1. Introduction
In their critics’ essays, Clark Wolf, Elizabeth Edenberg, and Helga Varden probe the foundations and applications of my theory of liberal dependency care (henceforth LDC) (Bhandary, Reference Bhandary2020).Footnote 1 From within the shared parameters of the liberal and social contract theory project, they urge me to develop LDC in new directions — as a form of subject-centered justice (Wolf), with a political liberal justification (Edenberg), and by embracing Kantian foundations for an account of ‘private right’ (Varden).Footnote 2 Lavender McKittrick-Sweitzer, too, in their editor's introduction, identifies LDC's continuities with, and departures from, other liberal, feminist, and social contract theory accounts.
More broadly, my critics raise three kinds of issues: (1) foundational theoretical principles of liberal social contract theory; (2) the potential exclusion of particular groups; and (3) the link between justice and state coercion. My response addresses these clusters of issues by emphasizing that LDC is an account of distributive justice, and, as such, it is also an anti-oppression account of justice, one that offers an assessment of the basic structure with remedies for that basic structure. Consequently, the sense of ‘justice’ my theory advances is not an account of state-sanctioned uses of force. LDC also does not offer an account of justice as a virtue of individuals, nor does it explain how we might achieve justice for victims who have been wronged, where this sense of ‘justice’ includes retributive justice or assigns law its proper role. Moreover, I show that my views as presented in Freedom to Care require the inclusion of the social groups of concern identified by Wolf and Varden.
Overall, I attempt to make some progress on these foundational matters in the liberal and social contract theory tradition, emphasizing LDC's commitments to diversity as well as the imperative of laying bare the terms of normative disagreement among varieties of liberalism. In doing so, I aspire to honour my critics’ robust and thoughtful engagements with my theory.
2. Reply to Wolf
Wolf aptly characterizes the guiding concern of my project as liberating invisible caregivers — and the social groups that are clustered within this category, such as women of colour — from having our substance grafted onto others as aspects of an unjust caregiving arrangement. In this context, he writes:
As Bhandary rightly points out, these claims to care are claims to be cared for by others who also have rights and needs, and who are also self-authenticating sources of claims. Like other rights, the right to dependent care must take its place among the competing claims of others. In circumstances of scarcity, the need for dependent aid may exceed the supply. In such circumstances, there is an obligation to ensure that needs will not be unmet. But this cannot be an excuse to violate the rights or deny the claims of providers. (Wolf, Section 3)
2.1 Cognitive Ability Does Not Correlate With Social Contributions
Despite his appreciation for LDC's theoretical protection for the rights and claims of care providers, Wolf urges me to embrace what he calls “subject-centered liberal justice” (Wolf, passim). This modification, however, would not preserve the merits of LDC that Wolf so clearly identifies. Instead, I show below that LDC's approach to social cooperation is more inclusive than he claims because people whose care is costly are included in my account. Nowhere in Freedom to Care do I exclude people who might give rise to costly needs for care; nor is their exclusion implied by the infrastructure of the account.
Wolf argues that John Rawls, Cynthia Stark, Samuel Freeman, and I all suffer from a commitment to David Hume's framing of justice as a mutually beneficial arrangement whereby each person must do her part, or the whole apparatus falls apart. Comparing distributive justice to the social virtue of justice, he invokes Hume's metaphor of an arch. Wolf writes:
David Hume compares the ‘artificial virtue’ of justice to a vaulted arch in which “each individual stone would, of itself, fall to the ground” if not held in its place by all the others. Hume continues, “nor is the whole fabric supported but by the mutual assistance and combination of all its corresponding parts” (Hume, 1983b, p. 305). Like stones in a vault, each just action takes its place to mutually support and be supported by the actions of others. In a simple arch, the whole edifice will fall if one stone (or one person) fails to do its (or her) part. Everyone must contribute, or the entire system will fall to ruin. Because he adopts this conception of justice, Hume must exclude non-contributors from the mix. Those too weak to bear their portion of our cooperative burden simply cannot be part of the arch created by the coordination of our just actions. (Wolf, Section 1, emphasis added)
However, whereas Hume might be read as linking rationality to the idea of social contributions, I do not. Wolf credits the idea that everyone must contribute to a scheme of social cooperation with the exclusion of people who need dependency care throughout their entire lives. But the claims of people who need care throughout their lives are not excluded from justice in LDC.
Because the account of fairness that I offer is about social arrangements, nothing follows from my work about how we should assess the life of any particular individual. It is not an account of exchange reciprocity.Footnote 3 Understanding justice as being fundamentally about the fairness of the system of practices, or of social cooperation, does not mean that people who require more care than they can contribute are excluded. It also does not mean that everyone benefits from each interaction. Wolf asserts, “On the standard interpretation of Hume's view, parts of which Bhandary endorses, individuals who lack underlying capacities that make mutually beneficial reciprocal relationships possible simply do not fall within the scope of justice” (Wolf, Section 2). However, Wolf moves too quickly here when he attributes this aspect of a Humean picture to me.Footnote 4 What I embrace from Hume is the idea of the circumstances of justice, which includes the idea that a background framing of moderate scarcity makes justice both necessary and possible. As a form of constructivism, LDC requires rationality in the modelling of agents in the original position, who subsequently arrive at principles to which everyone can agree. It does not, however, equate the model conception of persons in the original position with the set of people who count as those to whom justice is owed (Bhandary, Reference Bhandary2020, p. 40). Instead, the role of rationality is to ascertain what would be fair by thinking about terms everyone would accept.
The four principles of just caregivingFootnote 5 that I defend in LDC are fully compatible with meeting the needs of humans with disabilities, and these needs are included in my account of justice. The aspect of Rawls's thought that I embrace is the original position, through which we understand hypothetical acceptability. The original position includes the idea of rationally arriving at the principles of justice that one would accept from any social position. The use of rationality in that context does not describe the people to whom the principles apply. In no way do I think rationality is a precondition for social cooperation, nor is it a guarantee for social cooperation.
2.2 Toward a Realistic Historical Assessment of Social Harms and Contributions
A view that actually measured any one individual's contributions would also have to measure the harms that person causes society. Such a view would quantify social contribution in a way that assesses the person's contributions and the damage he causes. This is not my view, but thinking in this way makes it abundantly clear that rationality is neither necessary nor sufficient for a person to be a contributor to the benefits and burdens of society.
Consider white males, a social group by whom the dominant liberal tradition was authored. This social group has benefitted from the ongoing appropriation of the labour, bodies, and capital of others. Consequently, the paradigm person upon whom the idea of the ‘contributor’ has been founded is shaped by the assumptions and experiences characteristic of this social group. And because their way of life is premised on unjust distributions and relations of appropriation, those facts were invisible, according to the terms of social cooperation.Footnote 6
However, there is today mounting global awareness that the members of this small and powerful subgroup have not, in fact, been contributing what they claim. Therefore, political theories of distributive justice should not abandon a role for social contributions in an account of fairness — certainly not at the precise moment when we are making visible how little these men, who were previously paradigm social contributors, actually contribute. Instead, political theorists of distributive justice should assert that social contributions come in many forms. By including care, it becomes undeniable how much people of colour, women, and specifically women of colour contribute relative to white men. My account of social cooperation in LDC calls for a clear-eyed assessment of the ways individuals contribute, rather than the loss of the idea of contributions altogether, where contributions include and extend beyond material/productive, intellectual, and reproductive labour.
The theory of LDC provides a framework with which to identify distributive injustice, and it is one that can be used to identify, for instance, the caregiving labour slave families, servants, and colonial subjects provided to European whites. A theory that only evaluates what we all deserve or need, without also identifying where the labour comes from, loses the theoretical power to make comparative assessments and to identify the exploitation and differential burden of labour performed by people of colour. And maintaining scepticism about human nature in LDC protects people from being exploited within the terms of the theory itself; this is what the role of prudential self-interest, as represented by Rawlsian mutual disinterest in the original position, provides. Rationality's role in the original position is to evaluate terms for social cooperation when people do not know who they are in the resultant schema; this way of presenting peoples’ views protects us from inaccurate empathy and mistaken understandings of what other people need.Footnote 7
In the development of our systems of social cooperation, we must meet the needs of people with severe disabilities; this outcome is required under LDC's criteria to meet society's needs for care. In doing so, however, we must remain attuned to the complexities of speaking for others.Footnote 8 When technology develops to the point where people who are currently non-verbal can convey their values, the intimates of people who could not speak should be prepared to learn that their formerly disabled intimates valued otherwise than they believed. Thus, we must remain attuned to the fact that describing the needs of people with severe cognitive disabilities, and seeking to meet them in the absence of communication, will be an imperfect exercise.Footnote 9
Because the social contract tradition has paid too little attention to the tenuous links between rationality, social contributions, and social harm, let me now elaborate on the third consideration, social harm. Who creates the greatest needs for care? When considering Kenneth Arrow's bottomless pit, who is the target for the analysis? Wolf writes, “Bhandary, Arrow, and Stark are right to be concerned that people with expensive permanent needs might impose excessive social costs, including the burden of care imposed on those who provide it” (Wolf, Section 4).
However, excessive costs do not play a role in my own view. If I were to evaluate costs, I would include the sources of needs for care, and once we do so, masculinity will be responsible for a large bill, for norms of masculinity are the cause of many needs for care. For instance, males have higher rates of risk-taking, which lead to, among other things, higher rates of traumatic brain injury (TBI) in males than in females (Tamás et al., Reference Tamás, Kocsor, Gyuris, Kovács, Czeiter and Büki2019).Footnote 10 Because a person who suffers from severe TBI requires caregiving by others, when costs of care are included in a system of social cooperation's “account books” (Baier, Reference Baier1994, p. 8), we must also include the extractive and violent tolls created by masculine violence.Footnote 11 Therefore, if I were to evaluate conditions merely in terms of the crude analysis of financial cost, it would be clear that a person with disabilities does not cost society more than the abusive and able male who harms multiple people. But financial cost is not even a metric I articulate in my theory. My view, instead, tracks the distribution of benefits and burdens in our systems of social cooperation. As theorists of justice and as real-world participants in our societies, we will decide, for instance, that the ‘cost’ of taxpayer dollars supporting a person to work as an aid to a child with cerebral palsy so that he can participate in a junior high band is important and valuable. The fact that a theory of justice includes a Rawlsian concept of reciprocity does not bias against this outcome.
2.3 On the Relationship Between the Justice of Institutions and Just Persons
The essence of Wolf's alternative proposal is to include everyone as a self-authenticating claimant, and therefore as a subject of justice. If the concept ‘subject of justice’ identifies the set of people who are owed legal obligations, Wolf's proposal is attractive. And this is perhaps the dominant usage of the word ‘justice’ — one that follows Ronald Dworkin's formulation of liberalism as the minimal use of the law. My account of justice as distributive justice follows the Rawlsian usage, however. Whereas Dworkin's liberalism is about the limits of the law, Rawls's account is fundamentally an assessment of the distributive arrangement, or, even more foundationally, the fairness of the “system of practices” (Rawls, Reference Rawls1958, p. 169; see also Bhandary, Reference Bhandary, Bhandary and Baehr2021, p. 147). Furthermore, the way people are modelled for Rawls is not equivalent to Hume's discussion of them. And my view is not reducible to Rawls's view. What I adopt and revise from Rawls is the idea of hypothetical acceptability as a way of coming to identify principles to constrain social arrangements so that the society is acceptable from any social position. This is the idea of fairness that Rawls elucidates with his idea of the original position. Therefore, Wolf's critique of Hume does not apply to LDC.
2.4 ‘Disability’ Is Not a Natural Kind
Finally, let me address the meaning of the category ‘people with disabilities.’ Humans vary greatly in our capacities and abilities. We all have differing abilities, and what counts as a disability is dependent on the technology of a society, our access to it, and the prevalent norms of functioning. People who are characterized as existing within this group vary, as do people who are characterized as ‘able.’ Disability or ability are range properties relative to the social environment.Footnote 12
3. Reflective Equilibrium (Reply to Wolf and Varden)
A final component of my response to Wolf simultaneously takes up Varden's suggestion for a deeper Kantian engagement. The original position is not where two-level contract theory ends. Instead, two-level contract theory augments the Rawlsian idea of hypothetical acceptability with the claim that the theorist who imagines the original position needs an improved informational context, and that that context, for an anti-oppression theory, requires people in the real world to cultivate their autonomy skills. To supply a stronger defence for the place of people with cognitive disabilities within LDC, I turn to a deeper Rawlsian insight — the idea of reflective equilibrium and considered judgements.Footnote 13 It is a considered judgement that humans with all varieties of disabilities are full members of society to whom justice is owed. Therefore, the principles of justice that derive from two-level contract theory must reflect that fact.
The model-conception of persons in LDC's original position is an artificial device that plays a role in arriving at an idea of fairness that is unbiased by individuals’ social positions; it is not a way of setting forth criteria with which to identify who counts as a person with rights in the real world. In fact, the role of skills in two-level contract theory creates an indirect informational context for the hypothetical acceptability of the original position device, where the influence of the effects of these skills is mediated by the theorist as the person who gives content to the outcome of the modelled deliberation.
More generally, two-level contract theory creates new information and constraints for the account of reflective equilibrium. This deeper Rawlsian idea — reflective equilibrium — allows us to accommodate changing considered judgements. One such considered judgement is that all humans have claims to care. The justification for this judgement might be based on species, on the particular dependence of humans on other members of the human community, on social tradition, on relationships, on species-typical ability, or something else. Whatever the foundation, it is now a considered judgement that humans with disabilities of all types should be fully included in the community, and that people with disabilities have claims of right for protection and freedom.Footnote 14
4. Reply to Edenberg
Political liberals seek to justify political principles to all who share reasonable doctrines, and feminist political liberals have worked to show that political liberalism is compatible with — or, must incorporate — feminism. In her article, Edenberg analyzes “how well LDC fits into a broader political liberal framework, while still securing protection against oppression” (Edenberg, Section 1). The core of Edenberg's critique is her argument that a political liberal layer of justification for teaching autonomy skills is needed. More precisely, she claims that LDC is “missing a justification of the second level of the contract device addressed to those who do not already embrace Bhandary's central anti-oppression commitments” (Edenberg, Section 1). The focus of this objection is my proposal to teach autonomy skills to people in the real world. She argues that “autonomy … needs its own independent justification, addressed to a pluralist society, to demonstrate why these autonomy skills are necessary to ensure LDC's legitimacy” (Edenberg, Section 1). Moreover, she suggests that such a justification is, in fact, available. Although a political liberal defence of LDC's autonomy skills would be an important finding, here I will explain why I do not offer a justification in political liberal terms, instead excavating a subset of the values that underlie the political liberal project.
Before I offer this explanation, let me acknowledge that, in the current political context in the U.S., some aspects of public education are subject to intense scrutiny. Consequently, I agree with Edenberg that proposals to teach autonomy skills in a system of public education could elicit vociferous objections from people who embrace gender-conservative religious doctrines. And, although Edenberg and other philosophers I respect have devoted themselves to the task of evaluating whether — and showing that — gender egalitarianism can be compatible with political liberal justifications, I remain unconvinced as to whether political liberalism is the form of political justification that best meets liberal demands for legitimacy. I am not even convinced that it has the greatest practical efficacy to achieve a democratically stable society. Clearly articulating our value claims conveys greater respect for one another than the thin accounts of freedom and equality employed by political liberalism.
4.1 Why I Do Not Endorse Political Liberalism
Let me emphasize, though, that I am not rejecting political liberalism. Instead, I am presenting some of my reasons for refraining from taking a position in the political and comprehensive liberal debate in Freedom to Care. Most significantly, it does not seem respectful, to me, to not hold up one another as equals in discourse. Therefore, insofar as political liberalism requires that we present our claims as claims that are based in a shared public political culture, it demands that we maintain reticence about our real commitments in the political sphere.
In addition, political liberalism's concept of a shared public political culture is vague. Determining the content of public political culture requires assessing which claims count as claims to freedom. There is too much latitude in the interpretation of these claims to determine the content of this “shared public political culture,” particularly because our claims to freedom often conflict with one other.
4.2 Philosophy Is Not Diplomacy
Nonetheless, political liberalism might be effective as a political strategy. It is certainly endorsed by philosophers who are fair-minded and supportive of diversity. Here again, though, I am sceptical as to whether it really is politically effective. I suspect, instead, that real people care more when their way of life, and thus, their context for intelligibility, is changing than when a position is presented in terms that do not overtly conflict with their value claims. As I argued through my discussion of the value of “being at home” (Bhandary, Reference Bhandary2020, p. 180), the importance of the intelligibility of action is a source of some people's resistance to changing the social structure in the fundamental ways required to achieve justice such that the just society includes the caregiving arrangements.
The political liberal's attention to expressive disrespect is an important addition to the liberal tradition. But attention to expressive disrespect sometimes seems to require the view that it is disrespectful to say, ‘I disagree with you, and here's why.’ The difficult task of determining how to communicate disagreement is a problem for diplomats and politicians, but how we should communicate our disagreements should not be confused with discerning and articulating what those disagreements are in the first place.
Philosophers should not modify accounts of justice with attempts at diplomacy. In fact, in virtue of philosophers’ commitments to openly assessing facts and arguments in the search for truth, we are among the last people I would suggest to serve as diplomats. When we are debating — even debating about normativity — there is value in asserting the view without artifice.
In contrast, when the purpose of debate is to arrive at an agreement with real people, then diplomacy is needed. Diplomatic communication may include attention to how all parties can ‘save face’ or how parties who do not ultimately win the day may feel that they ‘remain standing.’ Thus, diplomatic efforts in politics may require that views are not rejected altogether. How to accomplish this task is a subject for psychologists and specialists in a number of other disciplines.
4.3 Which Disagreements Matter Today? Moving Beyond the Reformation
Political liberalism's goal of attaining stability in the face of disagreement is laudable, and necessary. But liberalism's focus on toleration was shaped by early liberal thinkers’ apprehension of the effects of religious disagreement in relation to the Reformation and its aftermath.Footnote 15 As a result, contemporary liberalism has inherited an approach to disagreement for which religious disagreement is the paradigm case. However, because the most pressing matters of disagreement in many nations today have to do with racial inequality, ethnic nationalist claims to territory, and the end of patriarchy, that paradigm should no longer be the one we use to understand disagreement. Therefore, the liberal tradition should not remain wed to framings that characterize all disagreement along terms initially formulated to address religious disagreements to prevent religious violence.
Instead, we need to evaluate the resistance to change prompted by movements toward racial and ethnic equality as we (hopefully) leave a global order of European white male domination. The concept of “being at home” (Bhandary, Reference Bhandary2020, p. 180) that I explicate as a source of the resistance of patriarchal men to my claims can elucidate the resistance that derives from an individual's embeddedness in a way of life and context of intelligibility. When a person's context of intelligibility is founded on injustice, the changes required to achieve a just society may be felt to threaten the very self of the beneficiaries of the old social arrangement. When a person's sense of self is threatened, it will not be surprising if that person fights to retain it.
4.4 LDC's Argument for Autonomy Skills
So, where do these considerations leave LDC's argument for autonomy skills? In Freedom to Care, I distill a set of valuable criteria from the political liberal turn:
First, the endorsement of autonomy cannot rely on the claim that reason is the ultimate source of value, for to do so will exclude religious believers. For the same reason, it cannot rely on the metaphysical claim that we are all radically free. Third, … it cannot require that people pursue lives of maximal individuality. Fourth, it cannot rest on the premise that ethical pluralism is true. Fifth, it cannot result in the expressive subordination of a population with a reasonable doctrine. (Bhandary, Reference Bhandary2020, p. 101)
Edenberg identifies that satisfying these criteria is not the only way that one can fail to be a political liberal. She writes, that, “Political liberalism's shallow justification eschews comprehensive commitments to any particular metaphysical doctrine and instead grounds the justification in minimal values derived from our common culture” (Edenberg, Section 4). Although it would certainly be a welcome result if a feminist political liberal were to show how my account of autonomy skills might be justified through political liberalism, LDC is not a form of political liberalism. Instead, the claim I defend is that LDC can endorse an account of autonomy while remaining culturally sensitive and applicable to a broad range of societies (Bhandary, Reference Bhandary2020, p. 99). The account of autonomy skills I advance does not require that people make any particular choices or that they live in particular ways. Moreover, my approach to justification does not assume that there is meaning in claims about a ‘common culture.’ What the culture means, and to whom, and from which vantage point, all vary greatly.
Autonomy skills are designed to enable persons who are oppressed gain an understanding of who they are — without requiring that they endorse or reject the practices that oppress them. Thus, these skills leave open that people might continue to embrace oppressive practices for selective beneficial consequences from these practices. For instance, as Uma Narayan argues, women might participate in patriarchal practices through an assessment of the relative benefits and costs of attempting change (Narayan, Reference Narayan, Antony and Witt2002, p. 420). Therefore, although I welcome the possibility that Edenberg might show that the argument for autonomy skills can be made in political liberal terms, LDC does not require that argument.
In the context of evaluating the width and depth of LDC's account of political justice, Edenberg identifies that I assert LDC's cross-cultural applicability. She is right. Two-level contract theory allows for variation in the principles of justice that various societies might endorse, but the core idea of fairness is one for which I see no reason to limit the scope. To fill in the content of the idea of fairness, people in the society need to have autonomy skills. We cannot know what people value unless they communicate those values in some way, and the pervasive presence of oppression across human history is a reason we should not demur about the value of autonomy.
Edenberg notes, as well, that I theorize about justice in a way that includes some distance from questions about state coercion. Nonetheless, she argues, the state may need to play a role in the following situations: “If a person's legitimate needs are not addressed, and this qualifies as an injustice, the state should play a role in ensuring that these needs are met. … the state can play a role in supporting legitimate entitlements to care” (Edenberg, Section 5). Here, Edenberg raises the possibility that my proposal to teach autonomy and caregiving skills could be forced upon parents. She writes, “The state may presumably step in to force parents to ensure that a minimal level of autonomy and caregiving skills are developed if this is part of the justification of the theory” (Edenberg, Section 5).
However, I do not endorse forcing parents to ensure that a minimal level of autonomy and caregiving skills are developed. Instead, I assume the presence of the types of policies currently in place in the U.S. allowing parents to opt out of selective aspects of public education curricula, such as ‘health’ or ‘personal development’ classes. Because I consider autonomy intrinsically valuable (although my arguments do not rely on it), I predict that enough people will value autonomy skills to secure the adequacy conditions for theorists that I specify as an essential component of two-level contract theory.
Finally, I insist on the distinction between distributive justice and justice as an account of the justified use of state violence. LDC is an account of distributive justice rather than a guide to the state's use of violence. Although Edenberg correctly identifies that education is, in fact, coercive in one regard, as is taxation to pay for educational reforms, there is a meaningful distinction between taxation used to pay for school curriculum and a police officer using deadly force. While refusals to pay taxes might eventually result in violence, the varieties, degrees, and proximity of likely force all certainly matter. To see this point more clearly, let's engage in an imaginative exercise where social organization takes on unfamiliar forms. In this world, people might be less concerned with controlling one another. One can imagine, as well, possible worlds where humans do not perpetrate so many violent acts. What must be remedied, and how it must be remedied, will depend on details about our social practices and on their failings. Clearly, incarceration, fines, and the death penalty are not ideal methods for governing. Thus, the ‘state’ should not serve as a euphemism for purportedly justified violence. When we imagine radically different societies, we should think, in an open way, about the forms of enforcement and regulation of individuals. Moreover, when we attend to the way racism has been built into the content and enforcement of laws, the idea that enforcement via violence-wielding authorities is the paradigm of justice becomes evidently specious. By insisting on this conceptual separation between justice as fairness, practical legitimacy, and legal coercion, though, I am not denying the urgency of questions about how to make institutions factually legitimate, as well as effectively legitimate — that is, legitimate in the eyes and real lives of the people who are subjected to their laws.
4.5 Justice Is Incompatible With Oppression
Edenberg shares my foundational commitment to fighting oppression. She writes that, while we share this commitment as feminists, “However, this is a deep value commitment, and one that may not be accepted by all reasonable comprehensive doctrines” (Edenberg, Section 4). It is necessary, she argues, to defend the claim that “foundational commitments to freedom and equality also require resisting oppression” (Edenberg, Section 4) — a task undertaken by a number of leading feminist political liberals.
I simply do not see why a theory of justice should make room for the idea that a specification of justice would support oppression. If freedom and equality do not entail anti-oppression, then they are versions of freedom and equality that only apply to some people, and that are predicated on the oppression of others. A theory of justice that does not supply the conceptual toolkit for identifying oppression is simply a rationalization of the status quo.
Throughout Freedom to Care, my concern is to develop a theory of distributive justice. A theory of distributive justice should provide a framework with which to identify, in order to rectify, injustices in our social arrangements that run afoul of basic constraints for social cooperation. Theorists who have occupied the most socially privileged positions in their societies have not found the varieties of oppression to be relevant.Footnote 16 But those of us, like Edenberg and myself, who are feminists because we have critical knowledge of our societies, recognize the complexity of oppression. I simply do not see how an inquiry into injustice can proceed without a deep engagement with the complex phenomena of racism and gender subordination (see Mills, Reference Mills2017, p. 214; Young, Reference Young and Young1990).
5. Reply to Varden: Principles of Justice and LGBTQIA+ Inclusion in LDC
By revisiting the idea of reflective equilibrium in Section 3, I have partially addressed Varden's urging that I engage more broadly with the neo-Kantian tradition.Footnote 17 And, with my insistence on the philosophical separation between the account of distributive justice and the role for justified violence and coercion in Section 4, I also offer a response to Varden's claims about coercion.
In this section, I will take on two main questions, both of which have generated further thought and will remain in my thinking in the future. Varden urges additional engagement with leading Kantian thinkers who have developed a different branch of the neo-Kantian tree than Rawls, pointing out that many were students of Rawls, and arguing that their neo-Kantian accounts fare better than his. In addition, she argues for a more robust anti-oppression engagement — one that makes explicit how LDC rejects the implementation of its principle to oppress LGBTQIA+ persons. I will address the inclusion of LGBTQIA+ persons first, and then, in Section 6, I will take some initial steps down the path of deeper engagement with contemporary Kantians. Varden writes:
I also worry that it is possible to use the arrow of care and the four principles in relation to a particular society where we combine them only with the socially prominent social categories. … After all, when we look for who cares for whom or which groups are dependent on which, then we need social categories to identify the ‘who.’ In some countries, there are officially no LGBTQIA communities: ‘such people don't exist here,’ and similarly dehumanizing or existence-denying things are said about many social groups. These groups often also are not cared for or do not receive care; quite the opposite — asking for care can be dangerous (as existing is deemed a crime). In other words, I believe Bhandary's theory would be strengthened if it were impossible for it to be wrongly or mistakenly applied so as to ignore those who occupy social categories that the powerful in a given society deny or oppress. It is not clear to me that this is currently the case, and given the prominence of cis language and straight lives in Freedom to Care, I worry that the text does not give sufficient resistance to such bad applications. (Varden, Section 5)
Although the inclusion of care needs for LGBTQIA+ needs is a logical extension of my claims, and part of the design of the arrow of care map concept, I think Varden is correct that, when a group is persecuted and subjected to marginalization and violence, there is a need to make explicit that the members of that group are included in the account of justice. So let me do that here with claims that are implicit in Freedom to Care but which I agree should be made explicit: (1) The existence of the care deficits for LGBTQIA+ persons is an injustice. (2) The principles of LDC apply to all people, including necessarily to LGBTQIA+ people and communities. (3) The needs for care of LGBTQIA+ persons are fundamental needs and, as such, must be included, for instance, in the survival baseline principle.
5.1 Arrow of Care Map
In fact, the arrow of care map is an abstract tool designed precisely for identifying the complexities of caregiving arrangements. The concept is based in my rejection of the idea that the heterosexual family should serve as the base unit for inquiring about caregiving injustice. The arrow of care map is an open concept, a scaffolding to employ for intra-cultural and cross-cultural analyses about groups. To arrive at a just society — one that is just with respect to care — we must evaluate the extent to which the care needs of LGBTQIA+ individuals are met, and this category — one that is indeed comprised of multiple categories — must be included. We must note as well that queer individuals intersect and overlap with heteronormative communities.
There should be iterations of the arrow of care map that are employed in ways that incorporate LGBTQIA+ identities, and a focus on identifying the harms experienced by members of these groups is essential. But it is also vital to identify inequalities within groups. Nonetheless, I am reluctant to be the person who applies LDC to the internal workings of these groups because meanings of care may differ when a person expresses care in a social context where they are embattled. Although I am part of the LGBTQIA+ community in certain respects, and I am an ally, my social location in a heterosexual marriage means that I do not experience daily marginalization.
One reason that I did not employ examples of LGBTQIA+ individuals in my discussions of care inequality and exploitation was that the focus of my analysis was to identify asymmetries and relations of privilege and resistance to change. Because queer communities are marginalized communities, I did not want to extend my theory's critical lens into these communities. For instance, a simplistic way to extend my claims would have been to point to members of same-sex lesbian parenting relationships, and to identify the enduring presence of a linkage between feminine coded behaviour, other-directedness in activity, and care exploitation. But inequalities in the distribution of labour might have different meanings in this context, and what is normatively salient may also differ. For instance, as Varden highlights, the social perception of being unworthy of care is perhaps a more salient type of injustice and inequality for the LGBTQIA+ community than is feminine-coded caregiving. An analysis that responds to differences in meanings and moral salience is precisely a virtue of the arrow of care map. In virtue of its abstract nature, it respects the variation across contexts and is ready to be employed in these various contexts by scholars working in these domains. For instance, the arrow of care map might be employed in relation to William G. Hawkeswood's (Reference Hawkeswood1996) ethnography of African American gay men in Harlem, which “reveals how Black gay men in New York City forge alternative familial networks …, creating a viable alternative to the biological family from which they have been excluded” (Arnold & Bailey, Reference Arnold and Bailey2009, p. 173).Footnote 18
Let me clarify, as well, that when I write about the social construction of the heteronormative family form, it is not a way of indirectly endorsing this family form as the best kind of family. It is, instead, a way of responding to the fact that this construction is a core social practice for the liberal philosophical doctrine, and that norms governing this social practice are integral to the social form. In Freedom to Care, boys get a whole chapter (Chapter 7) because socialization into a form of self-oriented boyhood remains perhaps the most significant obstacle to justice in caregiving arrangements — and this is the case in many countries and cultures across the world. Liberalism's understanding of the representative man occurs in a world where heterosexual-presenting boys and men are not expected to attend to the needs of others. Therefore, I subject them to a sustained critique because they are such a large part of the problem of unjust caregiving.
The arrow of care map is designed to address the ways in which important phenomena and injustices in care may be missing from longstanding liberal feminist and ‘mainstream’ justice analyses in political philosophy. My arguments for the necessity of abstraction are based on the idea that we do not know all of the oppressed categories, or which categories will be oppressed in the future. It is my hope that scholars in queer studies will apply the map concept to identify the nature of care arrangements within queer communities as well as the ways these community constructions interact with institutions that remain predominantly heteronormative.
5.2 Examples and Methodology
Varden's question about examples intimates a broader inquiry about the use of examples in philosophical methodology. When we use an example, the example might be interpreted as setting forth an exemplary case, one that is particularly worthy of discussion. Examples interpreted in this way might indicate that it is this case that is exemplary, and therefore the most valuable version of the kind it exemplifies. These examples, — let us call them ‘exemplary examples’ — are implicitly held up as examples of excellence or typicality, where their valuation implies that other cases are less valuable. The implied value of these cases is often linked to assumptions that other cases are disvalued to the extent that they depart from these cases. For instance, a white, heterosexual, upper-middle class set of parents with children, a nuclear family, with the financial flexibility to decide that one parent can stay out of the paid workforce, might be construed as the ideal family in the U.S., and family constructions that depart from this one might be considered less ideal to the extent that they depart from it.
A different kind of example might be fictional, such as a Barbie doll represented as a model of an ideal woman.Footnote 19 Charles W. Mills describes this kind of example as an ideal-as-idealized model (Mills, Reference Mills2017, p. 74). This kind of example does not identify an actual instance of a category, but rather, a (purported) ideal or, as in the case of Barbie, a visual and conceptual mechanism for regulating the behaviour of real people and valourizing those who come closest to its approximation. The heterosexual examples in Freedom to Care are used as objects of inquiry rather than examples to emulate. My vision of a just society is one in which LGBTQIA+ families are as central to the basic structure as are heterosexual families.
But there were reasons for proceeding as I did, and why I still think that including LGBTQIA+ examples in my explication of the theory of LDC is a theoretically and methodologically complex endeavour. I think it would be inadequate for me to simply change the language — in no small part, because the meanings of actions change when people are prohibited from participating in a domain of life. Moreover, queer relationships may have more freedom to experiment with different models of family life. Despite that freedom, manipulation and exploitation can also occur within queer relationships because narcissistic personalities, for example, can exist in any gender.
6. An RSVP to the Feminist Kantian Table
Although I am not a Kant scholar, I accept Varden's invitation to sit at the feminist neo-Kantian table. The constructivist tradition is one that cannot be wrested from its Kantian origins, and Onora O'Neill's articulations of that view are of enduring interest to me.Footnote 20 Without entering into the robust debates internal to the Kantian tradition, I will comment that LDC is likely compatible with Kantian approaches to the morality of interpersonal interactions.Footnote 21 Nonetheless, on matters of the justice of the basic structure, I prefer the Rawlsian hybrid approach. Rawls combines the Kantian idea of respect for persons with the model of human motivation as potentially self-interested, and this mixed view, in my neo-Rawlsian, rather than Rawlsian, form, thus supplies better resources to address the cooperative conflicts that Susan Okin (Reference Okin1989) and Amartya Sen identify (Bhandary, Reference Bhandary2020, p. 118). As I argue in Chapter 4 of Freedom to Care, a theory that models people as self-interested is better equipped to handle the forms and manifestations of self-interest than an account that locates self-interest as an obstacle to duty and morality. Therefore, I remain committed to modelling the circumstances of justice as I did, including to Hume's influence on Rawls. For an account like mine — one that tracks when people's legitimate needs for care are unmet — recognition of the possibility of scarcity is vital. Because real societies fall so far from regularly satisfying norms of equal respect, it is valuable to retain self-interest at the stage of modelling persons for the purpose of ascertaining what a just distribution would be — as Rawls and I do. Perhaps Kantians have a way to embrace self-interest with a more realistic understanding of human motivation, but according to my knowledge of Kant, he does not supply us with these resources.
7. Conclusion
With Freedom to Care, I have not answered all questions about justice. What I have done, instead, is offer an account of distributive justice, setting aside questions of enforcement, coercion, and the law. We should not equate transparency and the critical scrutiny of the fairness of our practices with an account of justified violence. Assessments of methods of law enforcement and punishment are separate subjects that are not already implied in an assessment of a just basic structure.Footnote 22 Correspondingly, the account of distributive justice in Freedom to Care is an assessment of the fairness of the basic structure, which I conceptualize as “the system of practices we use to organize social life.” We can assess the fairness of the basic structure — those institutions that impact our lives from the start — prior to asking how to remedy injustice and protect people from being harmed by others. Thus, it is meaningful to make these assessments in ways that are distinct from the task of designing laws, such as in family law.
As Edenberg concisely summarizes in her article, the core of LDC is its two-level contract theory and the arrow of care map. Although I specify a set of parameters for possibly just societies, the precise nature of these societies is something I leave open. It is my hope that the way I have modified the tradition of liberal social contract theory as a form of critical liberalism, and an anti-oppression theory, can create conceptual and theoretical possibilities to arrive at greater clarity about the nature of caregiving injustice experienced by members of different groups and individuals.
Acknowledgements
I am grateful for Lavender McKittrick-Sweitzer's dedicated work as editor of this symposium. Richard Fumerton read and offered valuable comments on the written version of this article, as did an anonymous reviewer; I thank them both. My thinking on these issues was enriched by discussion with Helga Varden, Elizabeth Edenberg, Tiina Vaittinen, Carol Hay, Christopher-Rasheem McMillan, and Elizabeth Rodriguez Fielder. The final revision of this article benefitted from Sooraj Saksena's questions about my use of the term ‘critical liberalism.’ Dialogue EA Jill Flohil provided indispensable editorial assistance for the entire symposium, and Cécile Facal provided vital assistance for the French abstract.
Open access
