Limiting heteronomy: an account of autonomy to deal with oppression

Abstract

Is it possible to limit heteronomy under oppression through critical self-assessment and self-transformation? I answer by testing available models of autonomy in light of their capacity to deal with the forms of heteronomy which typically characterise oppression. Drawing from Foucault’s analysis of power relations, I claim that there are significantly different ways of being oppressed in contemporary Western societies and that we need to account for this difference when answering if self-emancipation under oppression is possible. First, I look into paradigmatic examples of the two main strategies available in the literature on autonomy: Christman’s procedural account, and Stoljar’s and Oshana’s substantive accounts. I analyse the strengths of these accounts but conclude that, as they stand, they are ill-suited to problematize forms of (what I call) ‘subjection’, namely forms of oppression which affect agents’ “normal” developments qua subjects of different kinds. Crucially, Christman’s model lacks resources to problematize settled characters and values, while Stoljar’s and Oshana’s models cannot sufficiently account for resistance and transgression in oppressive environments. To find a way out of the impasse of the substantive-procedural debate, I turn to Foucault’s analysis of power. Foucault’s resources are useful both to problematize agency and self-relations as effects of social power relations and to distinguish between the different interferences that contemporary theorists would call ‘oppressive’. I argue that some forms of oppression qualify as Foucauldian ‘domination’, where power imbalances are frozen and irreversible through the (limited) margin of freedom available to the individuals living in those conditions. Other forms of oppression, however, can be likened to what Foucault calls ‘government of individualisation’, where practices of self-clarification and self-transformation can make agents less heteronomous vis-à-vis specific power configurations. I propose a two-tracked approach to autonomy: a revised procedural account for cases of ‘subjection’, and a substantive one for cases of ‘domination’.