Self-esteem research, arguably the largest field of research in the history of social science, has devoted much of its efforts to the idea that self-esteem causes a broad range of behavioral and social problems, but has failed to produce strong consistent evidence for most claims. However, this research has conceptual and methodological problems, including a limited understanding of the role of culture, and the assumption that global levels of self-esteem are the main causal mechanism of interest. This paper argues that self-esteem motivated behavior may be better understood as socio-culturally contextualized pursuits of valued identities, which are difficult to understand without considering their social and cultural conditions. Self-esteem is therefore at the intersection of culture and cognition, and it is argued that an interdisciplinary approach to self-esteem pursuits could be beneficial. A way to reconcile constructionist views of identity with a cognitive self is therefore suggested and discussed. By drawing on models of neuro-cognition, it is possible to think of a cognitive self as performed, context-dependent and emergent rather than fixed, internal and expressed. Finally, the paper discusses the social contingencies and consequences of self-esteem pursuits in relation to research issues such as aggression, stratification, crime, masculinity, and political attitudes.