From Martin Seligman’s Learned Optimism:
Point 1: Self-esteem is a symptom, not a prerequisite or cause.
“I believe that self-esteem is just a meter that reads out the state of the system. It is not an end in itself. When you are doing well in school or work, when you are doing well with the people you love, when you are doing well in play, the meter will register high. When you are doing badly, it will register low. I have scoured the self-esteem literature looking for the causality as opposed to correlation, looking for any evidence that high self-esteem among youngsters causes better grades, more popularity, less teenage pregnancy, less dependence on welfare … There is nothing of this sort to be found in the literature. Self-esteem seems only to be a symptom, a correlate, of how well a person is doing in the world.”
Point 2: Unwarranted self-esteem causes problems
Until January 1996, I believed that self-esteem was merely a meter with little, if any, causal efficacy. The lead article in the Psychological Review convinced me that I was wrong, and that self-esteem is causal: Roy Baumeister and his colleagues (1996) reviewed the literature on genocidal killers, on hit men, on gang leaders, and on violent criminals. They argued that these perpetrators have high self-esteem, and that their unwarranted self-esteem causes violence. Baumeister’s work suggests that if you teach unwarrantedly high self-esteem to children, problems will ensue. A sub-group of these children will also have a mean streak in them. When these children confront the real world, and it tells them they are not as great as they have been taught, they will lash out with violence. So it is possible that the twin epidemics among young people in the United States today, depression and violence, both come from this misbegotten concern: valuing how our young people feel about themselves more highly than how we value how well they are doing in the world.