Freedom as non‐domination or how to throw the agent out of the space of reasons

Abstract

This paper analyzes agency in Pettit’s republican conception of freedom. By understanding freedom intersubjectively in terms of agency, Pettit makes an important contribution to the contemporary debate on negative liberty. At the same time, some of the presumptions about agency are problematic. The paper defends the thesis that Pettit is not able to provide the sufficient conditions for freedom as non‐domination that he sets out to do. In order to show why this is the case and how we can address this shortcoming, a distinction is introduced between a thick and a thin intersubjective account of agency. It is argued that while Pettit’s freedom presupposes a thin account, he would need a thick account in order to elaborate not only the necessary but also the sufficient conditions of freedom as non‐domination.

Keywords:

• freedom
• non‐domination
• Pettit
• intersubjectivity
• Brandom
• agency

Acknowledgments

I wish to thank Niklas Möller, Sofia Näsström, Maeve Cooke, and Mark Haugaard for constructive comments on earlier drafts of this paper. I also thank Riksbankens Jubileumsfond (RJ) for financing my Pro Futura Fellowship.

Notes

1. Of course, on a poststructuralist account such as Foucault’s, an important criticism of both liberalism and Marxism is precisely that their power analyses are premised on freedom of some kind, namely, on the idea of a power‐free state of affairs where the agent is free or emancipated.

2. Arguably, there has been doubt about whether positive and negative liberty really are two different conceptions of freedom. Gerald MacCallum has famously argued that all conceptions of freedom embrace the same triadic matrix: they involve an agent (X) who is (is not) free from constraint (Y) to do (not do, become, not become) something (Z) (see MacCallum 1967). However, as pointed out by Pettit, it still seems to make perfect sense to talk about two different conceptions, as it all depends on the conventional matter of how ‘concept’ and ‘conception’ is to be understood (see Pettit 2003, pp. 388–389).

3. While the dominating party is always an agent, Pettit’s analysis is not restricted to personal agency, but might involve a collective or corporate agent as well. The dominated agent, however, is always a person or group of persons, never just a corporate body. In this essay, I will also limit my examples to personal agency.

4. Markell is critical of Pettit’s view of non‐domination as the ‘supreme’ political ideal, the overarching value under which other values and commitments are supposed to be subsumed. From the paradigmatic case of slavery, Pettit reconstructs the idea of ‘dominium’ as the subjection to arbitrary power, which turns problems of agency into problems of control. Markell’s point is that this leaves out those dimensions of agency that concern problems of involvement. So, rather than criticizing the ideal as such, Markell argues for the need to supplement the idea of domination with that of usurpation. From this perspective, a relationship could not only subject a person to arbitrary power but also deprive him or her of involvement in affairs that affects life (see Markell 2008, p. 11).

5. In a similar vein, Charles Taylor has famously argued that freedom somehow seems to need an exercise dimension (see Taylor 1997).

6. This observation is made by Kukla and Lance (2008, p. 9).

7. Here, I follow the discussion in Erman (2006).

8. The idea that second‐personal interaction inevitably involves mutual recognition is one of the central claims made by Darwall (2006).

9. For a discussion of power in terms of agency versus structure, see Hayward and Lukes (2008).