With the current emphasis within psychological social psychology on cognitive processes, and within sociological social psychology on the self, social psychology has largely ignored the concept of habit. Identity theorists (Burke, this volume; Burke & Tully, 1981; McCall & Simmons, 1978; Stryker, 1980, this volume; Turner, 1978; Turner & Billings, this volume) have used only the concepts of role-identity salience, role-person merger, and the like, which suggest conscious thought in relationship to action. The popular psychological theory of reasoned action (Fishbein & Ajzen, 1975; Ajzen & Fishbein, 1980) includes only attitudes, subjective norms, and intentions as predictors of behavior. Attribution theory (Kelley, 1967; Jones and Davis, 1965; Weiner, Russell, & Lerman, 1978) until very recently has assumed a very active, analytic “lay scientist” model of social information processing and action.

Habit as a concept was popular among the early theorists in sociology but fell out of favor with the rise of Watsonian behaviorism in psychology (Camic, 1986). I will use habit here to mean actions that “are relatively unmotivated” (Giddens, 1979: 218), actions in which “means-ends relations … are [from the actor's standpoint] not subject to argument” (Hartmann, 1939: 91). Following Camic (1986: 1,044) “the term ‘habit’ generally denominates a more-or-less self-actuating disposition or tendency to engage in a previously adopted or acquired form of action.” I am not suggesting that such behavior is mechanical or lacking in meaning, in the sense used by symbolic interactionists.