POINT

Aggression and self-esteem

By Ervin Staub, PhD
University of Massachusetts at Amherst

The following opinion piece is in response to 'The American way of blame,' the July 'Presidents column' by Martin E.P. Seligman. To read that column, visit www.apa.org/monitor/jul98/pc.html.

The traditional view of the relationship between self-esteem and aggression has been that low self-esteem gives rise to aggression. In his column, Martin E.P. Seligman proposed that it is high self-esteem that is related to aggression. There is increasing information on this issue and challenging the traditional view has been valuable. I will offer some considerations, however, to suggest that this relationship is quite complex and, as yet, far from clear. I will focus on children, because one purpose of Seligman's column was to question the value of programs that promote high-self-esteem in children.

Dan Olweus found that bullies do not have low self-esteem. John Coie and Kenneth Dodge, in reviewing research on aggression in children, reported that aggressive boys do not have low self-concepts and tend to blame others rather than themselves for 'negative outcomes.'

Several avenues may lead children, specifically boys, to become aggressive. Some boys are effective in getting what they want through aggression. It is possible that they have become aggressive through parental permissiveness, which is related to aggression, and possibly parental encouragement to be tough and aggressive.

More common are boys who are ineffectual aggressors, who in addition to being aggressive are angry, argumentative and disruptive. They tend to be rejected by peers but are unaware of this. Not knowing that they are rejected makes it difficult for them to improve their peer relations. Over time, they become more aggressive and less effective in conventional realms, such as school.

It seems likely that the origins of these boys' aggression is harsh, punitive parenting, the most researched avenue to aggressiveness. Its results include poor social skills, deficiency in academic work and poor performance in school. Aggressive boys see other people as hostile, especially to themselves. They respond to hostility they perceive or imagine by aggressive actions, which creates hostility to them.

On the face of it, it does not make good psychological sense that these children (and adults) would have high self-esteem. It is possible that some aggressive boys do have low self-esteem. Earlier research suggested such a relationship and in a dissertation I supervised, Andrew Theiss found that in college students low self-esteem was associated with greater self-reported agrression in response to provocation. Other aggressive boys are defensive and fulfill the need for seeing themselves in a good light by having a 'compensatory' self-esteem. Aggressive boys blaming others for negative outcomes, which is consistent with the Freudian view of projection of one's own problemamatic characteristics into others, supports the notion of some kind of defensive process. Since self-report and even inference from observation are far from foolproof, they may fake a high self-esteem.

But perhaps the answer lies not simply in level of self-esteem, but what self-esteem is based on. Children and youth will strive to gain a positive image of themselves. Many boys who become aggressive don't have the socially valued means to gain such a positive image: competence and good performance in school, good relations with peers. They organize their self-esteem, therefore, around strength, power, physical superiority over others. Their early experiences as victims, the models of aggression around them, and the culture's focus on male strength and superiority all facilitate this. It is how self-esteem is constituted, what self-esteem is based on, that may matter.

But the self-esteem of aggressive boys appears to be very vulnerable, fragile. Its maintenance requires specific circumstances and behaviors on their part. This again raises the question whether we can consider it a genuinely positive self-esteem. It is also maintained in ways that creates continuous trouble for them.

Their inadequacy in realms that tend to be the socially valued sources of esteem may continue to frustrate them. Because they have not gotten enough respect in their lives, respect becomes especially important. They try to gain identity and self-respect by congregating with other kids like themselves, whether in the classroom or as members of gangs. Gang culture, which focuses on respect and on boosting the self-esteem of members, helps to elevate their self-esteem. It is worth noting, however, that frequently self-esteem, whether high or low, may have no causal relationship to aggression. Instead, mistrust, perceptions of others' bad intentions and the feelings these generate and other factors have causal influence.

Would programs that attempt to directly raise the self-esteem of the kind of boys I focused on reduce or increase their aggressiveness? It would increase it if they come to feel more entitled and then in turn frustrated. It would reduce it if they come feel respected and cared about without having to show strength and power. Research on resilience shows the importance of children's connection to caring adults-teachers, counselors or friends of the family.

Neither self-esteem, nor established patterns of aggression can truly be changed without experiences that a.) make the world seem safer-partly by connection to caring, benevolent people-and b.) create greater self-awareness and awareness of other people and help aggressive children learn to effectively fulfill their needs and goals in constructive ways.

Ervin Staub, professor of psychology at the University of Massachusetts at Amherst, has studied helping and altruism, youth violence, and the origins and prevention of genocide and ethnopolitical conflict.

Counterpoint

Low self-esteem does not cause aggression

By Roy F. Baumeister, PhD
Case Western Reserve University

Martin E.P. Seligman and Ervin Staub have offered different views about the relation of self-esteem to aggression. I would like to offer my own hard-won views about this issue. I have devoted much of my career to studying self-esteem, partly because I shared the optimistic hope that it would hold the key to solving many social and individual problems in modern society, including aggression. Regrettably, the data have convinced me that those hopes were largely misplaced. High self-esteem helps people feel good but its ability to produce beneficial outcomes is small to negligible.

Seligman is correct to attack the view that low self-esteem causes aggression. In my own work, an extensive literature review (Psychological Review, 1996) and laboratory experiments (Journal of Personality and Social Psychology, 1998) have contradicted it. More generally, people with low self-esteem tend to be shy, modest, self-effacing, reluctant to take risks, unsure about themselves and likely to blame themselves for failure. Aggressive, violent people are not like that.

On the contrary, we have found that aggressive people have very favorable views of themselves. But, as Staub suggests, the relationship is complex. High self-esteem is a heterogeneous category that contains violent and nonviolent, nonaggressive people. The aggressive subset of people with high self-esteem is narcissistic: They regard themselves as superior to others, think they deserve special treatment and privileges, and are emotionally invested in being admired by others. They turn aggressive toward anyone who questions their inflated opinions of themselves.

What about hidden low self-esteem? Staub suggests that some people act confident but secretly suffer from self-loathing, and that aggression could derive from that low self-esteem. There are two arguments against this view. First, several researchers (including Olweus on bullies) have failed to find the hidden self-loathing behind the narcissistic or egotistical front. They conclude that aggressors do not typically have low self-esteem.

Second, the view is ultimately illogical. Suppose someone were able to show that aggressors have egotistical surface veneers but secret, hidden low self-esteem. What would the cause of aggression be? We know that overt low self-esteem doesn't cause violence. Hence the crucial factor would be the fact of being hidden. But that means to shift attention away from the low self-esteem itself and on to what is hiding it-which returns us to the narcissistic, egotistical veneer. Moreover, the therapeutic intervention to prevent violence would then focus not on altering the low self-esteem but on making it overt instead of hidden, because overt low self-esteem is nonviolent.

Staub is right to say we must consider what self-esteem is based on-but that line of thought leads back to supporting Seligman's call to dismantle the school self-esteem movement. All too often, this movement takes the form of uncritical self-celebration as an entitlement of being a human being, instead of applauding hard-earned achievements. Awarding trophies to all contestants or 'socially promoting' students who haven't learned the material is not conducive to well-founded self-esteem. In fact, these practices may cultivate inflated views of self and entitlements, which constitute the dangerous form of high self-esteem. I see nothing wrong with praising a child (or adult) for an outstanding or brilliant performance. I see plenty wrong with praising everyone even when the actual achievements are mediocre.

Can anyone benefit from self-esteem boosting, in school or therapy? Sure. Some people genuinely fail to recognize their abilities and achievements and might shortchange themselves. But these are a small minority. Many research findings show that most Americans already hold inflated opinions of themselves.

My profound disappointment with the benefits of self-esteem has been partly offset by discovering something else that does seem to work. Self-control, as in being able to regulate one's emotions, impulses, performance patterns and thoughts, has plenty of positive payoff, for the individual and society. Self-control problems are central to most problems in our society: teen pregnancy, drug abuse, violence, school failure, unsafe sex, alcohol abuse, money problems and debt, eating disorders, ill health, and more. My conclusion, therefore, echoes Seligman's call to discontinue schools' self-esteem programs. Instead of dismantling them altogether, though, I suggest we focus them on instilling the capacity to control, discipline and regulate oneself. Ironically, in the long run, that approach will even probably do better for self-esteem. Self-control permits the individual to discipline the self to achieve goals and to fulfill social and personal obligations. That creates a stronger basis for self-esteem than indiscriminate flattery.

Roy F. Baumeister holds the E.B. Smith Professorship in the Liberal Arts at Case Western Reserve University. His recent books include 'Evil: Inside Human Violence and Cruelty' (Freeman, 1997), 'Losing Control: How and Why People Fail at Self-regulation' (Academic, 1994) and 'Self-Esteem: the Puzzle of Low Self-Regard' (Plenum, 1993).



Read our privacy statement and Terms of Use

Cover Page for this Issue

PsychNET®
© 1999 American Psychological Association

APA Home Page . Search . Site Map