Introduction
The âcapability approachâ (CA) or âcapability perspectiveâ as it has been developed in the writings of Amartya Sen since his Tanner Lecture (âEquality of What?â) given at Stanford University in 1979 can be (and has been) interpreted in very different ways. Because Sen has used the approach in so many different contexts â from the conceptualization and measurement of development, to theories of justice and rights to issues about poverty, gender, social exclusion and disability, including empirical work â the literature on the approach has grown significantly as work in each of these contexts and areas has grown (see Alkire et al. 2008; Comim et al. 2008). Furthermore, the approach has been used by academics and practitioners in many fields and disciplines so that the literature is genuinely multi-disciplinary. The CA, while manifestly not a âtheoryâ in the sense relevant to theories of justice or morality â like John Rawlsâ theory of justice (Rawls 1972, 1993 inter alia) or utilitarianism â yields a range of insights in several contexts. And some have supposed that something akin to a theory â to rival others, including Rawlsâs theory â can indeed be developed out of it. In some ways, this seems to be precisely what some, including Martha Nussbaum who has developed her own âcapabilitiesâ approach, have tried to do thereby addressing the âlimitationsâ of a mere âperspectiveâ or âapproachâ by adding to the basic elements of Senâs approach various principles and ideas which make it closer to a theory even if still somewhat incomplete â whether in the context of gender justice or development. The fact that the approach is merely an âapproachâ has, for some, made it an inadequate response to the limitations Sen finds in the works of others â like Rawls and some utilitarians. But as can be seen from the quotation at the head of this chapter, Sen himself retains his view that the chief contribution of the CA arises from its focus which involves a change in the informational basis of social evaluation. There seem to be potentially significant differences amongst those who advance distinct versions of the CA and views of how it can be most usefully articulated and applied, with the most prominent being the versions of the approach which Sen and Nussbaum themselves have developed. In fact there seem â on the face of it â to be a range of approaches involved, not just one approach.
In this chapter I argue that the CA can be interpreted in very different ways, with its scope, ambitions and reach being drawn very differently according to each interpretation. Consistent with the passage from Senâs The Idea of Justice cited at the head of this chapter, the CA invites us to see things in a different way to other approaches by focusing attention on certain sorts of information. Those who advance or use the approach can then agree on a core of claims which they can develop differently. That avoids the danger of âschismsâ or disagreements amongst protagonists of the approach, which would distract from many possibilities for research and insight generated by a general if limited perspective. Sen in particular argues that it is important to recognize the limitations of the CA. While his body of work on capability has often been criticized for its incompleteness, my reading of his work (and of work in this growing field) supports the case for restricting the scope of the approach itself while allowing for very different ways of further developing and using it in different contexts. Nonetheless, I argue that some lines of argument Sen advances in suggesting that the approach is limited are not meant as criticisms of the approach which highlight its weaknesses. Rather the aim is to restrict and delimit its scope with a view to identifying its contribution. Furthermore, some of Senâs claims are not about the limitations of the CA, but about the limitations of the value of equality.
The chapter is structured as follows: the first section distinguishes various ways in which the approach can be interpreted; various possible criticisms and âlimitationsâ of the approach are discussed in the second section; and the third section concludes.
Interpreting the approach
It is important to recognize that there remains some confusion about what the CA is. The approach can be seen, on the one hand, as providing a critical perspective on other views which leads to a positive case for including certain types of information in normative evaluation while making suggestive remarks â but little more than that â on application. I refer to the approach in this form as constituting a âthin viewâ. In this basic form the approach makes critical claims about problems in John Rawlsâ theory of justice, and income-based views of poverty and development, for example, on the grounds that these focus on means: âsocial primary goodsâ (which are all purpose means) in Rawlsâ approach, and income in influential economic approaches to the assessment of progress or deprivation. It asks us also to look at what these means do for people, the actual lives or freedom they make possible.
This freedom is what Sen usually refers to as âcapabilityâ and the various beings and doings people can achieve are known as âfunctioningsâ. However, it is also clear that in this literature (at least) two different uses of the term âcapabilityâ have emerged and that these two senses can come apart (Qizilbash 2006: 21â22). On one interpretation of the term âcapabilityâ, it âreflects the alternative combinations of functionings the person can achieve and from which she can choose one collectionâ (Sen 1993: 31). Here âcapabilityâ is most naturally understood as an opportunity concept. On another more natural interpretation, capability refers to an ability or power to do or be something (Sen 2009a: 19). What I think of as Senâs âtechnicalâ sense of capability refers to the first sense: the freedom to achieve a collection of functionings. When Sen uses âcapabilityâ in the second sense (i.e. some specific ability or power) he is using it in a more informal sense. Recognizing this looseness in his use of the term âcapabilityâ, he writes that âthe CA is ultimately concerned with the ability to achieve combinations of valued functioningsâ and âyet it is often convenient to talk about individual capabilitiesâ â where âindividual capabilitiesâ are seen in terms of âthe ability to achieve the corresponding individual functioningâ (Sen 2009a: 233). Of course, the opportunity and power interpretations can, nonetheless, be run together, so that âcapabilityâ refers to âour ability to achieve various combinations of functioningsâ â and capability refers to both a power or ability and an opportunity â and it may be in bringing these two senses together that Senâs notion acquires its force.
As a thin view, the CA shifts our perspective â it asks us to look at information on capability and functioning alongside or instead of the means to achieve these, particularly where a focus on means alone can be deceptive. That focus can be misleading because of the different rates at which people can convert means into valued ends, where these rates reflect a variety of âconversion factorsâ. On the other hand, a focus on âutilityâ, where this might be understood as desire fulfilment or happiness can â according to Sen â mislead us for a quite distinct reason. Various underdogs â like the oppressed housewife, the hardened unemployed or the overworked labourer, etc. â might learn to find happiness in small mercies, or may satisfy their desires by cutting them down to realistic proportions. But that does not mean that they are doing especially well â that their lives are decent or good or that they have the opportunity to live good lives. The metric of âutilityâ can for this reason be misleading in assessments of the quality of life, development or justice. In these examples, the disadvantage of the relevant underdogs would arguably be better captured if one started from a capability perspective. These arguments, which are central to the thin view are, in Senâs earlier works on capability, supplemented with some comments on the use of various techniques for evaluation when there are several functionings which are valuable and people may differ in the functionings they value or the weights they give to these in some specific evaluative exercise. If people all agree on a number of functionings â even if they differ on the weights given to these â there can nonetheless be agreement on a range of judgements â one option will typically âdominateâ or rank higher than another if it has more of all relevant functionings. Sen refers to this technique as the âdominance partial orderâ (see, for example, Sen 1992: 46â49). And there may be a range of functionings â such as being minimally adequately nourished, or avoiding starvation â which are fairly uncontroversial, so that in an exercise such as the evaluation of poverty one may be able to go quite far (Sen 1993: 31, 48â49). Even if there is disagreement on weights, the possibility of some agreed judgements is not out of reach: if there is a range of weights on which people can agree, then even if they disagree on the precise weights to give to specific functionings, they may agree on a range of judgements. It is noticeable that in each of these cases when Sen is concerned with the evaluation of functionings, he is typically concerned with the possibility that people with different views might achieve agreement.
In his later writings, most notably in Development as Freedom and The Idea of Justice, Sen has responded to a range of worries about the fact that the CA is incomplete â inasmuch as it does not yield either some unique or specific list of functionings (or capabilities), which does not vary according to context, or indeed a complete ranking of social states. His views have developed in responding to these concerns. Importantly, he has argued for the importance of social choice and democracy in evaluations which can generate judgements which are acceptable (Sen 1999: 76â81) across society or nations according to the context. This is a central element of Senâs further development or articulation of the approach. I have termed this more extended development of the approach Senâs âthick viewâ, since it goes beyond the limited suggestions about valuation in his earlier work (Qizilbash 2007, 2011a, 2011b). It is important that in these writings Sen still leaves the approach incomplete. He does not suggest any formula for interpersonal comparisons of well-being or advantage or indeed for the selection of weights. Rather than doing that, he leaves these matters to public debate or reasoning if they are to be the basis of social evaluation. Nonetheless, that takes his approach well beyond its earlier incarnations and indeed raises further questions â what form of public reasoning does Sen have in mind? Might such reasoning itself not be âcorruptedâ by imbalances of power in society making it a problematic approach in contexts such as justice, including gender inequality, and indeed in the context of development more broadly? Sen has much to say about these matters but I will not go into it much further here (for further discussion see Sen 2009a and Qizilbash 2007). When people refer to the CA as including these further views they have in mind Senâs thick view. And while many of those who agree with Sen on the importance of foundational concepts such as capability and functioning, as well as on problems relating to resource- or âutilityâ-based evaluations â Nussbaum, for example, uses both foundational concepts and critical claims about alternative positions in developing her âcapabilities approachâ â on issues relating to valuation and social decision making, differences emerge (e.g. about the importance of articulating a list of functionings). What I have termed the âthin viewâ is, I suggest, the core motivational base of the CA. Once one begins to go further by âthickeningâ the approach, there is more scope for disagreement. That in part justifies Senâs concern in his earlier writings not to go beyond a general perspective which is broadly constituted by his thin view. If capability and functioning are the key âvalue-objectsâ which are central to the CA, then Sen (1993: 48) suggests that there is a case for âpausingâ prior to making the approach more âcompleteâ â along various lines which might include providing a specific list of functionings or weights. Nussbaumâs early and explicitly Aristotelian way of developing the approach (Nussbaum 1988, 1990, 1992, 1995a, 1995b) â as well as her later articulations of it (Nussbaum 1998, 2000, 2003, 2006, 2011) â were routes Sen might have taken. But he insisted that there may be disagreement on evaluation including both the grounds on which weights are chosen as well as on the actual weights used in evaluation, while there can nonetheless be âreasoned agreement on the general nature of the value-objectsâ (Sen 1993: 48). If reasoned agreement is something one is looking for, one might further claim that there is a strong case for associating the approach itself with the thin view.
There is a minor point I must add. In Senâs statements of the CA a distinction is made between well-being and agency â between those goals which involve pursuit of our own well-being and the broader range of goals which include but might extend beyond that pursuit, our agency goals. This leads Sen to make a fourfold distinction between: well-being freedom; well-being achievement; agency freedom; and agency achievement (Sen 1993: 40). In some exercises (e.g. involving the evaluation of the living standard) Sen thinks one might use the CA to evaluate well-being freedom or achievement leaving out agency goals. I introduce this distinction at this stage in part because Nussbaum has resisted it (see Nussbaum 2000: 14) and indeed in her more recent work writes that: âthe distinction [between well-being and agency] is obscure and not useful to one who, like Sen, has rejected (on good grounds) utilitarian notions of well-beingâ (Nussbaum 2011: 200). This distinction remains central in Senâs more recent statements of the CA (Sen 2009a: 286â290). Since the usefulness of the distinction is the subject of controversy and is not common ground between Nussbaum and Sen, I exclude it from the thin view.
Another way to distinguish various interpretations of the CA is in relation to its scope or focus. On one interpretation the approach narrows the informational space to capability and functioning â it narrows the informational focus of judgements. That focus seems key to the quotation at the head of this chapter â Sen sees the focus on specific information as key to the distinctive contributions which flow from taking a CA to various issues. The question is whether the approach actually restricts information in focusing on capability and functioning, in the way that âwelfarismâ (on Senâs own celebrated definition, see Sen 1979) restricts itself to information on well-being. If the CA argues not merely for a focus but also a restriction of information (i.e. an exclusive focus), the reading which results is a ânarrowâ interpretation of the approach. An alternative interpretation of the approach is that it does not restrict information in this way, but that it actually allows for other forms of information. It includes a concern with not just capability and functioning, but also other aspects of freedom which Sen distinguishes from what he calls the âopportunity aspectâ of freedom â notably the âprocess aspectâ of freedom which includes non-interference in the personal sphere and freedom of choice. Different themes in Senâs own writings on capability can lead one in the direction of thinking that he sees the approach as either ânarrowâ or âbroadâ. On the one hand, he argues that the CA is limited and in some contexts cannot take account of aspects of freedom that matter, such as process. In doing so, he seems to be adopting a narrow interpretation of the approach so that information on process is not included within the approach. But his general position is that capability is not all that matters and some exercise in social choice or public reasoning may decide on the relative importance given to opportunity as opposed to process (see, for example, Sen 1999: 77). If that is what one takes to be the âapproachâ it is not ânarrowâ â and indeed is best understood as a broad approach which allows for a variety of valued objects of which capability and functioning are only two amongst others. On this interpretation the CA broadens rather than limits the informational space of social judgements. It does not exclude information on resources, desire satisfaction or happiness and so on but can supplement it, by advancing important arguments for not restricting oneself to the metrics of resources and desire satisfaction. On this âbroadâ interpretation, the approach does not argue for an exclusive informational focus but rather a broadening of the informa...
