A broad definition of agential power

Abstract

Can we develop a definition of power that is satisfactorily determinate but also enables rather than foreclose important substantive scientific and normative debates about social and political life? I present a broad definition according to which agents have power with respect to a certain outcome (including, inter alia, the shaping of certain social relations) to the extent that they can voluntarily determine whether that outcome occurs. This simple definition generates a surprisingly complex agenda for substantive descriptive and normative inquiry. The proposed definition is partly developed through a critical engagement with Rainer Forst’s important recent account of ‘noumenal power.’

Keywords:

• Power
• agential power
• noumenal power
• critical theory
• Rainer Forst

Notes

1. A concept which, surprisingly, has received insufficient attention by political philosophers. For a recent illuminating survey see Morriss (2012). See also the classic treatments in Lukes (2005) and Morriss (2002).

2. Forst (2015a). A German version of the paper (‘Noumenale Macht’) appeared as chapter 2 of Forst (2015b). Forst’s account is also presented in Allen et al. (2014) (with a summary at p. 12).

3. I use ‘agential power’ to mark out a kind of power that can be had and exercised by entities in their capacities as agents. Agential power differs from other powers or dispositions (such as those held by non-agential entities – think about the power of water to dissolve sugar – or by agents when not operating agentially – as when my body has the power to displace water when it falls on a pond by accident). In this paper, I mean agential power when I use ‘power’.

4. Forst (2015a, p. 115).

5. Forst (2015a, p. 124).

6. Scanlon (1998, pp. 18, 19). See further Parfit (2011, vol. 1, pp. 34–37).

7. Forst (2015a, pp. 119–121).

8. The double significance of relations of justification (for purposes of description and normative assessment of social life) is addressed in the Introduction to Forst’s (2015b). For Forst’s powerful normative agenda (according to which we have a fundamental right to justification, and must engage in discursive exchanges in which we seek reciprocal and general reasons to justify the terms on which we live our social life), see Forst (2012).

9. Forst (2015a, pp. 114, 115, 125, 126).

10. Forst (2015a, p. 114)

11. Forst (2015a, pp. 114, 116).

12. Forst (2015a, p. 115).

13. Note that my objection here is not addressed by Forst’s emphasis on the fact that noumenal power can be exercised in many ways, including those that affect its subject’s emotions, bodily dispositions and so on. That reply may be enough to respond to an objection that NPA is too narrow in its account of how people are affected in the discursive and mental processes leading to actions. See Forst’s response to Allen et al. in (2014, p. 22). My objection is that there are dimensions of power which involve forms of influence that altogether circumvent the subjects of power’s discursive or mental life.

14. See Cohen (2001). For a critique of technological determinism that affirms a greater malleability of social (noumenal) frameworks, see Castoriadis (1998, Part I). The debates about the relative significance of physical violence and various forms of persuasion and consensus formation sparked by the work of Antonio Gramsci on hegemony are still relevant. See Gramsci (2000, sects. VI and X). For discussion, see Anderson (1976), Mouffe (1979), Przeworski (1985) and Wright (2015).

15. See Kant (1996). Particularly striking is Kant’s critical discussion of the vice of ‘servility’ (at 6:434-7).

16. Forst (2015a, p. 116, n. 17).

17. See Forst in Allen et al. (2014, p. 11).

18. Allen in Allen et al. (2014, p. 27). The criticism is hard to assess as it is phrased in quantitative terms (focusing not on whether, but on how much, responsibility it is appropriate to ascribe to oppressors and victims).

19. It is also useful to distinguish between a concept and a conception of power. The distinction between concept and conceptions comes from Rawls (1999, p. 5) and Hart (1994, pp. 160, 246). A concept can be shared across different conceptions. Even if they use ‘power’ in the same sense, different scientific theories may disagree about what mechanisms of power operate in a certain social context, and different normative theories may propose different views about the desirability of those mechanisms.

20. We can add time indices to make the account more explicit. We can for example talk about A having power at time t1 with respect to an outcome at tn – where tn coincides with, or comes later than, t1. Furthermore, it is common to add a counterfactual clause saying that the outcome of power exercise would not have occurred without that exercise. One should phrase this point carefully, however, as it could be that, e.g., if A had not exercised power over B to get B to produce O, O would have still been produced by B because, say, C got B to do it, or B decided to do it independently of anyone else’s prompting. What is crucial for the counterfactual is that O would not have arisen in the exact same way (the one whose description makes reference to the agent of power’s exercise of their power).

21. As a matter of degree, power is best characterized as involving scalar feasibility. I explore this point in more detail in Gilabert (2017). See also Gilabert and Lawford-Smith (2012) and Gilabert (2012, chs. 4 and 7).

22. I should add that although my characterization of power focuses on what agents can voluntarily bring about, it does not assume that the choices, intentions, etc. of those agents must be conscious. It is not ruled out by the definition proposed here that power be wielded unconsciously.

23. We can in this way capture the idea of ‘power with’. See Morriss (2012, p. 589).

24. Sen (2009, pp. 215–217).

25. Often power operates through communicative interaction. That interaction can involve respectful and concernful discursive argumentation of the kind explored by Juergen Habermas’s discourse ethics and politics. See Habermas (1990) and (1998). Communicative interaction can also display mechanisms of violence and domination. See Bourdieu (1991). Forst’s discussion of noumenal power nicely covers the whole spectrum of communicative power.

26. See Wright (2015).

27. There is also the view that subjects of power may themselves be, in some respects, constituted by exercises of power. This happens, e.g. in the contexts of socialization in the family or in educational institutions, in which many people’s beliefs, preferences and habits are formed. The topic of constitution is familiar from sociological research. See Lukes (2005). It is also a central element in Foucault’s inquiries about power. See e.g. Foucault (2000). For a discussion of how constitution is important for critical theory see Saar (2010).

28. So, even though BAP is restricted to intentional agents with respect to the power-wielder, it does not say that the subject of power must also be an intentional agent, and even when the subject of power happens to be an intentional agent, the account does not say that every power relation must affect the receiving subject as an intentional agent.

29. See note 14 above. Of course, another important topic concerns the normative dimensions of power relations between human agents and non-human animals and plants.

30. On the related notions of dynamic duties and dynamic power see Gilabert (2009, 2011, 2017).

31. Gilabert (2015, forthcoming).