ABSTRACT
The state is best understood, it is said, through local “practices” and the production and reproduction of its images (Gupta 1995: 375-6). Day-to-day interactions between the multitude of petty officials, local service-providers and social groups in the community shape the popular image of the state, and the arising discourse about the state feeds the subsequent practices. The state is more than its constituent organizations, instruments in use (for example monopoly of legitimate coercion), or the actors involved. Above all it is not an autonomous and monolithic entity above the society but, as described in Migdal’s “state-in-society” approach, is constitutive of the processes of negotiation and contestation over the rules of daily behavior, and reflected in the images as perceived among multiple actors (Migdal 2001: 11).1 This stress on the discursive nature of the state, and the fluidity of state images and practices as product of contestation in a specific time-space, finds echoes in Bourdieu’s discussion of social space and the relational (Bourdieu 1998a: 3-4) and Foucault’s “state as governmentality” (Burchell et al. 1991: 103). Three interrelated messages as regards research on “the state” as the
“relational” and “process” consequently emerge. First, more attention needs to be paid to the lowest echelons of the state organization where the majority of the people had their first-hand experience with the state, and where decisions made “high-up” in the state hierarchy are translated into practices in society.2 Second, research needs to adopt an approach that sees processes and actors on either side of the state-society boundary – which is essentially blurred and permeable-having similarly important, if different, roles in the shaping of “the state,” and society. Finally, given the fluidity and contingency of existing state practice and image, the researcher needs to be modest in the conclusion drawn from substantive observations obtained at a specified time-space. As Hirschman (1970: 339) warned more than thirty years ago, “the immediate effect of social analysis is to convert the real into the rational or the contingent into the necessary.” This may be a sin impossible to escape from, since any statement-or language-requires a degree of generalization. What is demanded is perhaps self-awareness on the part of the social analyst, and with it a difference in the kind of statements (and thus
the level of generalization) produced, and a modesty in what we as the analyst may imply for future action.3
