Beyond ‘having reason to value’: why we should adopt a procedure-independent and value-neutral definition of capabilities

ABSTRACT

Sen has famously defined the notion of capabilities as the doings and beings that we ‘have reason to value,’ which is still widely regarded within the capability literature as the correct or only definition of the concept of capabilities. In this paper, I argue that capability theorists should abandon Sen’s definition because it suffers from two issues - namely, procedure-dependence and value-ladeness - that make it unsuitable to encompass the many different applications of the capability approach and the capability terminology. In contrast, I argue that the concept of capabilities should be defined minimally as ‘the real freedom to achieve certain doings and beings’. This definition avoids the two issues identified with Sen’s definition: first, the procedure by which we select relevant capabilities can be tailored relative to the particular application and, second, can thus accommodate applications of the capability approach which also focus on non-valuable capabilities.

KEYWORDS:

• Capability approach
• Amartya Sen
• having reason to value
• capabilities
• value-neutrality

Acknowledgements

I am grateful to David Axelsen, Willem van der Deijl, Stacy Kosko, Matthias Kramm, Sophie Mitra and two anonymous reviewers for their comments on earlier drafts of this paper as well as to Ingrid Robeyns for challenging me on several aspects of my argument. The paper has also benefitted from comments by the audiences at the 2017 conference of the Human Development and Capability Association in Cape Town, South Africa, and the workshop on Foundational Issues in the Capability Approach held in Leuven on 6-7 March 2019 and organized by Nicolás Brando. I am also thankful to Stacy Kosko and Serene Khader for making available ahead of print their paper analyzing Sen’s concept of ‘having reason to value’.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 In the capability terminology, these various conditions are usually referred to as ‘conversion factors’ (Robeyns, 2017, p. section 2.6.3).

2 See, though, Gasper (2007, p. 341) who argues that Sen sometimes employs another definition of capabilities that emphasizes the alternative lives that are open to individuals: ‘A person’s capability is (definition I): the set of alternative n-tuples of functionings she could attain (‘capability set’), in other words, the alternative lives open to her, the extent of her positive freedom.’ In that sense, however, this definition of capabilities is akin to what I shall refer to as capability-sets –i.e. the collection of capabilities that someone can choose to realize – while I reserve the term ‘capability/-ies’ to particular doings and beings that a person can choose to realize.

3 While it is beyond the scope of this paper to make a comprehensive survey of how capabilities are defined within the capability literature, versions of the ‘having reason to value’-formulation are used within a number of prominent works on and introductory texts to the capability approach, both recently and in the past and within different disciplines, including Alkire (2002, pp. 4–10), Arneson (2010), Crocker and Robeyns (2010, p. 86), Deneulin and Shahani (2009), Entwistle, Cribb, and Owens (2018), Evans (2002, p. 56), Evans (2017), Hart and Brando (2018), Ibrahim (2017), Mauro, Biggeri, and Maggino (2018), Poveda and Roberts (2018), Robeyns (2016b), Svensson and Levine (2017), Venkatapuram (2011), Walker (2005), and Walker and Unterhalter (2007).

4 For a detailed discussion of how Sen’s critique of existing social choice theory, in which he calls for the expansion of the informational bases of economic analysis by taking subjective values into account, provided the groundwork for the capability approach, see Walsh (2003) and Baujard and Gilardone (2017). Putnam (2002) also holds up Sen’s work as an example of the collapse of the fact/value dichotomy. See also Sen (1985b) for a critique of utilitarian and resourcist measures of well-being.

5 For a discussion of adaptive preferences within the capability approach, see Teschl and Comim (2005).

6 It should be noted that it is an open question whether Sen’s (2000, 2007, 2008b, 2009a) work on identity can and should be seen as part of his work on the capability approach. For an example of how identity theory, including Sen’s, can play a role within the capability approach, see Kirman and Teschl (2006).

7 Robeyns (2017, p. 32) provides a helpful table of (some of) the many different applications of the capability approach.

8 It should be noted that asking this question might be unfair. After all, Sen arguably had a particular purpose in mind when defining capabilities as what we have reason to value. However, given that the ‘having reason to value’-definition is still widely used within the capability literature and in overview articles of the capability approach (e.g. Robeyns, 2016b), I do not think it unreasonable to question whether it would work as a general definition of capabilities. Moreover, as I will argue later, the purpose is not here to invalidate Sen’s work on the capability approach nor the notion of ‘having reason to value.’ Indeed, for particular capability applications and selections of capabilities, it might be useful to focus on what we have reason to value.

9 It could be argued that the perfectionist interpretation does not necessarily rely on a particular procedure, as it is possible to identify a perfectionist list of reasonable values without going through any procedure. Even if this is true, the perfectionist interpretation of ‘having reason to value’ is still subject to the second objection.

10 See, though, Carter (2014, p. 79) who argues that Nussbaum ‘clearly assumes the value-laden sense of functioning.’ In section 4, I return to Nussbaum’s definition of capabilities and whether it represents the value-laden conception of capabilities.

11 See, though, Carter (1999, 2014) who argues that the freedom to choose between realizing different capabilities is valuable in itself. If this is true, then the ability to choose between two brands of washing liquid would be a valuable choice.

12 For a similar argument, see also Nussbaum (2003, p. 45), Robeyns (2017, p. section 2.6.2), and Stewart and Deneulin (2002, p. 67).

13 Again, it should be noted that it might be unfair to criticize Nussbaum’s capability concept, given that it reflects a context of discussing the constitutional architecture for a decent society. However, Nussbaum (2011, pp. 18–19) herself does argue that she holds the definitive and final version of the capability approach. Hence, I believe, it is not a stretch to argue that she also holds that capabilities should be defined objectively as what promotes human dignity. See Robeyns (2016a) for a criticism of Nussbaum’s argument that she presents the definitive version of the capability approach.