Rights,goals,and capabilities

1. Introduction

In a consequentialist moral or political theory, the assessment of an action or policy depends on the implications of the action or policy. In particular, it depends on whether its performance or enactment will contribute to the realization of certain moral and political goals. One way of characterizing the distinctive normative perspective of the capability approach is to say that it constitutes a specific kind of consequentialist thinking. It differs from other types of consequentialism because capabilities figure prominently among the moral or political goals.

Consequentialism has often been seen as being at odds with deontological theories that emphasize the importance of rights. In a deontological perspective, rights are side constraints within which individuals should act and that may not be overridden – irrespective of the resulting consequences of that prohibition. Libertarianism, whether in its egalitarian or non-egalitarian variety, is a paradigm example of a deontological theory. On a libertarian account, the primary criterion when assessing social states is whether rights were violated when these social states came into being. The characteristics of the social states themselves are only of secondary importance, if important at all. Such deontological accounts have been criticized for neglecting other relevant moral considerations, for example, the impact a social policy may have on our well-being.

Challenging the opposition between consequentialism and deontology, Sen (1982, 1984, 1987) argues that the consequentialism the capability theory stands for is compatible with the importance we attach to rights. This is so because rights themselves can and should be included among the relevant goals of society. Rights that are also goals are called ‘goal rights’ by Sen, and those goal rights that are rights to capabilities ‘capability rights’. ¹ According to Sen, rights belong to the relevant goals of society and their realization or non-realization therefore forms one of the important aspects of an evaluation of social states. ²

In his account of the nature of capability rights, Sen uses an axiological perspective. This means the rights are discussed in terms of how they affect the goodness or badness of social states. Of course, rights also affect our deontic judgements, that is, our judgements about rightness and wrongness. Martha Nussbaum’s views on the relation between capabilities and justice are a clear representation of the latter, deontic approach to the notion of capability rights (2006, 2011). Nussbaum lists a number of capabilities the possession of which by each individual characterizes a minimally just society. Indeed, the items on Nussbaum’s list refer to the ‘fundamental entitlements of all citizens’ (2006: 155, 166; 2011: 70 ff.). Obviously, capability rights then also play a role in the subsequent societal choice of actions or policies based on that evaluation.

There has been quite a lot of attention given to the question as to which capabilities should be realized (an issue that is sometimes referred to as the ‘list issue’) and also to the question as to which capability rights should be formulated. Less attention, however, has been paid to the analysis of the nature of capability rights, that is, to the exact relationship between capabilities and rights. The aim of this article is to give a precise account of these capability rights and, by doing so, to provide a clearer picture of the kind of consequentialism that the capability theory represents. ³ This is needed because, as I will argue in the first part of the article, it is by no means clear what kind of right a capability right exactly is. In Section 2, I discuss the idea that a capability right is completely determined by its being a right to a capability. I will argue that such a definition is too weak since it allows for the possibility that such rights refer to negative rather than positive freedoms, something which does not square very well with the motivation underlying the capability approach. As argued in Section 3, this problem of the underspecification of capability rights can be avoided by stipulating that the rights are things we actually can realize – the right to a capability is itself a capability. However, that account is too strong and yields a plausibility problem: in a significant number of cases, it leads to counterintuitive conclusions.

The second part of the article (Sections 4 and 5) gives a rendition of the relationship between capabilities and rights that avoids the two problems. I first discuss the different ways in which rights can be said to exist, and then I defend an alternative account of capability rights. It is argued that to take the idea of capability rights seriously, capability theorists need to take a position on three rights issues. There is the issue of proclamation (which rights are to be protected), the issue of specification (what is the nature of those rights), and the issue of implementation (when can a right be said to exist). Since each of these questions can be answered by providing a list (namely, a list of generic rights, of so-called token rights, and of combinations of token rights, respectively), there can be said to be at least three additional ‘list issues’ that need to be addressed by the capability theorist. Using the distinction between the three issues as well as a point made by Nussbaum, Section 5 presents an account of capability rights that avoids both the problem of underspecification and the plausibility problem. In that account, capability rights are defined in a purely instrumental way: they are instruments for the realization of one or more capabilities. Whereas this view of the nature of capability rights does not suffer from the problems discussed, it raises doubts about the claim that the consequentialism of the capability approach can give more importance to rights than traditional forms of consequentialism.

2. Rights to capabilities

That the right to have a capability is an important goal because it is valuable (Sen’s axiological stance) or a matter of justice (Nussbaum’s deontic view) does not yet say anything about what kind of right it is. A capability right is defined in terms of its object (it is a right to a capability) not in terms of its type. ⁴ Thus, the right to have a capability could be merely a negative one demanding that others do not interfere in specific ways. It could also be a more positive right according to which others have certain duties to promote or realize your having the capability in question.

The distinction between positive and negative rights is, of course, just one possible way of distinguishing between different types of rights. Another way makes use of Kant’s famous distinction between perfect and imperfect duties. The proper rendition of the distinction between the two duties has been the subject of considerable controversy, but one common account is to say that a perfect duty is an obligation that can be discharged only in a specific way, whereas an imperfect duty allows for discretion in its enactment. The rights corresponding to a perfect duty would then be perfect rights; similarly, imperfect rights incur imperfect duties (compare Bagchi, 2008: 123). Another distinction between types of rights (indubitably the most influential one) was provided by Hohfeld (1964), who defined eight different basic types of rights. Different taxonomies have been derived, depending on the assumptions about the logical relations between these basic right types (see Kanger and Kanger, 1966; Lindahl, 1977; Van Hees, 1995).

In fact, we can go even further and combine different taxonomies of rights. Take the Hohfeldian claim right, which a person has with respect to some other person if the other person has a duty to act in a certain way. Applying the negative/positive rights distinction to it yields two new types of rights. With a negative claim right, for example the right not to be sexually harassed, others have a duty not to interfere, whereas positive claim rights, for example a right to an unemployment benefit, would impose a duty on others to provide certain services or commodities to you.

Ignoring feasibility constraints, the type and object of a right are independent dimensions: information about what a right is to need not imply anything about what sort of right it is. ⁵ Since there are different types of rights, there can also be different types of capability rights. Consider the capability ‘to be adequately nourished’. The capability refers to the actual ability and opportunity to have adequate nourishment. Under a very weak and rather cynical account of such a right, a starving person may possess a right to such a capability. That would be the case, for instance, if such a right were what in the Hohfeld classification is called a privilege. The right then refers to the fact that the person would do no wrong (is at liberty) if he were able to nourish himself adequately. Like any conditional, this can be true without the antecedent being satisfied. Another scenario yielding the same conclusion is to see it as a negative claim right. Having the right then means that others have a duty not to interfere with you having or exercising the capability if you were to have the capability; again, it does not imply that you do in fact have the capability.

The idea that capability rights can be of any type runs counter to the normative force driving the capability approach. I identify this as the problem of the underspecification of capability rights. An important motivation for the development of the capability approach was to move away from empty notions of ‘formal’ freedom to a ‘substantive’ account of freedom (compare Sen, 1984). Since the idea that negative freedom is too meagre an ideal is an important impetus underlying the capability approach, it is an understatement to say that the view that all capability rights could be negative rights would be rather odd. Indeed, Sen and Nussbaum have emphasized repeatedly that the rights refer to actual abilities and thus to positive freedoms. This means that on their view capability rights cannot be of just any type. Hence, there is more to the relationship between rights and capabilities than the fact that a capability right is a right to a capability. ⁶

3. Rights as capabilities

Sen writes that capability rights have correlating obligations, which refer to what ‘others can do to safeguard and expand’ the capabilities in question (2005: 152). The formulation suggests that many of the capability rights will be positive claim rights, that is, others have a duty to ensure that persons actually do possess the capabilities. ⁷ Clearly, such a view avoids the underspecification problem. Elsewhere, however, Sen (2004: 339–40) makes it clear that such an interpretation is too strong. Given Sen’s focus on how rights affect the assessment of social states, it is not difficult to see why he thinks so. As noted, other values also affect our assessments and they may in fact override the importance we attach to rights. Such overriding can occur in two ways. First, we can decide to drop a goal right from the list of desiderata if it is not compatible with other ideals. Thus, say, even though we may find it a valuable goal that all citizens have a positive claim right to proper medical treatment, the right to such treatment may be contested because it is assumed that it cannot (yet) be implemented. Sen rejects this way of dealing with conflicts between rights and other goals. In particular, he criticizes the underlying presumption that a capability right must be ‘wholly accomplishable’ (Sen, 2004: 347). Sen therefore opts for a second route, one in which the goal right is weakened rather than abandoned. In particular, the rights are taken to involve imperfect duties instead of perfect ones: the corresponding duty is to give ‘reasonable consideration to a possible action’, it is not an absolute obligation to undertake some particular action (Sen, 2004: 339). Clearly, to say that the state has a duty to consider seriously the implementation of a system of universal health care is much weaker than to say that it should actually implement such a system.

However, if feasibility considerations alone allow us to weaken the type of right involved, it could just as well be weakened in different ways, for example in terms of a mere negative right (others may not prevent you from getting the medical care you need) or a privilege (you are at liberty (do no wrong) to have medical care). In other words, if we follow Sen here, the problem of defining capability rights too weakly and thus the problem of the underspecification of capability rights may crop up once again.

From Nussbaum’s description of the relationship between capabilities and rights, it is clear that the problem is not likely to occur under her account. Though she views her capability approach as a species of the human rights approach, Nussbaum prefers to speak of capabilities rather than rights precisely because using the term ‘capabilities’ brings out more sharply that more is at stake than the protection of negative liberty. Nussbaum (2006: 286–8; 2011: 65) rejects as too thin the view that securing a capability right only requires the absence of interference by the state. Instead, she emphasizes the importance of the actual possession of the capability in question: capability rights are secured ‘only when the relevant capabilities to function are present’ (Nussbaum, 2006: 286). If the non-realization of x always is an indication of a violation of a right to x, then the right surely cannot be a negative one.

Nussbaum’s stance thus suggests capability rights are to be viewed as positive claim rights. Osmani nicely emphasizes both the positive nature of capability rights and their incurring of duties for others when he remarks that the obligation which capability rights entail for others ‘extends immediately beyond the constraint of non-violation’ (2005: 211). It is not just that right-bearers have a negative claim of non-interference upon others; there is a positive one, that the object of the right in question is ‘protected and promoted’. ⁸

Let us assume that capability rights can indeed be defined as rights to capabilities that are positive claim rights. Now consider a world in which those rights are secured. This means that an agent has the means to ensure that others fulfil any of their obligations towards her; for instance, because they can be enforced through legal means. Since rights to capabilities are taken to be positive claim rights, the realization of a right to capability x then means having the capability to have the capability x. After all, either a person possesses the capability already (others fulfil their duty to realize x) or she does not, but she nevertheless has the means to acquire it (can ensure that others fulfil their obligation to realize x). Hence, on this view capability rights are not only taken to be rights to capabilities, but also rights as capabilities. In the case in which an individual does not have the means to acquire the capability when others do not fulfil their obligations (that is, cannot enforce the fulfilment of those obligations), the right is violated. To quote Nussbaum again:

The right to political participation, the right to religious free exercise, the right of free speech – these and others are all best thought of as capacities to function. In other words, to secure rights to citizens in these areas is to put them in a position of combined capability to function in that area. (Nussbaum, 2000: 98).

It could be objected that this critique of the rights-as-capabilities view is based on a rather uncharitable reading of Nussbaum’s account of capability rights. In particular, Nussbaum does not claim that others should make sure that an agent actually possesses the ability to love (or laugh, care, and so on), but rather that the conditions under which people can enjoy such capabilities should be protected. ⁹ Suppose, for instance, the material circumstances one lives in are so dire that one lacks the time and the opportunity to develop strong emotional ties with others – the development of such ties is simply a luxury one cannot afford. If those circumstances could be improved in such a way that they no longer formed a preventing condition, then the fact that people do live in these dire circumstances is reason enough to say that the capability right is violated. Put differently, the capability right to love others, say, does consist of certain claims, but they are claims to conditions that are necessary rather than sufficient for the capability in question. The problem of the account of rights being too strong will indeed not arise with this interpretation. However, it does seem to reintroduce the problem of underspecification of capability rights. After all, if for at least some capability rights the absence of preventing conditions allows us to say that the rights are protected, then capability rights need not refer to positive freedoms and everything seems possible again. I shall argue that capability rights should indeed be seen as conditions for the realization of capabilities rather than as rights to capabilities and that one can do so in a way that avoids the problem of underspecification. To prepare the groundwork for that account of the relationship between rights and capabilities, we first have to turn to the analysis of the conditions under which a right can be said to exist or not.

4. The existence of rights

Following Dowding and Van Hees (2003), we can distinguish between three phases or stages in the creation of rights. The first step is that of the proclamation (or pronouncement) of some list of the rights that agents possess or should possess. The proclamation or pronouncement can be by means of a bill of rights, but may also be part of some moral or political theory. Whether we consider legal or moral rights, the rights in this step are generic ones. They are rights to perform types of actions or to realize types of states of affairs and do not yet refer to the particular actions (or states of being) that may fall within the scope of the right. Human rights are typically generic rights – they have generic actions (for example, free movement) or states of affairs (for example, equal pay for equal work) as their object.

The second step in the construction of rights consists of the specification of the various rights. Any act of specification has two elements to it: a specification of the object and of the type of the right. First, the rights-object is determined by the tokens that correspond to the generic action or states of affairs. Thus the token action of driving my car to work in the morning falls under the scope of the generic action of my moving around freely – it is an instantiation of it. Deciding which tokens are and which are not instantiations is not a linguistic exercise: it involves taking a normative position. In a purely physical sense, entering my neighbour’s house against his will is an instantiation of moving around freely, yet it does not fall under the scope of the generic right to free movement. Second, the type of right involved needs to be specified. My right to drive by car to work, for instance, does not involve any positive duties on the part of others to ensure that I do indeed have a car in which I can drive to work. It is not a positive claim right, but rather a Hohfeldian privilege: it entails that I do not have a duty not to take the car. By contrast, the right to education is usually taken to consist of positive claim rights: the state has an obligation to ensure that parents have the ability and opportunity to send their children to school. Clearly, different views can exist as to the nature of the token rights that a particular generic right involves. In western democracies, the right to work refers to the absence of constraints. In India, however, the National Rural Employment Guarantee Act (NREGA) guarantees rural workers employment for a period of 100 days, making the token right in question a positive claim right. ¹⁰

Assuming we know which generic rights have been pronounced, the existence of which we are therefore examining, and also assuming we know which particular token rights instantiate the generic rights, the third step is to assess whether the generic rights are secured. That is, we have to examine not only whether the generic rights exist in the sense of being formally proclaimed, but also whether they exist ‘materially’ (Dowding and Van Hees, 2003). ¹¹ This is what I call the issue of the implementation of rights. To establish such existence we have to address two questions. First, we need to know which of its various token rights are secured. If you are allowed to become a member of just one political party, the official state party, then we do not say that your formal right to join a political party exists materially. ¹² Second, it should be clear what it means to say that the token rights in question are adequately implemented, that is, when they exist materially. The answer to that question will depend on the type of the token right in question. If it is a liberty right, for instance, it exists if others will not interfere, or are not likely to do so, if you were to try to perform the action. If it is a positive claim right, on the other hand, the criteria for its existence will be more demanding. It means that you have the capability to perform the action, that is, you have the actual ability and opportunity to perform it. ¹³

To sum up, we can make a distinction between the formal and the material existence of rights. The first step is to proclaim some set of generic rights (the proclamation stage). In order to determine which of those rights also exist materially, we next need to ascertain the relevant token rights that instantiate each of the formally existing rights (the specification stage). The final step is to establish which combinations of token rights need to be protected (the implementation stage). Each of these steps can be made by providing a list of rights: a list of generic rights, a list of token rights that instantiate the generic rights, and a list of the various combinations of token rights that would allow us to say that the generic rights exist not only formally, but also materially.

5. Rights and capabilities

On Sen’s definition, a capability right has two aspects: it is a right to a capability and it is a right the realization of which is an important moral and political goal. The question that has been examined above was whether these two aspects together sufficiently explain the relationship between rights and capabilities. The answer was negative. In this section, I shall give an account of the relationship between goals, rights, and capabilities which forms a way out of this dilemma. To do so, I will redefine the concept of a capability right. Since it suffices to concentrate on the first two stages in the construction of rights, proclamation and specification, the question of their implementation is here ignored.

First, consider the proclamation stage in the construction of rights. In a consequentialist moral or political theory, the rights that are to be proclaimed will depend on the set of goals G. For a capability theorist, G will include capabilities. Indeed, discussions within the capability approach about ‘the list’ are discussions about which capabilities should and which should not be included in G. ¹⁴ For the purpose of this article, let us suppose the issue is resolved and thus that some such list of capabilities is available. Call these the capability goals. The (capability and non-capability) goals in G determine which rights are to be proclaimed and therefore limit the set of possible rights allocations. Clearly, different goals may require different means. Different theories of consequentialism in general, and different versions of the capability theory in particular, may thus yield different rights allocations. Note also that some particular list of goals may be compatible with more than one rights allocation. This is the case, for instance, if the goals cannot (yet) be realized simultaneously. Depending on which goals one does decide to realize, different (non-ideal) rights allocations can be proclaimed.

Now let us assume one such rights allocation R has been singled out as constituting the rights that are to be implemented. The rights in R form the rights that are proclaimed in what was described as the first stage in the construction of rights. Some of these rights can themselves be goals – these are what Sen called ‘goal rights’. Sen defined capability rights to be those goal rights (that is, elements belonging to both G and R) that have capabilities as their object. We saw that to define capability rights in this way yields problems in the second stage of the construction of rights, in their specification. The rights cannot be of any type, yet to say that they all are claim rights will be too strong. We saw that Nussbaum circumvents the problem by taking certain rights to be about the conditions under which people can have capabilities rather than as rights to the capabilities themselves. Whereas in many cases this is indeed intuitively much more plausible, the response seemed to reintroduce the problem of underspecification: it is not clear why some capability rights are to be interpreted as negative ones while others are positive claim rights.

We can, however, take up Nussbaum’s suggestion and apply it to all capability rights. ¹⁵ This means we drop the assumption that a capability right is a right with a particular object (namely, a capability) and instead define it as one with a particular objective (namely, the realization of a capability). Stated differently, a capability right is defined in purely instrumental terms. It is any right in R that is essential for the realization of one or more of the capability goals in G: if it is not included in R, some capability goals in G will not be realized. ¹⁶ With such an adjustment of the definition of capability rights, the dilemma between having an account of capability rights that is either too weak or too strong disappears. The problem of underspecification is avoided because capability rights are not defined as goals, but as instruments, that is, they are defined in terms of their role in bringing about the goal capabilities. This instrumental role determines their type. Some capabilities (say, the capability to have adequate shelter) will indeed require positive claim rights. However, not all capability goals will be realizable through positive claim rights. As we saw in the discussion of the capability to love, positive claim rights may not be effective for the realization of a capability goal and may in fact be in conflict with other goals in G. In those cases, the right in question will be weaker even though it still is a capability right. As a result, the plausibility problem – defining the rights too strongly – does not arise either.

6. Conclusion

The analysis is important for several reasons. First, it reveals that the relationship between rights and capabilities is more complex than it is sometimes taken to be. There has been a lot of discussion within the capability approach about ‘the list’ of central human capabilities. Perhaps there has even been too much discussion, for, as we have seen, to make sense of the relationship between rights and capabilities we not only need to know what the goal capabilities are (the ‘first’ list issue). The rights that should implement them yield at least three additional list issues: without a proclamation of the rights which are to lead to the realization of the goal capabilities, a specification of the relevant token rights corresponding to those generic rights, or knowledge of the combinations of token rights that allow us to say that a capability right is implemented effectively the relationship between capabilities and rights remains elusive. In fact, on the instrumental definition of capability rights proposed here, the stages of specification and implementation are of crucial importance. After all, to establish whether a right contributes to the realization of one or more capabilities, capability theorists need to focus on the effects of securing the various token rights it comprises. This presupposes both that it is known which token rights it encompasses (specification) as well as what it means to say that those token rights are properly protected (implementation).

Second, the analysis shows there may be more overlap with other moral or political theories than is sometimes suggested. Consider the securing and protection of negative freedoms. Having a right to such freedoms will undoubtedly play an instrumental role in the realization of some of the goal capabilities. Classic liberals, for instance, have defended civil and political liberties (freedom of speech, freedom of religion, and so on) with reference to the role these freedoms play in the realization of the capability to live a life of autonomy and dignity. On the definition given here, those rights would therefore also be viewed as capability rights, even though they refer to negative freedoms only.

Finally, despite there being overlap with other moral and political theories, the account also allows us to say more about the distinctive nature of the capability approach. Capability rights are not defined as a specific type of right, but in terms of their role in helping to realize the capability goals. Not all capability rights will be goals, and to the extent that they are not, rights thus are subordinate to the capability goals. Conversely, not all of the relevant social and political goals will be rights and there is no reason to suppose that the capability approach gives particular importance to the rights among those goals when a trade-off has to be made. Of course, whether this instrumental account of capability rights is normatively appealing is a very different issue that has not been addressed here. However, the analysis does raise doubts about claims that the capability approach gives more importance to rights than traditional forms of consequentialism do.