“Situation” Trumps “Disposition”- Part II

On Tuesday, we posted Part I of our reaction to the criticism that a number of social psychologists made of Situationist Contributor Phil Zimbardo’s latest book, The Lucifer Effect. (Phil Zimbardo’s compelling reply to that criticism is available here.) [To pick up reading directly from Part I, skip from here to below the three asterisks.]

For convenience, here again is the criticism:

We are concerned by the message that has been conveyed to the general public regarding the power of the situation to “trump individual dispositions” (“The Banality of Evil,” Observer, April 2007). In contrast to Zimbardo, we believe that there is actually little scientific evidence indicating that situations are more important than dispositions for explaining behavior. Indeed, researchers recently summarized over 25,000 studies and found that personality and situations contribute almost equally to various outcomes, and many studies demonstrate the complex ways in which people react differently to similar situations. Our concern is that Zimbardo has misrepresented the scientific evidence in an attempt to offer a purely situational account of the antisocial acts perpetrated at Abu Ghraib. The scientific consensus, based on existing data, is that people vary in their propensity for antisocial behavior and that environments transact with personalities. Some people are more likely to turn out to be bad apples than others, and this is particularly evident in certain situations.

As we began to argue in Part I, the critics have built their arguments on particular definitions of “disposition” and “situation” that we believe are ill-conceived. The thrust of their argument seems to be that if individuals behave differently in seemingly similar situations, then that proves the power of disposition (or personality). Part I tried to clarify definitions and show how the differing behaviors of subjects in the first version of Milgram’s experiment were entirely consistent with our position that situation trumps disposition.

That post ended with the assertion that there “is nothing about differences in individual behavior across seemingly similar situations that demonstrates that disposition is the driving force.” This post picks up by describing some of the many reasons why that’s the case.

***

To begin with, as Milgram would go on to discover, by manipulating the situational cues somewhat, he could elicit almost any ratio of compliance to non-compliance among his subjects. Where the pleas of the “student” were made more remote, for instance, closer to 100% of subjects would shock to 450 volts. Given that few people would attribute their unwillingness to shock (perhaps fatally) an innocent victim to the proximity of that victim, those findings suggest that situation trumps disposition. And the fact that 1/3 of the subjects were unwilling to shock all the way up to 450 volts in the initial rendetion of the study does not show that their disposition was greater than the situation, as much as it shows that situation didn’t call for greater compliance.

Furthermore, what seems like an identical situation rarely is. When people appear to have different dispositions, it is often the case that the situational analysis is insufficiently subtle. How do we know what situational influences the subjects in Milgram’s first experiment might have been susceptible to beyond those controlled by Milgram. Were some hungry or tired or hurried? Were some feeling fed up with people telling them what to do? Were others simply unaccustomed to taking instruction? Were some especially phobic about electric shocks? Were some in need of money and hoping to be selected for future experiments? Were some subjects implicitly less motivated to obtain closure and thus end the experiment? And so on.

The tendency to presume situational equivalence is itself a reflection of the fundamental attribution error (discussed here) and the difficulty humans have in seeing situational forces. To assume that the situation of each subject in Milgram’s experiment was identical is equivalent to assuming that the situation of each sibling in a family or racial or socioeconomic group is identical, a surprisingly common set of suppositions. They are implied, for instance, when people are incredulous that two siblings can “turn out so differently” or assert that the difference between those who succeed in the U.S. and those who don’t boils down to the presence or absence of motivation, character, work-ethic or some such other dispositionist quality. Although those suppositions may be common, even a tiny bit of reflection or observation reveals that they are incorrect.

Similarly, the fact that individuals do have a “personality”—seemingly stable ways of acting and interacting over time—does not prove the power of disposition over situation. It only indicates one reason why the “dispositionist model” continues to form the foundation of “common sense” and can be useful as a means of predicting some people’s behavior in a given situation. But as Professors Lee Ross and Richard Nisbett argued years ago in their important book, The Person and the Situation, people can have stable personalities precisely because they occupy stable situations. In addition, as many social psychologists have shown, a perceived “personality,” or personal narrative, or self-schema, carries its own self-fulfilling weight (for related Situationist posts, click here and here). In short, “personality” and “situation” are mutually constitutive not, as Zimbardo’s critics would have it, mutually exclusive. And once “personality” is understood as the consequence of internal and external situational factors, it’s meaning and significance changes from that commonly supposed by the dispositionist model.

It may be helpful if we come at that again. Situationists do not deny that there are important differences across individuals. In fact, situationist scholars are interested in identifying and understanding those differences. For instance, Situationist contributor John Jost and some of his co-authors have devoted much attention to understanding the “antecedents of ideology”—differences across individuals that influence their behavior, not simply in individual laboratory experiments, but across wide, real-life behavioral domains.

The key, however, is that those “dispositional” differences are actually situational as we would define the term — that is, they fall outside the naive psychology of stable, explicit attitudes or preferences and outside the self-schemas that individuals typically employ to make sense of their own conduct and that of others. They are, in other words, part of our “interior situation.” A person doesn’t say “I’m a conservative because I crave closure and structure and tend not to be open to new experiences.” Instead, she offers reasoned-based explanation for her ideological position, something like “I’m a conservative because conservative policies are best for the country’s future and protect those values that I hold dear.” But the implicit motives operating under the radar are often more influential than the stories we tell.

The fact that individual and group voting patterns are predictable does not mean that people behave according to the dispositionist model summarized above. Rather, the dispositionist model is the story we tell ourselves to make sense of behavior that is caused by situational forces within us (including the “antecedents of ideology”) and around us about which we are rarely conscious. (For related Situationist posts, click here and here. For a long article on the gap between the the dispositionist model and the situational character, click here.)

Professor Zimbardo’s critics seem motivated, in the end, by their “concern” that the public might forget that “[s]ome people are more likely to turn out to be bad apples than others.” In other words, they seem nervous that if situation is taken too seriously we might lose our ability to hold the people who engage in bad activities responsible by no longer conceiving of them as “bad apples.” We might lose our ability to make easy normative distinctions between good apples and bad apples—the sort of distinctions that provide the legitimating, normative punch behind everything from the individualistic (dispositionistic) criminal justice system to ideologies and blame frames that justify vast inequalities.

As we have already indicated, however, the evidence that there are “complex ways in which people react differently to similar situations” does not demonstrate that those differences are dispositional in any normative sense.

Our own concern is that those psychologists are misrepresenting situationism in an attempt to individualize responsibility for “antisocial acts.” They, like most people, want to be able to blame someone. It may be true as they assert that “that people vary in their propensity for antisocial behavior.” Nonetheless, it is at least as true that the naive, dispositionist, preference-driven attributional models that are so commonly used to explain behavior are wrong and that overlooked situational forces (both external and internal) are leading to conduct that, as with Milgram’s experiment, is both harmful and inconsistent with our explicit, shared values.

Sometimes bad apples need to be removed—in part because they are contaminating. But if a bad apple can harmfully effect the situation of good apples, then don’t we owe it to the “bad apple” to consider whether its condition itself reflects a contaminating environment? If we don’t like “bad apples” or their effects, shouldn’t we at least endeavor to examine the tree, it’s roots, the soil, the air quality, the parasites, the orchard, the toxins, and, in a word, the situation.