Foucault sees all of his work contributing to outlining the intricate and highly variable forms ofpower in discursive and non-discursive practices. McNeil (1993) notes that at the centre of all Foucault’s enquiries is his concern with the power – knowledge relationship. This can be seen to begin within his early methodological texts, notably The Archaeology of ‘Khowledge (1972), and carries through to his specific historical studies (case studies on madness, criminality and sexuality). However, there is a tension in Foucault’s work, as Ramazanoglu (1993:17) notes, between his later emphasis on the fractured and unstable nature of power which is both productive and accessible, and his earlier emphasis on a more stable and concentrated conception of power.

Peter Dews (1984) maintains that Foucault introduces and begins to elaborate his theory ofpower in his text Discipline and Punish (1977), establishing a theoretical distance from his more narrowly focused methodological concerns of the late 1960s. During the 1970s, he says, Foucault begins to develop a view of power which clearly indicates disillusionment with structuralism. His inclination is to play down the repressive and negative aspects ofpower and to present the operation of power as primarily positive and productive. This shift of emphasis in Foucault’s work creates tensions between a feminist and a Foucauldian concept of power. As Ramazanoglu (1993) notes, Foucault’s early work is concerned with domination and physical power, but he moved increasingly to a position which denied that power was repressive or resulted from a single source of domination or oppression. While feminists identify male power as repressive, Foucault moved towards a position which defined all power as productive.

In fact, Foucault does not present a ‘theory’ of power, but develops a series of methods for examining the operation ofpower. He describes his work as an ‘analytics of power’, a series of methods which make no claim to lasting or eternal truth. However, as Grosz (1990a: 87) indicates, although Foucault does not consider power either as a uniform or homogeneous thing, it is possible to extract a number of ‘methodological theses’ from his work to assist an examination of power relations. Grosz notes that, for Foucault, power is not possessed, given or seized, rather it is exercised and exists only in actions: ‘It is…a moveable substratum upon which the economy, mode of production, modes of governing and decision-making, forms of knowledge, etc., are conditioned’ (ibid.). As a result of this position, Foucault refuses to equate power with social structures such as patriarchy. His deconstruction of theories of power such as patriarchy or capitalism led him to emphasise the unstable ways in which power is constantly created. As Ramazanoglu (1993:5) indicates, ‘He conceptualised people’s experiences of domination and subordination as “effects” of power, rather than as proceeding from a specific source of power.’

For Foucault, negotiations or struggles within society are not essentially about the possession of power, but rather the contested terms of the deployment of power. Thus Foucault implicitly contests a notion of men’s possession of power over women. As Ransom (1993:129) indicates, this theory of power underpins Foucault’s pluralism; power is understood as plural, not operating on ‘a single trajectory’ or with reference to specific questions. Foucault understands power as ‘capillary, spreading through discourses, bodies and relationships in the metaphor of a network’ (ibid).

Foucault did acknowledge men’s exercise of power over women but denied that men hold power. Ramazanoglu and Holland (1993) argue that there has been insufficient analysis of the middle ground of power relations, i. e. between the micro-politics of everyday life and the consolidation of deeply entrenched male privilege throughout social life. Foucault (1988:103) himself states that the way in which power operates and is exercised was little understood. He suggests focusing on specific techniques of power to show how those in power arrive at particular decisions.

Fraser (1989:22) argues that Foucault’s empirical studies of modern societies focus on the nature of modern forms of power. She states that, for Foucault, modern power is unlike earlier power in that it is local, continuous, productive, capillary and exhaustive. Foucault maintains that the nature of modern power/ knowledge developed gradually in local piecemeal fashion largely in what he called ‘disciplinary institutions’ beginning in the late eighteenth century. Fraser (1989:24) argues that in describing power as ‘capillary’ Foucault meant that power did not emanate from a single source ‘but circulates throughout the entire social body down to even the tiniest and apparently most trivial extremities’. Fraser contends that this emphasis on the ‘capillary’ character of modern power denies a crude ideological critique of power and emphasises the ‘politics of everyday life’.

Foucault does not deny feminist concepts of women as an oppressed group, nor Marxism’s concept of the oppression of the working class. As Grosz (1990a) indicates, he ‘demassifies’ and localises the categories ‘woman’ and the working class so that these categories are no longer universal categories. Ramazanoglu (1993) argues that Foucault’s deconstruction of power releases feminism from rigid conceptions of universal patriarchy, racism or heterosexism. However, she goes on to note that, by seeing power as everywhere and at some level accessible to everyone, it could result in a lack of acknowledgement of women’s systematic subordination to other women, as well as systematic domination by men. Ramazanoglu makes the important point that using Foucault

means acknowledging the multiplicity of difference and claiming the end of ‘woman’ as a universal category. But it can also lead to a tendency to revert to speaking in abstracted terms of deconstructed ‘women’ because of the absence of.. .class, racism or gender as categories of power relations in his thought.

(Ramazanoglu 1993:10)