Identity entered the anthropological lexicon in the 1960s and 1970s, in work associated with the "Manchester School and influenced by the American sociological traditions of "symbolic interactionism and social constructivism. Classics of the genre, "Fredrik Barth’s Ethnic Groups and Boundaries (1969) and A.L. Epstein’s Ethos and Identity (1978) focus on ethnic identity, emphasising the contextual and creative construction of ethnicity in relation to particular political contexts. Central to the approach was the focus on boundary, rather than ‘content’ or essence of ethnic identity, which was picked up in the UK by Anthony Cohen, who focused initially on the symbolization of community boundaries (1985) and then later on the question of the self (1994). The American agenda developed through the emerging ‘identity politics’ of the 1970s, itself influenced by subaltern and post-colonial studies, which called for the recognition of difference based on race, ethnicity, gender, class, sexuality, etc. (Rosaldo 1989; Taylor 1994; Rowe 1998).
As Brubaker and Cooper (2000) suggest, ‘identity’ is both a category of analysis and a category of practice. As a category of practice it has an ambiguous role as both starting-point and end-result of political mobilization. Within nationalism, for example, whilst pre-existing identity is often the justification for claims to nationhood, identity also figures as a future project, or teleology (Mitchell and Ashby Wilson 2003). The same is true for other socio-political movements (Holland et al. 2008).
As a category of analysis, anthropologists have tended towards a ‘soft’ or ‘weak’ account of identity as contextually constructed or negotiated. Baumann and Gingrich (2004) have attempted to identify the different ‘grammars’ through which this construction takes place. They initially focus on three: orientalism, through which self and other are reciprocally essentia-lised; segmentation, through which processes of group fusion and fission emerge in relation to strategy and context; and encompassment, through which ‘otherness’ is co-opted as a form of sameness. That two of these models — segmentation and encompassment — are drawn from works that do not use the term ‘identity’ (Evans-Pritchard 1940; Dumont 1980), raises questions about the utility of the category. Indeed, Brubaker and Cooper (2000) have suggested that the vagueness, looseness and ‘softness’ of the category as it has developed, all but render it useless. Richard Handler (1994) criticises anthropological uses of ‘identity’ on the grounds of its ethnocentrism. Western notions of identity, he argues — which are those that inform the social theory of identity – presume qualities of sameness and boundedness that are not sustained across other societies’ understandings of person and group. However, even where ‘identity’ does not mean the same as it does in ‘the West’, it is still often a dominant political category; a category of practice. As such, it should demand our attention, through work that examines not only how identity is constructed in diverse contexts, but also what ‘identity’ means within different groups. This would entail a move from ‘politics of identity’ to politics of ‘ identity’ .
