ion. And we have to do this not only critically, in the necessary history of rural and urban capitalism, but substantially, by affirming the experience which in many millions of lives are discovered and rediscovered, very often under pressure: experiences of directness, connection, mutuality, sharing, which alone can define, in the end, what the real deformation may be. [39] If urban anthropologists are to take seriously these spaces in which images, social processes and economic structures come into play the spaces of everyday life in cities a more productive conception of power and culture is called for: one in which the individual and the state are not construed as analytic poles, requir? ing on the one hand a discourse on alienation, and on the other a discourse on planning, but as terms related and mutually constituted within a whole series of paradigms, practices and strategies. As Michel Foucault has persuasively argued, power may be best understood not negatively, by focusing on repressive institutions and powerful individuals, but positively, by tracing the pervasive mechanism by which it inserts itself into actions, attitudes, discourses and everyday lives. What is needed, he suggests, is "a study of power in its external visage, at the point where it is in direct and immediate relationship with that which we can provisionally call its object, its target, its field of application, ... where it installs itself and produces its real effects [40]". So, too, might culture be understood in terms of a productive deployment in terms of the strategic inven? tories it installs, the interpretations it enables, the practices it gives meaning to, the pos? sibilities it circumscribes. The avenues by which this double deployment proceeds in neither case adequately described as an imposition from top to bottom, or as a structural determination from bottom to top are fertile territory for an urban anthropology limited neither to holism nor to community studies. The city, from this perspective, can be approached indirectly and by way of such mediating technologies and institutions as industrial discipline, social housing and the family, where strategies of government, production and individual practice intersect. The city can be understood on the one hand as requiring or making possible new relations between culture and power, new penetrations of the latter into everyday lives, and new forms of resistance to that penetration. The City can be understood on the other hand as partially constituted by these new relations: transformed by the new collectivities, reproductive strategies, and ways of seeing that grow from, or in response to, subjectification, bureaucratic interventions and the waning of dependency relations. The attempt made by Jacques Donzelot to trace the origins and implications of a "policing of families" is instructive in this regard [41]. Donzelot isolates a shift from "government of the family," where relations between state and family mirror those between father and son, to "government through the family." The latter was accomplished, he argued, by means of a two? fold strategy: medical intervention in the bourgeois family, and a philanthropic interven? tion in the working-class family. In the first case, it was a question of enacting "protective" measures; in the second, it was a question of establishing a direct surveillance. Concerning This content downloaded from 157.55.39.215 on Wed, 31 Aug 2016 04:34:21 UTC All use subject to http://about.jstor.org/terms