The true self and the situation
You are probably already familiar with Darley and Batson’s (1973) study. Participants were students at the Princeton Theological Seminary. As part of the study, they were asked to give a short sermon in a nearby building. Half of the seminarians were told they were running late, so they’d better hurry to the building. The other half were told they had plenty of time, but they might as well mosey over. On their way, both the hurried and relaxed seminarians encountered a man slumped in a doorway, groaning.
The overall pattern of results won’t surprise you, but the strength of the effect might: relaxed seminarians were six times more likely to help the injured man. Even if they were on their way to deliver a sermon on the parable of the Good Samaritan, it didn’t matter—hardly any hurried seminarians stopped to help.
Take a moment to consider one of the hurried seminarians, rushing along, trying to figure out what he will say in his sermon. He notices the injured-looking man in the doorway, but he doesn’t stop to help. He just keeps walking. Question: what caused this callous behavior? Does it seem to arise from within the seminarian himself, or does it seem that something about the external situation caused him to ignore the man? The common thing to say, of course, is that the cause does not lie within the seminarian himself. After all, this very seminarian would have helped, had he been in the relaxed condition. So, it seems the situation is to blame.
It seems like a natural verdict! But what does it mean? In their excellent review article, Strohminger et al. describe recent research that suggests the natural verdict is importantly ambiguous: do we mean that the cause of the hurried seminarian’s callousness lies “on the sunny side of his epidermis,” or do we mean that the cause is not a part of his true self?
According to Strohminger et al., the true self is assumed (defeasibly) to be morally good. So, if the relevant distinction is between those actions which are caused by the seminarian’s true self and those which are not, it is easy to see why people would naturally attribute his callousness to the situation. But if this is so, it should also seem that the relaxed seminarians, nearly all of whom offer to help the victim, are not led to help by the situation. Rather, their kind actions should seem to arise from deep within themselves. Put another way, if the question “Is the action caused by the person or by the situation?” concerns the actor’s true self, we should expect to find a surprising asymmetry in how people explain good and bad actions. Actions that are perceived to be good should seem more person-caused than those that are perceived to be bad, even when they are otherwise exactly alike.
In my own studies (unpublished, as of this writing), I have consistently found this pattern in participants’ explanations for morally valenced actions. If you want to predict whether an observer will explain an action more in terms of the actor herself or more in terms of the situation, you can’t do better than to find out about the observer’s moral attitudes towards the action.
Why did the young woman decided to terminate her pregnancy? People who believe that abortion is morally evil are far more likely to say that she aborted because her boyfriend had recently broken up with her. By contrast, people who believe that abortion is morally permissible are much more likely to locate the cause of her decision within the woman herself. Why did the evangelical Christian man, who believes homosexuality to be immoral but is himself attracted to other men, give into his erotic urges? People who are disgusted by male homosexuality are much more likely to say that it had something to do with a stressful event he endured earlier that day; people with positive attitudes towards homosexuality think it’s because he’s gay. Why did the white woman decide to convert to Islam? People who believe that “Muslims are dirty” are much more likely to say it was because she was peer-pressured or because she came from a broken home; people with more positive moral attitudes towards Muslims prefer to say she converted because she’s a spiritual person.
These findings make good sense if it is beliefs about the true self that are relevant to whether people perceive actions as arising from within the actor or the situation. On the other hand, if the self is understood in a morally-neutral sense, as is common in social psychology, it is less clear why the data should pattern in this way. Think of it in terms of the Good Samaritan study. The experimenter and experimental conditions are equally ‘outside the skin’ of the relaxed and hurried seminarians, so if the epidermis-centric concept of the self is the relevant concept, shouldn’t both the kind and callous actions seem equally caused by the situation? 
The research described by Strohminger et al. suggests that ordinary people are largely unconcerned with the boundary between the sunny and meaty sides of the epidermis. By contrast, the list of psychological phenomena in which beliefs about the true self play an important role is growing quickly. Conceptualizing the person/situation distinction in value-neutral terms may therefore be unproductive for social psychology; in fact, it may be seriously distorting. To better understand how people perceive the causes of actions, social psychology should investigate the concept of the true self.
*** The person/situation distinction is often analyzed in terms of “causal covariation.” The rough idea is that an action is caused by those factors with which it covaries. On this view, an action is situationally caused to the degree that the actor would not have performed the action, had the situation been relevantly different. But this analysis also will not do: both the hurried and relaxed seminarians would have behaved differently had they been assigned to different experimental conditions. So this analysis, too, fails to capture the intuition that the callousness of the hurried seminarians, but not the kindness of the relaxed seminarians, is caused by the situation. In presently unpublished studies, I compare the effect of observers’ moral attitudes and the effect of their beliefs about covariation on how they explain morally-valenced actions. In the cases I’ve tested, the effect of covariation information is insignificant, both in absolute terms and by comparison to the powerful effect of participants’ moral attitudes.