342 The Social Self
the pathological outcomes connected with shame, such as
suicide and major depression (Tangney, Burggraf, & Wagner,
1995). Shame also seems to produce socially undesirable out-
comes such as, for some people, a complete withdrawal from
others. Other people, however, respond to shame with anger
(Tangney et al., 1992). The shift from shame into anger may
be a defensive effort to negate the global negative evaluation.
There is some evidence that this shift in emotions can lead
to violent outbursts (Baumeister et al., 1996). Kitayama,
Markus, and Matsumoto (1995) have proposed that the
movement from shame to anger reflects the independent self-
hood model common to Western cultures and may not occur
in cultures that emphasize more interdependent selves.
In contrast, guilt is more reparable and less socially disrup-
tive than shame. Guilt has a strong basis in relationships even
when no transgression is involved. For example, some people
feel survivor guilt because they have survived when others
have died or suffered. The term originated in studies of sur-
vivors of the Holocaust and the Hiroshima bombing (Lifton,
1967). More recently, survivor guilt emerged during episodes
of corporate downsizing, when people who kept their jobs felt
guilty while others were fired (Brockner, Davy, & Carter,
1985). In general, people may feel guilty when they outper-
form others (Exline & Lobel, 1999).
According to Baumeister, Stillwell, and Heatherton
(1994), guilt is mainly interpersonal and seems designed to
strengthen relationships. People may try to avoid hurting
close others because it makes them feel guilty. After a trans-
gression, guilt makes people seek to make amends or rectify
the situation in an attempt to repair the damage to the relation-
ship. It makes people change their behavior so that they will
not repeat the damaging behavior. It makes them try to live up
to the expectations of others. Feeling guilty is sometimes ben-
eficial to the relationship in and of itself, because guilty feel-
ings confirm that the person cares about the relationship (even
if the transgression made it appear that he or she did not care).
In addition, people sometimes exaggerate how hurt or upset
they are by another person’s actions, in order to make that per-
son feel guilty. The guilt makes the other person more willing
to comply with the wishes of the person who felt hurt. This
tactic can be used to redistribute power in a relationship: Guilt
enables otherwise powerless people to sometimes get their
way. Usually, the person who is hurt makes his or her feelings
and disappointment clear. If the other person cares about your
welfare, he or she will want to avoid hurting you, because
hurting you will make him or her feel guilty. Hence the person
will do what you want.
Baumeister, Reis, and Delespaul (1995) confirmed that
guilt plays an important role in close relationships. The
authors asked participants to describe their most recent
experiences of six different emotions, including guilt. These
were then coded for the level of interpersonal connection.
Guilt scored the highest of the six major emotions on interper-
sonal connection. That is, hardly any guilt stories referred to
solitary experiences or interactions with strangers; the over-
whelming majority of guilt stories involved partners in close
relationships, such as family members or romantic partners.EmbarrassmentSimilar to shame and guilt, embarrassment seems to be a mix-
ture of self and interpersonal concerns. Modigliani (1971)
linked embarrassment to the public self by showing that the
best predictor of embarrassment was a situational, perceived
loss of others’ good opinion. In addition, embarrassability
correlates more highly with public self-consciousness than
with private self-consciousness (Edelmann, 1985).
In an influential review, Miller (1995) argued that two
theoretical perspectives on embarrassment are predominant.
The first theory emphasizes concern over being evaluated by
others; to be embarrassed, you must first be concerned about
others’ evaluations. The alternative view invokes the un-
pleasant nature of awkward social interactions. In one study,
Parrott, Sabini, and Silver (1988) presented participants with
a hypothetical scenario in which someone refused a date.
People reported they would feel less embarrassed if the rejec-
tor used an obvious excuse than if the rejector bluntly
rejected them, even if the person’s rejection was equally neg-
ative. However, making an excuse may itself convey a posi-
tive evaluation, such as concern for the rejected person’s
feelings (Miller, 1995). Miller concluded that both perspec-
tives are valid; nevertheless, the concern over social evalua-
tion is the more common cause of embarrassment.
Blushingis one common sign of embarrassment, but peo-
ple sometimes blush even when there is no obvious social
evaluation. Leary, Britt, Cutlip, and Templeton (1992) con-
cluded that unwanted social attention is the most common
cause of blushing. In general, people blush as an appease-
ment to others after violating social norms. People hope that
looking embarrassed after a transgression will inform other
people that they feel remorseful. Apparently, embarrassment
is effective in minimizing negative evaluations. Semin and
Manstead (1982) found that subjects expressed greater liking
toward someone who was embarrassed after an accidental
transgression. When the target person was not embarrassed,
subjects did not like the person as much.Social AnxietySchlenker and Leary (1982) argued that social anxiety is di-
rectly linked to self-presentation. In their view, social anxiety
arises when someone wants to make a particular, desired