Care has come to occupy a central place in debates about welfare. In part this reflects a growing recognition of what Mary Daly and Jane Lewis called the ‘analytic potential’ of the concept of social care in the ‘study of welfare states and how they are changing’ (Reference Daly and Lewis2000: 282). As Fiona Williams has argued, the ways in which care is organised ‘provides a particularly significant lens on both the micro and macro dimensions of our lives’ (Reference Williams2003: 1). At the same time the question of how and on what terms care should be provided, supported and regulated has become a key issue in normative debates about the future direction of welfare reform (Lewis, Reference Lewis, Glendinning and Kemp2006a). In particular, Daly and Lewis pointed to ‘a stampede to emphasize obligations (rather than rights)’ that had been manifested in ‘the direct enforcement of parental responsibilities and more indirect encouragement of responsibility for the care of elderly relatives’ (Reference Daly and Lewis2000: 295). That ‘stampede’ has continued, in subsequent years, and has been augmented by a plethora of measures to combat anti-social behaviour and to promote New Labour's ‘respect’ agenda (Respect Task Force, 2006). These measures reflect Tony Blair's insistence that his government ‘would not hesitate to encourage or even enforce’ the obligation ‘to bring up children as competent, responsible citizens’ (Reference Blair1998: 12). This suggests a level of intervention that is at odds with the autonomy customarily granted to parents and other carers, and there is a clear tension between attempts to encourage people to accept their responsibilities and attempts to shape or to direct the ways in which they fulfill those responsibilities.

The focus of this article is upon the question that lies at the heart of these debates: on what grounds should those who provide informal care for dependent adults and childrenFootnote ¹ be held to warrant public support through welfare? What is the basis of their claim to welfare, and what obligations should they accept in return for support?

The article discusses two answers to this question. The first is Stuart White's argument that care work should be viewed as ‘civic labour: roughly speaking, labour that provides a significant service for, or on behalf of, the wider community’ (Reference White2003: 98). For White, it is a requirement of justice that those who claim a share of the social product should make a proportionate productive contribution in return. Carers, then, are entitled to claim public support because they fulfil the contributive obligation required by the principle of fair reciprocity. It follows, however, that their entitlement is conditional upon the satisfactory performance of their civic labour.

A second answer is provided by Eva Feder Kittay's principle of doulia. In essence, this holds that the carer who meets the needs of those dependent upon her is necessarily rendered less able to attend to her own needs or to pursue her own interests. This derived dependency of the carer creates an obligation upon the wider community to provide the support she needs to thrive. That support is properly understood as a form of compensation for the costs borne by the carer, and so it follows that it should take the form of an unconditional entitlement.

This article will examine these contrasting interpretations of the rights and accountabilities of carers. It will show that White and Kittay draw upon and articulate different understandings of the self, and of the nature of reciprocity and of obligations. The argument will be that it is Kittay's writings which best capture the relationality of care and the nature of caring relationships. It does not follow, however, that welfare conditionality should necessarily be rejected. On the contrary, the case for some forms of conditionality can be grounded more effectively in an ethic of care than it can in contractual ethics.

Care as civic labour

Stuart White starts from the premise that a good society is a ‘community of mutual concern and respect. More exactly, the good society is one in which individuals exhibit democratic mutual regard’ (2003: 27). His objective is to outline a conception of economic citizenship that is appropriate to such a society. At the heart of this conception is the idea of justice as reciprocity. This, ‘roughly stated’, holds that if ‘one willingly enjoys the fruits of one's fellow citizens’ labours, then, as a matter of justice, one ought to provide some appropriate good or service in return’ (49). White goes on to argue that the obligation upon citizens to make a productive contribution to the community is contingent upon the opportunities and resources available to those citizens. If a society fails to offer all citizens a ‘sufficiently generous share of the social product’, then those who are ‘unjustly disadvantaged by this have a proportionately reduced obligation to contribute’ (91). Where, however, a society is ‘acceptably (though not absolutely) just’, then reciprocity of this ‘fair dues kind’ is a requirement of justice’.

This, of course, raises the question of what constitutes such ‘productive service’, and it is in this context that White considers the status of care work. The critical issue is that of choice. Reciprocity-satisfying work is not to be equated with any activity that a citizen decides to pursue. It follows that if raising children or caring for dependent adults is to count as civic labour, then it must be seen as something more than ‘a personal lifestyle choice’ (109).

In the case of parental care work, White argues that this something more is the fact that all citizens have a vested interest in the creation and nurturance of the next generation. Not only will they be economically dependent upon future generations, but they need them to be brought up as responsible citizens. In the absence of public support for parents, the wider community can be said to be free-riding upon the efforts of those who raise children. White acknowledges that this argument can only be pushed so far. Pregnancy and childbirth are generally a cause for celebration and at some point the care work involved in parenting becomes a price that has to be paid for the enjoyment of a private good. Where the line should be drawn will depend in part upon the view that the wider community takes about the desirability of future population growth.

Of greater significance for the present article, however, is White's insistence that if parenting is to count as civic labour, then parents must be accountable to the wider community for the ways in which they bring up their children.

White puts forward a different argument in respect of those who care for, say, ‘an elderly infirm relative’. They are performing civic labour insofar as they are taking upon themselves the full burden of a community-wide obligation. More precisely, the care that they provide is a merit good; that is, a good to which people are entitled as a matter of justice. ‘Primary carers’, then, ‘can be seen as unofficial providers of a particular merit good, and as having a claim to share in the social product based on this fact.’ (112). Here again, however, those who claim support from the community for their care work must accept the ‘corollary of this: that the community is entitled to expect quality care work in return’. For both parents and other carers, ‘the logic of casting care work as a form of civic labour, entitled to public support, also implies that conditionality is perfectly legitimate’ (2004: 11).

Carers as dependants

Eva Feder Kittay presents a radically different interpretation of the rights and obligations of those she terms dependency workers. Her starting point is that dependency is a fundamental part of the human condition. We are all highly dependent upon others, in infancy, in frail old age, or due to illness or significant impairment. These inevitable dependencies give rise to ‘relationships with the most compelling bonds.’

Kittay argues that the significance of these dependencies has gone unrecognised in much feminist theory. In consequence, many feminists have failed to appreciate how the inequitable distribution of caring responsibilities is central to both the explanation of women's subordination and to its resolution. ‘Acknowledging human dependency’, she claims, ‘is indispensable for finding a route round these obstacles to a truly inclusive feminism’ (Reference Kittay1999: 19).

Kittay argues further that the neglect of dependency is still more marked within the tradition of political liberalism established by the work of John Rawls. She claims that Rawlsian liberals necessarily ground their theories on an assumption that all citizens are autonomous actors, who are equally free to make choices and to enter into exchanges in pursuit of their ends. The realities of care, however, render such a conception implausible, and repudiate the notion ‘that the founding obligations of a social order are derived from the voluntary association of equally situated and empowered individuals’ (73). An alternative framework, then, must begin by considering what reciprocity means for the dependent person and for those who care for her. It must recognise that equality will be unattainable unless those who are called upon to care for others can do so without losing the capacity to function as free and equal citizens. It is on these grounds that Kittay calls for a public ethic of care, which acknowledges the responsibility of the wider society to care for the caregiver.

It should be acknowledged at this point that Kittay is not alone in making these criticisms of Rawls. Martha Nussbaum, for example, has written extensively on the ways in which Rawls ‘idealizing fiction’ of people ‘who are fully cooperating over a complete life’ serves to exclude physical, sensory or learning impairments (Reference Nussbaum2006: 141). Moreover, such critiques have not gone unchallenged. Linda Barclay, for example, concedes that the ‘abstract individualism’ rejected by Kittay and Nussbaum pervades ‘a great deal of moral and political theory’ (Reference Barclay, Mackenzie and Stoljar2000: 60), but still contends that they overstate the distinction between the autonomy assumed by liberalism and the relationality of the ethic of care Footnote ². We are, she says, ‘quite compatibly both autonomous agents and deeply social selves’ (2000: 68). Similarly, Norman Daniels has argued that it is possible to extend Rawl's theory to support an account of justice and care. In debate with Kittay he has insisted that Rawl's principle of fair equality of opportunity can encompass the ‘social supports’ that are necessary to ensure that ‘such care does not burn out caregivers and is compatible with their continuing to pursue other goals in their lives’ (Daniels, Reference Daniels2002: 72; Kittay, Reference Kittay2002).

This is a complex and important debate, but it is not one that can be covered here. What is important for the present article is the way in which Kittay draws upon her understanding of reciprocity to argue that public support for carers should be viewed as a form of compensation, provided in recognition of the price the caregiver must pay if she is to meet her obligations to those dependent upon her. It follows that such support should take the form of an unconditional entitlement. The community is not employing the carer, it is making restitution for the costs she meets on its behalf. It does not ‘have an open ticket to intrude upon the relationship or to regulate the life of the dependency worker’ (1999: 144). Only if she has her own needs met in this way, can the caregiver enjoy equal citizenship.

Doulia and conditionality

White has dismissed the ‘compensation argument’ for supporting parental care work. In a debate with Anne Alston, he has acknowledged that parenting is a long-term and potentially burdensome commitment. ‘But so what? People make all kinds of potentially burdensome choices and society is not obviously obliged to compensate people for them’ (Reference White2004: 8).

The difficulty with White's argument here is that choice is a double-edged sword. It could equally be claimed that some people are choosing to provide voluntarily what he regards as civic labour. For Mead and Beam, for example, ‘the theoretical issue behind care work’ is not whether or not it is onerous but ‘whether obligations can be self-chosen’ (Reference Mead, Beem, Mead and Beem2005: 255). Strictly speaking, someone who is caring for a partner with a degenerative condition, or for a frail parent, or for a disabled child could walk away. The great majority, however, do not see their decision to remain as constituting a choice in any meaningful sense. It is simply the acceptance of a commitment that is inherent in the relationship itself. As Kittay expresses it, they are not free to choose, but nor are they coerced.

Moreover, Diemut Bubeck, Robin West, and others have argued convincingly that even if care work could be regarded as having been chosen it still has the potential to impose burdens upon the caregiver that distinguish it sharply ‘from other “chosen” paths of life’ (West, Reference West, Kittay and Feder2002: 93). Caring can be empowering and rewarding but it is still an activity that is essentially directed towards the needs of others. ‘Unlike exchange or contracts, the usual paradigms in liberal social or political theory, caring is not mutually beneficial, but is an asymmetrical transaction of material benefits’ (Bubeck, Reference Bubeck, Kittay and Feder2002: 167). As West argues, the nature of the caring task means that it should not be regarded as ‘employment at will’. It cannot be abandoned if a better opportunity presents itself, and nor can it easily be set aside if the carer herself is ill or exhausted. Moreover, as Lewis and Guillari note, the complexity of care relationships means that they cannot be fully commodified, and the responsibilities of the carer cannot be wholly relieved by the provision of child care, domiciliary care, or other forms of support.Footnote ³ All of this applies, ‘whether or not the decision to embark on the care giving path was voluntarily taken’ (West, Reference West, Kittay and Feder2002: 93).

So is this the end of the argument? Does the principle of doulia establish a compelling case for giving all care workers an unconditional entitlement to welfare? This, for example, was the position of those who joined Kittay in the Women's Committee of One Hundred to oppose the welfare reforms introduced in the United States in the 1990s. The focus of their opposition was the imposition of work requirements upon welfare mothers, but at the heart of their campaign was the demand, in Gwendolyn Mink's words, that welfare should ‘become an income owed to nonmarket, care-giving workers – owed to anybody who bears sole responsibility for children (or for other dependent family members).’ This would be ‘nationally uniform, paid automatically to categorically eligible care-givers, and impervious to the race, class and moral judgements society makes of individual women’ (Reference Mink1998: 135). In particular, it would start with a presumption that ‘the care-giver does do her job’ and is free to do so ‘as she deems appropriate.’ It would shield her from ‘intrusion by requiring government to establish a compelling justification – child abuse or abandonment, for example – to override the rights of parents’ (138).

Fiona Williams argues that the research evidence would support a similar ‘presumption’ in Britain, since it entirely disproves fears that there has been a decline in people's sense of commitment to each other and to their children (Williams, Reference Williams2004; Seaman et al., Reference Seaman, Turner, Hill, Stafford and Walker2006). She also points to the ways in which expectations of carers are highly differentiated by gender and by social class, while Tony Fitzpatrick is yet more dismissive of the generalised ‘care test’ that he believes is the logical conclusion of White's arguments.

Others theorists of care have argued that attempts to shape the behaviour of carers through welfare conditionality are doomed to fail since people cannot be compelled to care. Clement, for example, notes that ‘caring requires not just that one act in a certain way but that one feels a certain way, and one cannot be forced to have certain feelings’ (Reference Clement1996: 116). Similarly, Joan Tronto has argued that ‘any conception of a right to care must also forgive any particular individuals from any particular obligation to care’ (Reference Tronto, Hirschmann and Leibert2001: 77).

This is a formidable battery of arguments. Nonetheless, it is still argued here that some forms of welfare conditionality can be reconciled with the principle of doulia. This claim is made for two reasons. The first and most important is that not all dependency workers are excluded from the labour market or from civil society to the same degree. As Jane Lewis has observed, Kittay's ‘characterisation of care is undoubtedly accurate in the case of a profoundly disabled person, but may be less apt when describing the care needed by an early teenager’ (Reference Lewis, Paxton, White and Maxwell2006b: 156). Of course, Mink is right to claim that parenting and household management is hard work, and that we know this because ‘people who can afford it pay other people to do such work’ (Reference Mink1998: 134). It does not follow, however, that such work can be equated with caring for those with the greatest dependencies. Indeed, Kittay herself acknowledges that it is hard to ‘draw the temporal boundary of the dependency relationship between mother and child’ (1999: 196), and that the costs to the care giver ‘hang in part on the degree of the dependency’ (54). If this is so, then the strength of the claim to unconditional welfare also ‘hangs in part’ upon the degree of derived dependency for which she is being compensated. This is the critical weakness in Kittay's claim that ‘if one accepts the principle of doulia’ then welfare must become ‘a right issued forth from a principle of care and freely dispensed to all dependency workers’ (Reference Kittay1999: 118). On the contrary, the principle of doulia points to a continuum of dependencies, responsibilities and entitlements rather than a simple dichotomy between dependency workers and others.

The second reason why conditionality may still be justified is that Kittay draws too sharp a distinction between policies that provide care givers with help and support and those which seek to coerce them and to regulate their lives. If, as she herself notes, the wider society has a moral responsibility to ‘assure that care can be and is provided’ (2000: 535), then the discharge of that responsibility may require more than the payment of a cash benefit. In the case of children, for example, it may go beyond making sure that the ‘child is neither neglected nor abused, nor denied provisions of a fundamental sort’ (1999: 144). This is not a question of requiring a parent to have particular feelings (as Clement suggests), but of requiring her to ensure that the child attends school or does not harass the neighbours. Moreover, her obligation to do this – and the claim of the community upon her – reflects the profound influence of family and of parenting practices upon the life chances of children (Piachaud, Reference Piachaud2001; Reference Piachaud2005). As a recent IPPR report noted: if public policy is to give children a more equal start in life, it ‘must be more readily prepared’ to enter into ‘the private world of the family, although this is a step that must be taken with great care’ (Harker and Kendall, Reference Harker and Kendall2003: 3). In the same way her obligation to do all that she can to ensure that her child does not engage in anti-social behaviour is grounded in the impact such behaviour has upon communities and especially upon poor communities.

Moreover, there is evidence that measures such as parenting orders or rehabilitation programmes for anti-social families can provide an effective response to people who do indeed fall below a ‘certain threshold’ in respect of their own behaviour or that of their children.Footnote ⁴ It may be true, as Selma Sevenhuijsen says, that policies should adopt ‘an attitude of initial trust in the moral capacities of individuals’ (Reference Sevenhuijsen, Carling, Duncan and Edwards2002: 133). They should also, however, acknowledge that this trust would sometimes prove misplaced, and assert the right of communities to insist that parents, like everyone else, fulfil their wider social responsibilities.