.

2. Behavior described as obedient may result from various influences of an authority. These include respect for the authority's presumed expertise, awareness of the authority's potential exercise of reward and coercive power, or felt obligation to obey a legitimate authority. Milgram emphasizes the last of these, but empirical evidence challenges his account. I argue that obedience derives from obligation only under highly unusual circumstances, although obedience is often justified after the fact in terms of obligation.

4. The model of obligation-based obedience may convey a misleading implicit history and explicit psychology of the Holocaust. Although obligatory obedience to commands from above played an undeniable role in the Nazi regime, historical scholarship also documents the extent to which the direction and implementation of the Holocaust were forged from below.

Although space limitations make it impossible to treat these points as thoroughly as they deserve, I hope what follows calls into question prevalent assumptions about the nature and role of obedience to authority in the laboratory and in life.

Obedience as a Description Distinguished from Obedience as an Explanation

What does it mean to label studies in the tradition of Milgram's as "obedience" studies? This is a fundamental and persistent question in the literature on the Milgram experiments (e.g., Milgram, 1965a; Miller, 1986), one that echoes through discussions in the present issue (e.g., Hamilton & Sanders; Miller; Miller, Collins, & Brief). Most often, treatments of this labeling question treat obedience as a form of social influence and contrast it with alternatives such as conformity or compliance. However, I want to draw attention first to a more basic conceptual contrast: between the use of "obedience" as a description of behavior in the Milgram experiment and its use as an explanation of that behavior.

When an individual acts in a manner that is consistent with a properly authorized command, instruction, or rule, we commonly describe that behavior as obedient to authority. Milgram justified his use of the label similarly when he argued that obedience occurs when a person in a subordinate position in a hierarchical social relationship "does what another person tells him to do" (1965a, p. 58). In the context of the Milgram experiment, a subject's obedient behavior, operationalized as the number of serially increasing shocks that subject administers, represents an outcome consistent with the study's specified procedures (in most conditions) and the experimenter's ongoing instructions and prods. Most subjects in Milgram's experiments and in basic replications of those experiments (reviewed by Miller, 1986) were obedient in this descriptive sense.

What we call the influences responsible for why people come to act in a manner that matches the specifications of an authority's order or instruction represents a different kind of labeling problem from the descriptive one just considered. That requires a close analysis of whether individuals encountering authority act as they do because of the authority's general influence and, if so, due to distinctly authoritative features of that influence characteristic of obedience. In 1965, Milgram differentiated such descriptive and explanatory senses when he wrote, "to obey and to disobey, as used here, refer to the subject's overt action only, and carry no implication for the motive or experiential states accompanying the action" (1965a, p. 58). Nonetheless, the term "obedience" easily lends itself to assumptions about why individuals act as they do in the face of authority.

"Obedience" Explained by Obedience

For Milgram (1974), behavior is due to obedience to authority when it derives from an obligation to obey that is associated with occupancy of a particular type of role, that of an agent in a legitimate authority-agent relationship (see also Kelman & Hamilton, 1989). This state of perceived obligation, which Milgram termed the "agentic state," theoretically results in psychological attitudes (e.g., a heightened attention to the authority, a focused drive to fulfill instructions competently, and a shift of personal responsibility to the authority) that lead individuals to act as they have been ordered by an authority. According to Kelman and Hamilton (1989) and Milgram (1974), such obligatory obedience takes precedence over personal preference or, in Milgram's terms, normal autonomous functioning.

However, individuals may follow authority for reasons other than their belief that they are obligated to do so. They may be respecting the expertise of authority, what Kelman and Hamilton (1989, pp. 128-129) call professional authority. Or obedience may be due to a common source of influence: the authority's potential exercise of the concrete power to punish transgressors and to reward the obedient. Although Milgram's experimenter was an expert of sorts, he lacked concrete coercive power. This latter fact (but not the former) reinforces Milgram's argument that authority influence was due to subjects' felt obligation to obey.

Milgram's casting of authority has been broadly influential in social psychology, even though his "agentic state" terminology has not been widely adopted. When we invoke the notion of "blind obedience," for example, we are most often implicitly attributing behavior to a sense of role-based duty. Authority itself may promote the idea that people obey out of strong obligation, an ideological claim social psychological accounts may unwittingly reinforce. And individuals accused of wrongdoing may seek protection in assertions that they were just following orders or only doing their jobs - claims often presented as elements of obedience made blind by obligation.

Does empirical evidence support the assumption that obligatory obedience is centrally responsible for subjects' behavior in the Milgram studies? An important indirect finding challenges Milgram's analysis. Milgram believed agentic subjects would attribute responsibility for their actions to the authority they obeyed. In Milgram's view, "Unable to defy the authority of the experimenter, they [obedient subjects] attribute all responsibility to him" (1974, p. 8). However, Milgram's own responsibility clock data (1974, pp. 203-204) showed that obedient and disobedient subjects attributed the same amount of responsibility to the experimenter. Moreover, Mantell and Panzarella (1976) found no correlation between the amount of responsibility attributed to the experimenter and if and when subjects ended their participation in a Milgram-like study.

What also remains unclear within the purview of a parsimonious obligation account is why authority influence is so strongly attenuated by peer behavior, why even those subjects thought to be agentic resist authority influence so persistently (Modigliani & Rochat, this issue), and why authority influence depends so strongly on the immediate presence of the experimenter and on other binding factors. Moreover, although Milgram asserted that obedience "is maintained through the simple assertion by authority that it has the right to exercise control over the person" (1974, p. xiii), interactions represented in Milgram's film of his study (Obedience, 1965b) and in transcripts in his book (e.g., 1974, pp. 48, 51) suggest the reverse, namely that when the experimenter explicitly reminds subjects of their agentic obligations ("You have no other choice"), subjects are most likely to terminate their participation in the study.

Toward a Broad Social Psychology of "Obedience"

The two sets of influences discussed above illustrate that authority may substantially affect individual behavior for reasons other than an agent's sense of duty, via processes quite common to social behavior. In contrast, when we treat Milgram's studies primarily as obedience experiments, we may miss social psychology's larger relevance to this work and the findings' larger significance for social psychology.

History and the Social Psychology of Obedience in the Holocaust

When the president of the American Historical Association issued a controversial call to his colleagues to employ the tools of psychology - by which he largely meant psychoanalysis - in the discipline of history, he was acknowledging that historical accounts often contain implicit or explicit claims about human psychology (Langer, 1958). Similarly, when social scientists view historical events through the lenses of psychological studies and concepts, we advance implicit and explicit claims about history (see a related argument by Cherry, 1995). Unfortunately, the history psychology presents may offer undergraduates the only learned exposure to the Holocaust they receive. What I want to suggest below is that when our references rely on the model of obligatory obedience in the Milgram experiment, they do not adequately represent the behavior of perpetrators in the Holocaust nor do they prompt new social psychological understandings of that behavior.

What is it that psychology teaches about the Holocaust? In 1986, Miller reviewed then-current commentary relating the Milgram experiment to the Holocaust in psychology and cited excerpts from undergraduate texts on the topic. Overall, both sources showed psychologists emphasizing the banality of evil (in Hannah Arendt's oft-cited phrase), the role of "just following orders" or obligation-perpetuated obedience, and the normality as opposed to potentially pathological character of the perpetuators of evil. A review of current textbooks (e.g., Gleitman, 1995; Gray, 1994; Matlin, 1995; cf. Miller, this issue) reveals a similar stress on the role of "blind obedience" in the Holocaust, where the Milgram experiment is applied to make sense of why "good citizens obeyed the commands of Nazi authorities" (Marlin, 1995, p. 606).

Regardless of whether the Milgram experiment remains an apt symbol for duty-based obedience, there is little doubt that obedience to authority in all of its forms was a significant and potent force in the Nazi dictatorship. For example, obligations to obey were implicated in why German Order Police followed commands to massacre Jewish civilians (Browning, 1989), how Hitler Youth were indoctrinated (Schirach, 1981), and even, poignantly, why some Italian Jews reported to a detention camp (Levi, 1960). And the Nazis were ruthless in the exercise of coercive power, even if that power was not sharply directed against German soldiers or doctors refusing orders (e.g., Browning, 1989; Lifton, 1986).

What an emphasis on obedience slights, however, are voluntary individual and group contributions to Nazi ideology, policy, bureaucracy, technology, and ultimately, inhumanity. Historical scholarship recognizes this problem in the controversy between intentionalists, who view the Holocaust as the product of Hitler's plans and orders, and functionalists, who see the Holocaust as evolving from bureaucratic developments and rivalries, improvisation, individual and group initiatives, and other external conditions and forces (see Marrus, 1987, chap. 3, for a review). The top-down orientation of the intentionalist perspective melds well with social psychology's emphasis on obedience to authority; the functionalist perspective, which is prominent in historical scholarship, does not.

Staub (1989) commented that "Milgram's dramatic demonstration of the power of authority, although of great importance, may have slowed the development of a psychology of genocide, as others came to view obedience as the main source of human destructiveness" (p. 29). It is surprising, certainly, that social psychology, which was so strongly affected by the Holocaust (Cartwright, 1979), does not address the Holocaust more explicitly and richly than it now does. For example, two volumes of The Handbook of Social Psychology fail to index the topics of genocide or the Holocaust, even though the interested reader can find references to countless other "Special Fields and Applications" (but see Lindzey & Aronson's general disclaimer, 1985, p. v). Perhaps if we recognize that obligatory obedience addresses only a part of the tragedy of the Holocaust, we will consider anew why people were attracted to authority (Staub, 1989), how they perceived their circumstances and why they were unwilling or unable to alter them, how individuals responded to peer behavior and the social dilemmas they faced (e.g., Browning, 1989), and why so many people used the opportunity provided by the Nazis to further their own interests, even if that meant tolerating or contributing to genocide. Raul Hilberg concluded that "In the final analysis the destruction of the Jews was not so much a product of laws and commands as it was a matter of spirit, of shared comprehension, of consonance and synchronization" (1985, p. 55). Certainly social psychology, including Milgram's research, has more to contribute to understanding this and to meeting our obligation to Abraham Lewin and those who suffered with him.

Blass, T. (1991). Understanding behavior in the Milgram obedience experiment: The role of personality, situations, and their interactions. Journal of Personality and Social Psychology, 60, 398-413.

Gilbert, S. J. (1981). Another look at the Milgram obedience studies: The role of the gradated series of shocks. Personality and Social Psychology Bulletin, 7, 690-695.

Kelman, H. C., & Hamilton, V. L. (1989). Crimes of obedience: Toward a social psychology of authority and responsibility. New Haven, CT: Yale University Press.

Mantell, D. M., & Panzarella, R. (1976). Obedience and responsibility. British Journal of Social and Clinical Psychology, 15, 239-245.

Milgram, S. (1965a). Some conditions of obedience and disobedience to authority. Human Relations, 18, 57-76.

Milgram, S. (1965b). Obedience [Film]. (Available from New York University Film Library.)

Milgram, S. (1974). Obedience to authority: An experimental view. New York: Harper & Row.

Miller, A. G. (1986). The obedience experiments: A case study of controversy in social science. New York: Praeger.

Nissani, M. (1990). A cognitive reinterpretation of Stanley Milgram's observations on obedience to authority. American Psychologist, 45, 1384-1385.

Zimbardo, P. G. (1974). On "Obedience to authority." American Psychologist, 29, 566-567.

NEIL LUTSKY completed his Ph.D. degree in social psychology at Harvard University. He is Professor of Psychology at Carleton College, where he has taught since 1974. He has written on attitudes toward elderly persons and old age, stigma and stereotyping, personality processes, and the teaching of social psychology. Lutsky currently serves as a consulting editor to the journal Teaching of Psychology.

Article A18320025