The Self as an Interpersonal Actor 335researchers controlled for narcissism statistically, self-esteem
did not predict aggressive behavior. Thus it appears that
narcissism is the better predictor of interpersonal hostility.
This fits the view that aggression comes from only a subset of
people with high self-esteem, while other people with high
self-esteem are not aggressive.
Other research has found that narcissists are willing to
derogate others after receiving threatening feedback (e.g.,
Kernis & Sun, 1994). They react with hostility, denigration,
and aggression when they feel threatened (Rhodewalt &
Morf, 1998). In fact, a recent study found that men incarcer-
ated in prisons scored significantly higher in narcissism than
samples of male college students (Bushman, Baumeister,
Phillips, & Gilligan, 2001). Levels of self-esteem, however,
did not differ between the two groups. Thus narcissists tend
to be more personally sensitive to criticism, but insensitive to
how their behavior affects others. Like the research on self-
esteem presented earlier, these results suggest that inflated
self-views can often lead to poor consequences for interper-
sonal relationships.
Reflected Appraisals
Thereflected appraisalsmodel suggests that people learn
about themselves by interacting with others. People find out
what other people think of them and then internalize these
opinions into their self-views. In addition, information about
the self often is meaningful only in comparison to others, as
social comparison theory emphasizes. One is only fat or thin,
intelligent or stupid, friendly or hostile in comparison to
other people. In these cases and many others, self-knowledge
can grow only when people make these implicit comparisons.
Much of reflected appraisals theory stems from symbolic in-
teractionism (e.g., Mead, 1934). Mead’s theory argues that
most self-knowledge comes from social interactions. The
process of reflected appraisals (i.e., how other people’s ap-
praisals of you shape your self-understanding) is often de-
scribed with Cooley’s (1902) term the looking-glass self.
Using an antiquated term for a mirror, the looking-glass self
posits that other people provide the mirror through which in-
dividuals see and understand themselves.
Cooley (1902) argued that the self-concept consists of
“the imagination of our appearance to the other person; the
imagination of his judgment of that appearance, and some
sort of self-feeling, such as pride or mortification” (p. 184).
Thus our self-esteem is also heavily influenced by what
others think of us. Mead (1934) elaborated on this notion by
suggesting that the self is also shaped by our vision of how
a generalized other perceives us. The generalized otheris
basically the person’s whole sociocultural environment. If a
society has a negative view of children at a given time, for
example, children are likely to internalize this negative view
of the generalized other.
An influential literature review by Shrauger and
Schoeneman (1979) concluded that symbolic interactionism
was partially supported by data. The review gathered data
comparing self-concepts with the views of others. Although
these correlations were positive, they were rather small. Sub-
sequent studies have confirmed that symbolic interaction ef-
fects are significant but small (Edwards & Klockars, 1981;
Malloy & Albright, 1990). Even some of these weak links
can be questioned on methodological grounds, as noted by
Felson (1989).
On the other hand, Shrauger and Shoeneman (1979) found
that self-concepts were highly correlated with how peoplebe-
lievedthat others perceived them (and subsequent work has
replicated this conclusion). Therefore, there is a meaningful
link between self-perceptions and other-perceptions (although
the causal direction is unclear and probably bidirectional). The
discrepancy arises between how people actually perceive Bob
and how Bob thinks other people perceive him—but Bob’s
view of himself is quite similar to how he thinks others see
him. Thus others do shape self-views, even though people are
not always accurate about how others perceive them.
There seem to be two major reasons for these inaccuracies
(see Felson, 1989). First, people do not generally tell some-
one precisely what they think of him or her. The exchange of
interpersonal evaluations is highly distorted. People do not
want to offend or distress someone by an honest, negative
evaluation, and they are often afraid that the person they crit-
icize will no longer like them. (This is a legitimate fear; most
humans tend to like people who like them, and distrust those
who criticize them.) When refusing a date, for example,
people tend to give false and misleading explanations, often
resulting in their being unable to discourage further invita-
tions from the same person (e.g., Folkes, 1982—although
some of these explanations have become so popular that they
are now more easily understood as a genuine brush-off: “It’s
not you, it’s me.” Translated: “It’s totally you. You are the big
problem. I’m fine.”). Even when people are engaging in
deliberate self-presentation, they are not very accurate at
estimating the impression they actually make on others (e.g.,
DePaulo, Kenny, Hoover, Webb, & Oliver, 1987). Given the
dearth of honest and precise negative feedback from others, it
is not surprising that people’s self-views remain blissfully
unaffected by those concealed opinions and appraisals.
The other source of distortion is self-deception. Often,
people do not accept information directly into their views of