One conventional explanation of intergroup conflict is Social Identity Theory. That theory asserts that strong ingroup sympathies can give rise to outgroup antipathies which in turn fuel intolerance and conflict. While embraced by both macro- and microlevel analysts, this theory actually has not been widely investigated outside a laboratory environment. In this article, I test hypotheses linking group identities with intolerance, based on a 2001 survey in South Africa, a country where group identities have long been politicized. My empirical findings indicate that group identities are not useful predictors of South African intolerance. Indeed, for neither the black majority nor the white minority do ingroup identities activate very much outgroup intolerance. Moreover, group identities are positively, not negatively, correlated with holding a South African national identity. These findings, based on unusually broad indicators of both identity and tolerance, suggest that the causes of group conflict lie elsewhere than in group attachments.