Choice, Dependence, and the Reinvigoration of the Traditional Family
Kathryn Abrams(1)
If contractual marriage is a meeting of the minds, then we should be glad that Jeff Stake and Eric Rasmusen are not applying for a license. As a careful reading of their paper makes clear, they do not always agree on its central goal. One strain of the paper, which I identify with Stake,(2) is concerned with choice. The salient fact about marriage, and divorce, is that one size does not fit all. The advent of no-fault divorce has pressed all marriages toward a particular mold: one in which both parties are independent, self-supporting, and free to go without a finding of culpable action. But the American social landscape is in fact populated by couples with a range of different tastes, patterns, and aspirations, some of which are ill-served by this increasingly conventional norm. The second strain, which I identify with Rasmusen, is concerned with protection. He worries about the growing difficulties faced by those who prefer a traditional, gender-differentiated family, and, in particular, about the homemakers whose household-specific investments help to preserve it. The no-fault system signals to these women that investing strongly in the family makes them vulnerable to spousal departure. The law should instead serve these women and their families by facilitating the traditional family structure.
As with many longtime companions, however, a simple difference of opinion does not prevent Stake and Rasmusen from forming a union. Adept at accommodating each other's idiosyncracies, they have arrived at a framework for their argument that minimizes the difference in their ultimate concerns. They have given the rhetorical lead to Stake, by constructing a framework--and an argument--that emphasize choice: courts should be legislatively authorized to enforce private marital agreements regarding the terms of an ongoing marriage, the grounds for divorce, and the division of property on divorce.(3) Yet if Stake's concerns inform the rhetoric, it is Rasmusen who achieves the last word regarding substance. The couples most likely to benefit from this proposal will be those who elect a more "traditional" marriage: either by re-embracing a fault standard, or by structuring their marriage around a traditional, gendered division of labor, or both. The resources of state family-law systems, under this proposal, will help to reinvigorate the traditional family. Employing a "choice" framework, moreover, enables Stake and Rasmusen to achieve these results with far less argumentative exertion than they would be obliged to undertake were they to argue outright for a reinvigoration of traditional family roles.(4) They argue that choice has its virtues, either because couples' preferences in marriage are diverse, or because choice means an absence of legal or governmental coercion as to the form of one's marriage.(5) Then, given the virtue of a choice-based framework, they ask: Is there anything so objectionable about the traditional family that we should prevent couples from choosing it over the currently favored, dual-wage-earner, easy-exit model?
In this Response I will argue that Stake and Rasmusen have elicited the wrong answer by framing the wrong question. Instead of accepting their broad account of the virtues of choice, we should ask whether this proposal actually fosters choice, in at least two respects. First, we should question how choice is produced within heterosexual unions, where power relationships are complicated and often unequal. Second, we should ask how a legal proposal to enforce contractual choice will affect a social landscape that is shaped not only by legal rules, but by a range of other social influences. Undertaking these inquiries, I believe, will reveal this proposal to be something more than a neutral means of effectuating marital choice. The proposed regime appears likely to enforce many marital contracts that are the product of inequalities in bargaining power. Its most likely patterns of enforcement will also send a signal that women's growing autonomy in relation to family should be reconsidered, and that more institutional support should be given to family forms entailing greater dependency for women. Viewed in this light, the question is not "Is the traditional family form so objectionable we should prevent couples from choosing it?" It is, rather: "Is the traditional family form so valuable that we should risk these consequences in order to reinvigorate it?"
In the second Part of the essay, I take up this question, by moving from the authors' facial arguments for choice to their underlying arguments for protection. There are two parties whose welfare might arguably be protected by legal support for the traditional family form: women who make full-time investment in the family, and the children who benefit from her labors. In the first subpart of this Part, I argue that the benefits to children from legal support of the traditional family form may, in fact, be illusory. Children's needs in infancy and childhood are very great--greater than can readily be met under a two-worker model with minimal public or private support for the substantial responsibilities of parenting. Yet the re-enlistment of women to fill these roles, exclusively and in familial isolation, by erecting a legal framework that assumes a unity of interest between at least some subset of mothers and children, is a flawed approach to addressing these needs. It may deprive children of the benefits of having a second fully engaged parent, and leave women and children vulnerable to contingencies beyond marital breakdown. Its persistent concentration and privatization of the costs of children's dependence--and subsumption of this privatization under the rubric of choice--prevents society from rethinking and redesigning institutions to address this very pressing need.
In the final subpart of this Part, I return to the proposition that enforcement of contractual exceptions to the no-fault scheme will protect full-time homemakers. This claim is most difficult to gainsay, particularly in light of the authors' assertion that these women--some subset of whom genuinely prefer this role--have been disadvantaged under a no-fault divorce regime. Yet here too, the advantage is more illusory than the authors suggest. Not only are there plausible opportunities for inequitable exit for men who urgently wish to find them, but this legal protection signals to women that it is safe to assume a dependent position, in a society that values dependence--and interdependence--even less than it did in the heyday of the traditional family. A more durable commitment to such women requires that we revise our norms and institutions to reflect a greater valuation of interdependence. Such a goal cannot be achieved--and may be deferred--by premature proposals to secure dependency through marital contracts.
In Stake and Rasmusen's argument, much of the appeal of vindicating choice comes from the particular moment on which they focus. When a marital couple chooses their family form, the interactions that brought the bargain into being have been concluded, and the impact of this and similar bargains upon the legal and social world has yet to become clear. At this moment, the legal enforcement of individual choice appears both most innocuous (the government seems paradigmatically neutral between marital versions of the good life) and most promising (the strategy has the potential to enhance the well-being of a range of different couples). Yet the promise of this particular moment may be misleading. If one focuses on the period immediately prior to that moment, when two potentially unequal parties negotiate their "bargain," the legal enforcement of marital choice may look less like a vindication of the individual and more like a means of entrenching inequality. Similarly, if one focuses on the period after that moment, when the choices of particular couples produce a cumulative impact on legal and social norms shaping family structure, the proposal may seem less like a means of effectuating choice and more like a means of reinvigorating the traditional family form. In this Part, I examine the "prechoice" and the "postchoice" periods, to provide a more critical vantage point on Stake and Rasmusen's proposal.
Life is easy for Nat and Dot, the fictional couple whose plans and preferences help elaborate the authors' proposal. Through a magical identity of perceived interest and power, they appear to speak as one. They "agree before marriage that they want to commit themselves to each other in marriage to the same degree as traditionally expected by the law."(6) They also know that they want to "specializ[e] within the marriage": she is going to specialize in household production and he is going to specialize in market production--or, as Stake and Rasmusen tellingly put it, she is going to "learn[] how to cook Nat's favorite dishes" while he is going to "learn[] how to bring home more wages."(7) But who are "they," this preternaturally concordant couple, and how do "they" come to want what "they" want? While Stake and Rasmusen are eager to "lift the veil" on variation in preferences among couples, they are less inclined to scrutinize variation in preferences within couples. Yet given the seemingly infinite human variation in tastes, which expresses itself even among marrying couples, it seems likely that such differences will occur. More to the point, inequalities of power among cross-sex, or heterosexual, couples, mean that we may rightly be suspicious of the terms on which these differences are resolved.(8)
Let us return to Dot. If we assume simply that she is a woman, contemplating marriage to a man, what do we know about her preferences and her power to implement them, relative to that of her prospective partner? First, we know that she may experience a greater social and economic pressure to marry than does her male counterpart. Should Dot decide to eschew familial investment entirely and focus her efforts exclusively on the market, she faces continuing sex segregation and sex discrimination in the workplace, and a wage differential of more than twenty-five cents per dollar.(9) She also is, and has been, subject to a variety of social norms that emphasize the centrality of marriage to women's happiness and that characterize unmarried women in harsh and stigmatizing terms.(10) So marriage--for the moment, undifferentiated--may seem like a more attractive prospect to her than to Nat, who faces an uncompromised employment prospect and a stigmatically neutral, if possibly lonely, life as a "bachelor." The greater urgency that Dot may feel about marrying, in general, may erode her power to negotiate for a particular kind of marriage. But her problems do not end here. Dot may also be hampered by the fact that she may be less certain about what she will want in the foreseeable future than Nat is likely to be. The average woman contemplating marriage is younger than her prospective spouse,(11) meaning that she may have less experience of the world and may have developed less clarity about her tastes and goals. She is also, at this moment of social transition, confronted with a wider array of options.(12) She may invest primarily in the family; she may invest primarily in the workplace; or she may devise some more diversified portfolio of work and family investments.(13) She may have less information about some of her prospective roles than is available to her spouse: many women, particularly those without substantially younger siblings, may know little about the nurturing roles that they may be called upon to perform in the family, beyond the fact that they are supposed to like them. And she is also likely to be receiving multiple, conflicting social signals about the contours of an acceptable and satisfying life for a woman. Thus, Dot's uncertainty about what she wants, and what she should want, may also make her a less effective negotiator on her own behalf. As if these difficulties were not enough, Dot may also confront a liability attributable to her female socialization. If one credits the work of scholars from Carol Gilligan(14) to Carol Rose,(15) the relational orientation of women's socialization may make them more acquiescent negotiators than men.
These difficulties are not conclusive as to what preferences Dot will bring to the table, though some might make her less likely to invest in the workplace than her prospective partner. But they should shape our view of the final bargain that a couple seeks to have enforced by law. The fact that women bargain in the shadow of inequality, and in the shadow of ongoing social contention over women's changing roles, should counsel caution in treating marital choice as an authentic, univocal preference. Enforcing these choices uncritically(16) may exacerbate this inequality by formalizing or institutionalizing it, a prospect that should at least deter us from treating the legal facilitation of "choice" as an unequivocal social or political good.
If the proposal neglects the complex dynamics of marital bargaining that occur in the "prechoice" phase, it also neglects the complex dynamics of social and legal influence over marital structure that occur in the "postchoice" phase. We can begin to glimpse these dynamics by asking two questions. First, what kinds of couples will be encouraged to register their marital choices under the proposal? And second, how might we describe the field of social and legal influences in which the proposal will operate, and how is the proposal likely to interact with these influences? The answers to these questions suggest that the effects of the proposal are predictable ex ante: it seems likely to vindicate the goal of at least one of its authors, to reinvigorate the traditional, nuclear, gender-differentiated family.
What kinds of couples will be encouraged to register their marital choices under the proposal? The most obvious answer is "anyone" and "no one in particular." In the abstract, the whole purpose of facilitating the choice of familial structure is that anyone can vindicate their preferences. Moreover, it is this opportunity, rather than any particular result that might come of it, that is ostensibly the goal of the exercise. Yet this abstract answer may not suffice to describe the incentive effects of the proposal and its likely impact on the existing social landscape: both the details of the proposal and the legal background against which it operates suggest that it will be a decisively more attractive option to couples with some preferences than to couples with others. To begin with, couples who favor a two-wage-earner family structure and an easy-exit approach to divorce are unlikely to register their preferences under the proposal. These couples conform to the operative assumptions under the no-fault system, which would become the default approach (i.e., operative if no preference is specified) under Stake and Rasmusen's proposal. Less obvious perhaps--but equally evident on a close reading of the proposal--is the fact that few couples with more unconventional marital arrangements will be likely to register their preferences under the proposal.
In keeping with the Stake-inspired emphasis on choice, the proposal extends to gay and lesbian couples, celibate couples, multiple-adult families, and others the opportunity to contract regarding marital or family structure, grounds, and terms of marital or familial dissolution.(17) Yet in extending this legal opportunity, Stake and Rasmusen declare themselves unwilling to address the other legal barriers that have tended to marginalize these family forms. Ted and Alice, the ironically named celibate couple, may define their marriage for purposes of state family-law enforcement; yet Stake and Rasmusen make clear that the federal government is not obliged to recognize their union for social-security purposes.(18) More surprisingly, this state legislative proposal would not address those factors of state law, such as sodomy statutes,(19) that have institutionalized the marginalization of some unconventional couples or families. These omissions are likely to make resorting to the proposed statutory option far less attractive for unconventional couples. A celibate couple might derive some clarity, or some personal satisfaction, from registering their marriage as one that cannot be annulled for the refusal of one party to consent to sex. Yet if the proposed statute did nothing to mitigate the disadvantage to which this choice of marital form would be subject outside the domain of family law, celibate couples might not be flocking to the secretary of state's office.(20) The likelihood of nonparticipation might be even stronger in the case of gay and lesbian couples, who could secure no advantages outside the domain of family law, and might expose themselves to the possibility of sodomy prosecution if they registered their marital preference. Although the foregoing does not represent a comprehensive canvassing of the possibilities, it nonetheless seems clear that the most important beneficiaries of the proposal are likely to be the object of Rasmusen's original concern, traditional, gender-differentiated families. These couples could secure themselves certain protections that are not available under the no-fault regime, but would not subject themselves to stigma or ancillary legal disadvantage for so doing.(21) One might expect the proposal to produce a steady stream of couples who remove themselves from the no-fault regime by specifying a traditional gender-differentiated marriage which is possible to leave (if at all) only on a legal finding of fault.
But what is the social landscape on which this change will be wreaked, and how is this change likely to act on that landscape? Stake and Rasmusen focus on one aspect of that landscape, the legal movement to no-fault divorce. Viewed against the backdrop of this legal development alone, a proposal authorizing choice of marital form--even if that choice tended to support the traditional family--might look like a modest, liberalizing reform. Yet the proposal does not simply right a wrong produced by an arbitrary legal interjection into familial diversity. No-fault was itself the result of a series of changes in social norms, some of which related explicitly to gender and work.(22) These normative changes have continued to unfold, creating the environment in which the proposal will generate its effects. Although a systematic social historical analysis is far beyond the scope of this Response, I will highlight two developments not systematically analyzed by Stake and Rasmusen--developments which are coterminous with, and implicated in, the debate over no-fault divorce.
The first is a movement by many contemporary women to develop greater autonomy in relation to their families. This can be seen in the first instance by the rapidly increasing numbers of women choosing work outside the home, even during periods when they are rearing young children.(23) For some women this decision may reflect identification with family, for these decisions may be in part financially driven. But the decision to work outside the home also reflects a desire for greater financial independence--that may be encouraged by, but is not exclusively the product of, no-fault divorce--and a desire to develop a sense of competence and worth that is not exclusively derived from familial labor.(24) The movement of greater numbers of women into the workforce is not, moreover, the only example of this move toward greater autonomy in relation to family. The call by women for greater state intervention in the family--through legal vehicles such as marital-rape statutes or enforcement against spousal abuse--also reflects an effort to disentangle their individual interests from the marital and familial interests to which they have traditionally been assimilated.(25)
But if there has been movement by women toward greater autonomy in relation to family, it has not been smooth, unimpeded, or unidirectional movement. On the contrary, it has been hotly contested in a range of different contexts. This highly vexed, continuing controversy over women's roles, particularly in relation to family, is the second factor that bears on the effect of Stake and Rasmusen's proposal. The movement of women into the workforce has challenged, but far from eroded, the belief of many traditionalists that women should define themselves primarily in relation to the family. Women choosing labor-market employment have been beset by perturbed spouses,(26) conservative pundits,(27) and mainstream media decrying a war between working women and homemaking women,(28) and self-interested commercial producers, hopefully proclaiming the advent of "the new traditionalism."(29)
How does this proposal look different when we refocus its backdrop in this way? Given the current baseline of a no-fault regime, and given the couples most likely to seek exemption, it looks like a state-sponsored legal accommodation of a marital form that reintroduces dependency into women's familial lives. In this way, the proposal lends the support of the state to a family that provides a counterweight to women's increasing autonomy. And it adds the state-accommodated model of the traditional, gender-differentiated family to the range of influences contesting this move toward greater autonomy. For those concerned about women's opportunities and well-being, the question should be, not "Is the traditional family so problematic that we should prevent couples from choosing it?" but rather "Is the traditional family so advantageous that we should cultivate it through legal reform, even at the risk of perpetuating these effects?" To answer this question we should consider those arguments by the authors that sound not in choice but in protection.
The first object of the authors' protective impulse is the homemaking woman. She makes her family-specific investments early in the marriage, and these investments render her poorly equipped to take on the world of work; thus a no-fault regime makes her particularly vulnerable to spousal departure. However, many such women have glimpsed their precariousness under the no-fault regime, and have moved industriously to obtain greater security. Some have sought to buttress traditional commitments through affiliation with conservative religious communities; others have hedged their bets by investing in the world of work. Neither solution is wholly satisfying to Rasmusen (or, arguably, to Stake): resort to the moral authority of churches may impose social and religious constraints on otherwise irreligious individuals;(30) resort to the market may result in "selfish career building at the expense of family."(31) Their protection is, in the first instance, paternalist: it seeks to save smart women from foolish choices by rendering more secure the traditional, gender-differentiated role they had sought in the first place.
Yet the homemaker is not the only object of the authors' solicitude. She is worthy of protection at least in part because of the benefits her exclusive familial focus confers on a second object of concern: her children. In the course of establishing the traditional family as choice worthy, Stake and Rasmusen make a range of arguments ostensibly designed to demonstrate that the traditional family does not generate externalities or spillover effects harmful enough to justify barring it as an object of choice. But far from simply establishing that such arrangements "contravene public policy,"(32) these arguments reflect affirmative claims that distinguish the traditional, gender-differentiated family from its less traditional counterparts. Securing these advantages seems to be a central goal of facilitating traditional family formation. I will begin by scrutinizing the protection of this form as a means of securing the welfare of children. I will then turn to the protection the proposal affords to homemaking women.
The authors begin with a series of positive, societal effects generated by the traditional family form:
The presence of one spouse in the home should reduce the costs of police protection paid for by others and may even reduce the need for police protection for all neighbors, since there would be more monitoring during workdays. . . . . . . . In addition, other children benefit when a sick child stays at home with his or her parent rather than being sent to school or day care.(33)
But the heart of their claim is the benefits traditional families can confer on children. "Traditional roles allow parents to spend more time with their children,"(34) Stake and Rasmusen argue. "Beyond the issue of 'quantity time,' parents often do a better job educating and nurturing their children than temporary caretakers."(35)
Can such benefits really be claimed from a gendered division of labor in the family? The answer is far from clear. Though many parents, overwhelmingly mothers, derive satisfaction from greater involvement in the education and nurturance of their children, there is little evidence that suggests that preschoolers who go to day care, or school-age children who return home before their parents, experience emotional or developmental deficits, or underperform in subsequent life challenges.(36) Moreover, the "quantity time" issue is highly equivocal. Although Stake and Rasmusen refer to the time that children in traditionally structured families are able to spend with their "parents," these children are likely to see more of their mothers, and less of their fathers, than are children in families whose roles are less sex-differentiated.(37) There is a growing body of evidence suggesting that children in two-parent families do better when their fathers play a more engaged and substantial role in their lives.(38) When both parents in two-parent families are actively involved in child rearing, there is more opportunity for specialization in child-rearing investments, which the authors describe as a source of gains in familial welfare.(39)
Finally, as the last point suggests, the traditional family--and the nuclear family, of which it is the most sex-differentiated example--may be particularly problematic vehicles for securing the welfare of children. The needs of young children are very great; they must be met day after day, week after week, regardless of the stresses or acts of God to which their adult care givers may be subject. The tremendous constraint of the two-career family has highlighted how unsupported families, more generally, are in their efforts to raise their children. The "nuclear" form of the family, as feminist theorist Martha Fineman has observed,(40) allows little flexibility when parents must commit themselves to other roles; this form courts chaos or dissolution when a more sustained or serious disability befalls one or both parents. This recently highlighted vulnerability has made the nuclear family a subject of scrutiny and criticism. Fineman has argued, for example, that we must replace the nuclear family--a family organized around the "sexual dyad" and isolated from public or private support for the substantial responsibilities of parenting--with a more flexible family unit, organized around the "child-care giver dyad" and connected with a range of institutions designed to share the costs of young children's dependency.(41)
The authors' proposal takes these insights about the precariousness of the two-career family and moves in the opposite direction. It retains the nuclear family but attempts to avoid the precariousness, by facilitating a form that concentrates child-care responsibilities on a single parent working inside the home. Yet this answer is unlikely to be an adequate substitute for more searching efforts to provide social supports for the meeting of young children's needs. Not only do most couples require two incomes,(42) making the sex-differentiated-family form a luxury that few can afford. But concentrating the responsibility for children's dependence in one parent makes the children highly vulnerable to that parent's unexpected illness, death, or incapacitation, as well as to events, beyond departure, that may befall the primary wage earner.(43) If we reinvigorate the traditional family, as an alternative to challenging the privatization of child-rearing costs on to the nuclear family, we may short-circuit the search for social arrangements that acknowledge the social stake in, and spread the considerable costs of, meeting the needs of young children.
Yet, if the authors' proposal does not provide the optimal arrangement for children, it may provide a more secure situation for a subgroup of mothers: those women who, sincerely, and one might even say "authentically," prefer full-time child rearing to any work they might perform outside the home. The number of women who make this choice--indeed, who can afford to make this choice--is decreasing; moreover, they make this choice (and a range of others reconciling the calls of work and family) in the face of social and familial pressures so complex as to render the term "authentic" somewhat naive. Nonetheless, there are women who make this choice whose decision I would be prepared to endorse. When we view child rearing as work(44)--work that demands a particular set of tastes and skills--it seems entirely plausible that there would be some subgroup of women (and men) who prefer this kind of work to any other. There is also likely to be a group of people (probably mostly women) whose orientation toward the child-rearing role is so strong that they take deep satisfaction out of spending their "working" hours with their children, regardless of whether they feel particularly suited to the skills it demands. For either--and perhaps for other--kinds of reasons, these women feel consistently fulfilled by the traditional roles of wife and mother; these same women, according to the authors, are rendered more vulnerable by the advent of no-fault divorce.(45) The proposal offers one means of making these women more secure; if, as feminists, we value the welfare of all women, it is an advantage that must be seriously considered.
First, there is reason to question the authors' claim that these women have been disadvantaged by the move to no-fault divorce.(46) Feminist scholar Jana Singer has argued, for example, that homemaking women may have been less well-off under a fault system than earlier scholars had suggested. Although women may have been eligible for alimony under the fault system, they rarely received it, meaning that awards of alimony dropped only slightly for women as a group, and actually increased for women married more than five years, with the advent of no-fault divorce.(47) Moreover, the fault system permitted the courts to police the personal and sexual conduct of divorcing women, denying alimony where women departed from conventional sexual norms or failed to perform the roles of dutiful wife and mother.(48)
Second, even if the authors are correct about the fate of homemaking women, the proposal may provide less protection than they suggest. To begin with, there may be ample opportunities for men to extricate themselves from the bargain that was struck. Stake and Rasmusen are frankly unresolved about whether renegotiating the initial bargain is possible. If it is, these women's security is only as strong as their ability to prevent a full or partial reversion to the no-fault framework. If a woman's negotiating power in a traditional relationship decreases over time,(49) either because she has already supplied the essential performance or because of prevalent male preferences for less powerful, more physically and reproductively nubile women,(50) she is less likely to be able to avert this result, particularly if renegotiation is made a condition precedent to avoiding a custody battle on divorce.(51) If renegotiation is impossible, such women may remain married, assuming their husbands do not provoke a divorce according to fault standards (by intentionally injurious behavior).(52) Yet, one might legitimately question whether years of marriage to a man who no longer wishes to be married is recognizable as what a full-time wife and mother bargained for. And it is worth noting that such a bargain merely secures her, in her dependency, against unilateral, no-fault dissolution.(53) It does not secure her against the range of other factors, from spousal death or disability to a downturn in the working spouse's employment market, that might render a nonemployed woman with socially unsupported responsibility for small children vulnerable. The problem with the proposal, in the end, is not that it is disguised paternalism (as I suggest in Part I), or that it is false paternalism (as I suggest in Part II.A), but that it is bad paternalism. If one counts as substantial the probability of any of the foregoing developments, the woman who turns to the labor market or seeks the support of a religious community for herself and her marriage is likely to do better for herself than this proposal could ever do for her.
Perhaps the greatest danger of this proposal is its explicit message that female dependency can be made secure in a society where this is no longer feasible (if indeed it ever was). Security, when one has invested in one's family at the expense of one's own human capital, requires more than a formalized commitment on the part of a youthful spouse. It requires spouses who have been socialized to value familial work, not only when they know they are going to need it, but after it has been done. Beyond that--given that death, disability, or downsizing may make spousal valuation irrelevant--it requires a public sector prepared to undertake transitional training, a private sector willing to see the promise in a woman who has been out of the labor market for many years, and a commitment on the part of both sectors to help such a woman manage the dependency of her young children. It requires, in short, a society that values interdependence as well as autonomy, and this is not, at present, the society we live in. Our society, whose uncompromising valuation of autonomy has been challenged by everyone from difference feminists to civic republicans, values interdependence only for women and other subordinate groups and only so long as such interdependence--or should one say, dependence--serves the immediate needs of the most powerful autonomous members.(54) This tendency has only become more pronounced as women have increasingly embraced more autonomous roles. Moreover, the greater mobility attributable to national and global economic markets has attenuated the local bonds of family and community that once helped to support women and children in their dependency.
If we want to facilitate the choices of financially dependent women, we must rethink a range of social and institutional choices that value autonomy over interdependence. This might include, for example, recent welfare reforms that devalue and foreclose the choices of single women who seek to focus their efforts on child rearing through temporary financial dependence on public assistance, rather than on a male breadwinner.(55) Unless and until we are able to make such a commitment, a proposal to enforce individual marital contracts will be woefully inadequate, and perhaps dangerous to those whose path to dependence is eased by its assurances. It will remain a legal program oddly reminiscent of the circumstances of its authors: an uneasy truce between competing social goals, that may work for the occasional couple, but disserves everyone else.
1. Professor of Law, Cornell Law School. I want to thank Martha Fineman and Mary Becker for conversations on the subject of this Response.
2. Both my inference of a divergence in perspective, and my association of "choice" with Stake and "protection" with Rasmusen, are fueled by a section of the paper in which the authors, under the noms de plume of "liberal" and "conservative," openly discuss their disagreement about whether marital contracts should be subject to renegotiation. See Eric Rasmusen & Jeffrey Evans Stake, Lifting the Veil of Ignorance: Personalizing the Marriage Contract, 73 Ind. L.J. 453, 476-81 (1998).
3. They argue that the law should permit couples, at a minimum, to elect to have their marriage governed by the "traditional rules of marriage and divorce"--as recently attempted by Louisiana through the authorization of "covenant" marriages. Id. at 464. Legislatures might even go further, to permit couples to define their own marriages, by selecting from a range of legally prescribed options, or by crafting their own original terms. See id. at 464-65.
4. It is far from obvious to me that both authors would support an affirmative, collective decision to deploy legal resources to reinvigorate the traditional family. I suspect that Rasmusen would, while Stake might not. What I mean to point out by pursuing the analysis below is that advocacy of individual choice may sometimes contribute to the attainment of predictable social outcomes, outcomes that some proponents of choice may support outright, while others may not. If Stake is, in fact, the "liberal" that he claims to be, he may be getting more than he bargained for in this particular authorial union.
5. This latter argument is sometimes made rather obliquely. Stake and Rasmusen note, in the course of affirming their understanding that legal enforcement will shape as well as reflect people's preferences, that:
Indeed, one of the important preferences law shapes is the taste for law itself. The Bill of Rights develops the American taste for freedom from governmental controls of expression and religion. American marriage law sends the opposite message: it is up to society to define important familial relationships. Our proposed legislation might foster preferences for extending private control and diminishing governmental control in marital matters.
Rasmusen & Stake, supra note 1, at 465 (footnote omitted). They suggest, by analogizing to
the Bill of Rights, that choice, understood as freedom from governmental coercion, is a good
thing. They imply, though they do not argue explicitly, that marital law may be out of sync with
other American laws in that it does not facilitate such choice. Finally, Stake and Rasmusen
conclude that by facilitating choice in marital relationships, their proposal may foster a taste
for choice in marital relationships. They do not consider the possibility that, injected into a
legal system featuring a particular baseline or assumption about the structure of marital
relations, their proposal may foster not only a taste for choice, but a taste for a certain kind of
marital relationship. This is the possibility I consider infra Part I.
6. Id.
7. Id. at 466.
8. For an interesting and sophisticated economic analysis of men's and women's
respective bargaining power in marriage, albeit one that involves some assumptions different
than those at work in this paper, see Amy L. Wax, Bargaining in the Shadow of the Market:
Is There a Future for Egalitarian Marriage?, 84 Va. L. Rev. (forthcoming May 1998).
9. In 1993, the ratio of women's annual full-time earnings to those of men was
approximately 72%. However, the gap is even narrower for single women without children. See
June O'Neill, The Cause and Significance of the Declining Gender Gap in Pay, in Neither
Victim nor Enemy 1, 8 (Rita Simon ed., 1995).
10. A good example of the social valuation placed on marriage for women was the mass
popular and media agitation caused by a 1986 Newsweek article citing data suggesting that a
40-year-old college-educated woman would be "more likely to be killed by a terrorist" than to
get married. Eloise Salholz et al., Too Late for Prince Charming?, Newsweek, June 2, 1986,
at 54, 55. For a sample of the often hilarious commentary sparked by this article, see Susan
Faludi, Single at 30 . . . And Why Not? The Truth About That Newsweek Story Should Leave
Someone Blushing--And We Don't Mean the Bride, Chi. Trib. Mag., Dec. 21, 1986, at 18,
Leslie Pound, Against All Odds, Dallas Morning News, June 22, 1986, at 1F, available in
1986 WL 4324415, and Jack Smith, Eye-Contact Sports Looking for a Potential Husband? Try
a Jelly Doughnut with Sky-Diving Policeman, L.A. Times Mag., Nov. 16, 1986, at 4. Many of
these commentaries included flagrant examples of the devaluative stereotypes imposed on
unmarried women. See e.g., Faludi, supra, at 18 (contrasting cover stories in People magazine
on unmarried women--entitled "The New Look in Old Maids"--and unmarried men--entitled
"The Unmarried Man: Prime Catch and the Top 10 Holdouts"); Pound, supra, at 1F
(describing comments made to 36-year-old unmarried woman: "You're such a pretty girl. I
wonder why you have no husband," and "You're such a nice person--what's wrong with
you?").
11. The average American woman is 24 years old at the time of her first marriage, while the
average American man is 25.9 years old at the time of his. See U.S. Bureau of the Census,
Statistical Abstract of the United States 105 (116th ed. 1996).
12. The social transition referred to above--the breakdown of traditional gender roles in
the family and workplace--has also had some impact on men. But while men may now choose
to invest more strongly in the family than they have traditionally done, see Adele Eskeles
Gottfried et al., Role of Maternal and Dual Earner Employment Status in Children's
Development: A Longitudinal Study from Infancy Through Early Adolescence, in Redefining
Families 55, 59 (Adele Eskeles Gottfried & Allen W. Gottfried eds., 1994) (discussing
research suggesting that fathers tend to be more involved with children when mothers work
outside the home), most empirical evidence suggests that investment by men in the family has
not been substantial, even for those whose spouses also work outside the home. Howard
Hayghe reported in 1990 that in only 2.2% of American families did the husband specialize in
care of the family while the wife worked outside the home. See Howard V. Hayghe, Family
Members in the Work Force, Monthly Lab. Rev., Mar. 1990, at 14, 17 tbl.3 (reporting
statistics from 1988). While men have increased the amount of time devoted to household
tasks, they have not compensated for the tasks relinquished by, and still lag far behind the
hours put in by, women. John Robinson reports that between 1965 and 1985, in families with
children under age 5, women have cut back at least 10 hours per week on home-related tasks
(from 32 to 22) while men have increased only 5 hours per week (from 4 to 9). See John P.
Robinson, Who's Doing the Housework?, Am. Demographics, Dec. 1988, at 24, 27-28. These
findings mean that men, in practice, have a narrower array of roles from which to choose in
crafting a marriage.
13. These latter two patterns, which tend to be regarded as new, have in fact been followed
for generations by working-class women and women of color, who have not tended to have the
financial freedom to remain at home. See Alice Kessler-Harris, Women Have Always
Worked 15-20 (1981). These patterns have received more attention of late, because they have
been attempted by middle-class white women, and because they have been pursued by women
of all races and classes in greater temporal proximity to the birth of their children than has
previously been the case.
14. See Carol Gilligan, In a Different Voice 24-63 (1980). The relational modes of
cognition and problem solving that Gilligan attributes to women would appear to have
implications for the way they negotiate, particularly with men who do not share these modes
of thought and interrelating. However, Gilligan does not discuss the implications of her theory
for interspousal negotiations.
15. See Carol Rose, Women and Property: Gaining and Losing Ground, 78 Va. L. Rev.
421, 423 (1992) (using game theory and assumption that women have greater "taste for
cooperation" to explain why women may have less property than men and proposing possible
solutions).
16. Of course, contract law contains the usual exceptions for fraud, duress, and
unconscionability. However, the kind of inequalities to which I refer in this section are subtler
than the inequalities generally captured by these exceptional categories. My point is not that
we should look more critically at marital bargains, a goal which seems difficult to achieve
given the varied and often diffuse social influences that can contribute to the inequalities I cite.
My point is that we should look more critically at the kind of approach espoused by the
authors: one that assumes that we can treat a marital choice as the univocal preference of a
couple, and one that views the legal enforcement of private choices as a liberating strategy that
contributes in an uncomplicated manner to the well-being of a range of different actors.
17. See Rasmusen & Stake, supra note 1, at 489. It is arguable that even the terms in which
this opportunity to obtain the legal enforcement of contract is extended are stigmatizing to
these unconventional partnerships or families. The authors refer in this section to
"[h]omosexual partnerships, polygamy, kept mistresses, and other relationships." Id. Certainly
the use of the term "mistress" (particularly when combined with "kept") has acquired a
pejorative connotation; and the term "polygamy," while technically referring to the practice of
having more than one wife or husband at a time, is also associated with the patriarchal,
hierarchical, and legally discredited practices of the nineteenth-century Mormon Church.
Linking homosexual unions, at least in this initial presentation, with two forms of coupledom
that are subject to substantial social disapprobation may have both the intent and the effect of
casting aspersions on this form of union as well.
18. See id. at 490.
19. Far from addressing the tension between their proposal (which claims to facilitate a
range of marital unions and family forms) and sodomy statutes (which criminalize certain kinds
of sexual conduct, thereby casting aspersions on certain kinds of unions), Stake and Rasmusen
take at least a nonjudgmental and perhaps an acquiescent attitude toward the moral and
political claims behind sodomy statutes, in one place referring to enforcement under sodomy
statutes with the archaic and intolerant term "prosecution[s] for crimes against nature." Id.
20. This is particularly true given the fact that the chance of misunderstanding between the
marital couple would seem to be small in the case of celibate marriage. One could imagine,
though it might be a disturbing prospect, a couple entering into marriage without having
clarified their mutual expectations concerning the allocation of responsibilities for child
rearing. However, it is difficult to imagine a couple entering into marriage without having
clarified their mutual expectations concerning whether sexuality will be part of the relationship.
This means that the potential gains in the domain of family law that would be experienced by
a celibate couple in registering a choice of family form are likely to be quite limited.
21. The tendency of the proposal to encourage traditional, gender-differentiated marriages,
or marriages that opt out of the easy-exit default rules, or both, is likely to be exacerbated if the
proposal produces the effect on third parties that Rasmusen (as "conservative") in one section
predicts: that everyone from lending institutions to other family members will be more willing
to invest in families that assume these configurations. See Rasmusen & Stake, supra note 1, at
477. One could imagine that the ultimate result of the proposal might be--at least vis-à-vis this
range of third parties--to penalize those families who choose to remain within the default
framework.
22. See Lenore Weitzman, The Divorce Revolution 28-37 (1985).
23. As of 1992, three-quarters of married women with school-age children, six-tenths of
married women with preschool-age children, and half of married women with children under
the age of two were in the labor force. See Andrew J. Cherlin, Public and Private Families
298 (1996) (citing U.S. Bureau of the Census, Statistical Abstract of the United
States, 1993 (1993)). Daphne Spain and Suzanne Bianchi report that the labor-force
participation of single mothers with dependent children is approximately the same as that of
married mothers with dependent children. See Daphne Spain & Suzanne M. Bianchi,
Balancing Act 146 (1996). The latter figure, however, is not broken down according to the
age of the child.
24. See Lillian Breston Rubin, Worlds of Pain 169-73 (1976) (discussing sense of self
and independence that women gain from labor-force work).
25. See Frances Olsen, The Family and the Market, 96 Harv. L. Rev. 1497, 1530-39
(1986) (discussing changes in the law of marital rape, as well as initiatives to secure greater
legal intervention in the family in relation to spousal abuse).
26. See Arlie Hochschild, The Second Shift (1989). The case studies reported in this
classic reflect the dissatisfaction voiced by some men over their wives' labor-force
employment, particularly when such employment interferes with what these men understand
to be their household or child-rearing responsibilities. See Wax, supra note 7 (Aug. 1997
manuscript at 71-72) (discussing the husbands' view of the effects of spousal employment on
their welfare).
27. See, e.g., David Gelertner, Why Mothers Should Stay Home, Commentary, Feb. 1996,
at 25 (discussing a possible connection between women working and a decline in children's
health and behavior).
28. See, e.g., Kathryn Abrams, Gender Discrimination and the Transformation of
Workplace Norms, 42 Vand. L. Rev. 1183 (1989) (critiquing this strategy for dividing--and
conquering--women).
29. In 1988, Good Housekeeping magazine began its "New Traditionalist" campaign, which
sought to describe (and, arguably, create) its target audience. Good Housekeeping defined a
"new traditionalist" as "'a woman who brings a contemporary attitude to a traditional
lifestyle.'" Mary Hanson, A Question of Values, Milwaukee J., Aug. 2, 1992, at 1 (quoting
Alan Waxenberg, publisher of Good Housekeeping). Although Good Housekeeping has
endeavored to include working women among the ranks of this group, see id. at 4, others have
characterized the New Traditionalist as a "former lawyer or business executive who has ditched
her career for a life of cookie-baking and car pooling." Susanne Trowbridge, Wife, Yes;
Housewife, Never!, Baltimore Sun, Feb. 26, 1993, at 17A. Some critics have challenged this
image for making difficult choices look easy or natural, or the often stressful juggling of roles
look manageable. See Hanson, supra, at 1.
30. See Rasmusen & Stake, supra note 1, at 463. This claim strikes me as dubious and is
not fully supported by the evidence the authors cite. They offer some, albeit contested, data that
suggest that active participants within a number of religious faiths are less likely to divorce. See
id. at 463 n.47. Yet their evidence does not demonstrate that the support of religious
communities is being sought out by irreligious members of traditional families who would
otherwise have little interest in affiliating with these communities. A more plausible
interpretation of their data is that the preference for more traditional and/or more durable
marital bonds and the preference for deep affiliation with religious communities are mutually
reinforcing and tend to be found, simultaneously, in the same individuals. In addition, Stake
and Rasmusen's worry that this resort to the support of religious communities will
"conservatize[] the social fabric of society," id. at 464, seems specious, given that their
proposal's support for the traditional, gender-differentiated, nuclear family is likely to do
precisely the same thing.
31. Id. at 467. To be fair to Stake and Rasmusen, this phrase is taken, not from any overt
condemnation of working women, but from a section of the paper in which they argue that two-career couples might want to commit to a more enduring (i.e., more difficult to terminate)
marital relation because it would discourage them from "selfish career building at the expense
of family"--a form of behavior they assert is "equally selfish for men and women." Id. at 467
n.62; see id. at 466-69. However, it is not difficult, in the context of their paper as a whole, to
draw the inference that this charge is particularly applicable to the woman who decides to trade
her homemaking for a life in the market economy. The incentives provided by no-fault rules,
which "press both spouses to devote their time away from family and [pursue] cash income,"
id. at 467, have operated primarily on homemaking women, the only adult family participants
who were not devoting their time to pursuit of cash income in the first place. Moreover, the
range of arguments they offer to suggest that the two-career family disadvantages children, and
the larger society, see id. at 482-84, also make clear that they view this charge as applicable to
the career-oriented woman. See discussion infra Part II.B.
32. Id. at 484.
33. Id. at 483-84. I cannot resist expressing some skepticism about these unsubstantiated
positive externalities. The monitoring gains seem likely to be minimal at best: anyone who has
been charged with providing for the safety and entertainment of one or more infants, toddlers,
or preschoolers knows there is little energy or attention left over for such monitoring. I can
personally attest to at least one occasion when an apartment building down the street burned
virtually to the ground, and, engrossed in the care and feeding of my two children, I did not
even notice until the fire truck pulled up in front of my house to get access to a nearby hydrant.
As for the costs to other children from having the sick children of two-career parents attend
school or child care, the authors may have more of a point. However, in the preschool situation,
where there is undoubtedly the greatest problem with contagious diseases, the configuration
of the problem is less an externalization of costs onto innocent third parties than a prisoner's
dilemma or collective-action problem. Most children enrolled in preschool have two parents
who work or have some nonfamilial commitment that is not easily canceled on short notice.
These parents might arguably be best off if no one sent sick children to child care, but would
certainly be worst off if they kept their sick child home while other parents sent their sick
children to child care. Particularly given the difficulties of missing work, most parents resolve
the dilemma by sending children who are not dramatically ill, but may be slightly ill or not fully
recovered, to child care at one time or another. It is surprising, given that the collective-action
problem involved does not seem insuperable, that groups of parents do not get together to
lobby child-care centers to establish sick-child rooms, or failing that, pay a retainer to an
experienced child-care provider to be available for sick-child care to the group. It may be that
parents do not regard the costs of having a child who is sick more frequently as so high as to
prompt a collective response, particularly if a sick child can sometimes be sent to child care
anyway. It may be that working parents are sufficiently stressed and time-starved that they do
not have the leisure to look for and solve collective-action problems.
34. Id. at 484.
35. Id.
36. Most studies conducted outside the United States have found that the placement of
children in day care has either positive or no effects on children's cognitive abilities in
elementary school. See Anders G. Broberg et al., Effects of Day Care on the Development of
Cognitive Abilities in 8-Year-Olds: A Longitudinal Study, 33 Developmental Psychol. 62,
62 (1997). Most studies in the United States have reached similar conclusions. See Jay Belsky
& Laurence D. Steinberg, The Effects of Day Care: A Critical Review, 49 Child Dev. 929, 931
(1978) ("day-care experience has neither salutary nor adverse effects on the intellectual
development . . . of most children"); Margaret O'Brien Caughy et al., Day-Care Participation
as a Protective Factor in the Cognitive Development of Low-Income Children, 65 Child Dev.
457 (1994) (finding early enrollment in child care associated with higher reading recognition
and math scores for low-income children, although with lower scores for children from more
optimal environments). One recent study found that children whose mothers were employed
full-time beginning in their first or second year were significantly more noncompliant than
agemates without such early experience. See Jay Belsky & David Eggebeen, Early and
Extensive Maternal Employment and Young Children's Socioemotional Development:
Children of the National Longitudinal Survey of Youth, 53 J. Marriage & Fam. 1083 (1991).
This study has, however, been controverted. See, e.g., Sandra Scarr, On Comparing Apples and
Oranges and Making Inferences About Bananas, 53 J. Marriage & Fam. 1099 (1991);
Deborah Lowe Vandell, Belsky and Eggebeen's Analysis of the NLSY: Meaningful Results or
Statistical Illusions?, 53 J. Marriage & Fam. 1100 (1991).
37. See Rasmusen & Stake, supra note 1, at 484.
38. Greater paternal involvement with children in two-parent families has been correlated
with an increased maturity in children's adaptive and social behaviors. See Adele Eskeles
Gottfried et al., Maternal Employment, Family Environment, and Children's Development:
Infancy Through the School Years, in Maternal Employment and Children's
Development 11 (Adele Eskeles Gottfried & Allen W. Gottfried eds., 1988).
39. See Rasmusen & Stake, supra note 1, at 483-84. The authors argue that specialization
in intrafamilial investments should be viewed as an advantage, and that this advantage is more
likely to be secured if the parties do not have to worry that their investments could be rendered
useless by the dissolution of the marriage. See id. at 466-68. Specialization in child-rearing
investments (e.g., I learn about early elementary curricula, you learn about soccer and soccer
leagues), can be equally advantageous, with less worry that the investments can be rendered
useless by the dissolution of the marriage. Particularly in an era of joint-custody agreements,
many parents retain substantial responsibility for their children, even in the event of divorce.
So having particular areas of expertise in relation to children would be likely to remain
valuable--and perhaps even useful in reducing disputes over the division of responsibility or
labor--in the case of divorce.
40. See generally Martha Albertson Fineman, The Neutered Mother, The Sexual
Family, and Other Twentieth Century Tragedies (1995).
41. Fineman argues that we need to replace the nuclear family--or what she calls the
"sexual family," built around the dyad of the sexual couple--with a notion of family that puts
the child and care giver at the center and creates a range of relationships between this dyad and
other supportive public and private institutions. See id.
42. See statistics cited supra note 22.
43. Concentration of this responsibility would also leave children financially vulnerable to
the primary wage earner's unexpected illness or death, or any market events that tended to
eliminate or diminish his wage-earning potential. Particularly given the difficulties many
homemakers encounter in the transition back to labor-market employment, the lack of
specialization in the two-career family might permit more flexibility or adaptability in the face
of such circumstances.
44. For a thoughtful and provocative argument characterizing mothering as a particular kind
of work, see Sara Ruddick, Maternal Thinking (1989). Ruddick describes the practice of
mothering, or caring for dependents more generally, as creating particular habits of mind and
processes of thought in anyone who undertakes it. See id. at 23-27. However, inasmuch as it
requires particular kinds of acts and mental habits, one could also imagine that there are those
who are better suited to it at the outset, even though the work itself would ultimately shape
anyone who undertook it.
45. Stake and Rasmusen are not the only ones to reach this conclusion. Feminist family-law
scholars have offered similar observations, though the normative proposals they offer to
address this increased vulnerability are different. See, e.g., Weitzman, supra note 21.
46. For articles critical of a proposed return to fault-based schemes, particularly in relation
to their effects on women, see Ira Mark Ellman & Sharon Lohr, Marriage as Contract,
Opportunistic Violence, and Other Bad Arguments for Fault Divorce, 1997 U. Ill. L. Rev.
719, and Jana B. Singer, Divorce Reform and Gender Justice, 67 N.C. L. Rev. 1103 (1989).
47. See Singer, supra note 45, at 1106-07.
48. See id. at 1110-12 & n.43 (arguing that fault divorce, including alimony, was justified
for wife's failure to live in domicile chosen by husband, failure to prepare husband's meals,
or indifference to husband).
49. See, e.g., Rasmusen & Stake, supra note 1, at 466 n.59.
50. The fact that women's attractiveness as potential partners in a subsequent marriage, and
consequently women's negotiating power within an existing marriage, declines with their
advancing age is much discussed in a burgeoning literature applying economic analysis to
family law. See, e.g., Lloyd Cohen, Marriage, Divorce, and Quasi Rents; or, "I Gave Him the
Best Years of My Life", 16 J. Legal Stud. 267 (1987); Wax, supra note 7. My objection to this
literature is not with its conclusions, which strike me as plausible, but with the way these
conclusions are stated. Women's decreasing marriageability is described as a characteristic that
inheres in women, rather than as a result of dominant tastes among men, some of which bear
the inflection of various forms of sexism. The result of this emphasis is that men and women
may conclude that these are inalterable features of femaleness, rather than social norms that
may be altered over time by exposing their sexist underpinnings. Thus, in discussing women's
declining negotiating power, I have attempted to highlight the often sexist attitudes that have
contributed to this result.
51. Women may enter into a range of financially disadvantageous property settlements in
order to prevent challenges to their custody of their children. See, e.g., Jana B. Singer &
William L. Reynolds, A Dissent on Joint Custody, 47 Md. L. Rev. 497, 515-18 (1988).
52. See Rasmusen & Stake, supra note 1, at 477-78. Here, I would definitely have to credit
the perspective of Rasmusen and Stake's "liberal" who argues that criminal law is not enough
to sanction, let alone deter, this spousal abuse and related behavior.
53. Of course, other bargains are possible, but this is the paradigmatic bargain that Stake
and Rasmusen discuss in connection with the protection of the traditional family.
54. A related argument was offered by Peter Kramer in an op-ed response to Louisiana's
law establishing "covenant marriage." See Peter D. Kramer, Editorial, Divorce and Our
National Values, N.Y. Times, Aug. 29, 1997, at A23. Kramer argues that Louisiana's return
to a fault-based option for marriage is "insidious [because] . . . contrary to claims on its behalf,
it is out of touch with our traditional values: self-expression, self-fulfillment, self-reliance. [It]
invites couples to lash themselves to a morality the broader culture does not support . . . ." Id.
Although Kramer observes that our society does a "half-hearted job" of teaching mutuality and
interdependence "which is to say that we still teach these skills mainly to women," id., he
critiques the statute in terms of what it asks of couples, not dependent women. See id.
55. See, e.g., Margaret A. Baldwin, Public Women and the Feminist State, 20 Harv.
Women's L.J. 47, 100-05 (1997); Martha L. Fineman, Images of Mothers in Poverty
Discourses, 1991 Duke L.J. 274, 282-84.