But a spate of recent articles suggests that the tide may be turning.
When Senator Robert Torricelli failed to admit wrongdoing as he resigned, Andrew Sullivan's opinion piece in Time magazine (October 7, 2002) blamed "the sheer, blinding brightness of the man's self-love" on the self-esteem movement. An article by Erica Goode in the New York Times (October 1, 2002) proclaimed that "'D' students . . . think as highly of themselves as valedictorians, and serial rapists are no more likely to ooze with insecurities than doctors or bank managers." Worse, the writer said, some people with high self-esteem are likely to respond with aggression if anyone dares to criticize them: "Neo-Nazis, street toughs, school bullies . . . combine preening self-satisfaction with violence."
In the pages of the New York Times Magazine (February 3, 2002), psychologist Lauren Slater maintained that the self-esteem movement has produced a "discourse of affirmation" that ladles out praise regardless of achievement. She concluded that self-appraisal and self-control need to take the place of self-esteem in psychotherapy. In the Christian Science Monitor (October 24, 2002), conservative commentator Dinesh D'Souza said of self-esteem that "unlike honor, it does not have to be earned."
Most such media critiques draw on the well-publicized research findings of the same three social psychologists: Roy Baumeister, Jennifer Crocker, and Nicholas Emler. But, as we shall see, these psychologists rely on mistaken conceptions of self-esteem and on flawed research methods.
Psychologists against Self-Esteem
Roy Baumeister, a professor of psychology at Case Western Reserve University, is the academic psychologist best known for claiming that "D" students, gang leaders, racists, murderers, and rapists have high self-esteem. Examining empirical studies on how murderers and rapists respond to self-defining statements, Baumeister and his colleagues have pointed out that these individuals consciously believe they are superior, not inferior—a belief that, Baumeister says, is characteristic of high self-esteem.
Baumeister does not claim that high self-esteem necessarily leads to aggression; in order to do so, it must be combined with an ego threat (a challenge to one's high self-appraisal). In a study that has gotten less media attention, Baumeister and Brad Bushman tested this hypothesis experimentally. Participants were given the Narcissistic Personality Inventory, which contains such items as "If I ruled the world it would be a much nicer place," and the Rosenberg Self-Esteem Scale. (See below for more about these questionnaires.) The ego threat was a strong criticism of the participant's intellectual competence. Participants were given the opportunity to aggress against the people who had criticized them, by delivering a blast of noxious noise. (Since this was a social psychology experiment, the noise was not really delivered to the critic.) What the results showed was that the narcissism measure, not the self-esteem score, predicted the strength of the aggressive response (the intensity and duration of the noise). But because those who scored high on the narcissism questionnaire also tended to score high on the self-esteem scale, it looked as though some people with "high" self-esteem are aggressive when their sense of self is threatened.
The research of Jennifer Crocker, a professor of psychology at the University of Michigan, has indicated that deriving one's self-esteem from certain "external" contingencies, such as appearance, is associated with potentially destructive behavior, including alcohol and drug use, and eating disorders. Crocker and her colleagues conducted a study with applicants to graduate programs who based their self-esteem on academic competence. They found that such students showed greater increases in self-esteem on days of acceptance and greater decreases on days of rejection. The stability of self-esteem is an important area of investigation because several studies have found that people whose self-esteem is unstable (that is, fluctuates substantially on a daily basis) are more emotionally reactive to everyday events. They are more likely to become depressed when confronted with daily hassles and are more prone to anger when their self-esteem is threatened.
Crocker's findings have led her to conclude that the pursuit of self-esteem has significant costs. Crocker has gone on to contend that self-esteem ought to be non-contingent: not based on any source at all. If people value themselves positively without conditions or criteria, Crocker maintains, they will be less likely to suffer from problem drinking, maladaptive hostile reactions, and depression.
Nicholas Emler, a psychologist at the University of Surrey, is a researcher whose work has garnered extensive media attention in Great Britain. He also believes that high self-esteem is a source of trouble. His 2001 monograph Self-Esteem: The Costs and Causes of Low Self-Worth
reviews a wide range of published research, concluding that low self-esteem is not a risk factor for delinquency, violence against others, or racial prejudice. On the contrary, he suggests, high
self-esteem is the more plausible risk factor. Relying on Baumeister's and Crocker's evidence about the pitfalls of self-esteem, as well as other research, Emler asserts that people with high self-esteem are more likely to engage in risky pursuits, such as driving too fast and driving drunk. Lastly, Emler finds little evidence that self-esteem and educational attainment are associated, since even failing students can show high self-esteem on questionnaires.
The Objectivist Conception of Self-Esteem
Is what these researchers are saying true? Has the public been led to believe that high self-esteem is a cure for personal and social ills when empirical evidence makes it out to be a cause of those ills? Not unless self-esteem consists in feeling good about oneself, or in feeling superior to others, or in taking pride in superficial attributes like appearance. The research suggests very different conclusions if one interprets it in light of the concept of self-esteem that Ayn Rand proposed and Nathaniel Branden subsequently refined and elaborated. Genuine self-esteem has two dimensions of self-evaluation: (1) an evaluation that one is competent to deal with life's basic challenges (self-efficacy) and (2) an evaluation that one is worthy of happiness (self-worth). Self-worth encompasses the conviction that one is deserving of success, love, and friendships, and the acceptance of positive feelings—such as pride and joy—as "natural" and proper to one's existence. Moreover, according to Branden, these evaluations of competence and of worth, if they are to be secure and enduring, need to be based on objective standards. The standards for adult self-esteem include self-reflective and independent thought; taking responsibility for and authentically asserting one's thoughts, beliefs, values, and actions; pursuing meaningful life goals; and adhering to moral values that are based on reason. Branden calls these objective standards of self-esteem "pillars," meaning that they are foundational to self-esteem. If we act in ways that meet these objective standards, our self-esteem will necessarily rise; if we fail to act in these ways, or betray these standards, our self-esteem will drop. In sum, these sources of self-esteem are internal to the person; they depend on self-directed psychological processes that are under each person's control.
We note that Branden's conception of self-esteem is not an isolated one. Two clinical psychology theorists, Richard Bednar and Scott Peterson, have proposed a similar model in their book Self-Esteem .
For a person's self-esteem to improve, they say, he must confront anxiety with acceptance and realism. This requires coping directly with the unwanted thoughts and feelings that precipitate anxiety, such as pain, embarrassment, shame, and fear. But such realism is possible only if (1) a person perceives himself as responsible for his thoughts, feelings, and behavior; (2) he is able to accept the beliefs and feelings motivating the anxiety; and (3) he is able to disclose his authentic thoughts and feelings to others in an appropriate context. These three self-directed processes correspond to Branden's pillars of self-responsibility, self-acceptance, and self-assertiveness.
Branden makes a sharp distinction between genuine self-esteem and pseudo-self-esteem. Pseudo-self-esteem relies on "external" sources, such as being admired or approved by others, social status, or physical appearance. People tend to put their reliance on external sources to the extent that they are lacking in the self-directed psychological processes that constitute internal sources of self-esteem. Because external sources are not under our direct control, they cannot realistically enhance our feelings of competence. Self-esteem that depends on them is therefore insecure and under constant threat. Branden's notion of pseudo-self-esteem, then, can help explain how some people who have high scores on self-esteem questionnaires can respond to an ego threat with aggression, as in Bushman and Baumeister's experiment. They are people with pseudo-self-esteem; their feelings of competence and worthiness are "on the line" when challenged, and so they are more prone to defend against this threat by lashing back at the source of criticism.
On theoretical grounds, the foregoing accounts make a good deal of sense, but how can stable, realistic self-esteem that draws on internal sources be measured?
The procedures in common use for measuring self-esteem are highly vulnerable to criticism. Virtually all self-esteem research has relied on answers to a handful of self-report questionnaires. The one most relied on (Emler calls it the gold standard) was developed in 1965 by sociologist Morris Rosenberg. It seeks responses to statements such as "On the whole I am satisfied with myself," and "I feel that I have a number of good qualities."
This procedure of asking people to introspect or engage in self-report is a legitimate one. But introspective data are of no value if the person is not answering truthfully, or if he is out of touch with the feelings he is supposed to report, or if he lacks an understanding of the matter at issue. All of these problems are to be expected in measuring self-esteem.
Consider the matter of truthfulness. According to Nathaniel Branden, conscious feelings of low self-esteem are so hard to bear that people suffering such feelings have an incentive to put up psychological defenses against them. On this view, a person who is compensating for feelings of inefficacy or unworthiness might well assert, "On the whole I am satisfied with myself."
In addition, people's answers to questionnaires can be influenced by "social desirability," and certainly the growth of the self-esteem movement has led to self-esteem's becoming a socially desirable quality. Perhaps this explains why American college students frequently show average scores of 70 out of a possible 90 on the Rosenberg Self-Esteem Scale. By contrast, according to Hazel Rose Markus and Shinobu Kitayama, East Asians who express favorable opinions of themselves are likely to be criticized for immaturity, or excessive pride, or disregard for family and community. Scores on the Rosenberg Scale from China or Japan might therefore be misleadingly low instead of misleadingly high.
To make matters worse, when the usual questionnaires are used to measure self-esteem, researchers cannot check the evaluations against reality. When self-report questionnaires can be crosschecked, they are often found to be seriously inadequate. For example, many American high school students who say they are good at mathematics perform poorly on math tests. On the Objectivist conception of self-esteem, this is a critical measurement flaw, for genuine self-esteem must be realistically based.
Researchers are just beginning to explore alternatives to these self-report questionnaires. One inspiration for their efforts is the Implicit Attitude Test, which was developed by researchers who were concerned that self-reported racial attitudes might be unreliable when it is socially undesirable to admit certain prejudices. Since 1995, several implicit measures of self-esteem have been proposed. Unlike explicit measures, these do not ask people to report on their self-esteem. Indeed, "implicit self-esteem" can be defined as an automatic, non-conscious self-evaluation.
One promising method of getting at implicit self-esteem involves recording reaction times to computer-based word associations, which is the basis of Anthony Greenwald's Implicit Association Test (IAT). It is assumed that participants are not aware their self-esteem is being assessed during the test. Thus, responses can be seen as more accurate indicators of a person's self-esteem. At this point, the evidence in support of the IAT is limited but growing.
Walter Foddis and his colleagues have tried an implicit measure of self-esteem using sentence completions, for example, "My self-esteem depends on…." The completions, retrieved quickly from memory and subjected to little or no conscious review, are the "implicit" material here. Participants' sentence endings are scored based on how closely they correspond to Branden's pillars of self-esteem. One of the advantages of this approach, as compared with the standard questionnaires, is that participants are not being given ready-made socially desirable statements to evaluate, such as "My self-esteem depends on taking responsibility for my own behavior."
Foddis et al. have contrasted those who scored high on both explicit self-esteem (as measured by self-report) and implicit self-esteem (as measured by sentence completions) with those who scored high on explicit self-esteem but low on implicit self-esteem. Those with high explicit but low implicit self-esteem responded more poorly to criticism and endorsed fewer statements that they were worthy of being told they were loved. Even though they reported high self-esteem, participants who fell into this category resembled those who scored low in explicit self-esteem. Such data support the notion of pseudo-self-esteem. Among those with high explicit but low implicit self-esteem, the more extreme a person was in generating external sources of self-esteem in sentence completions, the more likely he would respond with defensiveness and hostility when criticized.
Much more work remains to be done in assessing implicit processes in self-esteem, but the evidence for their usefulness in self-esteem research continues to grow. The most pressing problem is the lack of agreement among the different implicit procedures currently under development.
Moderate positive correlations exist between scores on self-reported self-esteem and scores on the Narcissistic Personality Inventory (NPI). (A narcissistic person harbors grossly inflated opinions of his competence and his worth, regarding himself as superior to others.) Because the NPI involves self-reporting, it is subject to some of the criticisms that have been raised against the Rosenberg Scale. Researchers who sharply distinguish self-esteem from narcissism worry about the correlation between the two tests and seek better theoretical models and better measurement procedures for both phenomena.
Carolyn Morf and Frederick Rhodewalt have put forward a new model of narcissism that supports the effort to distinguish high self-esteem from vanity, self-absorption, and ego-inflation. Morf and Rhodewalt suggest that a defining characteristic of the narcissist is overdependence on social sources to affirm a grandiose sense of self. The narcissist needs other people, but only because of their instrumental value in bolstering his sense of self. Under the narcissist's grandiose exterior, therefore, is a vulnerable sense of self that is easily threatened and must be constantly supplied with affirmation.
Morf and Rhodewalt cite several studies that show how narcissists undergo instability and fluctuations in their explicit self-esteem because their self-image is both grandiose and vulnerable. For instance, in studies contrasting participants who scored high and low on the NPI, those with high scores displayed greater day-to-day fluctuations in self-reported self-esteem than did less narcissistic individuals. The narcissist's daily self-esteem was also more highly correlated with positive or negative social interactions: it depended on whether the narcissist received acceptance and whether the interactions made the narcissist feel "like himself." In this regard, a narcissist can be understood as a person with pseudo-self-esteem, who relies more on external sources for his self-esteem than internal sources.
Empirical findings suggest that other characteristics of the narcissist include viewing his abilities and accomplishments as superior, bragging to elicit positive reactions, lying about his past to make himself look better, and derogating anyone who gives him negative feedback. In short, the narcissist keeps looking for ways to obtain validation of himself, but the external validation is never enough to convince him of his own adequacy. Morf and Rhodewalt's model of narcissism ties in nicely with Branden's argument that those who do not have genuine self-esteem strive to "fake" it. The narcissist fakes it by lying and bragging to solicit admiration from others, and by downrating anyone who fails to be admiring. Morf and Rhodewalt marshal substantial empirical support for their model but acknowledge the need for more research on the relationship between psychological processes and interpersonal behavior. Focusing on sources of self-esteem could further illuminate what is going on in narcissism.
Sources of Dispute
Fundamental differences in philosophy are one major reason that the subject of self-esteem is so much debated. What a person thinks about human nature will deeply affect his view of the matter. Thus, Baumeister seems to believe that every person either views himself as superior to others or wishes to. In his 1993 book Self-Esteem: The Puzzle of Low Self-Regard, he wondered why anyone would have low self-esteem. In a 2001 commentary, he opined that the reason narcissism fascinates psychologists "may well be that narcissists indulge the cravings that most people have."
A different strain of thinking in social psychology, which can be seen in Crocker's work, holds that human selves are socially constituted, so that your evaluation of yourself must derive from the way that others evaluate you. Within social psychology, this notion of the socially constituted self was the leading idea of two early twentieth-century thinkers, George Herbert Mead and Charles Cooley; they in turn owed intellectual debts to the communitarian pragmatism of John Dewey and to the still earlier nineteenth-century followers of G.W.F. Hegel. From this point of view, not only is much of what you do socially constituted, but your individuality is not fully real; thus, there is no way you could evaluate yourself without relying on others' evaluations of you.
Moving beyond academic social psychology, into clinical practice and education, one finds widespread distrust of the idea that self-esteem should be contingent on a person's self-directed functioning. Self-esteem should not have to be realistic, according to this outlook; it should not have to depend on how you are using your mind.
Today's clinical practice owes large debts to the humanistic psychologists of the mid-twentieth century. For example, Carl Rogers held that the therapist must show "unconditional positive regard" for the client. But where Rogers intended to encourage self-acceptance on the part of the client, others have gone much further, concluding that we should each have unconditional positive regard for ourselves.
Albert Ellis, for instance, is a prominent therapist who has argued that accepting your feelings and your actions as your own is not enough; genuine self-acceptance means never "rating" yourself at all. For Ellis, self-esteem is always unstable, ever vulnerable to threats: if you rate yourself on your competence or your worth, you may admire yourself as a hero today, but you will surely look down on yourself as a "worm" in a week or two.
But it is in America's teachers colleges and public schools that the movement for unconditional self-esteem has found its true home. Today's educational practices encourage children to think of themselves as special regardless of what they do; and the practices encourage teachers to overpraise students' work and shield them from criticism. Only on the assumption that self-esteem must not be contingent on thought and action could such practices be seen as promoting it.
This use of self-esteem to motivate "affective education" is driven by the anticognitive culture of colleges of education. When it comes to these anticognitive practices in the schools, we must of course agree with the critics of the self-esteem movement. Sullivan and D'Souza and Baumeister and Emler are right to object that parents and teachers should not be ladling out indiscriminate praise to children. But then no one with an adequate understanding of self-esteem would endorse such practices.
Those who believe that we human beings cannot avoid evaluating ourselves, and that high contingent self-esteem drawn from internal sources is part of healthy psychological functioning, need to engage the academic researchers more fully.
Such engagement will have to take place on at least two levels. First, we must expose and rebut the philosophical and moral beliefs that are in play. As we have noted, researchers in academic psychology are animated by deep-seated beliefs about the nature of human beings and the nature of a good human life. After all, when the data collection is finished, the results have to be evaluated against some theory of what is being measured. We, looking at the data, have objected to the standard self-report questionnaires because they cannot distinguish genuinely high self-esteem from defensive substitutes. But Emler frankly accepts the existing measurement procedures—because the relationships between measured self-esteem and other variables make sense to him.
Taken alone, however, the effectiveness of this philosophical approach will be limited. Most psychologists are still being trained in the positivistic tradition, which equates science with collecting and analyzing data. They distrust "philosophical" arguments; what they find convincing is supportive data from established, narrowly defined programs of empirical research. Notice what happened when Nathaniel Branden directly challenged Roy Baumeister during a discussion of self-esteem on National Public Radio ( Neal Conan's "Talk of the Nation," February 4, 2002
). Baumeister indicated that high scores on self-esteem questionnaires could reflect a realistic self-evaluation not easily threatened, or a narcissistic bubble vulnerable to bursting: "You can think well of yourself because you accurately appreciate what you're good at. You can also think well of yourself just 'cause you're a conceited snob. And the self-esteem is the same in either case." Branden asked whether there could be any value in a conception of self-esteem that throws together compensatory grandiosity and self-trust grounded in mindful and self-responsible living. But Baumeister was unmoved by the argument. Self-esteem is what the studies he has reviewed are about; it is what the established questionnaires measure. Branden's merely theoretical objection was of no avail.
Consequently, there is a vital need for focused empirical research programs on self-esteem. Initially, these programs will have to concentrate on improving the measurement of self-esteem and distinguishing it more clearly from narcissism. Better measurement procedures will, in turn, lead us into inadequately explored aspects of genuine self-esteem, and that, we believe, will require further elaboration from all of the competing theories, including Branden's.
Once the measurement problems have been alleviated, and self-esteem's nature has been better elaborated, the connections between self-esteem and various behavioral outcomes will need rechecking—and not just the links with our success in school or our employment history or our ability to refrain from crime or drug abuse. Self-esteem also needs to be related to issues that matter to counselors—and clients—such as the quality and durability of our intimate relationships, or our ability to overcome misfortunes and difficulties in life.
We are not recommending empirical research of this kind just to relieve academic psychologists of some of their theoretical muddles or to convince them of facts that clinicians already know. Genuine empirical questions remain about the nature, functioning, and development of self-esteem, and better psychological research will give us a chance to answer them.
This article first appeared in the July/August 2003 issue of Navigator magazine, published by The Atlas Society. Navigator was the predecessor to The New Individualist magazine.