Against Convergence Liberalism: A Feminist Critique

Abstract

Convergence liberalism has emerged as a prominent interpretation of public reason liberalism. Yet, while its main rival in the public reason literature—the Rawlsian consensus account of public reason—has faced serious scrutiny regarding its ability to secure equal citizenship for all members of society, especially for members of historically subordinated groups, convergence liberalism has not. With this article, we hope to start a discussion about convergence liberalism and its (in)ability to address group-based social inequalities. In particular, we aim to show that given the core features of the view and real-world pluralism, the policies needed to secure gender equality and protect equal citizenship for women will not be justified. We make our case by considering various inequalities that are due to the gendered division of labor and potential convergence liberal responses.

1. Introduction

Convergence liberalism has emerged as a prominent interpretation of public reason liberalism. Like other accounts of public reason, convergence liberalism is concerned with political legitimacy. The fundamental question for modern political philosophy on this view is whether persons with diverse and conflicting values and beliefs can find terms of social cooperation that are justified for all and that, when followed, sustain social stability.² Yet, while its main rival in the public reason literature—the Rawlsian consensus account of public reason—has faced serious scrutiny regarding its ability to secure equal citizenship for all members of society, especially for members of historically subordinated groups, convergence liberalism has not.³ That is, the question as to whether convergence liberalism has the tools to sufficiently redress group-based social inequalities such as those based on gender, race, sexuality, or mental or physical differences has been largely unexplored.

With this article, we hope to ignite a larger discussion about convergence liberalism and its (in)ability to address group-based social inequalities. Here we aim to show that, given the core features of the view and real-world pluralism, the policies needed to secure gender equality and protect equal citizenship for women will not be justified. We make the case by considering various inequalities that are due to the gendered division of labor and potential convergence liberal responses. The crux of the matter is not just that convergence liberalism will not require the adoption of certain laws and policies needed to adequately address gender inequality under certain conditions, but also that the view permits certain laws and policies central to gender equality to be blocked altogether from the set of those eligible for adoption, making it impossible to use the law and democratic processes for correction.

The implications of our critique are that in a real-world society organized in accordance with convergence liberalism, (1) the norms and practices that maintain and perpetuate women’s inequality are left in place and, so, women will continue to face various forms of oppression and (2) gender egalitarian ways of life are more costly than sexist ones. These implications make convergence liberalism unacceptable as a political framework for those concerned with meaningful equal citizenship for women and other historically oppressed groups.

In our analysis, we will not take up the question of whether Rawlsian accounts can address concerns about the gendered division of labor. While we think that Rawlsian accounts have better resources for addressing substantive forms of gender inequality, mounting a defense of that claim is a separate project.⁴ Our focus here is simply on developing a critique of convergence liberalism. Thus, in this paper, we only aim to lay out what we take to be the insurmountable challenges convergence theorists face when it comes to securing equal citizenship for women and other caregivers.

The article proceeds as follows: we begin with a brief description of convergence liberalism. After that, we discuss some forms of oppression that women face due to the gendered division of labor and why particular kinds of policies are needed for gender equality. Then we argue that given its demanding public justification principle and the persons to whom justification is owed (the moderately idealized counterparts of real-world persons), the policies needed to address gender inequality due to the gendered division of labor will not be justified. After we make the basic case against convergence liberalism, we discuss various ways that convergence liberals might try to show that the policies we argue are needed for gender equality may be justified. The strategies include (1) recognizing the fact that convergence liberals hold that the law may be used to address unjust social norms in some cases, (2) considering that people often have wildly different reasons for the same law and policy, and (3) varying the object of justification in convergence theory, as what is justified in a society depends, to some extent, on the object of justification. We end by considering the most likely response of convergence liberals to our argument: that our critique is sectarian and that our view permits a kind of authoritarianism in which some persons are subjected to law and policy that is not justified for them. We aim to address this worry by arguing that any liberalism capable of securing free and equal citizenship for all members of society must be sectarian to some degree and this means that some persons will be required to comply with rules that they do not (or would not if moderately idealized) accept.

2. Convergence liberalism

Although convergence liberalism has a number of defenders, Gerald Gaus (2011) and Kevin Vallier (2014, 2019) are its prime exponents, and, so, their work is the touchstone for our exposition.⁵ Convergence liberalism begins with the claim that persons are not naturally subject to others’ authority (Vallier 2014, 31–32) and have moral liberty “vis-à-vis one another to act as they see fit” (Vallier 2019, 85).⁶ As such, persons may do as they please or, at least, as their beliefs allow when not interfering with the agency of others. Their actions, however, must be justified for others when and only when their actions interfere with others’ agency. Importantly, the moral equality of persons on this view is expressed by the claim that no one “has more justificatory power than anyone else” (2019, 88).

The public justification principle is the standard that must be satisfied when persons’ actions must be justified for others. According to this principle, a moral rule is justified if and only if each person subject to the rule has a sufficient reason given their own evaluative perspective to comply with the rule (Gaus 2011, 263; Vallier 2019, 4). That is, a rule is justified for a society if and only if it is justified for each member of the public. Of course, sometimes, persons do not converge on the same rule despite the fact that they each have sufficient reason for some rule or other. For example, the content of sex education for children may be an issue for which there is not a rule on which members of the public converge. Nonetheless, each person may have a sufficient reason for some rule or other. In such cases, if each person has a sufficient reason for the same decision procedure for selecting a rule on a matter, then a rule can be determined via the decision procedure (Gaus 1996, 223–26; Vallier 2014, 99–100). Yet, there may not be convergence on a decision procedure either. What then?

In his later work, Gaus suggests the following: if enough persons have a sufficient reason for a particular rule and are complying with it, then someone who views a different rule as optimal may still have sufficient reason to comply with the rule followed by others. This is because that person may value acting “on moral requirements that are embraced by others, so that in her interactions she can make moral demands that respect their equality and moral freedom” (Gaus 2011, 389–409, quoting 399; Vallier 2019, 110–12). So, as Gaus says, convergence may be reached even under conditions of deep disagreement. Nonetheless, for many matters, convergence on a rule may not obtain as some persons or a single person may place higher value on following their own conscience, or retaining their freedom from coercion, or any other sufficient reason from within their evaluative perspective.

To determine whether a rule is justified, the set of justificatory reasons for each person subject to the rule must be considered. Each person’s justificatory reasons include only reasons that are “psychologically accessible” and “morally motivating on reflection” (Vallier 2019, 89). The set of such reasons are intelligible reasons and specified as “A’s reason Ra is intelligible for member of the public P if and only if P regards A as entitled to affirm Ra according to A’s evaluative standards” (91). Only intelligible reasons are relevant to the determination of the justification of a rule.

A person’s intelligible reasons are constrained by the cognition criterion and the morality criterion, and, so, real-world persons are moderately idealized for the sake of determining their intelligible reasons.⁷ The cognition criterion stipulates that the relevant reasons are those that “ordinary persons can reach through reflection,” and this criterion concerns both the reasoning powers and information we take our representations of persons to have. Such moderate idealization represents persons as “boundedly rational” and in possession of the information required for real persons to make decisions under ordinary conditions (Vallier 2019, 99; Gaus 2011, 244–58).

The morality criterion limits justificatory reasons to “recognizably moral reasons, or at least reasons the person in question can see as moral rather than as immoral” (Vallier 2019, 90–91). Such reasons flow from or are part of a person’s evaluative standard. Convergence theorists hold that reasonable evaluative standards “recognize that people are interested in moral relationships with one another” and that evaluative standards premised on sadism, masochism, or immorality are excluded as justificatory reasons (96). Still, the threshold for what counts as a moral reason is minimal, and reasons are not excluded as immoral merely because they are, say, premised on views that posit a natural hierarchy based on gender or race.

A person’s set of justificatory reasons may include considerations that count both for and against a rule, in favor of different rules, or in favor of a rule and in favor of no rule. Ultimately, a rule is justified for a person if and only if they have a sufficient reason for it given their evaluative standard. A sufficient reason is one that is “not rebutted or undercut by other reasons” (91).

Key to determining whether a rule is justified in a society is the identification of an eligible set of rules for the issue at hand. Eligible sets are those with rules that no moderately idealized person has an intelligible reason to reject (110). This means that no one has a “defeater” for any rule in the eligible set; that is, each rule in the set is viewed by each person as better than no rule at all for the issue under consideration. If a rule is excluded from the eligible set, it is not a candidate for adoption. It is, of course, possible that for some matters the set will be empty because there will be defeaters for all candidate members of the set.

The set of rules that are justified for all those subject to them are the rules of social morality. Sometimes, the rules of social morality are not enough to adequately guide social interaction (132) and, so, legal rules are needed given problems such as the uncertainty and ineffectiveness of moral rules (134–38). Convergence theorists argue that liberal institutions, in particular, are justified for moderately idealized members of the public in our real-world conditions (3).⁸ In short, they claim that a liberal regime can secure and protect equal basic rights for persons so that they can pursue their projects and plans (compatible with others doing the same); any more extensive regime would subject persons to rules that they (or their moderately idealized counterparts) do not endorse. Still, they do not think that convergence accounts of justification guarantee liberal institutions; whether liberal institutions are justified depends on the beliefs and values of persons in a particular society (Vallier and Muldoon 2021).

Given evaluative pluralism, the justification of the rules for social morality as well as of constitutional and legal rules is challenging. For any rule to be justified, it must be justified from the diverse moral perspectives of all those subject to the rule. Hence, convergence theories are said to have a classical tilt (Gaus 2011, 497–529), and this means that it is to be expected that the sorts of liberal regimes that are justified will have fewer rather than more rules and policies.

Having the basic elements of convergence liberalism in place, we now turn to develop the basis of a feminist critique of the view. Again, we aim to show that given the central features of convergence liberalism and views of real-world persons, the kinds of policies needed to address gender inequality due to the gendered division of labor are not justified.

3. Feminist critique

3.a The gendered division of labor and the fundamental interests of citizens

As Eva Feder Kittay (1999), Martha Nussbaum (2004, 2006), and many other feminists stress, caring for others is a large part of human life and socially necessary work. All human beings are dependent on others for emotional and physical care over the course of a life. For example, children need emotional care to develop socially and morally, and they need physical care (e.g., help with bathing, grooming, and eating) for their basic needs to be met. Most adults need emotional support and caring relationships with others for social and emotional health and to be able to pursue their view of the good (Brake 2012). Many adults need physical care at various times in adulthood, given illness or accident. Of course, some adults need physical care from others throughout adulthood for their basic needs to be met. Those who need emotional and/or physical care from others are dependent upon them. Most care work is performed by women; it is not properly valued or compensated, and those who perform it are unjustly disadvantaged relative to others because they perform this work.

It is worth noting that in modern liberal democracies most work is gendered. For example, paid work is gendered.⁹ This is so despite decades of antidiscrimination law in education and employment, including sexual harassment law, and despite the growing number of women with advanced degrees.¹⁰ Unpaid work—both domestic work and caring labor—is gendered, too, both in terms of how much unpaid work is done by whom and what type of work is done.¹¹

The situation of women in the United States illustrates the kind of gender inequality that can occur due to the gendered division of labor when a state lacks adequate policies for addressing the labor of caring for dependents. For example, women in the US earn less than men (about 82 percent of men’s weekly earnings in 2017),¹² and mothers earn less than fathers (71.4 percent of father’s earnings).¹³ Women of color, who face unique forms of oppression due to the intersection of race and gender, are the least well-compensated members of the labor force. All this is due, in part, to the fact that in the US most jobs either lack paid leave, lack sufficient paid leave, or lack the kind of flexibility needed for persons who are the primary caregivers of others. Further, there is not sufficient access to affordable daycare, aftercare, or eldercare programs for dependents. Many workers have little control over their schedule and must work additional hours at the will of their employer or as workload demands. Many well-paid jobs require availability for travel or relocation. Given gender norms, women more than men are expected to bend to meet the needs of others, and they feel pressure and responsibility to do so. And, so, women more than men often choose lower paying, more flexible jobs or part-time work. Or, when women work full-time jobs, which assume they do not care for dependents, women’s careers often stall—they take longer to reach promotions or settle for different and often less prestigious, less powerful, and less financially lucrative tracks. Some women leave the labor market altogether. For single mothers, balancing caregiving and paid work is incredibly difficult. Many single mothers and their families live in poverty in the US.¹⁴ Caregivers in relationships (whether married or not) often become financially dependent on their intimate partner and then have less bargaining power in their relationships. As a result, many caregivers tolerate emotional and physical abuse, adultery, and demeaning spending allowances.

In short, the gendered division of labor is central to women’s continued inequality and oppression. It perpetuates women’s exploitation, stigmatization, and marginalization. Young’s account of the faces of oppression is helpful for illuminating this fact. Young says that exploitation “occurs through a steady process of the transfer of the results of the labor of one social group to the benefit of another” (1990, 49). Women disproportionately perform care work without pay or perform such work in underpaid care jobs. This labor benefits those able to participate in the labor market as they rely on caregivers’ unpaid or underpaid work; caregivers’ labor is exploited when not sufficiently reciprocated (Brake 2020). Now consider the stigmatization¹⁵ that women face when they participate in the labor market. As caregivers, they are judged as distracted, unreliable, and uncommitted when compared with men, and they receive lower ratings on performance evaluations, are offered lower wages, and are passed over for jobs and promotions (Williams 2010, 92–93).

Finally, consider marginalization. Young defines those marginalized as those “the system of labor cannot or will not use” (1990, 53). Certainly some, such as the elderly or some persons with disabilities, fit this account. However, we think the idea of marginalization also accurately captures the situation of those who are systematically disadvantaged or burdened with systematic barriers when they try to combine participation in spheres of social life central to citizenship (the labor market, political sphere, civil society, and the family) as compared with others (Watson and Hartley 2018, 189–211; Fraser 1997, 48). Insofar as the labor market is structured in ways that do not support workers as caregivers of dependents and insofar as women are the primary caregivers of dependents, women are disadvantaged or unjustly burdened in their ability to participate relative to others. Many must work anyway to survive or better support themselves and their dependents. But the demands that women and other caregivers face as they participate in a labor market that doesn’t recognize their caring responsibilities means that they will likely struggle to meet the demands of caring for dependents. The consequences are especially bad for single parents, again, most of whom are women. Insofar as distinctive spheres of social life allow persons to access goods and social relationships, which are incommensurable or can make unique contributions to a life, marginalization from one or more of such spheres is incredibly burdensome and unfair (Fraser 1997, 48). And, when such marginalization is based on group membership, such as gender, it is oppressive.

Central to understanding the practices and institutions that sustain these forms of gendered oppression is that they occur in the context of a society in which gender norms encourage women to make choices regarding work and family life that disadvantage them relative to men (Chambers 2008, 117–57). Social and political equality requires that caregiving is viewed as socially necessary and valuable work, that caring labor is not exploited, and that caregivers are not stigmatized as labor-market participants. Further, those who perform the work of caregiving must not be disadvantaged relative to other citizens with regard to their ability to participate in the spheres of life central to citizenship (including the labor market and the political sphere). It is among the fundamental interests of persons as free and equal citizens that such forms of oppression are not perpetuated through policy or the lack thereof and that the social conditions of freedom and equality for all are in place (Watson and Hartley 2018). To be clear, it is not just women and other caregivers for whom these matters are fundamental. Society, as a system of cooperation over time, requires the rearing, education, and support of children as future citizens, without which societies would dissolve within one generation. Thus, all citizens have an interest in securing the social conditions under which children receive the care necessary to become cooperating members of society. Treating children as an expensive taste or hobby of some individuals fails to grasp this basic fact.

Below we emphasize that specific types of policies are needed to secure these fundamental interests of citizens and to address gender inequality.¹⁶ Such policies include particular forms of paid family leave, access to affordable dependent care, work-time regulations, and caregiver allowances for some especially burdened caregivers. As we explain below, together these policies protect caregivers of dependents from oppression through the recognition of care as socially necessary and valuable work by supporting caregiving, and through policies that ensure that caregivers are not disadvantaged relative to others in important spheres of social life, including the labor market. Incentivizing men’s care work is essential to achieving equality, as gendered caregiving work stigmatizes women (even when such work is recognized as valuable) and disadvantages women in the labor market, political sphere, and in civil society more broadly. We call the policies needed to address women’s oppression due to care work and its gendered nature Care Equity Policies (hereinafter CEPs). And, we aim to show that in a society organized by principles justified in accordance with the demands of convergence liberalism, the needed CEPs will not be justified given real-world pluralism. This means that in such a society (1) women would face continued oppression and (2) sexist ways of life would be less costly than gender egalitarian ones.

Importantly, as noted, only particular policies will protect women and other caregivers from oppression and secure the fundamental interests of citizens. Gina Schouten stresses this in her recent work on the gendered division of labor. Consider paid family leave. While this type of program is crucial to addressing gender equality, not just any type of paid leave succeeds. Schouten says:

The duration of leave must be calibrated to avoid setting parents back at work. Very long subsidized leave risks parents falling behind their peers in terms of job marketability and disincentivizes the hiring of parents or those deemed likely to become parents … Leave should be flexible to promote continued labor-market attachment among caregivers … Other things equal, pay replacement rates should be high, because low wage replacement rates discourage leave-taking among higher earners. (2019, 45)

The specifics matter for CEPs because not all policies that support caregivers in some way will address all the problems that caregivers face or the gendered nature of these problems.

The kinds of policies we highlight address the problems of exploitation, stigmatization, and marginalization, as we will explain. And, together they recognize caring labor as socially necessary and valuable, mitigate against pernicious aspects of the gendered division of labor, and incentivize men’s caring work. Consider how. CEPs recognize caring labor as socially necessary and valuable work because labor market participants are entitled to some paid family leave, to work-time regulations to help with caring, and to affordable dependent care options. CEPs address the exploitation of caring labor (and dependency that often occurs as a result) because caregivers in the labor market will not be disadvantaged relative to other workers and because there will be incentives for men to engage in more caring labor. This will reduce unreciprocated benefits related to caring labor among adults. CEPs combat the stigmatization caregivers in the labor market now face, too, insofar as the model of the labor market participants will be changed: labor market participants will be assumed to be caregivers of dependents.¹⁷ Finally, CEPs preclude the marginalization of caregivers insofar as persons are able to combine labor market participation with caring for dependents, except in circumstances of short-term paid leave or circumstances of unusual caring demands (children or other family members with permanent and considerable care needs). CEPs address oppression due to the gendered nature of the division of labor, as they enable caregivers to participate in the labor market as equals with others (insofar as possible) and in other spheres of social life, and they incentivize men’s caring labor. The state, uniquely, is in a position to make sure that all persons have their interests met related to these issues, and without state intervention, women’s interests will not be secured or will always be subject to the whims of those in power. Much more could be said about the fundamental interests of persons as they relate to the problem of the gendered division of labor and about the connection between particular policies and addressing these interests. However, here our goal is just to make plausible that certain interests are among the fundamental interests of persons as equal citizens and that specific laws and policies (CEPs) are needed to protect these interests.

3.b The basic case against convergence liberalism

We now turn to the limits of convergence liberalism when it comes to addressing some of the fundamental interests of citizens related to the gendered division of labor. First, consider the protections for citizens that convergence liberals think will be justified in a real-world society. They claim that given the views of moderately idealized members of the public and their public justification principle, most standard liberal rights and institutions will be justified, including freedom of thought, freedom of speech, freedom of occupation, rights to bodily integrity, personal property, and association (Vallier 2019, 203). Further, they recognize agents need income and wealth to exercise their agency, and, so, they claim that positive rights of agency demand that society be arranged so that persons have access to these resources (Vallier 2019, 204; Gaus 2011, 358–59).

This standard package of liberal rights, however, does not protect the interests of caregivers of dependents nor does it protect the interests that all citizens have in caregiving practices. As feminists have long observed, none of the rights that are part of the standard package of liberal rights addresses the fact that persons are vulnerable and needy and will need care from others over the course of a life, or the fact that many persons will need to provide care to others and have various interests as a result. In short, the standard liberal package does not address the fact that caregiving is socially necessary and valuable work for the continuance of society. One might suggest that the resources that persons will have access to as part of their agency rights will help address the interests of those who provide care to others to some extent. Such help, though, will be insufficient to protect the particular interests of caregivers, some of which are noted above, and to address how the gendered division of labor oppresses women through stigmatization, marginalization, and various forms of exploitation. Important here is that for some women the problem is not that they lack access to some option or some way of life that offers sufficient income, wealth, and capital. Rather, for some women, the problem is due, in part, to the set of options that women face and the social conditions in which they make choices.¹⁸ As the standard liberal rights package is not adequate for addressing the interests we’ve identified, the challenge for convergence liberals is whether policy that addresses the gendered division of labor and caring for dependents is justified given their commitments and real-world pluralism.

Let’s consider further that the gendered division of labor in modern liberal states is due, in part, to different choices that men and women make about paid and unpaid work given the options available to them. Some of the options available to women under current conditions include: forgoing parenting (“childless women, including married and unmarried, earn 93 cents on a childless man’s dollar” [Budig 2014]); single parenting, which often considerably decreases a family’s resources and may result in poverty; or parenting with a partner and cutting back on work outside the home, and, so, becoming financially dependent on and vulnerable to another. So-called “leaning in” is a real option for only a few (the wealthy), and equally sharing care work is costly for families given the current structure of the labor market. Each of these options as well as the choices persons make are further complicated by gender norms and expectations about caregiving and paid work.

Insofar as the gendered division of labor is due in part to the different choices of men and women, convergence liberals will not address the inequalities the gendered division of labor creates. To see this, consider that regarding primary rights to assistance and the poor. Vallier says:

Primary rights to assistance, while publicly justified, can be qualified by the presence of citizens who believe that such rights are sensitive to facts about what persons deserve. Many people believe, and may believe even if moderately idealized, that the able-bodied poor are, by and large, responsible for their poverty. If so, then poor persons do not automatically merit coercively financed assistance from those who have concerns about their deservingness. To take an extreme case, if middle-class John gives all his money to his church to live a life of poverty, many will believe that he does not have a right to assistance since he freely donated his income and wealth. (2019, 205)¹⁹

According to Vallier, then, we should not expect convergence on a policy to support helping those whose disadvantage is chosen. Some people think that individuals should be held responsible for the consequences of their choices even if their choices leave them in poverty, and the public justification principle requires that policies be justified for all members of the public. This shows it is convergence liberalism’s demanding standard of justification paired with some persons’ beliefs that will often block state action to help the disadvantaged.

Now consider choice, the gendered division of labor, and convergence liberalism. Some people believe that if women choose lower paying and/or more flexible jobs, choose to work part-time, or “opt-out” of the labor market to perform unpaid caregiving work, then they are responsible for what results—be it poverty or financial dependency. Others think that parenting or caring for dependents is an expensive taste for which individual persons are responsible. Still, others think that certain policies would serve as objectionable incentives for persons to reject traditional gender roles that are central to the way of life they value. These views simply reflect the kind of evaluative pluralism that characterizes modern liberal states. Further, we stress that convergence liberals, like other public reason liberals, think evaluative pluralism is inevitable. Persons who are reasoning well may endorse different worldviews for a variety of reasons because, for example, theories are always underdetermined by the evidence and people who make no formal mistakes in reasoning disagree about how to interpret evidence as well as how to weigh and order values (Vallier 2019, 20–21; Rawls 2005, 54–58). The combination of inevitable, diverse evaluative perspectives paired with convergence liberalism’s demanding public justification principle entails that the kinds of policies needed to adequately address gender inequality will not be justified and may be blocked altogether from the set of those eligible for adoption.

While convergence liberalism isn’t the only version of liberalism that falls short of addressing gender inequality due to the gendered division of labor, the reason it fails is important to understand and can be traced to the conception of moral persons and the interpretation of the freedom and equality of moral persons at the heart of the view. Again, convergence theorists are concerned with developing an account of social morality for moral persons, and they claim that a liberal regime will be justified given the uncertainty and ineffectiveness of mere moral rules. Moral persons, on their view, are simply those with the capacity for moral autonomy or moral agents. The theorists claim that “[t]o respect another as a free moral person is to acknowledge that her reason is the judge of the demands that morality makes on her” (Gaus 2011, 15). It is not that convergence theorists think that all reasoners are “equally correct.” They stress “the lack of authority of another’s judgment over one’s view of the demands of morality” (15). Further, they claim that moral equality is simply a matter of all persons being “equally authoritative interpreters of the demands that morality places on one” (15). Convergence theorists derive their demanding public justification principle from these claims.

Yet, it is precisely the thin notions of moral personhood, freedom, and equality adopted by convergence liberals that leave the view devoid of any resources for adequately addressing the oppression of women due to the gendered division of labor. The interests of moral persons qua moral persons on this view are the interests of persons as moral agents. There is no acknowledgment that moral persons (or developing moral persons) need care or need to care for others, and moral patients are not, as such, moral persons on this view.²⁰ Freedom merely has to do with not being subject to another’s interpretation of the demands of morality, and the notion of equality is a thin, essentially formal, conception based on an agent having no more justificatory power than any other agent. Neither the conception of freedom nor equality is concerned with group-based forms of oppression that constrain persons’ lives and undermine their status as equals. The substance of these notions of freedom and equality is inadequate. As a result, given the view of moderately idealized members of the public and the public justification principle, the policies needed to address gender inequality due to the gendered division of labor won’t be justified, and the public justification principle won’t even block profoundly sexist defeaters for policies aimed at addressing oppression. The morality criterion, after all, is minimal and only blocks reasons that are based on views like sadism. So, this criterion certainly won’t block many sexist reasons given that gender roles are central to many persons’ worldviews. Hence, given the views of moderately idealized members of the public, convergence liberalism will not adequately address gender inequality due to the gendered division of labor.

3.c Objections

In response to our argument, some convergence liberals might appeal to various considerations to try to show that their view can adequately address gender inequality due to the gendered division of labor. We will consider three types of strategies convergence liberals might employ to this end, including that convergence liberals think that the law can be used to address unjust social norms in some cases, that people often have quite different reasons for the same policy, and that convergence theorists can vary the object of justification, which makes a difference to what is justified. We argue, though, that none of these strategies work and that they fail for similar reasons.

Objection 1: Convergence liberals have resources for addressing unjustified or coercive social norms, and, so, they can adequately address unjustified gender norms that result in women’s oppression due to the gendered division of labor.

To the extent that persons’ options and their choices are the result of social norms as well as the practices and institutions that emerge due to these norms, one might think that convergence liberals have resources for addressing some of our concerns. For example, Vallier holds that the rules of social morality are a kind of social norm. Gender norms are social norms, and many (if not all!) gender norms- including gender norms related to caregiving- are not justified for all those subject to them. Importantly, Vallier recognizes that just because a social norm is not justified doesn’t mean it won’t be followed, and he says: “[i]f a bad or unjustified rule is in equilibrium, then it will resist change because few have an incentive to unilaterally deviate from it” (Vallier 2019, 134). He also endorses (134) Gaus’s view that “only the political order is apt to be an adequate engine of moral reform as it can move us to a new equilibrium much more quickly than informal social processes” (Gaus 2011, 437). Hence, both Gaus and Vallier think that legal rules may be used for justified moral reform, and, so, perhaps legal rules could be used to address unjustified gender norms.

Yet, it is still the case that law or policy that is enacted for the sake of justified moral reform must be justified from the perspective of all moderately idealized members of the public. So, even if it is recognized that gender norms are not justified social norms, given the views of some moderately idealized members of the public, there is no reason to think that there would be convergence on the kinds of policies needed to address unjustified gender norms and address oppression due to the gendered division of labor.

Relatedly, some persons view gender norms as coercive in a sense, and, so, in need of public justification.²¹ One might also think that this consideration can help convergence theorists address our critique. To see why not, first consider a challenge made to convergence liberalism by James Boettcher concerning the justification of property rights and coercion (2015, 202–3). Boettcher claims that convergence views are unclear about the conception of coercion they employ, and citizens may have diverse conceptions. Insofar as members of the public may hold a view of structural coercion like Jeffery Reiman’s, they may object to private property regimes that do not require substantive redistribution;²² other members of the public may hold a view of coercion such that they may “see all redistributive regimes as inferior to having no regime at all” (203). Hence, given different views of coercion, no property regime may be justified in a society, and this would make convergence theory unacceptably incomplete (203). Recall that it only takes one member of the public with a defeater for a policy or law to fail to be justified.

Vallier responds to this challenge by claiming “we can formulate a socially eligible set of proposals from evaluative standards with different notions of coercion.” And, he argues that when considering the reasons persons have for some proposal for a property regime or other as opposed to no property regime at all, we need to consider “how disastrous it would be if free and equal persons could not converge on property regimes that allow them to view each other’s actions as part of a shared enterprise of living together on equal terms” (2016, 260). Essentially, Vallier declares that convergence on a mutually justified property regime will emerge given how essential such rights are to each person’s own pursuits.

Even if this reply works in the case of property regimes, a similar reply is not available when it comes to our challenge that convergence liberalism lacks the resources to generate the kinds of policies necessary for protecting the interests of women and other caregivers of dependents. Start with the fact that when it comes to views about gender and work (or the distribution of caregiving work), there won’t be any justified moral rules. That is, moderately idealized representations of real-world persons will not converge on any such rules. Feminists and other egalitarians may hold that the work of caregiving should not be gendered at all and advocate for CEPs, but others do and will reject such policies. They may think that, by God’s laws, morality’s, or nature’s, women are the proper caregivers of children. And, as a result, they may think that social policy such as paid family leave, work-time regulations, or policy aimed at supporting affordable dependent care provisions is unnecessary because, after all, women have no discreet interest in both labor market and caregiving labor. They may also think incentivizing women’s participation in the labor market is harmful to traditional family values. Other people may think that given disagreement among people about gender and social roles, persons should just make their own decisions about labor market participation and family life and that it would be wrong or unfair for the state to enact policies that incentivize gender egalitarian lifestyles. Thus, given the views of moderately idealized members of the public, there will not be convergence on policy related to supporting caregiving or dismantling the gendered division of labor. CEPs will be blocked from eligible sets of social rules altogether.

We want to stress that Vallier’s response to the property regime challenge rests on the claim that each person clearly has a sufficient interest in there being some property regime or other in society: persons need to know that the investments of their labor will be fruitful and protected. Let us grant that legal rules are needed for this. Hence, given their various interests, moderately idealized members of the public will converge on a property regime, even if that regime is no one’s most favored. It is also the case that in any society that persists over time, children must receive care because caregiving is socially necessary work. While many (like us!) will regard it as unjust if such work is disproportionately performed by members of a particular social group to their disadvantage, others won’t. As we noted above, some people don’t think that gendered caregiving norms are unjust at all but, rather, good. Hence, while some people may believe that particular policies are needed to address gender inequality due to the gendered division of labor, those policies won’t be justified given real-world pluralism. People with different views about gender and work differ radically about what would be a disastrous result.

Also relevant here is that convergence theorists think that there is an order of justification when it comes to legal rules. The order of justification “specifies the order in which issues are ‘settled’ and this ‘settlement provides a background for further justification’” (Vallier 2019, 210; Gaus 2011, 275). For example, Vallier argues that agency rights are settled first followed by associational rights, jurisdictional rights, procedural rights, and international rights. Particular rights can be reviewed at any time, but when they are assessed, it is against a background of settled rights. So, particular law and policy that aims to address gender inequality due to the gendered division of labor would be assessed against a background of basic liberal rights. This just adds to our case that members of the public won’t converge on caregiving policy insofar as the standard package of liberal rights is already secured.

Objection 2: Convergence liberalism will not block needed state action. People with different worldviews can have different reasons for supporting the same policies, and people will often prefer some policy on an issue to no policy at all.

One might suggest that we have not sufficiently emphasized an important consideration that convergence liberals may raise in response to our concern that their view together with the views of moderately idealized real-world persons blocks (much) state action when it comes to addressing the gendered division of labor. In particular, some might stress that people with different worldviews may have quite distinct but sufficient reason from their own evaluative perspectives for the same rule or policy, and this may be the case even if that rule or policy is not either’s most preferred policy. What is required is that each person’s total balance of reasons for some policy be positive.²³ For instance, some conservatives and some feminists might favor restricting pornography in some way, albeit for very different reasons. And it may be that the restrictive policy on which there is convergence is not either’s preferred policy regarding pornography, as neither’s preferred policy is justified for all members of the public. Still, both may prefer some policy of restriction, given the balance of each’s reasons, to no policy at all.²⁴ The same may be true for different groups when it comes to some policies that would address the gendered division of labor.

For example, some feminists’ may favor paid leave policies that incentivize dividing care evenly between all parents or legal guardians of a child, but they prefer paid maternity leave to no paid leave for caregiving at all. Some conservatives might prefer other policies that incentivize women to be the primary caregivers of children over the course of several years to short-term paid maternity leave, but they may prefer such paid maternity leave to no incentives at all for women’s caretaking.

We think that this response is problematic for two reasons: first, given the diversity of worldviews there are and the high standard of justification, there isn’t any reason to think that there would be convergence on any caregiving polices at all, whether those policies are the CEPs we claim are needed or other policies such as paid maternity leave or caregiver stipends. Some people regard parenting as an expensive taste; others think we should not do anything to support/incentivize parenting given the climate crisis and the additional strain that more humans would put on Earth. And, while some social conservatives may prefer maternity leave in order to incentivize women’s caretaking, others may think that such a policy is altogether unacceptable, as they would not want to incentivize or make easier women’s labor market participation in any way. A policy is not justified if even one moderately idealized member of the public prefers no policy on some issue to any policy at all.

Second, policies on which there might be convergence that would address the gendered division of labor in some respects will not be adequate and might make some problems worse since addressing the forms of oppression that women face due to the gendered division of labor requires the adoption of carefully crafted policies. Otherwise, policies may have the effect of further entrenching gender norms. For example, Schouten notes that “[i]f [family] leave is allocated to the household as a unit, then one member can take the full amount, and social norms and household economic incentives will generally favor women taking the bulk of it” (2019, 45). Such policy, then, would not address and might make worse the stigmatization that women face in the labor market because gender norms and household economic incentives will result in women more than men taking the leave allocated for the household. So, women will continue to be the primary caregivers of children. In the labor market, they will face statistical discrimination, and when participating in the labor market, they will be stigmatized as distracted, unreliable, and uncommitted as compared to other workers. This policy has the effect of making gendered parenting arrangements less costly than gender egalitarian ones, and it won’t help address some important problems women face due to the gendered division of labor. Hence, even if we suppose that members of the public would converge on some policy, there is no reason to think that it would address women’s oppression due to the gendered division of labor, and it may make matters worse.

Objection 3: Convergence liberals need not adopt particular polices as the object of justification. By changing the object of justification, they can address oppression due to the gendered division of labor.

In our discussion up to this point, we have mainly focused on the possibility of a particular policy being justified given the demands of convergence liberalism. We have not considered if convergence liberals might be able to address our concerns by varying the object of justification—that is, what is to be justified. Convergence liberals, though, can adopt different objects of justification and that can make a difference to what is justified for members of the public. Consider two ways one might vary the object of justification from policy-by-policy evaluation. First, instead of focusing on policies for some particular issue, we might consider sets of policies for an issue understood more broadly. Second, instead of taking the objection of justification for convergence liberals to be fine-grained rules, one might take them to be principles (understood as standards for assessing or designing rules). We will briefly discuss both ways one might vary the object of justification and explain why neither move would address our concerns.

Suppose, first, that a convergence theorist claims that the object of justification should be sets of policies as opposed to individual ones. For example, notice that in the discussion above, Vallier stressed the importance of a society having a property regime. Property regimes are sets of policies. Arguably, when it comes to property law, a good case could be made for thinking about sets of policies as opposed to using a policy-by-policy evaluation. Here we assume that all members of the public would prefer some property regime to none at all. Suppose each property regime is constituted by a set of policies. For an individual, it may be that their favored property regime will not be justified for all, but given the total balance of their reasons, they have sufficient reason for another regime that is justified for all. So, perhaps, a policy that an individual would have had a defeater for in a policy-by-policy evaluation would be justified for a person given the person’s balance of reasons for a set. Hence, this way of varying the object of justification can make a difference in convergence theory.

Now consider dependency-care regimes understood as sets of policies for supporting and caring for dependents. Just as with property regimes, for an individual, it may be that a person’s favored dependency-care regime will not be justified for all, but given the total balance of their reasons, they have sufficient reason for another dependency-care regime that is justified for all. So, perhaps, a policy that an individual would have had a defeater for in a policy-by-policy evaluation would be justified for a person given the person’s balance of reasons for a set. This means that it is possible that in a policy-by-policy evaluation, an individual could have a defeater for paid family leave that incentivizes parents to share caretaking but that when dependency-care regimes are evaluated, the individual may have sufficient reason for a set of policies that includes such a policy. Fair enough. Still, though, dependency-care regimes have to be justified for all members of the public. When we consider whether a regime is justified for a person, we must consider the regime’s “tendency to employ … political power in such a way that she would be coerced to do things that, when she evaluates what she is being forced to do, she will have not have sufficient reason to endorse”(Gaus 2011, 493).²⁵ As a result, even if one could argue that all members of the public prefer some property regime to none at all, we don’t think that can be argued in the case of dependency-care regimes given different types of conservative, religious, and natural rights-based worldviews. And, even if all members of the public preferred some type of dependency-care regime, there would be defeaters for all but those that provide rather minimal support for dependents and their caretakers. There is no reason to think that a move to sets of policies as the object of justification would allow convergence theorists to show that the kind of dependency-care regime needed to address gender inequality due to the gendered division of labor would be justified for all members of the public.

One might claim that even if there are defeaters for sets of policies that concern adequate dependency-care regimes, there might be sets of policies that concern the welfare interests of members of the public (e.g., policies that include provisions for assistance to the poor) and/or sets of policies that address resources for persons’ agency interests. Sets of policies that focus on welfare interests and/or resources for agency interests might provide a safety net to keep families out of poverty or even provide a basic income. Of course, as we noted above, Vallier thinks that some members of the public will object to policies that are not sensitive to persons’ choices (e.g., like having children or gambling); so, they may also reject sets with policies that are insensitive to persons’ choices. But even if in a set-by-set analysis, a set that includes either welfare assistance for families or a basic income for all members of the public is justified, it won’t be sufficient. Certainly, such sets of policies would provide caregivers with some needed resources, may allow them to avoid some financial dependency on a partner, and may give them more bargaining power in their relationships. However, such sets would not address women’s stigmatization in the labor market or their marginalization in various spheres of social life. Such sets would also not address the fact that labor market work is insufficiently sensitive to workers’ responsibilities as caretakers of dependents or the lack of affordable, high-quality care options available to labor market participants. So, such policies won’t address the gender inequality due to the gendered division of labor and may make it worse insofar as gender norms encourage women to make choices regarding paid and unpaid work that disadvantage them relative to men in terms of power and resources.²⁶

Second, instead of taking the objection of justification for convergence liberals to be fine-grained rules, one might take them to be principles (understood as standards for assessing or designing rules). One might even combine the two ways of varying the object of justification and take the object of justification to be a set of principles. Principles are general in their nature, and, so, one might expect convergence on particular principles or sets of principles to be more likely. However, there is not any good reason to think that varying the object of justification to principles or sets of principles would result in a liberal regime in which the CEPs we claim are needed for gender equality are adopted. Further, it may be difficult to determine the principles or sets of principles on which there would be convergence given both the diversity of real-world views and cognitive limitations.²⁷ Related, in Gaus’s discussion of constitutional structures as the object of justification, he says that in persons’ evaluation of different constitutions, “Members of the Public will ask themselves: ‘how likely is it that this constitutional structure will enact laws outside the eligible set?’” (Gaus 2011, 493–94). That is, with convergence theory, members of the public are always concerned with being subjected to unjustified authority and would be inclined to reject any principle (or set of principles) that might leave them subjected to policies not justified for them. As a result, we don’t see how varying the object of justification from policies to principles or sets of principles or constitutions would make any difference to our evaluation of convergence liberalism’s inability to address the problems of the gender inequality that are due to the gendered division of labor.

4. The sectarian challenge

As we noted in the introduction, our aim is to challenge convergence liberals to show that their view will secure equal citizenship for all members of society, particularly those from historically subordinated groups. We limited our scrutiny of convergence liberalism to one aspect of gender inequality, and we showed that given the features of convergence liberalism and the beliefs and values of moderately idealized members of the public, convergence liberals will not adequately address some significant and enduring forms of gender inequality that are due to the gendered division of labor. We argue the problem is not just that the view will not require the adoption of certain laws and policies needed to adequately address gender inequality given certain conditions, but also that sometimes, the view allows that certain laws and policies central to gender equality be blocked from the set of those eligible for adoption. This means that in a real-world society that adopts convergence liberalism the (1) norms and practices that maintain and perpetuate women’s inequality are left in place and, so, women will be oppressed and (2) sexist ways of life are less costly than gender egalitarian ones.

We now want to address the worry that convergence liberals will claim that the kind of argument we are making is misguided. Convergence liberals aim to show how equal moral relations are achievable through social and legal rules justified for all members of the public and their account, as they say, takes diversity seriously with respect to both real-world pluralism and the challenges it poses for social life (Vallier and Muldoon, 2021). Again, they build their theory from the view that no one is naturally subject to another’s authority and that people have moral liberty to act as they want or as they believe they should unless they interfere with others. People are equal in that no one has more justificatory power than anyone else, and, so, they claim a rule is justified if and only if each person subject to it has sufficient reason given their own evaluative perspective to comply with it. As a result, convergence liberals will claim that our argument is sectarian in the sense that our view of gender equality is based on a particular worldview that some reject. They may also charge that our view permits a kind of authoritarianism in which some persons are subjected to law and policy that is not justified for them.²⁸ Indeed, convergence liberals think it is a virtue of their view that it doesn’t require the adoption of any particular laws and policies (or even the inclusion of any particular policies in the eligible set for addressing some issue), and so they do not even claim that their view guarantees the protection of basic human rights under all conditions. Of course, they claim that that is not cause for concern because “liberal arrangements have special properties that will support them over other political systems” in most circumstances (Vallier and Muldoon 2021). We have argued that the same can’t be said in the case of policies needed to end women’s oppression due to the gendered division of labor.

In response to the charge of sectarianism and authoritarianism, we think that all liberal theories offer some interpretation of the normative requirement that the state treat all citizens with equal concern and respect (Dworkin 1977, 180). Our argument aims to show that convergence liberalism’s interpretation of what equal respect and concern demands is impoverished. As we stated when making the basic case against convergence liberalism, the view lacks the resources to secure equal citizenship for all given its conception of personhood, its formal notions of freedom and equality, and the resulting public justification principle. That is, given real-world pluralism, a society structured in accordance with convergence liberalism will not actually treat all citizens with equal concern and respect. With many other feminists, we hold that liberal states should recognize that persons are needy and vulnerable in the course of a life and, at least sometimes, need various forms of care and, so, caring work is socially necessary and valuable. Further, liberal states should protect those who provide care from various forms of oppression and, insofar as possible, make sure that those who provide care to others have the ability to participate in central spheres of social life without being disadvantaged because they are caregivers. We have claimed that particular policies are needed to fulfill this desideratum. Clearly, liberal states do not currently provide all the protections that we claim are needed, and they would not even if they were arranged in accordance with the demands of convergence liberalism given the views of moderately idealized members of the public. In any liberal state, securing equal citizenship and protecting the vulnerable may demand that some persons comply with rules that they do not (or would not if moderately idealized) accept.

While convergence liberals will claim that our view is sectarianism and unacceptable as a result, we think that the real issue is not whether a liberalism is sectarian but whether it is unacceptably so. Liberalism needs a foundation such that it guarantees for all members of society the social conditions of free and equal citizenship, where free and equal citizenship is, to some degree, a substantive ideal. That makes liberalism sectarian to some extent. It may be that in some circumstances, the social conditions of free and equal citizenship cannot be realized, but the ideal is still the goal. It may be a neat trick to show that some familiar moral rights and/or liberal rights are justified for all members of the public in most conditions, but not all worldviews of those in modern democratic states are compatible with an interpretation of equal concern and respect for all persons such that members of historically subordinated groups will be protected from oppression.

Further, by their own standards, convergence liberals are to some degree sectarian. They offer an account of ‘social morality’ where they posit that persons are interested in moral relationships with each other, and they constrain the reasons that count in a person’s justificatory set to those that a person sees as moral rather than immoral given that person’s evaluative standard. They further exclude evaluative standards based on sadism or masochism. So, even on their view, some reasons are blocked from counting in a person’s justificatory set on substantive grounds. This means that insofar as convergence theorists are offering a minimally moralized account of rules for social interaction, their view is to some degree sectarian too. So, it actually isn’t the case that convergence theorists oppose all sectarianism. They just have a different view of what counts as unacceptable sectarianism, and one that allows them to make that the charge of sectarianism in a wide range of cases.

For convergence theorists, anyone who aims to impose a policy on others that some member of the public has a defeater for (consistent with the morality criterion) is authoritarian and their view is unacceptably sectarian. However, this position is the very object of our critique. And the fact that our view is unacceptably sectarian by the lights of convergence liberals doesn’t show our view actually is unacceptably sectarian or that we are not concerned with sectarianism infringing upon freedom. It does show that we disagree with convergence liberals about what should count as unacceptable sectarianism though.

Hence, our argument is based on a conditional: if you think it is plausible that the interests of persons that we discuss are important for equal concern and respect for all members of society, then convergence liberalism is unacceptable. Of course, one can reject the antecedent, but we think it is hard to do that and claim to care about gender equality.

5. Conclusion

We have argued that convergence liberalism will not adequately address some forms of gender inequality that are due to the gendered division of labor and that this makes convergence liberalism unacceptable. Further, we think our critique shows that any account of liberalism must be based on substantive conceptions of the person, freedom, and equality in order to adequately address the forms of inequality faced by historically, oppressed groups.

Among other things, the COVID-19 pandemic forcefully exposed the persistence and perniciousness of gender inequality with regard to the gendered division of labor, even among those otherwise privileged by race or class. Economists and other social scientists have documented that nearly 3 million women dropped out of the US workforce as a result of the incompatibility of their caregiving responsibilities and participation in the paid labor market.²⁹ Moreover, it is estimated that women’s disappearance from the paid labor market resulted in $800 billion of lost income globally over the course of the pandemic.³⁰ Many academic women have noted their progress towards tenure has been seriously interrupted, and women in STEM fields, already burdened in many ways compared to men, are exiting or failing to promote at higher rates.³¹ The pandemic simply exposed what was already there: vast and structural forms of inequality on the basis of gender, in large part due to women’s social role as caregivers. Any form of liberalism that will not adequately respond to such facts of gender inequality ought to be rejected on that basis alone.