This article is for use only of students enrolled in GPS 520: Advanced Social Psychology for purposes associated with this course and may not be retained or further disseminated.
Leary, M. R. (1999). The social and psychological importance of self-esteem. In R. M Kowalski and M. R. Leary (Eds.), The social psychology of emotional and behavioral problems, (p. 197-221). Washington, D.C.: American Psychological Association.
THE
SOCIAL AND PSYCHOLOGICAL IMPORTANCE OF SELF-ESTEEM
MARK
R. LEARY
Psychologists have been
interested in the topic of self-esteem since the earliest days of the
discipline. From the seminal writings of William James (1890) to contemporary
social psychology and clinical practice, research on self-esteem can be found in
virtually every area of human psychology. A recent scan of PsycINFO (American
Psychological Association's literature database) using the keyword self-esteem
yielded 13,585 articles published in the 30-year span between 1967 and 1996-an
average of more than 450 articles per year. This does not include the 881
chapters and books published since 1987 or articles that may deal with
self-esteem but do not have the term among their keywords or in their abstracts.
The literature on self-esteem is truly voluminous.
Despite
varying interests and orientations, most psychologists -both researchers and
practitioners-share three fundamental assumptions regarding self-esteem on which
virtually all of their work on the topic rests. These assumptions involve the
motive to protect self-esteem, the benefits of high self-esteem, and the effects
of raising self-esteem. Specifically, most psychologists-whatever their area of
formal training-would likely endorse some version of the following three
assumptions:
Assumption
1: Human beings are motivated to
preserve, protect, and occasionally enhance their level of self-esteem.
Assumption
2: High self-esteem is typically
more desirable than low self-esteem because it is associated with many
psychological benefits.
Assumption
3: Raising low self-esteem improves
psychological well-being and produces desirable changes in people's behavior.
On the
surface, these three assumptions appear relatively noncontroversial. Most
people, psychologists and laypersons alike, probably regard them as empirical
facts. However, the validity of these assumptions is much less well established
than is often believed. As seen later, many questions surround these
assumptions, and there are reasons to suspect that they are in serious need of
qualification.
My goal in this chapter is to
critically examine these fundamental assumptions, primarily from the
perspectives of social and clinical psychology. Although these assumptions have
considerable research support, the case for each is not as airtight as it may
first appear. I explore several issues that challenge these seemingly
self-evident assumptions and then consider the utility of a reconceptualization
of self-esteem-sociometer theory-for helping to resolve many of these questions
and for integrating what is known about self-esteem. The chapter concludes with
a look at the implications of sociometer theory for understanding and treating
the emotional and behavioral problems that have traditionally been associated
with low self-esteem.
THE
ASSUMPTIONS
Perhaps the most fundamental assumption
that undergirds psychologists' understanding of self-esteem is that human beings
are motivated to preserve, protect, and occasionally enhance their level of
self-esteem. Virtually all of the major theorists have concurred with James's
(1890) assumption that people want (or need) to feel good about themselves. For
example, Allport (1937), who is often credited with founding the area of
personality psychology, suggested that a person's "most coveted experience
is the enhancement of his self-esteem" (p. 169). Greenwald (1980), a social
psychologist, characterized the human ego as inherently
"totalitarian." just as a totalitarian regime controls information to
maintain a particular, desired image of the government, people control
information to maintain favorable images of themselves. Clinical and counseling
psychologists of a variety of theoretical persuasions have embraced this theme
as well. For example, Branden (1969, 1983), a clinician who helped to popularize
the importance of self-esteem to the general public, argued that self-esteem is
a fundamental human need; an inspection of the literature reveals dozens of
other writers who echo this theme. As Markus (1980) observed, the "notion
that we will go to great lengths to protect our ego or preserve our self-esteem
is an old, respected, and when all is said and done, probably one of the great
psychological truths" (p. 127).
Given
differences in their scientific and professional goals, social and clinical
psychologists have approached the self-esteem motive from slightly different
angles. Social psychologists have been interested primarily in three issues: the
kinds of events that heighten people's concerns with maintaining their
self-esteem, the ways in which people deal with real and imagined assaults to
their self-esteem, and personality variables that moderate people's reactions to
self-esteem threats.
Experimental
studies within social psychology typically induce threats to participants'
self-esteem by leading them to expect or experience either failure (or other
unflattering feedback about themselves) or social rejection. Research shows that
these sorts of experiences do indeed "deal a blow" to self-esteem and
heighten people's motivation to enhance the positivity of their feelings about
themselves. For example, after failure, people tend to make self-serving
attributions--explanations that reduce their apparent responsibility for
negative events (Blaine & Crocker, 1993; Bradley, 1978). People who have
experienced failure or rejection also behave in ways that make them feel good
about themselves, for example, by associating with people who are successful (Cialdini
et at., 1976) and distancing themselves from people who fail (Snyder, Lassegard,
& Ford, 1986). In addition, they compensate for their shortcomings by
presenting particularly favorable impressions of themselves on dimensions that
are unrelated to the unflattering feedback they have received (Baumeister &
Jones, 1978).
People do not
always wait until their self-esteem has been damaged to take ego-defensive
action. When they are concerned about a portending failure, people sometimes set
up impediments to their performance-a self. created handicap that provides a
plausible reason for failure that will minimize the impact of future failures on
their self-esteem (Berglas & Jones, 1978)-or offer pre-emptive self-serving
attributions, pointing out in advance that factors beyond their control will
likely interfere with their performance (DeGree & Snyder, 1985; Pyszczynski
& Greenberg, 1983).
Of course,
people differ markedly in their motive to maintain selfesteem and in how they
react to self-esteem threats, and social-personality psychologists are
interested in individual difference variables that moderate their reactions. The
most widely studied personality variable in this regard is trait self-esteem.
People who score low on measures of trait self-esteem tend to respond
differently to failure and rejection than people who score high on measures of
trait self-esteem (Baumeister, Tice, & Hutton, 1989; Tice, 1991, 1993).
Clinical
psychologists' interest in the self-esteem motive involves ways in which
people's efforts to maintain their self-esteem underlie emotional problems and
dysfunctional behavior. The self-esteem motive has often been discussed in the
context of ego defense -people's inability or unwillingness to recognize and
acknowledge undesired aspects of themselves. Allport (1961) wrote that "the
mechanisms of ego-defense are sly devices by which we try to circumvent
discomfort and anxiety. These self-protective strategies are common, but they do
not by any means constitute the normal person's entire repertoire of adjustive
reactions" (p. 29). The neurotic individual, Allport argued, over-relies on
ego-defensive reactions, resulting in anxiety and inadequate coping. Clinical
tore suggests that people who cannot sustain their self-esteem in socially
acceptable ways sometimes turn to maladaptive means of doing so. Problems as
diverse as domestic violence, prejudice, juvenile delinquency, and substance
abuse have been attributed to dysfunctional efforts to protect or promote one's
self-esteem (Mecca, Smelser, & Vasconcellos, 1989). Overall, then,
considerable theory and research in social and clinical psychology suggest that
the motive to maintain self-esteem is a pervasive and powerful influence on
human behavior.
The second
assumption is that high self-esteem is more desirable than low self-esteem. Most
psychologists would agree that people who have high self-esteem are better off
than people with low self-esteem and that "the absence of a healthy sense
of self-appreciation seems to be one of the basic warning signs of a
dysfunctional personality" (Bednar, Wells, & Peterson, 1989, p. 1).
This notion also pervades the culture more generally; for example, parents and
teachers appear to assume that children fare better if they have high
self-esteem.
The strongest
support for Assumption 2 comes from research showing that high self-esteem is
associated with psychological well-being whereas low self-esteem tends to be
associated with various psychological difficulties (for reviews, see Leary,
Schreindorfer, & Haupt, 1995; Mecca et al., 1989; and Mruk, 1995). Low
self-esteem is associated with greater depression (Hammen, 1988; Smart &
Walsh, 1993) and anxiety (Coopersmith, 1967; Strauss, Frame, & Forehand,
1987) than is high self-esteem. People with low self-esteem tend to be more
lonely than those with high self-esteem (Vaux, 1988) as well as more socially
anxious and shy (Leary & Kowalski, 1993). They are also more prone to
experience eating disorders (Katzman & Wolchik, 1984; Shisslak, Pazda, &
Crago, 1990), to join deviant groups (Tennant-Clark, Fritz, & Beauvais,
1989), and to abuse alcohol and other drugs (Cookson, 1994; Vega, Zimmerman,
Watheit, & Apospori, 1993).
Clinical
problems aside, social psychological research also shows that people with high
self-esteem tend to behave in more socially skilled, adaptive, and prosocial
ways than those with low self-esteem (Batson, Bolen, Cross, &
Neuringer-Benefiel, 1986; Berkowitz, 1987). They, are also less likely to
conform to obviously incorrect opinions offered by other people and more likely
to stand behind their principles (Coopersmith, 1967; Janis & Field, 1959).
High self-esteem is also associated with more appropriate levels of
self-disclosure, with leadership emergence in small groups, and with better
social skills. In many cases, the correlations between self-esteem and these
various measures of emotional and behavioral problems are small (see Dawes,
1994). Even so, given these findings, one can see why low self-esteem "is
an assumed condition in virtually all contemporary models of disordered
behavior" (Bednar et al., 1989, p. 1) and why most theories of mental
health consider high self-esteem an essential aspect of a healthy ego (Taylor,
1989).
This
assumption, buttressed by strong empirical support, has also gained widespread
acceptance in the general American culture. In part, this has been because of a
plethora of popular books touting the virtues of self-esteem. One popular author
asserted that he could not "think of a single psychological problem-from
anxiety and depression, to fear of intimacy or of success, to spouse battery or
child molestation-that is not traceable to the problem of poor self-esteem"
(Branden, 1984, p. 12). Along the same lines, a best-selling book on how parents
can foster self-esteem in their children states that "die key to inner
peace and happy living is high self-esteem" (Briggs, 1975, p. 26).
Following on
the heels of the idea that high self-esteem is better than low self-esteem is
the third assumption: Raising low self-esteem improves psychological well-being
and produces desirable changes in people's behavior. This assumption pervades
clinical practice, societal and educational interventions, and popular books on
self-esteem. Regarding self-esteem as a causal agent is the basis of clinical
and societal interventions that try to solve personal and social problems by
elevating people's self-esteem. Within clinical psychology, several systemized
approaches have been offered for raising clients' self-esteem in individual
counseling, therapeutic groups, and psychoeducational workshops (e.g., Bednar et
al., 1989; Bums, 1993; Frey & Carlock, 1989; Mruk, 1995; Pope, McHale, &
Craighead, 1988). Within a broader societal context, efforts have been made to
raise the public's self-esteem as a way to reduce such social problems as
juvenile delinquency, drug abuse, and teenage pregnancy (Mecca et al., 1989).
Furthermore, the self-help section of most bookstores is littered with books
that offer to raise one's self-esteem (or the self-esteem of one's children,
friends, or loved ones). The clear message is that high self-esteem is better
than low self-esteem and that raising one's self-esteem will bring about
improvements in one's life.
PARADOXES,
CHALLENGES, AND QUESTIONS
Had I not
cautioned the reader at the outset that there were good reasons to wonder about
the validity of these assumptions, the chapter could have ended here with the
resounding conclusion that the three assumptions are well supported by empirical
research and that the efficacy of their clinical implications is indisputable.
Yet lurking among these suppositions and their relevant findings are conceptual
paradoxes, empirical challenges, and unanswered questions that call the
assumptions into question and demonstrate gaping holes in psychologists'
understanding of self-esteem. In this section, I examine several issues that
pose difficulties for these assumptions and that should lead one to wonder how
well self-esteem is understood after all.
According to
the assumptions, people are motivated to preserve, protect, and occasionally
enhance their self-esteem, presumably because high self-esteem confers a variety
of benefits and adaptive advantages that low self-esteem does not. In support of
this, studies clearly show that people think and behave in ways that protect
their self-esteem from various onslaughts and that high self-esteem people
appear to fare better than low self-esteem people.
However, if
one accepts this assumption at face value, a paradox immediately arises: Many of
the tactics that people use to protect their self-esteem involve distortions in
their perceptions of reality. I can save my self-esteem in the face of failure
or rejection, for example, only at the cost of concluding--perhaps
erroneously--that I was not to blame. This conclusion is paradoxical because
most models of psychological well-being assume that adjustment is associated
with maintaining close contact with reality. People who distort reality are seen
as more-or-less dysfunctional, depending on the degree of the discrepancy
between their beliefs and reality. So if one accepts the assumptions, one must
conclude that a fundamental psychological motive (to maintain self-esteem) leads
people to misconstrue reality (presumably an undesirable, maladaptive response)
in the service of promoting adjustment (as seen, high self-esteem is typically
considered a keystone of mental health).
This paradox
is the subject of a broader debate involving the relationship between
"positive illusions" and psychological well-being (Colvin & Block,
1994; Taylor & Brown, 1988, 1994). Without entering the fray, let me mention
a couple of points relevant to the present discussion. From the standpoint of
adaptation and coping, it would seem that accurate perceptions-both of the world
and of oneself--are most, beneficial to a person's well-being. People are more
likely to survive and to succeed when they perceive their abilities,
characteristics, and worth accurately. It is difficult to identify any long-term
benefits to feeling better about oneself than one "deserves" to feet;
in fact, as seen in the next section, self-deception has several noteworthy
drawbacks. Needless to say, it is difficult to understand how a motive that
artificially reinforces flagging self-esteem in the absence of actual success or
goal accomplishment is beneficial, either psychologically or pragmatically,
except perhaps as a way of making the individual feel temporarily better. In the
long run, self-deception works against the individual's best interests;
narcissism is a prime, if extreme, example.
In an effort
to resolve this paradox, Baumeister (1989) suggested that there may be an
optimal margin of illusion; by seeing themselves as only slightly better than
they really are, people can "reap the benefits of illusions while avoiding
most of the negative consequences" (p. 182) Although the case can be made
for the occasional benefits of self-deception, data suggest that either over- or
underestimation sometimes may be beneficial (Baumeister, 1989; Baumeister &
Scher, 1988). If this is so, slight distortions of reality may sometimes be
beneficial, but negative illusions may be as helpful on occasion as positive
ones. Little basis exists for assuming that positive self-illusions are more
beneficial overall than negative selfillusions; as others have noted, the case
for or against positive illusions must await further research (Tennen &
Aftleck, 1993). In any case, the self-esteem motive that promotes
self-enhancement is not the unequivocal advantage to well-being that some have
supposed.
Contrary to
Assumption 2, research shows that high trait self-esteem is not always good nor
is low trait self-esteem always bad. I already discussed the extensive data
showing that high self-esteem is correlated with more positive outcomes than low
self-esteem. All of these are well-documented findings; although one can dispute
the results of any particular study, the body of evidence demonstrates certain
desirable correlates of high self-esteem.
The problem
is that researchers and practitioners have been myopic when it comes to equally
strong evidence showing that high self-esteem is sometimes associated with
negative psychological outcomes as well. For example, people who hold
excessively positive feelings about themselves are prone to take excessive
risks, particularly when their ego is threatened (Baumeister, Heatherton, &
Tice, 1993). This tendency is clearly dysfunctional; people who overestimate
their ability and worth can present significant dangers to themselves and to
others. (Do you want to ride. in a car with a driver who overestimates his or
her driving ability?)
People with
high self-esteem also are inclined to engage in nonproductive persistence,
beating their heads against the proverbial "brick wall" in the
self-deluded belief that they will eventually succeed (McFarlin, Baumeister,
& Blascovich, 1984; Shrauger & Sorman, 1977). They also tend to
externalize their failures, which blinds them to opportunities for
self-improvement (Dawes, 1994) and creates social difficulties when others
become aware of their self-serving attributions (Forsyth, Berger, &
Mitchell, 1981).
Baumeister,
Smart, and Boden (1996) reviewed an extensive body of evidence showing that
contrary to what is often supposed, high-not low-self-esteem is associated with
excessive aggression and violence. Studies of violent offenders show that they
are egotistical rather than self-deprecating. Research on childhood bullies
leads to the same conclusion; contrary to popular conceptions, bullies do not
have low self-esteem (Olweus, 1994). Many, although not all, rapists also have
inflated views of themselves (Scully, 1990), as do violent members of youth and
adult gangs (Jankowski, 1991). Among university students, people whose
self-esteem is high but variable (showing considerable day-to-day fluctuations)
are particularly angry and hostile (Kernis, Grannemann, & Barclay, 1989).
Baumeister et at. pointed out that many of the most inhumane actions throughout
history were perpetrated by people who, by virtue of their own sense of
superiority, regarded themselves as entitled to manipulate, dominate, and harm
others.
If one wants
clear-cut evidence that high self-esteem can be dysfunctional, one need not look
any further than the narcissistic personality. Despite narcissists' grandiose
self-views and unmitigated arrogance, some theorists have suggested that
narcissists actually have extremely low self-esteem and that their behavior is a
defense against a painfully negative view of the self (Kohut, 1971). However,
this psychological conceptualization of narcissism appears to arise, in part,
from an implicit acceptance of Assumption 2. Given that high self-esteem is
inherently beneficial, the argument goes, anyone who behaves as dysfunctionally
as a narcissist must have very low self-esteem. However, the research evidence
seems to refute this argument and shows narcissism to involve high self-esteem
(Emmons, 1984; Raskin, Novacek, & Hogan, 1991a, 1991b). Narcissists appear
to feel very good about themselves-far better than they deserve to feel--and
their behavior is the consequence of believing that they are better than other
people and, thus, duty entitled to dominate, manipulate, and exploit others
(Leary, Bednarski, Hammon, & Duncan, 1997).
Finally,
aside from the fact that self-esteem is not the unqualified blessing that it is
often assumed to be, the empirical relationships between high self-esteem and
psychological well-being (or, conversely, between low self-esteem and
psychological difficulties) are generally weak, (Dawes, 1994). Statistically
speaking, these correlations may be reliably different from zero, yet they do
not indicate a strong link between self-esteem and various emotional and
behavioral problems. In commenting on the extensive report of the California
Task Force to Promote Self-Esteem and Personal and Social Responsibility,
Smelser (1989) concluded that "the news most consistently reported ... is
that the associations between self-esteem and its expected consequences are
mixed, insignificant, or absent" (p. 15).
Even so,
enough connections have been documented between self-esteem and psychological
difficulties to suggest that raising low self-esteem improves psychological
well-being and produces desirable changes in people's behavior (Assumption 3).
However, the problem with this conclusion is that it is based on the premise
that self-esteem is causally related to behavior and emotion. It is said that
low self-esteem "causes" avoidance or poor coping, that high
self-esteem "leads to" positive changes, or that self-esteem
"affects" happiness and well-being. However, to my knowledge, not one
shred of scientific evidence exists to support the idea that self-esteem has any
causal influence on behavior or emotion whatsoever.
Psychologists
are taught at an early stage in their education and then repeatedly reminded
that the fact that two variables are correlated with one another-even highly
correlated-cannot be taken to mean that one causes the other. Yet in the case of
self-esteem, many psychologists have disregarded this dictum. The data show that
high self-esteem correlates with many indexes of psychological well-being
(although, as seen, the data are neither as strong nor as consistent as often
supposed). Yet these correlations do not tell anything about whether low
self-esteem causes psychological difficulties or whether raising self-esteem
causes positive psychological change. Given that virtually all of the research
relevant to differences between low and high self-esteem people is
correlational, no causal inferences can be drawn.
No matter
what might be said about particular conceptualizations, studies, and
interpretations, theory and research convincingly portray self-esteem as an
important psychological entity. Yet it is not at all clear from existing
evidence what self-esteem does or why it is important to human functioning.
Few writers
have addressed the question of self-esteem's function. Most theorists and
researchers implicitly appear to assume either that people need self-esteem for
its own sake or that people seek self-esteem simply because it creates positive
feelings. The first perspective- that self-esteem is inherently important-is not
an explanation at all because it begs the question of why do people need to feet
good about themselves.
The second
perspective -that people seek self-esteem because it is emotionally
rewarding-likewise skirts the issue of function. Some have suggested that the
positive feelings of high self-esteem promote well-being because people are more
likely to take action and persevere in the face of obstacles when they feet
confident, secure, and in control (Greenwald, 1980; Shrauger & Sorman, 1977;
Taylor & Brown, 1988). This is undoubtedly true, yet the argument fails to
account for the negative feelings of low self-esteem. Is it really beneficial to
have a mechanism that deflates one's feelings Of confidence, security, and
control, particularly in response to events such as failure and rejection, where
proactive, remedial action is often needed?
So one is
left with the question of what purpose does self-esteem serve? Emotion theorists
have insisted that emotions are functional-t-hey warn, motivate, energize,
reward, and punish a person-and their behavioral expressions serve important
interpersonal functions by communicating that person's affective state to others
(Frijda, 1986; Izard, 1977). Granted, people sometimes purposefully override the
natural functions that emotions serve (by taking mood-altering drugs, e.g.), but
at heart, emotions are functional. What then do the feelings associated with
self-esteem do?
For readers
who are drawn to evolutionary arguments, one can state the functional question
even more strongly. The universality of the self-esteem motive suggests that it
(whatever it really is) is an inherent aspect of human nature. All normal people
beyond a certain age react emotionally to events that seem, at least on the
surface, to threaten their self-esteem and engage in behaviors that maintain
their self-esteem. Of course, cultural differences exist in the kinds of events
that shake people's self-esteem and in the esteem-preserving tactics that people
use, but clearly self-esteem is a potent psychological process worldwide. Such
universality suggests that the self-esteem system is a basic aspect of human
nature and likely a product of human evolution, but how and why would a
mechanism for promoting self-esteem have evolved? Is there any adaptive
advantage to seeing oneself more positively than one ought? Would not a general
motive for self-accuracy be more adaptive than one for self-enhancement? Without
denying the occasional benefits of self-deception discussed earlier, organisms
undoubtedly fare best in the evolutionary game when they accurately perceive
their own personal resources and challenges.
One of the
few explicit efforts to address the function of self-esteem comes from the
proponents of terror management theory. According to the theory, self-esteem
serves to buffer people against the existential terror they experience at the
thought of their own death and annihilation (Solomon, Greenberg, &
Pyszczynski, 1991). Thus, people are motivated to maintain their self-esteem
because it helps them avoid the paralyzing terror they would otherwise
experience. Considerable experimental evidence supports aspects of terror
management theory. For example, making mortality salient appears to heighten
people's concerns with self-esteem, and high self-esteem does, in fact, lower
people's anxiety about death (to name one of many threatening things; Greenberg
et at., 1992). Yet the data do not yet support the strong argument that the
function of the self-esteem system is to buffer existential anxiety, and at
least a few studies fail to support aspects of the theory (Sowards, Moniz, &
Harris, 1991). Even so, terror management theory must be credited for providing
an account, albeit a controversial one, of what self-esteem does.'
Bednar et at.
(1989) also offered a functional explanation of self-esteem; they suggested that
self-evaluative processes provide "a basis for continuous affective
feedback from the self about the adequacy of the self" (p. 112). This
affective feedback--self-esteem--varies as a function of whether the individual
is coping with or avoiding a psychological threat. Coping leads to high
self-esteem; avoidance leads to low self-esteem. In turn, the level of
self-esteem affects the probability of subsequent coping versus avoiding
responses; high self-esteem increases coping, and low self-esteem increases
avoidance. The difficulty with this perspective, however, is that the feedback
loop is dysfunctional when people are already coping poorly. Decreasing
self-esteem would signal inadequacy, thereby leading to further avoidance,
followed by even lower self-esteem, greater avoidance, and so on. As Bednar et
al. (1989) noted, "the psychologically weak will become weaker with the
passage of time" (p. 133). Such a feedback system might be functional if
changes in self-esteem reflected a person's true resources for effective coping
because a poorly coping individual might be better off avoiding than engaging
the threat. But given that self-esteem is only weakly tied to one's
"true" ability to deal with one's environment, such a system would
have questionable benefit.
Readers who
were formerly unacquainted with the complexities of the existing literature on
self-esteem may be surprised by the "mess" this last section revealed.
Given the vast volume of theory and research that invokes the construct of
self-esteem, one might have imagined that most of the fundamental issues would
have long been resolved and the central questions answered. I think that the
current state of the area itself is attributable, in part, to psychologist's
noncritical acceptance of fundamental assumptions that are on surprisingly shaky
logical and empirical ground. It all seemed so plausible-so conceptually
tidy-that most psychologists (myself included) did not stop to question the
basic premises. Once one wades into the quagmire of paradoxical claims,
empirical challenges, and unanswered questions, one cannot escape the conclusion
that the assumptions under which one is operating are, at worst, wrong headed
or, at best, in need of qualification or revision.
SOCIOMETER
THEORY
My own
interest in self-esteem began with the question, Why are people motivated to
maintain their self-esteem? My sense was that if one could concoct a viable,
functional explanation of the self-esteem motive, one might begin to make sense
of the paradoxes, challenges, and questions that plague the literature. The
outcome of this work is what is called sociometer theory.' The details of
sociometer theory are available elsewhere (Leary & Baumeister, in press;
Leary & Downs, 1995; Leary, Tambor, Terdal, & Downs, 1995), so I confine
my discussion to points that are most relevant to the issues raised above and
that have implications for understanding the social and psychological importance
of self-esteem.
The basic
premise of sociometer theory is that the self-esteem system is essentially a
subjective indicator or gauge that monitors the quality of one's relationships
with other people. Upward changes in state self-esteem signal an improvement in
the degree to which one is socially included or accepted by other people,
whereas downward changes in state self-esteem signal a deterioration in the
degree to which one is included or accepted.
Human beings
evolved as social animals because they could not survive and reproduce without
the support and protection of other human beings. Because brain systems that
promoted social living conferred a distinct adaptive advantage (How long would
you last alone on the Serenghetti Plains of Africa armed with only a spear?),
motivational systems evolved that prompted people to seek out the company of
other people, to live in social groups, and to form social bonds (Barash, 1977;
Baumeister & Leary, 1995). Maintaining these vital relationships depended
not only on the individual being motivated toward sociality but also on the
assurance that one would be accepted by other people in one's social group.
Failure to be accepted and socially included by at least a few other people
would leave the individual isolated and without the protective affordances of
group living. Given that social acceptance by the clan was vital and that
rejection was tantamount to death, a system evolved to monitor the degree to
which others responded to the individual in an accepting or rejecting fashion.
This system,
which operates more or less continuously at a preattentive level (because a
person cannot consciously think about other people's reactions all of the time),
monitors others' reactions for cues indicating disinterest, disapproval,
avoidance, or outright rejection-any signal that the person is not adequately
valued and accepted as a member of the group. When such cues are detected, the
individual is alerted through the induction of negative feelings-just as many
other systems induce negative affect when threats to well-being are detected.
Furthermore, the system induces a motive to restore one's acceptance through
whatever behaviors seem appropriate to the social context.
Thus,
according to sociometer theory, the self-esteem system is a sociometer that
monitors the social environment for threats to one's social inclusion, induces
negative self-relevant affect--a loss of self-esteem when such threats are
detected, and motivates the individual to attend to his or her interpersonal
relationships. Subjective feelings of self-esteem function in this process as a
psychological readout of one's inclusionary status. Having sketched the basics
of sociometer theory, I now consider how the theory makes sense of the three
assumptions described earlier.
Sociometer
theory casts the self-esteem motive in a considerably different light from
previous conceptualizations. In fact, the phrase "self-esteem motive"
becomes somewhat of a misnomer because, strictly speaking, people are not
motivated to maintain their self-esteem at all. Instead, their goal is to
sustain an acceptable degree of social acceptance and to avoid social rejection.
Subjective self-esteem is the internal, psychological indicator that guides
their efforts to promote acceptance and ward off rejection.
According to
the theory, the kinds of events that threaten self-esteem and induce people to
protect and restore their self-esteem are best understood as events that affect
a person's perceived acceptance and inclusion by other people (inclusionary
status). Events that threaten a person's self-esteem are precisely those things
that if known by others, would likely lead them to devalue their relationship
with the person. As seen earlier, the events that are most likely to lower
self-esteem (and which experimental social psychologists use precisely for that
purpose) involve rejection (which, by definition, indicates social exclusion)
and failure (which is often associated with relational devaluation and possible
rejection because other people do not like "losers"). However, events
that raise self-esteem are those that increase a person's perceptions of being
accepted and included -achievement, recognition, compliments, admiration, and
the like.
Empirical
support for this notion comes from a series of studies (Leary, Tambor, et at.,
1995). In one study, participants indicated how they thought other people would
react if the participants performed each of several behaviors and then rated how
their own self-esteem would be affected if they performed each behavior. As
sociometer theory predicts, participants' ratings of their self-esteem closely
mirrored their expectations regarding how others would respond vis-à-vis
acceptance and rejection. In a second study, participants described a past
interpersonal encounter and then rated how included or excluded they felt in the
situation as well as how they felt about themselves at the time. Again,
perceived inclusion--exclusion correlated very highly with self-feelings, as
sociometer theory predicts. The strong link between social inclusion-exclusion
and state self-esteem was replicated in two laboratory experiments as well.
Participants who believed that other participants had decided to exclude them
from a laboratory group demonstrated sharp decreases in state self-esteem.
Furthermore,
when one looks at the behaviors that researchers have interpreted as ways of
preserving self-esteem, these behaviors are actions that one would expect to
halt the slide of relational devaluation, if not actually increase social
acceptance. When people make self-serving attributions, offer excuses,
"scapegoat," self-handicap, and engage in other ego-defensive
behaviors, they not trying to preserve self-esteem per se but are trying to
protect their inclusionary status.
Within
sociometer theory, individual differences in trait self-esteem can be
reconceptualized as individual differences in perceived inclusionary status.
People with relatively high self-esteem tend to believe that they are generally
acceptable individuals and that other people value their relationships with
them. People with relatively low self-esteem walk through life assuming that
they are less acceptable and that other people value their relationships less if
not expressly devalue them (Leary, Tambor, et al., 1995, Study 5).
Viewed in
this way, the correlates of low versus high trait self-esteem reflect the impact
of feeling generally rejected versus accepted, respectively. Because being
accepted and valued results in more positive psychological outcomes than being
rejected and devalued, one generally finds positive relationships between high
self-esteem and well-being and negative relationships between high self-esteem
land several psychological difficulties (Leary, Schreindorfer, et al., 1995).
However, from the standpoint of sociometer theory, self-esteem has no causal
impact on the link between self-esteem and psychological well-being. Self-esteem
and psychological well-being are coeffects of the degree to which people
perceive they are included, accepted, and valued. Thus, the causal agent in
these relationships is not self-esteem but perceived social acceptance or
rejection.
Sociometer
theory also can help one understand those instances in which high self-esteem is
associated with negative outcomes. For example, in the case of the link between
high self-esteem and aggression (Baumeister et al., 1996), people who already
feel highly included may not worry about the interpersonal implications of
behaving in a socially inappropriate or undesirable fashion. Similarly, the
egocentric, exploitive behavior of narcissists is what one would expect of
people who believe their superiority entities them to exploit others with no
fear of serious interpersonal consequences. Indeed, narcissists often seem
genuinely surprised when others react negatively to their egocentric behavior;
they seem to take others' acceptance of them for granted.
Sociometer
theory does not dispute that interventions designed to enhance self-esteem often
produce measurable psychological benefits, as the advocates of self-esteem
programs maintain. However, the theory suggests that the mediating processes are
different from what are typically assumed. Specifically, improvements in
well-being that accompany enhanced self-esteem are mediated by changes in
people's perceptions of the degree to which they are accepted, included, and
valued by other people rather than by heightened self-esteem per se. A large
body of research attests to the psychological benefits of being accepted and
valued. People who experience deficits in their sense of belongingness suffer a
wide range of emotional and behavioral problems (for a review, see Baumeister
& Leary, 1995).
An
examination of programs designed to enhance self-esteem reveals that these
interventions typically include features that would be expected to increase real
or perceived social acceptance. For example, Frey and Carlock's (1989) model
stresses that the facilitator should consistently offer sincere positive
feedback to the participant. Among other things, participants learn to affirm
the positive qualities they see in themselves and others. Clearly, such an
approach not only makes one's own desirable attributes salient, but it also
changes one's behavior toward others in ways that increase the likelihood of
social acceptance.
Pope et al.'s
(1988) program, designed specifically for children, takes a more behavioral
approach, but it too would be expected to increase social acceptance. After a
careful analysis of the participants' self-esteem in several domains, exercises
are undertaken to improve "performance" in deficient domains (e.g.,
social relationships, academics, or physical appearance). Among other things,
the participants team to solve social problems, develop social and communication
skills, and increase self-control. From the perspective of sociometer theory,
Pope et al. are helping children improve their chances of social acceptance,
thereby producing changes in how acceptable they feel.
Many clinical
interventions and much popular advice explicitly or implicitly stress the
importance of interpersonal acceptance and positive social relationships in
raising self-esteem. Considerable research supports the speculations of early
symbolic interactionists (Cooley, 1902; Mead, 1932) that people's self-concepts,
as well as their self-esteem, are related to their perceptions of how they are
perceived by other people-what are often called reflected appraisals (Felson,
1993; Shrauger & Schoeneman, 1979). For example, participants who learn that
other people hold negative impressions of them or that others do not want to
interact with them show drops in their state self-esteem (Leary, Haupt,
Strausser, & Chokel, 1998; Leary, Tambor, et al., 1995). Likewise, research
in developmental psychology clearly shows that children incorporate the
reflected appraisals of their parents and peers into their self-concepts
(Harter, 1993) and that self-esteem is strongly related to the quality of the
relationships that exist between children and significant others in their lives
(Coopersmith, 1967). Not surprisingly then, accepted children have higher
self-esteem than rejected children.
Sociometer
theory helps to explain why self-esteem is an inherently social phenomenon.
People's self-esteem is based on how they think others evaluate them because
that is precisely how the system was "designed" to operate.
Furthermore, the theory explains why self-esteem correlates more highly with how
people believe they are viewed by others than with others' true perceptions of
them (Felson, 1993; Shrauger & Schoeneman, 1979) and why self-esteem is
correlated highly with people's performance in domains that they think are
important to other people (Harter & Marold, 1991). Contrary to theories that
conceptualize self-esteem as a purely personal evaluation of one's own
characteristics, research suggests that people's self-esteem is far more
sensitive to others' reactions to them than to how they see themselves.'
On the
surface, sociometer theory may seem to suggest that all threats to self-esteem
reflect interpersonal events that have implications for the individual's social
acceptance and that people's efforts to deal with such threats always involve
behaviors that are directed toward other people. However, such an extreme
interpretation flies in the face of evidence suggesting that purely private
events can affect self-esteem and that people sometimes deal with such threats
in cognitive rather than behavioral ways. How does sociometer theory account for
these "private" threats to self-esteem?
First, events
do not need to occur in public for them to affect one's relationships with
others (and, thus, involve the sociometer). All individuals have done things in
private that if found out by others, would undermine their relationships. To
protect people against such indiscretions, the sociometer must alert people to
the possible interpersonal repercussions of private behaviors, so that they will
refrain from doing them, conceal them more carefully, or take anticipatory
measures to minimize the interpersonal damage that will occur if others should
learn about them.
Second, some
writers have suggested that private self-deception does serve an interpersonal
function (Neese & Lloyd, 1992; Trivers, 1985). One is more likely to succeed
in convincing others of one's ability, virtue, and worth if one believes in them
oneself. In this view, private self-esteem maintenance helps people behave in
ways that enhance their social desirability and acceptance by others.
Third,
sociometer theory does not deny that people may engage in purely cognitive
self-deceptions that serve no purpose other than to make them feel good. These
positive illusions may contribute to a transitory sense of well-being by
reducing anxiety, depression, and a sense of helplessness when more direct means
of improving one's lot are not readily available (Taylor & Brown, 1988). The
fact that people may do this does not argue against sociometer theory as much as
it suggests that people can "bypass" the sociometer system if they
desire. Just as the fact that people sometimes eat when they are not hungry does
not indicate that they lack a system to regulate food intake, the fact that
people sometimes engage in cognitive strategies simply to improve their mood and
feelings about themselves does not imply that the self-esteem system is not, at
heart, a mechanism for maintaining a minimum degree of social inclusion.'
Sociometer
theory provides an alternative perspective on the social and psychological
importance of self-esteem. Self-esteem is important socially because it is
involved in the maintenance of interpersonal relationships. The self-esteem
system helps people to avoid rejection by alerting them to potential exclusion
and motivating behaviors that promote acceptance, thereby enhancing the quality
of their interpersonal lives and increasing the chances of them obtaining a
variety of social affordances. Psychologically, self-esteem is important because
social acceptance promotes psychological well-being.
IMPLICATIONS
FOR CLINICAL AND
COUNSELING
PSYCHOLOGY
In this final section, I briefly
explore some implications of sociometer theory for understanding and treating
emotional and behavioral problems that implicate self-esteem. The most important
point is that whatever a client's presenting difficulty, low self-esteem should
rarely be regarded as the underlying problem. Low self-esteem may signal the
nature of the problem--a sense of relational devaluation or rejection-but low
self-esteem itself is not the cause of the client's emotional distress or
dysfunctional behavior.
Given that
self-esteem is not the problem, raising self-esteem is rarely the solution.
Sociometer theory suggests that the focus should be on helping clients to
cope-through either behavioral or cognitive means-with their feelings of
relational devaluation. This point may seem to fly in the face of the documented
efficacy of clinical interventions that include efforts to raise self-esteem.
However, as suggested earlier, the effectiveness of such approaches is not
mediated by enhanced self-esteem. Instead, the clinical strategies that
practitioners use to raise self-esteem often involve procedures that increase
real or perceived social inclusion.
From the
standpoint of sociometer theory, raising self-esteem artificially- that is,
without a corresponding increase in perceived inclusion-is particularly ill
informed. As callous (and possibly elitist) as it may sound, sociometer theory
suggests that some people ought to have low self-esteem. People who consistently
behave in destructive and inappropriate ways that lead others to ignore, avoid,
or reject them have low self-esteem because their sociometer has accurately
detected a low degree of inclusion. To try to convince these individuals that
they are in fact valuable, worthy, wonderful people may dissuade them from
taking action to deal with the real problem (Dawes, 1994) as well as lead to
confusion and anger. ("if I'm so wonderful, why does everybody avoid
me?") Such an approach tries to override the sociometer's natural signals
that the person's social acceptance is in jeopardy-somewhat akin to trying to
convince a driver to ignore the fact that the fuel gauge of the car is on empty.
A more
complex situation arises when a normal and socially acceptable client receives
undue rejection from important people in his or her life. Such an individual
might well have low self-esteem not because of any personal shortcomings but
because others do not sufficiently value their relationships with him or her.
Critical or uncaring parents, abusive spouses, and egocentric friends may induce
low self-esteem in people who do not deserve it.
The clinical
approach in such cases would be to help unfairly rejected clients understand
that the self-esteem system is, by design, an indicator of others' reactions to
them and, thus, their low self-esteem is nothing more than an accurate
reflection of how others have treated them. At the same time, they must see that
their self-esteem is not, as they likely assume, an index of their true worth as
an individual. If they can view their low self-esteem as a product of other
people's weaknesses and shortcomings rather than their own deficiencies, they
may feet badly about how they have been devalued by significant others yet learn
not to make the unwarranted leap to self-deprecation. At the same time, they may
be encouraged to pursue new relationships in which they are valued.
Some
practitioners have denied that clinical interventions should focus on helping
clients to obtain more positive responses from others to raise self-esteem.
Bednar et al. (1989), for example, reflected this sentiment