Jump to ContentJump to Main Navigation
Minimizing MarriageMarriage, Morality, and the Law$

Elizabeth Brake

Print publication date: 2012

Print ISBN-13: 9780199774142

Published to Oxford Scholarship Online: May 2012

DOI: 10.1093/acprof:oso/9780199774142.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 17 January 2019

Challenges for Minimal Marriage

Challenges for Minimal Marriage

Poverty, Property, Polygyny

Chapter:
(p.189) 8 Challenges for Minimal Marriage
Source:
Minimizing Marriage
Author(s):

Elizabeth Brake

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199774142.003.0009

Abstract and Keywords

Chapter 8 takes up the problem of implementing the marriage law of ideal theory in a non-ideal world. It addresses concerns that minimal marriage would worsen the lot of the vulnerable, especially women, by eliminating anti-poverty marriage promotion, mandatory alimony and property division protecting the economically dependent, and permitting gender-structured polygyny. Liberals could consistently support transitional provisions for benefits and property division, but liberalism can, and should, also address such problems through legal vehicles other than marriage, especially through education and default rules of financial fairness. Political liberalism holds a number of insufficiently recognized tools for feminists, including rectification, neutrality, and the status it assigns to goods of care and self-respect.

Keywords:   poverty, marriage, marriage promotion, feminism, polygamy, alimony, divorce, property division, ideal theory, liberal feminism

So far, I have been considering what justice would imply in an ideally just society. But we do not live in one. In nonideal circumstances, it may be unjust to implement the results of ideal theory. Sexism, racism, heteronormativity, and amatonormativity must be addressed by any political theory aspiring to relevance. Ideal theory has been criticized for its failure to attend to the most challenging problems of contemporary liberal democracies. Even worse, it has been argued that ideal theory is structurally unable to address these problems. In this chapter, I address specific challenges marriage reform faces in actual societies and show how liberal egalitarianism can consistently address transitional problems of poverty, property division, and polygyny. The example of minimal marriage offers a partial response to feminist critics of liberalism by showing that the consistent application of liberal principles can yield a marriage law that no longer arbitrarily privileges some members of society. Liberal egalitarian principles, consistently applied, hold unappreciated possibilities for radical reform.

“Ideal theory” in ethics refers to theories that construct or derive moral principles through idealizations, such as the idealized autonomous rational contractor of Rawls's original position or Rawls's assumption of strict compliance.1 Assuming ideal circumstances precludes generating principles directly responding to injustices such as slavery, segregation, the Trail of Tears, sexual harassment, domestic violence, and so on. Noting these ideal theoretical constraints, Charles Mills asks, “How in God's name could anybody think that this is the appropriate way to do ethics?”2 Indeed, ethical or political theory that fails to address actual injustices is poorer for that omission, risking, as Mills charges, becoming ideological.

Rawls's theory of justice has been criticized for its inattention to real injustices. For example, Seana Shiffrin suggests that it is “surprising that the principles of justice (p.190) do not directly protect against racial discrimination.”3 It has been suggested that his theory is systematically unable to address such issues. Rawls's characterization of the contractors as autonomous results in prioritizing the liberties and thereby precluding state interference in “private” sources of oppression.

However, the theory must be able to address actual injustices. Even were Rawls's two principles of justice implemented tomorrow, existing inequalities would continue to create further inequalities. Some individuals will always fail to comply, leading to the need to design new institutions to prevent and correct injustice. In any case, the principles are not likely to be implemented; reform comes piecemeal, and the ideal will always remain aspirational. Liberals need to be able to give a consistent account of how to respond to actual injustices.

Principles of justice derived from idealizations can be powerful tools for change. Despite their flaws, they reflect an ideal of moral equality and a fair bargaining situation unavailable in actuality and for this reason can suggest significant possibilities of reform. But taking the ideal as guide requires an account of how to move from the actual toward the ideal. Where reforms could cause new harms or injustices, there is reason for concern.

I. Poverty

This section addresses concerns that minimal marriage would exacerbate poverty for women and children by ending “traditional” marriage promotion and benefits. As noted in the Introduction, U.S. federal law addresses the poverty of single mothers through marriage promotion policies such as state commissions and proclamations, divorce law, marriage “education,” state tax policies and cash assistance, Medicaid policy, vital statistics, and youth education. Temporary Assistance for Needy Families legislation, which bankrolls such policies, addresses the correlation between poverty and single motherhood by trying to get mothers to marry: It aims to “end the dependence of needy parents on government benefits by promoting… marriage.”4

Incentives to adopt promarriage policies have been effective. Since this legislation was enacted, a number of states have revisited no-fault divorce law, considering legislation to abolish it. Louisiana, Arizona, Arkansas have introduced “covenant marriage,” an option imposing longer waiting periods, and in which “a no-fault divorce cannot be granted unless both parties agree… and no children are involved.”5 With “school-based marriage education” and “abstinence-until-marriage education,” “states are also targeting the marriage message to youth.”6 States receiving abstinence education funding must provide educational programs adhering to “an abstinence-until-marriage message” as defined in the Social Security Act, teaching that abstinence is socially, psychologically, and medically (p.191) beneficial, that “sexual activity outside the context of marriage is likely to have harmful psychological and physical effects.”7

It might be argued that minimal marriage would not only end such marriage promotion, it will decrease “traditional” marriages and thereby exacerbate poverty. But “traditional” marriage promotion is an inefficient antipoverty program. Trying to address the poverty of single mothers through marriage is like trying to shove an escaped elephant into a cage it has outgrown. The conditions that, according to historian Stephanie Coontz, have led to higher rates of divorce and lower rates of marriage—women's economic independence, birth control, and the idea that marriage should be emotionally satisfying—are enduring.8 At least one-third of U.S. children are now being reared outside marriage. The inefficiency of marriage promotion as an antipoverty program is even more obvious when one considers that many two-parent families are in poverty, and poorer families have less to divide on divorce.9

Furthermore, marriage promotion policy ignores sources of poverty in unrelated injustices. Justice as well as efficiency demands that society address economic disadvantage to caregivers, the drop in working-class real wages, racism and the legacy of slavery and Jim Crow, the effects of racial profiling, the lack of decent affordable housing, and the gendered division of labor. For example, as discussed in 5.iii, the African American Healthy Marriage Initiative targets African-Americans but seems to ignore extraordinarily high incarceration rates for African-American males. And marriage promotion seeks to alleviate female poverty through marriage rather than job training and subsidized child care. Indeed, its object is that poor women secure a breadwinner, not better-paid employment: A program funded under this legislation in Allentown, Pennsylvania, offered job training programs to men only.10 Justice and efficiency require addressing root causes of poverty, such as unequal pay and racial discrimination.11

Another concern may be that minimal marriage will increase the number of single parents. Single parenting is correlated with poverty. But marriage promotion is a poor solution: It promotes female dependency, making women and children vulnerable. The economic costs of single parenting must be weighed with the detrimental effects of high-conflict and abusive marriages. Given widespread abuse and violence within marriage, and the additional harms of high-conflict marriages, women and children may often be better off outside marriage. Further, marriage promotion benefits only children whose parents marry. Benefits should be directed at children in poverty, not married parents. Minimal marriage, in contrast, would help single parents by increasing their marital options.

A more pressing concern is that, in an ideal liberal egalitarian state, minimal marriage would bring no health care entitlement or other financial benefits. But immediate abolition of marital health care, pension, and other benefits—with (p.192) no alternative provision—would harm many. This might seem to pose a dilemma for liberals: Distributing health care and other benefits through marriage, as Card argued, unjustly excludes the unmarried and subsidizes the married at their expense. How can a transitional stage retaining benefit entitlements be justified? Why should single taxpayers or employees subsidize spousal health insurance and other benefits? In the case of dependency frameworks, the state can provide an answer, explaining why parental leave and other parental support do not discriminate against nonparents: Dependent children provide the state's future, and justice requires that children receive care. But why should a spouse who is capable of employment, but chooses not to work outside the home, receive financial benefits from the state or third parties?

A somewhat unsatisfactory response, considered already in Chapter 5.ii, is that while such benefits unjustly exclude the unmarried, providing health care or pensions unjustly to some comes closer to the universal provision that justice requires than unjustly not providing them at all. A transitional stage in which minimal marriage continues to carry such benefits would continue to exclude some unjustly, but it would help many others. This reflects the compromises that implementing ideal theory demands.

But we can give a more satisfactory answer as to why such benefits should be maintained for some time. First, the state has induced reliance on these benefits. People, and not only economically dependent wives, have made choices on their basis, and it would be unjust—tantamount to a violation of contract—to remove them. Entitlements are not fixed a priori, but are constructed through law and policy.

A second response applies specifically to dependent wives: Because their choices have been shaped in a system of amatonormative, patriarchal, and state-sponsored oppression, protecting women against the effects of these choices is a matter of rectification. These women might have been better off had they been encouraged to seek economic independence, not dependence; but given that the state has incentivized and promoted them into dependent marriage, it owes them rectificatory compensation. Given that, as late as 2001, some abstinence-education curricula have taught children gender-structured domestic roles (see Chapter 5.i), a long transition will be required to outlast the effects of state-sponsored gender hierarchy and amatonormativity.

These reasons also justify extending such benefits to minimal marriages, even when they do not involve dependent wives. Imagine that Anna receives health care entitlements through her current marriage. If she cannot receive this benefit in a new minimal marriage with members of her care network, her choice to leave her current marriage and enter the new one is unduly restricted—unduly in light of the state's role in inducing reliance and shaping her choices through amatonormative pressure. (My point, to be clear, is not that Anna's choice was politically unfree due to such pressures, but rather that the state owes her compensation for having unjustly (p.193) sponsored discriminatory role pressures. It ought not to have interfered with her choices in this way.) Had she not been induced to rely on marriage to provide this benefit, she might have sought to be able to provide it for herself, and hence be able to enter the new marriage without losing healthcare. These are reasons of justice to extend the benefits throughout minimal marriage, not only to current recipients like Anna, but to the next generation, who have begun to form their preferences in light of existing structures.

While minimal marriage is not any more of an anti-poverty program than marriage promotion, it would benefit the worst-off more by making benefits more widely available. Current marriage promotion aims to increase women's economic dependence on men and so may exacerbate abuse. In contrast, minimal marriage, assuming that for a transitional period it will offer benefits such as healthcare, allows women more marriage options and thus greater bargaining power.

II. Property and Divorce

Marriage reform must create good exit options. Exit options affect whether the outcome is just, when a marriage ends—whether the resulting situation comports with equality—and they affect justice within a marriage while it endures—as we saw in Chapter 5.1, lack of exit options facilitates abuse as well as power inequality. The need to ensure good exit options is challenging for minimal marriage because, in an ideal society, minimal marriage would not include default marital property arrangements that protect dependent spouses on divorce. The reason for this is that the law, recognizing the diversity of relationships in society, cannot assume a dependency relationship between married people. But property division and alimony or spousal support is the main means of assuring that economically dependent wives have good exit options. However, I argue that these protections be removed from marriage and replaced with universal default rules.

Minimal marriage would contractualize property arrangements, meaning that spouses could determine property division on divorce through individual contract. However, some feminists have argued against contractualizing marital property because marital property law protects women. As discussed in Chapter 5.1, Okin documented how women become economically vulnerable through marriage when they subordinate their economic independence to child care and household work. Such choices disadvantage women on divorce, and these disadvantages are compounded by the greater likelihood that children will remain with the mother after the divorce, bringing her greater financial costs. In light of these inequalities, Okin argues that in the transition to a gender-neutral society, divorce law protects economically vulnerable women in gender-structured marriages. Contractualizing marital (p.194) property would eliminate such protections. Okin argues for a legal requirement that all spousal earnings be equally owned by both parties, and for mandatory alimony and property division on divorce. She argues that because the pattern of such choices systematically affects women, fair equal opportunity requires correcting the accruing disadvantages. In Rawls's later work, he too justifies spousal support and property division by appeal to the political value of women's equality in structuring family law.12

Other feminists have pointed out that, given existing social inequalities, women are not in a fair bargaining position to negotiate marriage contracts. Martha Minow and Mary Lyndon Shanley write that

one of John Stuart Mill's great insights in The Subjection of Women was his observation that the decision to marry for the vast majority of women could scarcely be called “free.” Given women's low wages, scarcity of jobs, and lack of opportunity for higher or even secondary education, marriage was for them a “Hobson's choice”: that or none. Even the “I do” of someone very much in love and desirous of marriage does not in-and-of-itself guarantee freedom.13

Mill's insight suggests that given background inequalities, social disempowerment, and limited alternatives, contract does not assure women's free, in the sense of unconstrained, consent. If a woman's male partner earns more and has greater social power (by virtue of gender role hierarchy), and both expect him to be the primary breadwinner, a woman settling a marriage contract may be pressured to agree to terms she does not want: the higher-earning and more socially powerful party will have greater bargaining power. Freedom of contract is compatible with pressure to make disadvantageous choices. Nor, as Minow and Shanley point out, does prenuptial contract address dependencies arising within marriage over time. Insofar as women are likely to become economically vulnerable in marriage, they will be better protected economically by mandatory property division and alimony than by contractualization.

These concerns may justify a transitional stage retaining alimony, on liberal principles of equal opportunity, and for the reasons given in the last section for retaining benefits (the points regarding induced reliance and compensation apply here). But while mandatory alimony and property division address the inequities arising from gender-structured marriage, they do not address social pressures and gender roles that create these inequities. Furthermore, by assuming economic dependency in marriage, law may encourage women to become dependent.

There are other weaknesses to relying on alimony to protect the vulnerable. One concerns efficiency. The amount of money received and the percentage who receive it are particularly low for poor women. Marriage contracts, particularly alimony (p.195) provisions, have tended to be less well-enforced than other contracts.14 Also, mothers earn less than childless women (“mothers earn about 70 percent of the mean wages of men, and childless women earn 80 to 90 percent”), so pursuing policies targeted to the worst-off like raising the minimum wage, providing daycare, and Anne Alstott's “caretaker resource account” would help the worst-off more than alimony.15

There is also a problem regarding the grounds for interpersonal obligation in mandatory alimony.16 Addressing systematic gender discrimination in employment, wages, and social pressures through alimony risks injustice to individual men. If the reason for alimony is equal opportunity, why should individual men be held responsible for the inequities of the social system? This is especially pertinent because the husbands of the neediest women are likely to be poor themselves. Mandatory alimony based on equal opportunity needs to be sensitive to the position of both parties. However, in nonideal circumstances, the balance of reasons may favor burdening a well-off husband with the costs of his ex-wife's job training over allowing her to enter poverty.

But if overall justice does require such transfers, then default rules governing property division on exit from financially dependent relationships can be enacted independently from marriage.17 Separating such rules from marriage would extend protection to those outside marriage. If the rationale of property division is protecting the vulnerable against inequalities arising in relationships, there is no reason to limit protection to the married. The duration of the relationship, the extent of the dependency, the increase in economic inequality, and so on, are more relevant to determining whether support is owed (but not gender, for reasons I make clear below). Enacting default rules addressing inequalities within financially enmeshed intimate relationships does not assume dependency in marriage, but applies directly to cases of dependency. Separating such rules from marriage law may gradually alter the expectation that women make themselves dependent in marriage.

Such default rules would apply involuntarily. This allows the separation of minimal marriage rights and financial obligations suggested at the end of Chapter 7.iii. Minimal marriage entitlements will be voluntary—voluntary entry will be efficient because people will be motivated to apply for these entitlements as the need arises. Support obligations, however, would be imposed by default, so that they would protect vulnerable members of unregistered relationships. This would offer exit options from unmarried economically dependent relationships as well as dependent marriages.

The worry might arise that imposing such default rules involuntarily would interfere with freedom of contract and fail to treat adults as responsible for their choices. If a woman, due to religious views or her lower earning capacity, wishes to make a “vulnerability contract” in which she cedes rights to support while planning to forgo paid work, should such a contract be enforced when parties separate? (p.196) Should women's choices to prioritize child care over paid work be treated as private choices for the costs of which they are solely responsible? Can parties opt out of default rules?

Liberal egalitarianism does not require unregulated free contract and so can admit involuntary mechanisms to protect against dependency. Provisions preventing contracts from eventuating in one party's impoverishment are compatible with liberal egalitarianism. Freedom of contract is not a basic liberty, and the difference principle restricts it.18 Default rules can be justified, as Rawls suggests, by political values such as women's equality.

But default rules can also be justified by other legal mechanisms, many of which already exist in law. Rather than enacting specific laws for relationships, it might be possible to address dependency through general principles derived from contract law. Induced reliance might require that when one party makes herself dependent on another while reasonably relying on his continued support, he is obligated to aid her in becoming self-sufficient if the arrangement ends. Fineman argues for abolishing marriage and shifting relations now governed by marriage to contract, labor, and tort law. (Spouses are still exempt from U.S. labor law protections, despite the fact that they may work for one another.) Fineman argues that compensation for contributions to the other party's career might be justified by appeal to mechanisms independently available in law: “the interests of a cohabitant who contributed to the accumulation of wealth for the other, even if she did not have a contract, would be protected to some extent by default and equity rules. The general regulatory rules found in equity (such as unjust enrichment or constructive trust), partnership, and labor law could provide rules for decisions in disputes involving sexual affiliates.”19 Default rules might be justified through independently justified legal principles (thus obviating the need for specifying gender).

Applying these general provisions might be even fairer than a law targeted at dependency. Rules correcting economic dependency might not compensate the unpaid labor supplied by “traditional” wives. But provisions inherent in contract can. Support liability on grounds of opportunities forgone and contributions to the other's career might be justified by appeal to induced reliance and verbal contracts. Fineman also suggests that abolishing marriage would actually expand the possibilities of legal recourse under tort and criminal law; she envisions new tort law addressing harms that occur between sexual intimates. But this would also be compatible with minimal marriage.

Property division and support are crucial to provide exit options from marriage. The threat of losing benefits, such as health care and pensions, can also constrain exit; for this reason, I argued in the last section that such benefits be retained for a transitional period. While these benefits can constrain the choice to stay married, (p.197) removing the benefits actually reduces options overall. Extending them to minimal marriage increases women's marriage options and hence their bargaining power

A final note: As discussed in Chapter 2, the more restrictive the exit options, the more pressure to maintain a marital commitment. It might thus be thought that protecting exit options would reduce commitment to marriage. But, as argued there, commitment to abusive or exploitative relationships should not be protected. Moreover, internal commitment is not guaranteed by external pressures. Marriage may be a poor precommitment strategy, encouraging emotional laziness. Protecting exit options might improve marriages by giving spouses reason to keep each other happy—a husband who knows his wife can leave will be more motivated to share in domestic labor, for instance. Ensuring adequate exit options protects women and children, but it may also improve marriages.

III. Polygyny

Gender-structured polygyny presents another challenge for liberal feminism. It is important to be clear on terminology. Polygamy includes both polygyny—one husband and multiple wives—and polyandry—one wife and multiple husbands. Polygamy may be seen, as Emens suggests, as a type of polyamory, but polyamory takes many additional forms, ranging from a small interconnected group to a molecular structure with multiple nonoverlapping relationships to the “V” or spoke structure typical of polygyny (in which the husband is the center and maintains a separate, exclusive relationship with each wife).

Polygyny presents a problem because, as Thom Brooks reports, women in polygynous marriages suffer low self-esteem, depression, and lower marital satisfaction, at a higher rate than women in monogamous marriages.20 (Because my argument for allowing suboptimal parenting structures in Chapter 6.iii applies here, I set aside effects on children that do not violate justice.) Andrew March questions these data; indeed, we should keep in mind that correlation is not causation. It is possible for polygyny to benefit women, as Elizabeth Joseph, a working lawyer and polygynous wife, claims. Joseph praises the arrangement in feminist terms as building strong relationships between women and providing a way to share domestic labor (albeit with the other sister wives, not the husband!), which, in her case, facilitates combining motherhood and a law practice.21 And egalitarian or non-gender-structured polygyny involving a group of bisexual women and one man, on an equal footing, is possible.

However, if egalitarian forms or cases such as Joseph describes are relatively rare, polygyny creates an apparent tension for liberal feminists. This is because political liberalism requires, as I argued in Chapter 7.ii, recognizing the variety of different (p.198) relationship forms if any are recognized. For the same reasons that the liberal state should not discriminate between same-sex and different-sex marriages, it should not discriminate between monogamous or polygynous, polyandrous, or polyamorous marriages.22 But if polygyny does harm women, this seems to present a dilemma for liberal feminism. One response, suggested by John Rawls, is to argue that polygyny conflicts with the political value of women's equality and so there is public reason against it.23 However, this strategy faces a weakness.

The problems attributed to polygyny are all also problems of patriarchal monogamous marriage! If the choice to enter polygyny is said to be unfree due to social pressure and coercion, this can be said of monogamous marriage. John Stuart Mill said as much: “[I]t must be remembered that this relation [polygyny] is as much voluntary on the part of the women concerned in it, and who may be deemed the sufferers by it, as is the case with any other form of the marriage institution; and however surprising this fact may appear, it has its explanation in the common ideas and customs of the world, which, in teaching women to think marriage the one thing needful, make it intelligible that many a woman should prefer being one of several wives, to not being a wife at all.”24 If polygyny is said to harm women, this can also be said of gender-structured male-female monogamous marriage and cohabitation.25 Although polygyny may, as Brooks claims, be correlated with greater harms than monogamy is, the gap may disappear if we focus on monogamy in small patriarchal religious communities such as those within which polygyny tends to be located in the United States. In any case, as Susan Maushart reports, women in monogamous marriages suffer “more nervous breakdowns, inertia, loneliness, unhappiness with their looks; more insomnia, heart palpitations, nervousness, and nightmares; more phobias; more feelings of incompetence, guilt, shame, and low self-esteem” than unmarried women.26

By parity of reasoning then, this argument against polygyny suggests we should cease to recognize male-female monogamy. Brooks tries to bar this conclusion by arguing that polygyny is structurally inegalitarian as monogamy is not. But, first of all, in some communities—and in U.S. law within living memory—monogamy is structurally inegalitarian in its spousal roles, just as polygyny is: It gives the husband the dominant role. It might be said that not all monogamous marriage is patriarchal—but not all polygyny need be. In addition to cases such as Joseph's, egalitarian polygyny is possible.

Brooks gives two reasons for the claim that polygyny is structurally inegalitarian. First, it is asymmetrical in that the husband can bring a new wife into the marriage without other wives’ consent: Only he “can choose who will join or leave the relationship through either marriage or divorce.”27 But in minimal marriage, the wives could legally take on other spouses, building a molecular structure. There is no legal (p.199) asymmetry. Wives could choose to marry, or divorce, sister-wives. Brooks would presumably respond that they might not be empowered to do so in fact, despite having the legal right. But why not? If the answer is social pressure to marry and remain married, lack of exit options, or economic dependency, all of these features pertain to exclusive monogamy, in which a wife wanting to take on another spouse, or leave, may be constrained in doing so for these same reasons.

Second, Brooks argues that polygamy is structurally inegalitarian because it discriminates against gays and lesbians due to its gender-structured nature. However, as we have seen, many forms of relationship are possible. Minimal marriage allows all-women or all-men adult care networks or group marriages, or groups composed of lesbians, gays, and bisexuals. When all these options are on the legal table, the fact that some involve sex difference does not make their recognition discriminatory. Polygamy, like male-female monogamy, is only structurally discriminatory against gays and lesbians if same-sex marriage, including group marriage, is unavailable.

The problem with polygyny is not the form of relationship—as Joseph shows, the form itself can be relatively benign for women—but the injustices sometimes found in the community: Child abuse, underage marriage, lack of adequate education, social coercion, and indoctrination. However, the first three of these are crimes, and they can occur within monogamous marriage, too. Children should be adequately protected against abuse, and—legally—they should receive an adequate education. These problems should be dealt with through criminal law and social work, not by excluding this form of marriage. Furthermore, robust safeguards ensuring that consent is voluntary can be designed.28

Not only does denying marital status to polygyny treat polygynists unfairly (in the absence of compelling reason for differential treatment), but it threatens to deny recognition to polyamorous groups or care networks. It is not clear how these could be distinguished in law, except by specifying the polgynous structure. But as this structure can be benign, it would be more just to exclude specifically patriarchal or harmful marriages. But this leads back to the parity of reasoning argument against male-female monogamy—should all gender-structured marriages or marriages in which women suffer depression, and so on, be deprived of recognition? Should marriages within religious traditions that subscribe to gendered spousal roles be deprived of recognition? These consequences are simply illiberal. Law should not enforce such hierarchies, but it should not make marital recognition contingent on full egalitarianism between parties. Such choices between consenting adults should be respected.

There is a notable exception. In my view, law should interfere in family arrangements—whatever the form of the marriage—when they threaten children's developing self-respect. But given that children need continuity of care, that their self-respect may also depend on family pride, and that preventing certain (p.200) parent-child communications is impossible while children remain with the parent, the balance of reasons suggests that protecting children's self-respect is best done through public education, of both children and prospective parents.

A liberal state should ensure education fosters children's developing self-respect and respect for others. Self-respect is a primary good, and respect for others contributes to state stability. The primary good status of care, and its connection to self-respect or self-esteem, also implies requiring relationship education. Teaching children that different relationships types exist and are treated equitably by the state protects the developing self-respect of children who are or will be sexual minorities as well as inculcating respect for difference. While parents in closed religious communities may protest, parental rights cannot allow deforming children's future autonomy and self-respect; religious freedom does not give the right to jeopardize “perhaps the most important primary good” (self-respect) or devastate children's life chances. Such education must give children knowledge and skills enabling them to depart from their closed religious communities.

Finally, in most U.S. states, individuals are free to cohabit polygynously anyway.29 While critics charge that recognizing these arrangements as marriages will encourage them, increasing harms to women, it can be responded that recognizing polygyny would give multiple wives protective rights, such as alimony and property division on divorce. However, under minimal marriage, recognition would only give wives the more limited rights of minimal marriage; benefits would presumably be capped for efficiency, and spousal support and property division would apply independently under default rules, as discussed above.

In sum, the gravest harms associated with polygyny in closed religious communities should be addressed through education, robust consent procedures, and criminal law enforcement. Its other problematic features are not sufficiently different in kind from existing male-female monogamy to justify differential treatment. Finally, prohibiting, or failing to recognize, polygyny risks penalizing other, egalitarian, forms of group relationship. Even if, as Brooks claims, polyamory tends to patriarchal polygyny, there are still, as we have seen in Chapters 4 and 5, significant numbers of people in nonpatriarchal group relationships or networks requesting legal and social recognition.

IV. Liberal Feminist Resources for Nonideal Theory

The viability of a liberal feminist position on marriage is important because of the concerns of some feminists that liberal feminism is untenable. Catharine MacKinnon argues, repurposing Marx, that liberal freedoms protect patriarchal power: While they appear to have no gender bias, in fact they favor men. The supposedly neutral state is male.30 In part, she means that laws are perverted in their application, as (p.201) when, under antipornography laws, customs officers impound feminist books, but not violent pornography. But she also means that apparently gender-neutral laws are subtly discriminatory. Liberties are protected in exactly those areas that foster women's oppression: Freedom of speech protects pornography and hate speech; “privacy” protects unequal divisions of domestic labor, domestic violence, and exclusion of health coverage for abortion and contraception. The liberties are held equally, but in practice, their exercise systematically benefits men more. If ending oppression requires interference in areas protected by liberties of free speech, free association, freedom of conscience, and freedom of religion, then liberalism appears systematically unable to do so.

The state indeed has a double standard, when it refuses to “intervene” in parental teachings that degrade women while distributing abstinence curricula that interfere with other parents’ egalitarian aspirations for their children; in defining allowed uses of “private” reproductive functions and contraception, while withholding funding from abortion and contraception due to their “private” nature. Marriage has been the primary means of constructing privacy oppressively, with its private sphere historically protected from justice, thereby facilitating rape, abuse, and exploitation. Theorists such as Rawls have continued to resist the full extension of liberal principles into the “private”; while he acknowledges that arrangements within the family should not issue from or lead to injustice, he also relegates the internal regulation of the family to “natural affection and goodwill” and suggests that parents have rights to infuse their children with gender-structured teachings.31 Thus, feminists are right to be suspicious of a marriage law demarcating a “private” zone within which the state may capriciously intervene or not, or for the state to define who counts as a family and who does not.

These problems connect to the criticism of ideal theory. Liberal freedoms are derived in an idealized context, featuring idealized rational, autonomous agents, imagined under idealized circumstances. But, as exercised by actual agents who are imperfectly rational and subject to social pressures and constrained choices, these freedoms can contribute to oppressive outcomes. Actual injustices affect choices; the threat of stranger rape may lead women to choose monogamous relationships for perceived protection, workplace discrimination may make the choice to stay home easier, and the historical exclusion of women from certain fields may make those fields seem naturally masculine. Because liberal theory focuses on the individual in isolation, it fails to see how social structures may be systematically disempowering. While liberal egalitarianism can attend to economic inequalities between the worse-off and better-off, it does not track how these inequalities correspond to group membership on the basis of sex and race, or inequalities in power such as gendered inequalities in leadership roles.

(p.202) Its inability to address inequalities between persons with unequal attributes is the basis of MacKinnon's criticism of the liberal conception of equality:

It cannot recognize that every quality that distinguishes men from women is already affirmatively compensated in society's organization and values. … Men's physiology defines most sports, their health needs largely define insurance coverage, their socially designed biographies defined workplace expectations and successful career patterns, their perspectives and concerns define quality in scholarship, their experiences and obsessions define merit, their military service defines citizenship, their presence defines family. … For each of men's differences from women, what amounts to an affirmative action plan is in effect, otherwise known as the male-dominant structure and values of American society.32

Rawls's principle of fair equality of opportunity, for example, implies that men and women with similar abilities should have fair equal opportunity to access those positions: “[T]hose with similar abilities and skills should have similar life-chances.”33 But the principle does not correct for inequalities arising when boys and girls, differently encouraged by parents, teachers, peers, and society, develop different abilities.

Furthermore, liberalism protects choices that, as discussed in 5.i, systematically disadvantage women, such as the gendered division of domestic labor. Liberalism protects choices based on “adaptive preferences,” those developed in an oppressive context and responsive to the conditions of oppression. Examples are women's preferences to give an inordinate amount of attention to their appearance, to be a nurse rather than a doctor, or to stay home with children rather than compete in the workplace. These preferences are as “authentic” as any, so that they are protected under liberalism, even though their effect may be to disadvantage women.34 The feminist criticism might be restated thus: Although liberal egalitarianism has resources to extend protection to dependent wives and wives who can prove contributions to a husband's career, it fails, theoretically, to distinguish disadvantages resulting from oppression from disadvantages resulting from bad choices, laziness, and so on. It can thus not address oppression, because it reads its effects as “choices.”

However, I have shown a number of ways in which liberalism can respond to these charges. Protections against the consequences of adaptive preferences can be justified as rectificatory compensation and under induced reliance when the state has itself supported those preferences and provided incentives to develop them, as it has in the oppression of women, nonwhites, and gays, lesbians, and bisexuals. While Rawls does not take rectification as a central topic of justice, there is no principled reason why a liberal egalitarian account of rectification could not be given. Furthermore, (p.203) the argument that material caregiving is a primary good, and that the state should provide structures to protect it and those who do it, justifies dependency frameworks providing state support for caregivers and protecting their equal opportunity. Political liberalism, I have argued, requires such policies, which are also crucial to protect women. Finally, education must protect children's self-respect.

Liberalism can do much more to address oppression. Ideal theory itself, as the defense of minimal marriage shows, can prompt radical change by modeling how the world could be—if the state were truly neutral, or if we truly bargained as equals. Liberalism also has other resources for addressing oppression, which may guide minimal marriage law's implementation in a nonideal context.

First, Rawlsian liberalism requires nonideal circumstances to be considered at the legislative stage. Ideal legislators, only partly behind the veil of ignorance, can address real-world oppression. Legislators do not know their own positions, but know “general facts about their society,” including facts about past and present oppression. They must also take into account prospective inequities. In framing legislation to implement the difference principle, “the full range of economic and social facts are brought to bear.”35

Second, fair equal opportunity, implemented at this stage, can address ways in which the basic structure—including institutions of family and work—treats men and women differently. For example, employment practices penalize parents, and women bear the bulk of parenting, even if they have the same abilities and career aspirations as comparable males. Because “reproductive labor is socially necessary labor,” parenting is part of the basic structure, to which the principles apply.36 Fair equal opportunity implies that parents should be compensated, as with caregiver resource accounts, or employment be made more compatible with parenting, as through subsidized child care, and that men as well as women should be entitled to take parental leave.

Third, the first principle of justice requires that the social bases of self-respect should be distributed equally—indeed, they may be the most important primary good. This has tremendous implications. Racism, sexism, and other forms of discrimination undermine self-respect; even if adults can see themselves as free and equal citizens while inhabiting hierarchical roles in private, children cannot. Education protecting children's self-respect would include feminist and antiracist education teaching children their equality as citizens and bringing the wrongness of discrimination to light.37

Fourth, the requirements of rectification are also demanding. Major structures of oppression—sexism, racism, heterosexism, amatonormativity—have been constructed and enforced by the state. The state oppressed women by defining the private sphere and pressuring private choices. After World War II, propaganda encouraged (p.204) women who worked outside the home during the war to quit their jobs and return home. Married women began to achieve full legal rights and to escape gendered legal responsibilities only in the 1970s. Current U.S. marriage promotion and abstinence-until-marriage education promote gender roles and amatonormativity. These policies clearly violate neutrality and public reason. For example, state media campaigns “extol the virtue of marriage” in Arkansas and Oklahoma; West Virginia provides cash payouts to married couples; and welfare workers are “encouraged to discuss marriage with their clients.”38

Critics of liberal feminism allege that it cannot distinguish women's choices under oppression from other costly choices. But it can. Current distributions reflect past state injustice. What differentiates the “traditional” wife's choice from the beach bum's is the history of state incentives pressuring her to make that choice: Until the 1970s U.S. state law required wives to carry out domestic duties and defined them as “helpmeets”; earlier in the century, law allowed employers and educators to exclude women; federal monies still support curricula that teach gender stereotypes. Governments encroach on citizens’ rights when they use their coercive power and authority to influence such choices, a fortiori when the pressures are inegalitarian. True, the choices made are politically free, but the state owes compensation for the costs incurred by women due to its unjust pressure to take on gender roles. The state is and has been instrumental in promoting oppressive norms and depriving women of full equal rights—as it has racial minorities, lesbians, gays, and bisexuals. It is not enough to repeal unjust laws or end patriarchal, heteronormative, and amatonormative propaganda—the state must provide compensation.

Given this history, it is proper for the state to use law to reform attitudes. In contrast, Tamara Metz, in her argument for abolishing marriage, defends a prohibition on state intervention aimed at affecting beliefs, as opposed to behavior. But, for example, civics education for schoolchildren justly attempts to affect beliefs about race, in part as compensation for past state racism. Where the state has sponsored racist beliefs, simply ceasing to do so without compensatory action aimed at correcting those beliefs is not fully just. By parity of reasoning, since the state has fostered discrimination against same-sex couples, polyamorists, group marriages, care networks, and friends, the state should compensate for such injustice by fighting discriminatory beliefs and attitudes—for example, by recognizing these relationships as marriage.

Finally, neutrality and public reason are powerful tools for reform. MacKinnon shows that the actual state is nonneutral and that false “neutrality” has served gender bias. Past non-neutrality requires rectification that corrects its costs. Thus, heterosexist curricula might be rectified through curricula treating all relationships equally, through providing support groups and apologies for those harmed by heterosexism, and so on. The proper implementation of neutrality and political liberalism would (p.205) remedy bias by excluding as reasons for policy comprehensive doctrines that incorporate biased views. Taking political liberalism seriously in light of feminist social theory has far-reaching implications often unrecognized by liberals and feminists. It requires the state to root out its own sexist, heteronormative, and amatonormative assumptions. Minimal marriage is one example of the extensive change which that would require.39

V. Concluding Remarks

I want to conclude by emphasizing the feminist attractions of minimal marriage. Unlike current marriage, it involves informing prospective spouses of their rights, the terms of the agreement, and its implications. Arguably, equal opportunity and rectification for past discrimination require educating women about their potential economic vulnerability in gender-structured relationships. Information about the likely consequences of their choices might lead women to resist exploitative relationships.

Second, and more distinctively, minimal marriage gives women more marriage options, increasing their bargaining power. Along these lines, economist Gary Becker argued that polygamy, in a context of liberal rights, increases women's bargaining power.40 His ideal models do not take account of the pressures on choice that affect women's bargaining power in a gender-structured society. However, the idea is more convincing with regard to minimal marriage: The increased marriage options of minimal marriage would open alternative, potentially more egalitarian, relationship models to women and therefore increase their bargaining power in negotiations with men.

Finally, minimal marriage denormalizes heterosexual monogamy as a way of life. In this respect, I consider my position responsive and sympathetic to lesbian and queer critiques of marriage such as Claudia Card's, Paula Ettelbrick's, and Drucilla Cornell's. By extending marriage to all caring relationships, minimal marriage affirms difference. It does not mark some relationships as “legitimate.” Its rationale is to support the caring relationships individuals choose, not to distinguish among them. But minimal marriage will do more than abolishing marriage to combat amato- and heteronormativity because it makes new options salient.

This has a further implication. Social pressures surrounding different-sex monogamy contribute to women's economic vulnerability by promoting “traditional” wifehood. Minimal marriage removes state endorsement from “traditional” marriage, and over time this will change people's aspirations. I have drawn attention to how state marriage promotion reinforces oppressive social pressures; political liberalism, properly implemented, might combat them.

(p.206) “Traditional” marriage promotion is problematic because it takes a single type of relationship as good for everyone. But my argument has suggested a need to recognize different kinds of caring relationships and secure their social bases. Policies supporting caring relationships extend to education, in order to support children's developing self-respect, to remedy the effects of discrimination, and to prepare children for diverse caring relationships as adults. They could also lead to policies supporting unpaid caregivers in the home. On my argument, care is a matter of justice within political liberalism, and this is perhaps the most powerful of the strategies surveyed here for making society fairer for women.

Notes:

(1.) Rawls uses “ideal theory” in a narrower sense to refer to his assumption of strict compliance; see Theory: Revised, pp. 7–8. In the broader usage I follow Mills, “Ideal Theory,” though he distinguishes different senses of ideal theory.

(2.) Mills, “Ideal Theory,” p. 169.

(3.) Shiffrin, “Race,” pp. 1645–6. She also notes that Rawls does not address discrimination on the basis of gender, sexual orientation, and disability.

(4.) “State Policies to Promote Marriage,” p. 1; see Introduction, fn. 23. See also the 1996 U. S. Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), Title I, Section 101, Findings. TANF comes under PRWORA legislation.

(5.) “State Policies,” p. 8. Georgia has similar legislation. Such legislation has been pending or under consideration in at least thirteen other states since 2001.

(6.) “State Policies,” p. 22; in 2002, forty-nine states had accepted funding for abstinence-only education, although the number declined in subsequent years.

(7.) Social Security Act, Title V, Sec. 510.

(8.) Coontz, Marriage, Part Four; cf. Garrison, “Cooperative Parenting,” p. 268.

(9.) Alstott, No Exit, p. 8; cf. Cave, “Harm Prevention”; Vanderheiden, “Why the State.”

(10.) Legal Momentum, Annual Report 2005.

(p.223) (11.) On the point that poverty is attributable to social problems other than divorce, see Okin, Justice; Vanderheiden, “Why the State”; Young, “Mothers”; Fineman, Autonomy Myth, Chapter 3.

(12.) Okin, Justice; cf. Rawls, “Idea,” pp. 792–793.

(13.) Minow and Shanley, “Relational,” p. 11, their citation of Mill omitted; cf. Shanley, Just; Pateman, Sexual Contract; Hartley and Watson, “Political.”

(14.) See Shultz, “Contractual,” pp. 232–240; Weitzman, “Marriage,” pp. 1185, 1194–7.

(15.) Alstott, No Exit, p. 24.

(16.) Lucinda Ferguson, oral presentation, Society for Applied Philosophy annual conference, 2007.

(17.) Sunstein and Thaler, “Privatizing,” p. 384; Stark, “Marriage,” p. 1522, suggests precluding increases in economic inequalities during marriages.

(18.) See Rawls, Theory: Revised, pp. 53, 242; Rawls, “Idea,” pp. 792–793.

(19.) Fineman, Autonomy Myth, p. 134.

(20.) Brooks, “Problem,” pp. 111–112. See also May, “Liberal Feminism,” for criticism of polygyny.

(21.) March, “Is There,” pp. 258–259; Joseph is discussed in Emens, “Monogamy's Law,” pp. 314–317, 332–334.

(22.) Calhoun, “Who's Afraid”; Mahoney, “Liberalism”; March, “Is There.”

(23.) Rawls, “Idea,” p. 779; Brooks, “Problem.”

(24.) Mill, On Liberty, pp. 160–161 (my italics); see Baum, “Feminism.”

(25.) See Chapter 5.1; Calhoun, “Who's Afraid”; Maushart, Wifework.

(26.) Maushart, Wifework, p. 5.

(27.) Brooks, “Problem,” p. 116.

(28.) March, “Is There,” p. 258.

(29.) March, “Is There,” p. 266.

(30.) MacKinnon, Toward, pp. 161–162, see also 157–170, 195–214; and MacKinnon, Feminism.

(31.) Rawls, “Idea,” p. 790.

(32.) MacKinnon, Toward, p. 224.

(33.) Rawls, Theory: Revised, p. 63.

(34.) In this paragraph I follow Levey, “Liberalism”; see also Walker, “Liberalism.”

(35.) Rawls, Theory, pp. 199–200.

(36.) Rawls, “Idea,” p. 788.

(37.) Rawls, “Idea,” at pp. 787–794. See also Okin, Justice, and “Political Liberalism”; Nussbaum, Sex and Social, pp. 81–117.

(38.) “State Policies,” p. 16.

(39.) Cf. Brake, “Rawls and Feminism.”

(40.) See Becker, A Treatise, “Polygamy and Monogamy in Marriage Markets,” pp. 80–107.