Abstract
Berger and Luckmann (1966) claim that it is ""inadvisable to speak of 'collective identity'"", and the very term 'identity' had been also questioned. This criticism might be appropriate in the social scientific theories and discourse; nevertheless, individuals often do refer to themselves and others in plural pronouns and group-related categories.
Some of the recent influential perspectives on identity - identity as something fluid, fragmentated, virtually non-existent - are in striking discrepancy with the reality of everyday life, which is built on the assumption that people and groups (should) maintain certain important characteristics over time. In my paper, I will show that there are two basic sociological perspectives on the collective identity formation.
First, collective identity is seen as something growing from within: out of interpersonal relationships on the micro-level, based on empathy, similarity and identification between ""I"" and ""You"" (e.g. symbolic interactionism, phenomenology, social psychology). Second, collective identity is explained from outside, as something that arises from the inter-group relationships, processes of essentialization and categorization, based on perceived difference between ""Us"" and ""Them"" (e.g. cultural and political sociology).
Although these two strands of thought seem to be divergent, my proposition is that they correspond with two basic mechanisms of identity construction, similarity and difference, and must be seen as interdependent. Finally, I argue that the risks of reification (""self-evidentializing"" the collective identity) as well as over-theorizing (""rejecting"" collective identity) can be minimalized by focusing on actual linguistic realizations of collective identities (as pronouns, categorization etc.) in social interactions.