Handbook of Psychology, Volume 5, Personality and Social Psychology

Anomalous Prosocial Acts 475

to the empathy-evoking need without helping. Because em-
pathic arousal is a result of witnessing the need, either termi-
nating this need by helping or terminating exposure to it by
escaping should reduce one’s own empathic arousal. Escape
does not, however, enable one to reach the altruistic goal of
relieving the victim’s need. Therefore, the aversive-arousal
explanation predicts elimination of the empathy-helping rela-
tionship when escape is easy; the empathy-altruism hypothesis
does not. Results of these experiments have consistently
patterned as predicted by the empathy-altruism hypothesis
and not by the aversive-arousal reduction explanation, casting
doubt on this popular egoistic account (see Batson, 1991, for a
review).
A second egoistic explanation invokesempathy-specific
punishment.It claims that people learn through socialization
that additional obligation to help, and thus additional shame
and guilt for failure to help, is attendant on feeling empathy for
someone in need. As a result, when people feel empathy, they
are faced with impending social or self-censure beyond any
general punishment associated with not helping. They say to
themselves, “What will others think—or what will I think of
myself—if I don’t help when I feel like this?” and then they
help out of an egoistic desire to avoid these empathy-specific
punishments. Once again, experiments designed to test this
explanation have failed to support it; the results have consis-
tently supported the empathy-altruism hypothesis instead
(Batson, 1991).
The third major egoistic explanation invokes empathy-
specific reward.It claims that people learn through socializa-
tion that special rewards in the form of praise and pride are
attendant on helping a person for whom they feel empathy.
As a result, when people feel empathy, they think of these
rewards and help out of an egoistic desire to gain them.
The general form of this explanation has been tested in
several experiments and received no support (Batson et al.,
1988, Studies 1 & 5; Batson & Weeks, 1996), but two varia-
tions have also been proposed. Best known is the negative-
state relief explanation proposed by Cialdini et al. (1987).
Cialdini et al. suggested that the empathy experienced when
witnessing another person’s suffering is a negative affective
state—a state of temporary sadness or sorrow—and the per-
son feeling empathy helps in order to gain self-rewards to
counteract this negative state.
Although this egoistic alternative received some initial
support (Cialdini et al., 1987; Schaller & Cialdini, 1988), sub-
sequent research has revealed that this was likely due to pro-
cedural artifacts. Experiments avoiding these artifacts have
instead supported the empathy-altruism hypothesis (Batson
et al., 1989; Dovidio, Allen, & Schroeder, 1990; Schroeder,
Dovidio, Sibicky, Matthews, & Allen, 1988). It now seems

clear that the motivation to help evoked by empathy is not

directed toward the egoistic goal of negative-state relief.

A second interesting variation on an empathy-specific

reward explanation was proposed by Smith, Keating, and

Stotland (1989). They claimed that rather than helping to gain

the rewards of seeing oneself or being seen by others as a

helpful person, empathically aroused individuals help in order

to feel joy at the needy individual’s relief: “It is proposed that

the prospect of empathic joy, conveyed by feedback from the

help recipient, is essential to the special tendency of empathic

witnesses to help....Theempathically concerned witness to

the distress of others helps in order to be happy” (Smith et al.,

1989, p. 641).

Some early self-report data were supportive, but more

rigorous experimental evidence has failed to support this

empathic-joy hypothesis. Instead, experimental results have

once again consistently supported the empathy-altruism

hypothesis (Batson et al., 1991; Smith et al., 1989). The

empathic-joy hypothesis, like other versions of the empathy-

specific reward explanation, seems unable to account for the

empathy-helping relationship.

A Tentative Conclusion

Reviewing the empathy-altruism research, as well as related

literature in sociology, economics, political science, and biol-

ogy, J. A. Piliavin and Charng (1990) concluded that

There appears to be a “paradigm shift” away from the earlier

position that behavior that appears to be altruistic must, under

closer scrutiny, be revealed as reflecting egoistic motives.

Rather, theory and data now being advanced are more compati-

ble with the view that true altruism—acting with the goal of ben-

efiting another—does exist and is a part of human nature. (p. 27)

Pending new evidence or a plausible new egoistic explana-

tion of the existing evidence, this conclusion seems correct.

It appears that the empathy-altruism hypothesis should—

tentatively—be accepted as true.

Implications of the Empathy-Altruism Hypothesis

If the empathy-altruism hypothesis is true, the implications

are wide ranging. Universal egoism—the assumption that all

human behavior is ultimately directed toward self-benefit—

has long dominated not only psychology but other social and

behavioral sciences as well (Campbell, 1975; Mansbridge,

1990; Wallach & Wallach, 1983). If individuals feeling

empathy act, at least in part, with an ultimate goal of increas-

ing the welfare of another, then the assumption of universal