Main authors discussed: Gilbert Harman, John Doris, C. Daniel Batson

Recent work by both philosophers and social psychologists suggests that agents lack a moral character comprised of robust, global moral dispositions. According to situationists like Harman and Doris, most individuals overestimate the role dispositional factors play in explaining an individual’s behavior, and underestimate the role the situation plays. Drawing on experimental results in social psychology, situationists claim that our behaviors are best explained not be globalist conceptions of moral character traits, but rather by situational factors. While a number of virtue ethicists have attempted to call into question situationists’ criticism, Miller takes a different tack.

According to Miller, “there is an impressive array of evidence from the social psychology literature which suggests that many people do possess one or more robust global character traits pertaining to helping others in need. But at the same time, such traits are a far cry from the traditional virtues like compassion” (248). Miller thus argues “that we should adopt an intermediate position between the situationist proposals of Harman and Doris and the leading responses being offered by virtue ethicists” (248).

Section 1 of Miller’s paper briefly summarizes the situationists’ criticisms of the traditional globalist account of moral character traits. Global character traits are both consistent across a wide variety of trait-relevant situations and relatively stable across time. Character traits so understood are supposed to play a central role in: (i) explaining trait-relevant behavior, and (ii) helping to accurately ground predictions of such behavior in the future. But extant studies by social psychologists suggest that character traits are not consistent in the way globalists hold.

Miller focuses on helping traits, and argues in section 2 that, contrary to the claims of situationists, the empirical results are compatible with the widespread existence of one or more global character traits. Nevertheless, these ‘global helping traits’ (of GHTs) are “rather different from how compassion has traditionally been construed” (250) by globalist virtue ethicists. GHTs are dispositions to help others thought to be in need, and are sensitive to a number of different psychological factors, including background beliefs, normative convictions, feelings, moods, and embarrassment. Focusing on empirical studies related to the impact of mood on helping behavior, Miller writes:

When it comes to mood states, for example, we should expect that, other things being equal, many people with elevated levels of positive affect will be such that they help others in situations ranging from picking up dropped papers to volunteering time for charity organizations. On the other hand, we can expect that, again other things being equal, many people without elevated levels of affect will be such that they do not exhibit these and other forms of helping behavior (253).

And this is exactly what one finds in the relevant studies. (Miller is careful to note that while the studies are compatible with GHTs as he defines them, the studies do not yet provide direct evidence for the existence of GHTs.) The various factors that influence the expression of a GHT can thus add together to reach an ‘activation threshold’, which then results in helping behavior.

Miller briefly outlines, in section 3, what he takes to be the central features of empathy. X empathizes with Y’s feeling F if

(i) X tries to imagine what Y feels in the situation, rather than what X would feel like if X were in the situation which Y is in,

(ii) X experiences something similar to Y’s feeling F, and not necessarily F itself,

(iii) X believes Y is feeling F (or something similar to F), and

(iv) X’s empathizing with Y’s feeling F is conceptually distinct from X’s sympathizing with Y (258).

In section 4, Miller examines a number of recent social psychological experiments on empathetic behavior. In a 1982 study by Toi and Batson, college students who empathized with a fictional fellow student, Carol, were significantly more likely to agree to help than were control students who were not lead to empathize with her (71% vs. 30%). This study, as well as other similar studies, is consistent with the account of GHTs developed in section 2.

As a result, Miller claims in section 5 that the existing psychological data does not support the situationists’ claims that robust character traits do not exist, nor does it support the situationists’ claim that local character traits do exist. “The empirical data suggests that a significant number of people with raised levels of empathetic emotion would help in a wide variety of circumstances, while many of those without such raised levels would not. So there seems to be more structure at work than a fragmentation model of character would lead us to believe” (262f). However, nothing suggests that those individuals who engage in helping behaviors do so on the basis of the right motivating reasons, as traditional virtue ethics claims.

In section 7, Miller outlines two remaining worries that virtue ethics must face. First, the social psychology data suggests that most people do not seem to exhibit compassionate behavior when they have the opportunity to do so, even when the helping tasks are not particularly strenuous. So it looks like the majority of people lack the disposition to exhibit compassionate behavior when an opportunity arises, which could be taken as an embarrassment for virtue ethics. Second, if the virtue ethicist claims GHTs are in fact widely possessed, but merely are only relatively infrequently triggered, “we will need a story about how more people can instantiate the virtue of compassion…. But here the psychological results seem to put up a significant obstacle to such a proposal” (270) insofar as they suggest that the “empathy induced motivation to help is extremely fragile” (272). Miller concludes that while “the (widespread) existence of a traditional character trait like compassion might be in trouble, … it is premature to conclude from such a claim that no global traits whatsoever exist in this area” (272).

Reviewed by Kevin Timpe
University of San Diego