I. Just Bad Luck?

Mary’s experience is not unusual, and it is likely to become more common as more people live longer. We know that levels of emotional strain for caregivers can be very high. Apparently, men and women find different things stressful. Men find it stressful to have to handle difficult parental behavior; women react to the interference with their work and to the decline in the quality of their relationship with their parent. Although men make up about 25 percent of parental caregivers in the United States, it is clear that in the majority of families women carry the heaviest burdens. ^{Footnote 1}

The emotional, physical and material costs, and the general loss of value to Mary’s life, are regrettable. As a widespread phenomenon it is something about which we ought to try to think clearly, but we also need to ask exactly what sort of issue it is. In particular, is it something that raises questions of justice? And if it does, can standard principles of justice contribute to how we should respond to these issues? Let me begin by suggesting five ways of construing Mary’s situation:

1. 1. We could say that Mary has a lot of bad luck. There is a certain serendipity about whether people’s family lives go well or badly, but it cannot be anyone else’s responsibility to sort such things out, and the cost to Mary’s life is not something that should be conceived of as in any way a harm or injustice that requires some rectification or compensation.

2. 2. In contrast, luck egalitarians believe that, while people should bear the costs of what they choose to do (option luck), brute bad luck should be compensated for. Having a parent who is falling apart might be construed as brute bad luck: people’s lives go less well as a result, and that is something that should be responded to and compensated by society.

3. 3. We might see this case as one involving justice, but less as an ideal of an equal distribution of burdens and more as a threshold to cross — a degree of sufficientarianism. ^{Footnote 2} In an advanced industrialized society, the state should seek to avoid certain bad outcomes: People should not die of malnutrition, they should not be harming themselves while the balance of their mind is disturbed, and they should not be without basic protections and services to ensure their security and well-being. One such “threshold” that such a society might aim to cross is ensuring that the elderly can continue to function independently for as long as possible, and that there are services to support them in doing so, so that they are not left vulnerable, insecure, or burdensomely dependent on their families. The sufficiency threshold in relation to Mary might be one concerned to avoid (some level of) burdensome dependence on her by Helen. There are many issues I cannot deal with concerning the extent of the state’s direct responsibilities to Helen. My focus here is what we might think of as an independent responsibility to Mary in relation to the burden of care that she carries. How to identify that threshold and how to meet it raise further difficulties: even if the state has something like a sufficientarian responsibility to Mary, it might not be a responsibility that the state is best equipped to meet in practice, because of issues relating to individual autonomy and agency, family integrity, and issues of the accountability of administrative systems.

4. 4. In the case I have sketched, Mary is an only child, but if she has siblings there are issues about the responsibilities these family members may have to their parents, to each other, and to Mary in so far as she becomes the person who bears the major responsibility for their mother. There seem to be issues of fairness and justice here (at least intra-familial justice). These might not be issues that can be addressed by the state (perhaps because it inevitably lacks the detailed information on which to act); or they might not be issues that should be addressed by the state (perhaps because the state would need to be too intrusive; or because it should be a matter for voluntary decisions; or because it is a private matter over which the state should not claim jurisdiction). Even if it is not a matter for the state, we might hold that issues of justice in the distribution of benefits and burdens within the family exist.

5. 5. There are further issues about what level of responsibility it is reasonable to expect adult children to take for their parents, and what level of care they are expected to provide for them. There are also issues about what responsibilities the aging themselves have to prepare for their declining years and end of life decisions, and who should bear the cost if they fail to do this.

I use the term “realism” in my title and start with a concrete case rather than with principles of justice because I want to argue that the more sensitive our thinking is to the historically and contextually shaped realities of the problems faced by Mary and her family, the less plausible it is to think that ideal theories of justice can provide a determinate response. Starting with the concrete case certainly does not mean that no normative considerations are relevant, but my concern is to point to the difficulty of moving between the ideal and the concrete, and to argue that ideal theory will inevitably fall short in directing our response to the problems Mary faces, in large part because the ideal has multiple and contested dimensions. ^{Footnote 3}

II. The Burdens of Care

Each of the above reactions has some appeal and captures elements of intuitions that are widely shared, at least in Western Europe and North America. This suggests that several considerations and principles are in play and that this is a classic case of circumstances that are suboptimal in terms of justice but are marked by multiple competing dimensions of value. Following Sen’s account, this makes it difficult to see how to move from an ideal sense of what justice demands to the precise case because advances along one axis of value are likely to involve compromises or retreats on other axes. ^{Footnote 4} Multiple conflicting dimensions are one major obstacle to ideal theory delivering judgment- and action-guiding conclusions in concrete cases. ^{Footnote 5}

Mary’s case is intended to highlight a number of problems of justice linked to aging and the provision of care for the aged, and to identify features of these situations that resist idealization. One such feature is the depth of the emotional engagement and psychological burden of familial caring roles. Most people find it hard to deal with aging parents — hard even to have to come to see them as problems that must be “dealt with.” Becoming the person with whom the buck stops with respect to a parent necessarily changes the parent-child relationship. There is some role reversal, dependency reversal, and the child has to cope with aspects of their parent’s life that both sides may regard as extremely personal and intimate. They may have to experience their parent with a degree of physical intimacy that reverses their earlier relationship; they may have to take control of areas of their parent’s life to which they were not previously privy (such as financial and familial matters); and past issues of authority and power, and attachment and resentment are inevitably reopened as the relationship changes. Those who take up these burdens may find the demands of health and social care professionals both challenging in themselves (to attend appointments, provide support, and so on), and additionally stressful because they involve implicit expectations of the caregiver (and of his or her relationship with the parent) that they find hard to meet. Mary’s situation is not just about physical burden, financial loss, or even the loss of time (although each may be extensive), it also involves emotional conflict and loss, associated with having to experience one’s parent in disturbing and distressing ways, and with losing elements of that parent-child relationship and thereby elements of one’s given relationship to the world.

These burdens arise whenever a family member comes to be responsible for a parent in this way. They are not the only burdensome relationships we may face as siblings, spouses, or parents, but in these roles different reasons and emotions are likely to be involved. For example, in marrying and having children we accept certain responsibilities attaching to those decisions. But we do not choose our parents — especially as dependents. At least three significant dimensions to child-caring-for-parent relationships seem clear: the fact that they occur in a relationship which is in some senses unchosen, yet deeply emotional; that they involve the inversion and, on various dimensions, the disordering of an existing emotional relationship in ways that raise new psychological and practical challenges; and that these new demands in the relationship to one’s parent(s) are combined with the judgment by the child that these demands are ones that they have some responsibility to meet.

One response to Mary’s case would be to insist that I have mischaracterized the situation: we might better see Mary’s relationship with her mother in a narrative arc, from Mary’s birth to her mother’s death. What it is to be human is to have relationships that are deep, enduring and complex, that change over time, and in which the balance of “giving” evolves and often gradually reverses. To focus on Mary’s distress, loss, and “costs,” is to fail to see this as an essential part of the relationship and its quality.

This objection surely has some weight. We may retain concerns about the equity of the distribution of burdens and benefits (both between and within families), but this sort of wider narrative weakens the claim that Mary suffers brute bad luck. Indeed, the opportunity she has to share this final intense period of her mother’s life might be seen as an opportunity that others may lack.

The expectations of caring for elderly parents in certain cultures clearly help to structure people’s experiences of and their responses to these situations (although even positive framing does not necessarily mean that relationships are necessarily just, nonexploitative, or nonburdensome — burdens often fall heaviest on daughters and daughters-in-law). In some cultures it is relatively easy to make the physical adjustments necessary to include a parent in a more communal existence, and there may be strong normative support for doing so. There may be Western societies where this is true, but it is not commonly the case. Rather, in many places in Western Europe and in North America, evidence points to the costs that people bear, in terms of depression, anxiety, and a range of psychophysical conditions, linked to the responsibilities of caring for elderly parents. It is an especially acute problem in modern, cosmopolitan, individualist societies and cultures. It is so largely because of how our conception of the individual has evolved, the aspirations that people have for their lives and their work, the levels of geographical and social mobility, the kinds of relationships that they have with each other and with their parents, and the ways in which space is privatized and occupied. Each aspect complicates the “traditional” familial management of aging. ^{Footnote 6}

Moreover, we cannot simply insist that Mary’s case is mischaracterized because of the embeddedness of people in their relationships and the depth of their obligations to each other, because these components are often as much a part of the problem as they are a solution. It is because we are both modern individualists and endowed with psychologies and a humanity that tie us closely to others and generate senses of obligation and responsibility that we find ourselves assuming the burdens of parental care on grounds we cannot wholly rationalize, and with costs we find hard to anticipate or bear. Many of us want to do it but are ill-equipped to do so. Moreover, our modern individualist expectations are ones that those closest to us, such as our parents, often expect us to have. Given our values and expectations, we suffer when we don’t do it and when we do. After a certain point in parental aging, our lives go worse when we try to integrate their needs into our own; and for that very reason, our own parents often do not want us to do what we feel we ought to do. That generates additional burdens — not least because we rarely take what they say at face value!

Elements of this problem have probably existed for some time, but they may have been borne more easily because they were shorter-term problems in the past. The dramatic rise of longevity means that caring for someone may involve ten or twenty years of one’s life, in a way that relatively few experienced forty or fifty years ago. Moreover, increasing longevity in the West has been coupled with much greater social and geographical mobility, increasing incidence of family breakdown, single parent families, and families in which all the adults are in full time work. Each feature may compound the problem individuals face in respect of their parents.

On this account, while some may bear such burdens with equanimity, most people have trouble resolving the conflicting demands they face without stress and guilt. Their distress and sense of burden are symptomatic of their world: they suffer substantial costs because of the way it and they are constructed. Moreover, our ideas of justice and individual autonomy and self-realization are an integral part of that construction. The sense that to end a career is a cost, to restrict one’s geographical mobility is a constraint, that our elderly parents have a claim on us, and so on — in each case we read the situation as involving a responsibility that arises as a matter of brute bad luck, even if (indeed, because) we also feel moved to assume that burden.

Feminist theorists rightly point out that this is a burden borne most frequently by women. But in demanding equality and the right of individuals to choose and shape their own lives, they reinforce constructions of the lives we believe we should lead that ignore the depth of the emotional connections and senses of responsibility that exist between children and their failing parents. These disrupt our lives, leaving us unable to be what we think we ought to be: both the good child and the successful, independent agent. ^{Footnote 7} Feminism helps us see that such “domestic” problems cannot be understood wholly as matters of private concern and that these burdens are systematically unevenly shared, with women bearing the major load, often through structural disadvantage reproduced through the family. But even if these burdens were more evenly spread in gender and class terms, they would still be burdens, there would still be inequalities in their distribution and effects, and we should still feel that where people lack opportunities for their lives to go equally well, then we have a problem of injustice.

Not every adult feels this sense of responsibility toward their parents, and family members may feel differential senses of responsibility and obligation. This may be a function of differential emotional depth and responsiveness, or of differential vulnerability. But it is also the case that how we respond (and believe we should respond) is framed by the way these problems have evolved in Western societies and by the conflicting demands Western norms make on us. That this is so should contribute to our sense that these challenges are to some degree a collective problem.

Of course, not every parent-child relationship generates these emotional ties; and not everyone responds as we judge they ought. The matrix for responses and evaluation might be set out as follows:

That is, while we condemn those who are unresponsive to parents who have done much to warrant their care and concern and think of them as lacking an appropriate emotional response to their parents’ condition, we absolve those who were not so fortunate. Some people are lucky (perhaps somewhat praiseworthy) in being able to respond to their parents’ needs without a sense of burden, and others might be praiseworthy for caring despite their parents’ previous treatment of them (although we might wonder about their critical faculties and autonomy). But in many Western societies it seems that a very considerable proportion of the population will fall into the middle category — caught by the conflicting imperatives of the needs of self and other. ^{Footnote 8}

The burdens are not simply a function of resource problems. We cannot assume that, if only there were enough money, these problems would go away. Clearly, money can help in various ways. Nonetheless, the conflicting principles, emotional engagements, and personal aspirations that we have ensure that these situations cannot be resolved by money. Moreover, many practical (if partial) solutions are not close to hand: our societies have not prepared for this; in very few areas do we have the right types of housing mix and service provision that can support graduated independent living for the aged (for example, reducing their dependence on cars for shopping, providing easy access to apartments above the ground floor, enabling them to live in some proximity to their wider families, and ensuring responsiveness to changing levels of mobility and capacity for self-care, and so on), or neighborhood settings in which high levels of trust and security can enable wider community support. ^{Footnote 9} Similarly, more flexible arrangements at work for those who have aging parents might reduce some of the strain that caregivers experience. Some of these things could be established, but in very few places do they currently exist. Things could be done to alleviate the stresses of care in the long run, but our current world is not one in which they will be eliminated, in part because we want other things as well, and our wanting those things is as deep-seated a feature of the modern psyche as the desire to care for one’s parents.

Even if the long term could be made better, it cannot, on this account, be made burden free. Part of the stress of caring involves the experience of an inverted relationship, the breakdown of boundaries of privacy, and the interruption of the life of the adult child by unpredictable demands on their time, resources, and energies (emotional and otherwise), and a basic sense of loss. Further issues arise from the need to make judgments about the degree to which, and the conditions under which, the parent can cope and their judgment can be permitted to stand. That involves the assumption of a default responsibility for a person who is now imperfectly autonomous and who may disagree deeply about how far their capacity has diminished. One common issue is the decision about whether or not it is safe for them to continue to drive, and who should make that decision. This is just one of a wide range of decision points: we should respect Helen’s wishes — but the strain and emotional cost in these situations is often a function of iterated tussles over the exact degree of her capacity in relation to what she wants to do. How much short-term memory loss is compatible with independent living? How much depression is compatible with leaving people to themselves? The burden for Mary and for any family member who has primary responsibility for a parent is partly generated by having to negotiate these boundary issues. Moreover, initiatives that are future orientated (that set things up so that the situation will be more manageable when it deteriorates in five years, for example), rely on the elderly person being able to share the assessment that they need to be implemented now. Having extremely well-appointed and imaginatively run sheltered housing and care and nursing homes is very beneficial, but only if people want to go to them. People rarely want to go to them when they do not need them, and when they need them there may be quite long periods in which they resist recognizing (or cannot recognize) that they do. In these situations, the role of the family is to negotiate the territory around what level of choice and decision-making the person can be permitted.

Even when a parent enters residential care, the burdens do not disappear. There are questions of who visits, when, and how often, who listens to complaints and assesses their validity, who negotiates such complaints with the staff, who is to be consulted by the medical staff, who carries power of attorney and how is it exercised, and who is to be involved in end of life decisions, and so on. Such decisions cannot be wholly surrendered to professionals without emotional cost and a sense of failing in one’s duty; and it is not a strategy that professionals should (or generally do) encourage, for good reason (in terms of their own accountability, and in terms of the needs of the elderly person — including the need for an advocate). As such, a perfect, nonburdensome, situation cannot be reliably ensured for most people, not least because our sense of filial duty encompasses a responsibility to respond to our parent’s concerns and to assess the extent to which those concerns are ones that can or ought to be responded to by others — thereby exercising a supervisory sovereignty over the parent that in itself involves some indefeasible practical and emotional costs.

Additional issues are raised when dementia affects the person’s personality and memory, with the children being caught in a situation of being cut off from their parent and yet still in some respects responsible for ensuring their care. Should we treat the child who visits a severely demented parent as indulging in some kind of self-indulgent behavior, since their care cannot in any respect be reciprocated or recognized? Or should we treat this as merely the end of a continuum of the burdens of care — that the child remains responsible for ensuring that the best that can be done is being done, even though there can be no parental responsiveness, because there is no one else who can provide the level of assurance that we believe is required? Or, perhaps the child visits as a way of coming to terms with the changed reality? In both cases, visiting likely involves a burden, but should we file this away wholly as a case of option luck?

People’s senses of burden in relation to the aging of their parents are best understood as deeply “situated” — rooted in the particulars of their practical context and their often deep emotional and personal relationships, and framed by sets of societal norms, moral principles, and emotional ties that make demands that are in competition with each other. Because some of these conflicting demands occur at a deep emotional level (or trigger deeper emotional responses), there is a limit to how far principles that make idealizing assumptions about optimal solutions can be brought to bear on such issues. Part of the point of setting out Mary’s experience is that we have to appreciate its particularity even if we also acknowledge that, at a more abstract level, it captures an increasingly common set of experiences. The problems that she experiences are deeply embedded in her day-to-day life and in the history of her relationship with her parents. These profoundly shape the way she reasons and judges herself and others. Just as Elizabeth Anderson has warned against reading people’s experience for them in terms of concepts such as brute bad luck, so should we resist the idea that these are part of the evaluative language of those caring for their parents. ^{Footnote 10} In devising responses to Mary’s situation, then, we need to work in large part from where she and Helen sit and from the way they differently construct their situation. That their problem is structurally framed by changes in longevity and in societal norms and expectations in the twenty-first century does not mean that it is identical to the problems others experience. Similarly, while we can identify dimensions on which problems and issues might develop or become especially burdensome, we need to know the particulars of Mary’s past and present situation and the way in which such possibilities are in fact playing out for her and her mother in order to assess the burden she bears and how best she can be supported.

This might be a more general phenomenon — a splitting between rationality and abstract principle on the one hand, and the reality of deeply lived experience on the other. Montaigne, for example, felt that tension over and over again: “ . . . no matter how great a man’s wisdom, he can never grasp, through his judgment alone, the cause of another’s grief in all its intensity; his understanding is always enhanced by actual presence, when the eyes and ears have a share in it, organs that cannot be stirred except by incidental details.” ^{Footnote 11} I am unclear how far to press this point as a general claim. Principles of justice and fairness rely on the possibility that they can be recognized by all. But it seems wrong to see our quotidian, emotional, and embedded lives as a distracting detail in any account of what such principles enjoin. It is from that detail that we construct our lives and the principles we strive to realize have to be constructs rooted in that detail if they are to have the necessary validity and motivating force. Because of this, particular combinations of principles will inevitably fall short of wide generalizability, unless rooted in relevantly similar conditions of generation. Furthermore, because the values, principles, emotions, and commitments that have a grip on the minds of many people in the West who face these kinds of situation tend to make conflicting demands, people’s agency lies, not in following a particular principle, but in negotiating their way through the irreducible conflicts generated by the responsibilities that they acknowledge. As members of these societies, we should be able to recognize these conflicting demands, but, insofar as we really acknowledge the depth of their grip — their integral connection to people’s sense of what it is for them to be the person they are — then we ought also to acknowledge that there can be no generalizable way of resolving those conflicts.

III. Intra-Familial Justice

What if Mary has siblings? In many families the distribution of responsibilities is often deeply inequitable. This may be a function of family members’ differential emotional vulnerability or susceptibility, their bargaining positions, or a degree of brute bad luck. The powerful emotional consequences of being the person with whom the buck stops also affects one’s relationships with one’s siblings. Consider the case in which Mary is the primary caregiver because she is the sibling closest geographically, although she also has a brother and a sister. Their existence might be a component of the costs that Mary bears. When they visit occasionally, to the evident pleasure of Mary’s mother, and tell Mary what she ought to be doing to ensure a better quality to their mother’s life, Mary feels humiliated and resentful. Their comments underline her sense that they enjoy a relationship with their mother that the daily round of physical care has compromised for Mary; and they highlight that Mary’s burdens have rendered her emotionally isolated from them — because they cannot see the world from her point of view. If Mary is a single child these problems do not arise. But if there are siblings, things become more complex.

We need not assume any great fault on these siblings’ parts: perhaps one lives in the mid-West, the other in California, and Mary is on the East coast. Helen wanted to live near one of her children and she chose Mary. Does it matter why? It might have been because Mary lived in a town with which her mother was familiar, and neither of the other children did; or because Helen had most confidence in her; because she had always had a good relationship with her; or because she was seen as having fewer important things to do than her siblings. Some of these might be good reasons; at least one is certainly bad. But, in many respects, Helen’s decision is merely a compounding factor to the experience: the more questionable the motives, the more complex the emotional impact on Mary is likely to be. But even simple motives, like proximity to where she was previously living, produce many of the same outcomes and effects. Moreover, whatever the reason Helen chose Mary, our story assumes that Mary has ended up living with that choice and has “taken on” the responsibilities in relation to her mother. Indeed, Mary may have sought the responsibility and encouraged her mother to be near her. There are, again, many possible reasons for doing so — continuing a particular familial role, wanting to claim an exclusive relationship with one’s mother, not trusting one’s siblings, sheer convenience, a familial role of “the responsible one,” and so on. The idea that Mary “accepts” the responsibility requires care: although there are cases in which people actively choose such roles, they tend to do so under dramatically incomplete information. Moreover, in many cases, becoming the primary caregiver dawns on people only slowly, and at different rates between parents, children, and public authorities, doctors, and so forth. The nature of the exact demands is likewise understood gradually. Only as Helen’s capacity begins to deteriorate, is Mary’s commitment made concrete and its exact demands on Mary’s character and on her relationship with her mother made manifest. The impact of aging and consequent personality changes can dramatically influence what the initial “bargain” turns out to mean in practice, and people may find that they have backed into a situation (or been backed into a situation by their parent(s)’ decisions or by those of others) that they cannot now change, thereby adding to a sense of burden, helplessness, and unfairness, albeit linked to a desire to care for their aging parent(s).

The nature of the emotional burden that Mary bears is complex: she has to manage a change in role that her siblings do not; and she is likely to find it increasingly difficult to communicate her frustrations and difficulties with them because they lack awareness of the situation she faces on a daily basis. They may see her as unresponsive to suggestions that, while made in good faith, carry the implication that she is responsible for things that she feels unable to deliver on. Moreover, the burden she bears does not appear in a vacuum. It arises in a set of family relationships that already have certain dynamics and tensions that can exacerbate her sense of isolation, or of being “dumped on,” or of her needs, family, or career being treated as of intrinsically less worth than that of her siblings.

What do Mary’s siblings owe her in this situation? There is an intrinsic problem in ensuring an equal, consensual distribution of burden, even if everyone is extremely well motivated. One common case is where the dynamics of the situation create unfairness without intent, and in ways that are difficult for those involved to see — indeed, Mary’s own view will inevitably be partial. From her siblings’ points of view, if they are not conveniently close, it makes more sense for Mary to call more regularly. For example, let us assume that Mary’s brother Frank lives further away, so that his journey takes three times as long as Mary’s. In terms of the distribution of time and effort, if Mary is going to her mother three times a week, then Frank should go once (and stay longer) — since it will take as much time for him to do that as it takes Mary to do her three trips. But this is a classic case (common in childcare), where the cost in terms of time is not the only measure, and where time equivalents may produce uneven burdens. The person who is there most frequently often necessarily ends up taking responsibility for a substantial proportion of the quotidian round. Even if Mary actually spends no more time with her mother than Frank does, because she has most frequent contact the range of tasks that fall to her will dramatically outweigh those that fall to Frank. This does not presume any ill will on Frank’s part. Indeed, the more sensitive he is to not interfering with the arrangements that Mary makes in the week, the more he re-enforces her responsibility for those arrangements; the more he ignores them, the more he adds to Mary’s labor.

The situation is one in which the agency of A must become the responsibility of some B. Splitting responsibility between plural Bs threatens incoherence, where the costs of that are deemed by all involved to be unacceptably high. Only in relatively ideal and transparent situations could Mary and Frank share ultimate responsibility. Even if they are in constant communication they would need some way of deciding cases where they disagree, and what is likely to emerge is a primary holder of responsibility — or someone who has to assume sovereignty in relation to Helen. Even dividing tasks requires some locus of decision to determine whether an ambiguous task is in category M or category F.

I want here to emphasize the difficulty, not to claim the impossibility, of any equitable, mutually acceptable system — just as real equality in child care burdens is difficult (but perhaps not wholly impossible) to achieve. Mutual acceptability is important because it is not difficult to imagine cases in which people share the burden but add to it with the tension and discord generated by conflict over who is doing what and when. These situations are genuinely tricky and thus more burdensome. Of course, Frank can experience the same loss of the parent-child relationship if he is actively involved, and that may be the basis for giving Mary a sense that her experience is understood and shared. But it is also not difficult to see that any sort of “grit” in Mary and Frank’s relationship is likely to surface in the constant friction associated with someone needing to take charge of the situation, and with the inevitably different judgments made about how to handle things. Things are more complex in the three siblings case; and we have said nothing about the capacity of one’s parent(s) unintentionally to fan flames — telling Frank to leave it to Mary; praising Frank’s selflessness and devotion but not Mary’s (since that generation of mother works on the assumption that daughters do this sort of stuff, while sons have other responsibilities!); and so on. ^{Footnote 12}

In many cases, those less involved tend to underestimate the problems faced by the primary caregiver, suggesting off-the-peg solutions for them. If Helen feels isolated, then it makes sense that she should be taken out more, encouraged to join local groups, and so on. But these “solutions” are to a problem conceived independently from the practical and emotional context in which it arises. To be taken out more, means for Mary to badger and cajole her, exacerbating her sense that she is taking on a role that underscores her mother’s dependence on her and her own corresponding sense of responsibility. Such proposals are not wholly wrong, but they can dramatically exacerbate Mary’s sense of failure and resentment. Above all, they fail to start from where Mary is, and fail to engage with her perspective, and also fail to engage with how Helen sees the situation.

IV. On Restitution and Compensation

Could we make things better for Mary? There are many dimensions to this question and I will sketch only some that might not be quite so obvious. One concern is that, as the population ages and the problem grows the state might simply have to reject the view that the wider family costs and ramifications of parental care can be counted as within its remit and say that this is something that families will have largely to manage for themselves. Only when things break down could it step in to ensure that certain minimum levels of physical care and security are met. Rather than aiming for full compensation or restitution, a society may accept collective responsibility for only a narrow component of provision that, overall, might include:

1. i. help with some of the direct costs borne by Mary in caring for her mother;

2. ii. assistance in meeting the practical burdens of care — for example, either directly by providing additional home support for Helen, or through relieving Mary by providing holiday breaks or offering respite care for Helen. And in the family context, having siblings who will step up when she is able to go away and to take over the burdens similarly helps Mary’s life go better. Each may also reduce the sense of isolation, and thereby reduce the alienation from other family members;

3. iii. recognition of Mary’s contribution and burden. In the familial case, but also where Mary is an only child, a crucial component of the benefit that practical support provides lies in its recognition of the problems that Mary faces. A major cost for many people is their sense of guilt at not being able to do more for their parent, and for experiencing their parent as a burden. By acknowledging the challenges they are facing, and giving permission for Mary to feel this as a weight that she can share, this provision may in itself contribute to Mary’s well-being. That said, it is something that needs sensitive handling — Mary may not find it easy to articulate her needs and feelings in order to take advantage of such support. This “recognition” dimension to the way in which we respond to Mary is important: even if we cannot solve her problem, those with whom she interacts, either in services or in the family, can acknowledge the challenges she faces and can help her to identify and manage those dimensions that she finds most enervating and difficult, enabling her to express her perspective and her often conflicted feelings about her situation and about what she is required to do, and to have these respected by others. But this presumes that the capacity for empathy among family members is not derailed by existing hierarchies and the emotional baggage in their relationships.

These suggestions do not pretend fully to compensate Mary’s burdens. But that may itself be important: one dimension of the problem Mary faces is her own concern to be a part of a relationship that is not conceived of wholly in terms of cost-benefit, equivalence, or restitution. That means that any solution must recognize that people often need acknowledgement of, and help with, but not the elimination of their burdens, and that forms of practical support can also often have a recognition dimension.

These issues are often not handled well; but they might be managed better. Moreover, while some psychological insight might help, this may be a situation in which elements of realist political theory might be helpful as a guide, especially where (in the area of theories of recognition and cultural difference) it theorizes bringing together people with various amounts of conflicting emotional commitments and resentments, where the aim is to construct something like a modus vivendi over terms of engagement, to allow care to be managed without differences breaking out into war, and without replaying familial patterns of dominance, subordination, and exploitation. In this sense, the polity offers a metaphor that treats the family (and encourages the family to treat itself) as a unit in which there is equal standing between agents marked by conflicting interests and various dimensions of difference, and couples this with recognition for those who are most damaged by its structures of inequality and power. Even here, we are concerned with a realist “political” process — something that will be good enough for those involved, and that allows them to remain on terms with each other. Something, that is, which sustains civility in an arena where conflicts are deep, multidimensional and often partly zero-sum. ^{Footnote 13}

The aim of a realist-type process in the family context is not the eradication of or full compensation for burdens, or the rectification of past inequalities and their concomitant costs. It is more a negotiated compromise than an emancipatory or transformative one, and it is partly a matter of recognition. It addresses some distributional issues in a context in which a lot more is going on, but where the aim is to resolve a limited set of issues, and to do so, in part, by acknowledging the participants’ distinctive positions and experiences. This is a realist familial politics for a fractured world — and for the fractured and conflicted individuals that such a world produces. It makes no pretense to being fully reasonable, or ideal. There is a great deal it cannot resolve but it aims to provide a means by which family members can work together on a limited range of common objectives. That is an essentially “political realist” task: identifying workable compromises between people at odds with each other. This does not mean that values and ideals have no place: these proposals track conceptions of proximately equal well-being, autonomy, respect for persons, sufficiency in terms of basic needs and conditions for security, and so on. But greater clarity on the implications of such principles is a function of looking at how we make them concrete for these people in this particular context. The point of starting with Mary and Helen just is that, in this case, we are forced to face concrete issues that more abstract constructions tend to obscure, and we come to see the extent to which our own conceptions of justice and human flourishing are themselves deeply embedded in the construction of the problem (valuing autonomy, expecting reciprocity, and so forth), and cannot stand as a wholly external solution to them.

Is it helpful to use the language of justice to describe these situations? Modern conceptions of justice start from a fungibility model of justice in which the elements that compose an account of justice can be ordered within the same conceptual space and often along a single dimension (although that may itself have plural strands — like an account of primary goods). We are encouraged to conceive of justice as distribution on particular metrics — resources, welfare, capabilities, access to advantage, and so on — and a key presumption, perhaps modeled on the scales of justice, is that an injustice can be corrected. In contrast, one thought that drives this paper is that the multidimensional interconnected character of some areas of life are such that it is very difficult to see what would count as equality between those involved. This is one reason why compensation and rectification may not be appropriate goals; recognition and support might be the best that we can do.

V. Conclusion

There are a whole range of other problems relating to aging and families that I cannot deal with here. In sketching the broader field we might note the issue of exactly what responsibilities the aging themselves have, especially given the uncertainties of a life’s final phases; there are questions about how far families can help the aged take end of life decisions when their faculties have already begun to deteriorate — a matter that is distinct from that faced by people facing terminal illness that will at some point disable their functioning; ^{Footnote 14} there are questions of how families manage the distribution of parental goods when siblings have different and often competitive emotional attachments to things, as well as issues about fair inheritance after markedly different degrees of involvement in end of life care. That is not an exhaustive list — it merely indicates issues that often fracture wider family relationships in the last years of parents’ lives or in the wake of their deaths. None is easy to solve, because they share many of the features of the case on which I have focused: multiple dimensions of value that conflict; difficulties in separating the emotional and the rational in people’s attachments and responses; and the repeated disruption by the familial past of compromises and arrangements that express a shared sense of fairness. Addressing these issues also becomes a matter of reaching for rules of thumb and rough principles from the exigent details of each particular case, and negotiating solutions that imperfectly realize only some of a range of possible values.

That sense of imperfect balancing of competing considerations is evident if we return to the four types of response I indicated at the beginning of the essay. Each account seems to have some weight:

1. 1. Mary has had bad luck in ending up caring for her mother, and her life has gone worse than it otherwise would have. But that (Mary’s situation) is not something to wholly compensate Mary for, in part because it is not something we can fully compensate her for (nor is it something Mary sees in such terms). It may be the case that Helen has failed to plan and to act responsibly and that Mary is the person with whom the buck now stops. For the state to get involved risks moral hazard, reinforcing poor planning and irresponsibility; it may simply be too expensive or informationally infeasible for the state to pick up all such cases. That reinforces just how bad Mary’s luck is, but we should resist the thought that the sins of the parents should be borne by their children.

2. 2. Mary might be seen as exercising option luck — she wants to look after her mother and the costs of doing so are freely chosen — but even where she embraces the role, not everything that follows from her doing so should count as option rather than brute luck, because of the irreducible uncertainty as to exactly what she is taking on. If Mary gives up her job to take care of her mother, we would need to be clear whether feasible alternatives existed before concluding that this was a choice being exercised, rather than her believing that she had no alternative but to assume the responsibility! Tax breaks, provision of caregiver benefits, the creation of services to support the elderly and to provide respite for those who assume such burdens are ways in which the state can respond to the burden carried by Mary in providing care, which recognize that the caregiver is bearing a burden that is (in some degree) un-chosen, costly, and one that the state might otherwise have some (perhaps sufficientarian) responsibility to meet. There clearly are components of option luck: Mary wants a good relationship with her mother, to be part of the process of care, and to be able to provide comfort and support, but the character of Helen’s decline makes it difficult to do these things without incurring some degree of “brute-luck burden,” and we have to reflect more critically about the idea of “option-luck” when the choice is between a rock and a hard place.

3. 3. What level and kinds of provision should the state make for those who are rendered vulnerable in virtue of age and sickness? We might say that it is not tolerable in a civilized society for the elderly to be exposed to risk, insecurity, exploitation, or a deeply suboptimal way of life, and we might endorse a wider societal responsibility for Helen. In Europe, we tend to start with the State; in the United States, people tend to start with insurance and the individual’s responsibility to plan for their future. While there are various options, it is clear that, at some level, many see it as a collective responsibility to ensure that it is reliably done in ways that protect people’s interests. If that is right, it seems plausible to claim that there are issues of justice for the state in Mary’s case, even if it is only a case of it acting to enable her to undertake the role in ways that are consistent with the provision that the state makes for all who are most vulnerable.

4. 4. Finally, there is an issue about what Mary’s family should be doing for her. Even if they cannot fully compensate her, they can ameliorate her situation, and there is some sense that, at least within the family, something like “recognition” is called for so as to acknowledge the burdens she bears, and relieve them in whatever ways they can.

There are many different considerations that relate to Mary’s case, and these partly add up to the conclusion that on most standard accounts of justice there will inevitably be some uncompensated burden that Mary bears. In many theories of justice it is problematic to say that there remains some injustice. Modern theories of justice tend to “see like a state” (or from a similarly single moral perspective), insist on a public/private divide, taking justice to be a matter for the public domain and separating it off from the private world of emotional costs. Hence the rather fractured response to Mary’s case, where the public and the private are not easily separated, and in which conflicting values exist with no obvious way of trading-off between them. Mary’s case is messy. As such, the case counts as evidence against the thought that we need to get the ideal principles straight to address concrete issues of injustice and the maldistribution of burdens in society. If “realism” is to offer an alternative, however, it must grapple both with the general historical context and its features — which Bernard Williams emphasizes — and with the details of the range of cases that emerge from that context, where the reality is that any solution inevitably leaves some element of burden. ^{Footnote 15}

Nothing I have said here denies that there are values or principles or that some states of affairs are better than others. Indeed, I rely on such elements to make judgments that respect the individual’s perspective, that treats their emotional distress at their parent’s decline as something indicative of their humanity that should be responded to (in a way, for example, that someone’s visceral hostility to migrants should not), and that accords Helen’s perspective as much weight as we can (although I have been able to say little about this). Where I depart from ideal theory is in denying that there can be some way of reconciling all the different dimensions of value in a single metric. Rather, we face a world in which we constantly have to weigh competing values, guestimate probabilities, commit to some preferences over others, compromise with others’ competing views, and resign ourselves to the fact that the best we can do is not the best we can imagine but very much a flawed and inadequate runner-up about those involved will have differing views. Of course, if we strip out the details and complexities of cases we can find principles, but we will lose the connection with what moves us in our world.

Mary’s situation is regrettable. It can be partly alleviated, but not wholly; and what should be done will depend in part on the individual preferences and opportunities of those engaged in the process. Mary may rightly have a sense of bad luck, or unfairness, or injustice. Sometimes things might be done that will diminish that — perhaps to the point that she will not have such thoughts. But that may not mean that brute bad luck has been eliminated; only that she has found a way of living with it. Relatedly, people live with injustice and sometimes it no longer bothers them. Sometimes that’s a sign of resignation; sometimes an acknowledgement that they have done as much as they could and that is good enough, even if it is not wholly fair or equitable. On this view, as societies, we have to think, not about ideal justice, but about how we can help people live together in ways in which deep injustice is limited, in which fate does not cut too deep into the arc of individual lives, and in which there is enough agency and opportunity available to people for them to feel that, even if their fate is relatively poor, they have nonetheless been in a position to make something of it, and have thereby made it their own.