Converted or Unconverted: To Whom Shall We Preach CONVERTED OR UNCONVERTED: TO WHOM SHALL WE PREACH? AMYL. WAX* Feminist legal scholarship has given insufficient attention to key analytic approaches that figure centrally in other areas of legal scholarship. This failure of feminist scholarship to make use of methods that have enriched other scholars' work has hurt the journals that specialize in publishing feminist works. As participants at this conference have complained, feminist law journals are generally marginalized and ignored by all but a small coterie of scholars. Yet, in providing an outlet for work on topics of special interest to women-work that other journals cannot or will not accommodate-feminist journals play a role similar to that of other journals devoted to specialized subjects. Because the issues that feminist legal scholars address are of central concern to society as a whole and affect legal developments in many areas, feminist law journals have the potential to survive, thrive, and command the respect and attention of academics from a broad range of fields. The fate of feminist publications is not inevitable. This essay argues that there is nothing wrong with feminist law journals that reforming feminist legal scholarship will not cure. There are three methodological approaches to which feminist legal scholars g ive short shrift and which ought to figure more prominently in their work. These are economic analysis, empirical social science, and evolutionary theory, including evolutionary psychology. The first, economic analysis, is not easy to define because it is not monolithic. In general, economics attempts to identify the costs and benefits of rules for social actors and to do so systematically and quantitatively. This method assumes that social actors, including persons and organizations, seek to satisfy preferences in their interactions with one another on the market and in other institutional settings. The second method, social scientific empiricism, recognizes that assertions about the existence and extent of social problems, the identity of their causes, or the desirability of potential cures, cannot be established a priori as a matter of theory only. This method is committed to investigating social conditions and evaluating proposed policies by collecting factual evidence and analyzing data using rigorous, statistically sound procedures. The final approach that feminist legal scholars unduly ignore, or even denigrate, is evolutionary analysis. Evolutionary theory seeks to offer a scientifically grounded account of human psychology and behavior as an outgrowth of competition over time * Professor of Law, University of Pennsylvania Law School. Converted or Unconverted among individuals and groups for scarce resources. The rubric covers methods ranging from conventional Darwinian explanations to efforts to understand complex group interactions using game-theoretic models developed by economists and social scientists. Although feminist legal scholars do occasionally address work i n economics, social science, and evolutionary theory, their aversion to these methodologies is more a matter of omission than commission and therefore is not easy to document. Familiarity with the literature suggests that there are few, if any, feminist legal scholars active today who are willing to employ these approaches directly in their own work. Nor do many engage in active, sustained, and analytically sophisticated discussion of scholarship coming out of these fields. In contrast, feminist scholars outside law are not so dismissive of these methodologies, and many work productively within them. Feminist economics is a burgeoning field, with a new journal to showcase its products.' Women social scientists have long focused fruitfully on feminist concerns, including women's labor market participation, work and family, reproductive and marital behavior, and domestic violence, among others. Female evolutionary psychologists are strikingly numerous and prominent within that field.2 Moreover, the avoidance of evolutionary models is not a uniform feature of legal academia; (mostly) male law professors have applied the evolutionary paradigm to behavioral and social phenomena that are central to feminist concerns.3 In their own work however, feminist legal academics all too often fail seriously to engage this scholarship or to enrich their own discussions by employing these methods of analysis directly.4 There are several reasons why economics, social science, and evolutionary models ought to figure more prominently in feminist legal ' See Feminist Economics, a journal published by Routledge, Taylor and Francis Group. See also the works of Nancy Folbre, Shelley Lundberg, Robert Pollak, Paula England, Jane Humphries, Claudia Goldin, and Francine Blau, among others. 2 See Steven Pinker, The Blank Slate: The Modem Denial of Human Nature 342 (2002) (noting prominence of women in the field of evolutionary psychology). 3 See Pinker, supra note 2; see also Anne Campbell, A Mind of Her Own: The Evolutionary Psychology of Women (2002); Martin Daly & Margo Wilson, Homicide (1988); Sarah Blaffer Hrdy, Mother Nature: A History of Mothers, Infants, and Natural Selection (1999); Randy Thomhill & Craig T. Palmer, A Natural History of Rape: Biological Bases of Sexual Coercion (2000). Also see the works of Kingsley Browne and Owen Jones. See, e.g., Kingsley R. Browne, Sex and Temperament in Modem Society: A Darwinian View of the Glass Ceiling and the Gender Gap, 37 Ariz. L. Rev. 971 (1995); Kingsley R. Browne, Women at War: An Evolutionary Perspective, 49 Buff. L. Rev. 51 (2001); Owen D. Jones, Time-Shifted Rationality and the Law of Law's Leveraze: Behavioral Economics Meets Behavioral Bioloey 95 Nw. U. L. Rev. 1141 (2001). 4 F or a n example o f a n a rticle t hat does address t he application of evolutionary theory to feminist subjects, see Katharine K. Baker, Biology for Feminists, 75 Chi.-Kent L. Rev. 805 (2000). 20031 Columbia Journal of Gender and Law scholarship. First, skilled use of these methods to address problems pertinent to women--or, at the very least, to discuss the academic work of others that bears on these issues-is essential if feminist scholarship is to lose the taint of parochialism and gain the respect and attention of a wider scholarly audience. Feminist scholars must decide whether their mission is to preach to the converted or to the unconverted. If the mission is perceived as shoring up support among those who already agree with the central tenets of the feminist agenda, then rigorous engagement with different approaches and contrasting perspectives is unnecessary and may even get in the way. But the refusal to engage important scholarly results and to confront and wrestle with potentially uncomfortable conclusions emerging from other fields makes it too easy for those without sympathy for a feminist outlook to dismiss feminist scholarship as unworthy of serious engagement or consideration. If the goal is to influence mainstream scholars who think about social problems of concern to women, then attention to evidence and arguments generated by those working outside the feminist camp is essential. Second, the quality of feminist scholarship could benefit greatly from a willingness to test its arguments with rigorous methodologies and learning from other fields. Preaching to the converted is an invitation to sloppy thinking, illogic, and disregard of empirical reality. Laxity in method leads all too often to error in result. If the ultimate goal of feminist scholarship is to devise a program of social reform that will actually improve the lives of women, then it is important to start from an accurate picture of the social problems to be solved and the likely effect of proposed reforms and solutions. Feminists' prescriptions are worse than useless unless their description of the status quo is accurate, their diagnosis of the causes of objectionable conditions is correct, and there is evidence that the recommended remedies will actually work. If truth is not on feminists' side-if advancing the feminist agenda requires ignoring, hiding, and distorting the facts, engaging in slipshod reasoning, or disregarding careful scholarly work that relies upon rigorous and well-tested methods-then perhaps the agenda deserves the label of naked ideology and warrants the skepticism, dismissal, and marginalization of which feminists complain. The final reason that feminist legal scholars should pay attention to economics, social science, and evolutionary psychology is that, if properly employed and understood, these methods can help support feminist assertions and advance feminist priorities. And even if direct support is not always forthcoming, careful consideration of results that seem to challenge feminist ideas can help scholars rethink and refine their work in ways that ultimately make it more persuasive. There are many examples in which economics, social science, and evolutionary theory point to arguments in favor of feminist priorities or suggest opportunities to strengthen feminist positions. One area of feminist concern is the status of persons who care for others in our society. Feminist [Vol. 12:3 Converted or Unconverted theorists, including legal scholars, have long called for a collective commitment to financial support for caregivers and dependents. Martha Fineman, for instance, suggests that marriage be abolished as a legal category, to be replaced by caretaker-dependent units entitled to a government allowance.' In support of this claim, Fineman and others argue that caregiving is a form of work, and that domestic work is valuable and no less entitled to compensation than conventional forms of market labor.6 The call for more generous social support for caretakers, although a centerpiece of the feminist agenda, does not command uniform support from analysts outside feminist circles. Nor does it have much purchase with the citizenry as a whole, as evidenced by its lack of success in the political arena. (Indeed, recent changes in welfare programs move, if anything, in the opposite direction). If feminists are to gain ground on this priority, they must persuade others of the justice of their cause. It is impossible to argue coherently for this position, however, without taking economics seriously. The call for collective support for caretakers runs up against the objection that it proves too much. If mothers receive a social subsidy, then why not artists, gardeners, volunteer workers, and others who engage in effortful exertions outside conventional labor markets? Economic analysis provides a framework for dealing with this objection. Neoclassical economics starts from an assumption that markets work. If the outcome of a transaction that trades services for compensation is Pareto-superior in that it generates a net increase in utility for all concerned, then that transaction will ordinarily take place. That is, if a service or product provides value, someone will offer to pay for it and pay what it is worth. Because the market is regarded as a good measure of value, economists call upon feminists to justify collective compensation for caretakers by explaining why there is insufficient market demand for the services, or why the price of those services is too low. For economists, there are two alternatives: either t he service i s not a s valuable a s claimed, o r there i s some kind o f market failure that impedes transactions that would reflect the benefits those services confer on others. The framework these options set out provides a potential roadmap for building a compelling case for collective support of caretakers. Ethic-of- care feminists' call for support cannot succeed among the "unconverted" unless it confronts these alternatives systematically and thoughtfully, either 5 See, e.g., Martha Albertson Fineman, The Neutered Mother, the Sexual Family, and Other Twentieth Century Tragedies (1995). 6 See id.; see also Anne Alstott, No Exit: What Parents Owe Their Children and What Society Owes Parents (2003); Nancy Folbre, The Invisible Heart: Economics and Family Values (2001); Eva Feder Kittay, Love's Labor: Essays on Women, Equality, and Dependency (1999); Amy L. Wax, Something for Nothing: Liberal Justice and Welfare Work Requirements, 52 Emory L.J. 1, 30-35 (Winter 2003) (discussing claims made for care work). 2003] Columbia Journal of Gender and Law by showing that caretakers' poor showing on markets stems from market failure (rather than from the meager value of services provided), or by explaining why these alternatives fail to exhaust the universe of possibilities or are too narrow or misleading. Consider first the claim that caretaking services, such as a mother's care for her child, are not as valuable as claimed. It is useful to tease out who benefits from a mother's work and how much. The provider herself benefits: she receives what economists term "psychic consumption value." The child who is cared for enjoys the caretaker's attention, a ffection, n urture, and m atenial support. Finally, t he public at large, including parents and non-parents alike, is better off when children are brought up as law-abiding, worthy, and productive citizens. Finally, in modern Western societies, the benefits that accrue to the adult population from parents' labors in producing and rearing children is amplified by the operation of public old-age pension systems (such as our own Social Security), which rely on young workers to generate the revenues that support the elderly.7 Because retirees receive Social Security payments whether or not they have invested any personal effort in rearing the next generation, Social Security effectively allows non-parents to "free ride" on parents' childrearing efforts. The lack of any binding social or legal mechanism for compensating parents for the benefits they bestow on society as a whole generates what economists call a "positive externality." Economic theory predicts that these positive externalities will operate to create an undersupply (or, alternatively, an undervaluation) of the service- here, caretaking work-that produces them. Something less than an optimal or efficient amount of these services will be supplied . In other words, the positive externalities, or uncompensated benefits, generated by caring work represent a market failure that arguably justifies intervention in the form of public financial support for caretakers to help correct inefficient distortions. 7 Although Social Security is believed to be an "insurance" program financed by the employment taxes deducted from retirees' paychecks and "saved" for their retirement, that characterization is an illusion both in law and practice. The retired elderly today receive benefits far in excess of the actuarially sound return on their past contributions to the Social Security program. Benefits payments are financed on a pay-as-you-go basis by current workers' social security taxes. Although many elderly believe that the taxes they paid during their many years of work have created an entitlement to Social Security benefits that is equivalent to the contractual obligation incurred by private annuity funds, the Supreme Court has expressly repudiated such a parallel entitlement. See Flemming v. Nestor, 363 U.S. 603 (1960). There is no ineluctable legal right to a particular level of benefits or "return" on lifetime contributions to the Social Security program. Id. at 609-10. 8 See Amy L. Wax, Caring Enough: Sex Roles, Work and Taxing Women 44 Vill. L. Rev. 495, 512-16 (1999); see also Paula England & Nancy Folbre, Who Should Pay for the Kids?, 563 Annals Am. Acad. Pol. & Soc. Sci. 194 (1999); AmyL. Wax, Is There a Caring Crisis?, 16 Yale J. on Reg. 327, 338-49 (1999) (reviewing Shirley P. Burggraf, The Feminine Economy and Economic Man: Reviving the Role of the Family in the Postindustrial Age (1999)). [Vol. 12:3550 Converted or Unconverted Yet another source of market failure that fuels the case for public subsidy or intervention is the benefit that children themselves receive: children cannot pay on the spot for caretakers' services. At most, they can make a promise of delayed compensation in the form, for example, of support for parents in old age. Such promises are notoriously difficult to enforce, and the informal norms that bound children to these duties in the past have eroded under the pressures of modernity. The Social Security system is designed to improve upon the failures and uncertainties of the traditional system by collectivizing care for the elderly, but the program has the perverse effect of favoring non-parents over parents in the process. Results obtained from empirical econometrics also buttress claims for assisting caregivers. Using statistical techniques to compare jobs with similar levels of qualification and demands, economists have found evidence that care work is undervalued even in the formal labor market setting.9 These observations suggest that a market failure might be operating, with the result that potentially valuable services are undersupplied. In light of the potentially detrimental consequences of these distortions, it makes good economic sense to take steps to correct the imbalance through regulatory intervention or subsidy, both for reasons of efficiency and fundamental fairness. Although there is much room for argument about the degree and type of support that is justified, the position that parents deserve some kind of collective aid has much to recommend it. As noted above, one alternative to advancing ethic-of-care claims by identifying a market failure is to challenge the "markets or market failure" paradigm by arguing that these alternatives take too narrow a view of the realities of social life or distort the terms of discourse. Feminists may bristle at applying economics to argue for social support for caretakers on the grounds that the exercise indulges assumptions-such as that all persons operate as "rational" economic men (and women) motivated by pure self- interest-that are fundamentally antithetical to an "ethic-of-care" approach. Rather, caretaking is about giving and selflessness. It is about what is right, fair, and just. It advances a moral vision of what we owe one another and how we should live as a group--a vision about which economics has (or should have) little to say. This objection, although ably advanced by some scholars and undeniably important to sorting out the ethic-of-care debate,'0 does not ultimately carry the day. Economic analysis remains useful in clarifying the assumptions implicit in feminists' calls for changes in policy and for assessing the real-life consequences of recommended reforms. Moreover, it 9 See, e.g., Paula England et al., Wages of Virtue: The Relative Pay of Care Work, 49 Soc. Prob. 455 (2002). 10 See Elizabeth S. Anderson, What is the Point of Equality?, 109 Ethics 287 (1999); see also Martha C. Nussbaum, Disabled Lives: Who Cares?, N.Y. Rev. Books, Jan. 11, 200 1, at 34. 2003] Columbia Journal of Gender and Law is a mistake to assume that economics excludes consideration of altruistic motives. A complete picture of human behavior requires that these be included in the package of tastes and preferences that comprise individual "utility functions."'" Indeed, much cutting edge work in economics is now devoted to describing and analyzing behaviors that cannot be explained as narrow expressions of selfishness or self-interest.1 2 On the other hand, systematic investigation and empirical evidence suggest that pure altruism rarely operates as the sole engine of action in any sphere of human existence. Rather, sound policy depends on developing an accurate, empirically valid picture of the mix of other-regarding and self-regarding considerations that underlie human decision-making. Unfortunately, the case for subsidizing caretaking does not make itself and must take account of real world motivations in which self-interest is not all or none. If ethic- of-care claims are simply an appeal to pure altruism--to the duty to benefit the next generation without any hope of return-then they should be identified as such and defended on their own terms. But then ethic-of-care feminists cannot avoid the hard work of advocating their position in the face of competing claims for public regard and resources. Feminists must explain why we should care about children to the detriment of other worthy goals. Economic analysis provides a systematic framework for testing feminists' intuitions in this area as well as persuading those with differing views. The second methodological approach to which feminist scholars should pay more attention is evidence-based social science. Empirical legal scholarship, which is often conducted by law professors with interdisciplinary, specialized training, is on the rise but remains unusual. Legal scholars rarely present original experiments or data they have generated themselves, as they often lack the resources and training for first- line empirical work. Increasingly however, legal scholars have paid more 1 See, e.g., Ernst Fehr & Simon Gachter, Altruistic Punishment in Humans, 415 Nature 137 (2002) (showing that cooperation "flourishes if altruistic punishment is possible," id. at 137); Ernst Fehr & Simon Gachter, Cooperation and Punishment in Public Goods Experiments, 90 Am. E con. Rev. 980 (2000) (providing evidence that punishment occurs, despite personal costs, and intensifies the "more an individual negatively deviates from the contributions of the other group members," id. at 993); Karl Sigmund et al., The Economics of Fair Play, 286 Sci. Am. 83 (2002) (explaining human altruism as a function of "achieving biological success in complex social networks," id. at 87); Lynn A. Stout, In Praise of Procedure: An Economic and Behavioral Defense of Smith v. Van Gorkom and the Business Judgment Rule, 96 Nw. U. L. Rev. 675 (2002) (economic analysis of other- regarding preferences); Lynn A. Stout, Other-Regarding Preferences and Social Norms, Georgetown Law and Economics Research Paper, No. 265902, and Georgetown Public Law Research Paper, No. 265902 (March 2001), available at http://papers.ssrn.com/sol3/. For recent work in the field of empirically grounded behavioral law and economics, see, for example, Symposium, Empirical Legal Realism: A New Social Scientific Assessment of Law and Human Behavior, 97 Nw. U. L. Rev. 1075 (2003). '2 See, e.g., discussion infra note 18 and accompanying text. 552 [Vol. 12:3 Converted or Unconverted careful attention to information generated, gathered, and analyzed by social and natural scientists, and have become more sophisticated and systematic in their use of available observations to analyze the wisdom and likely effects of legal rules. How might greater attention to social science data and methods enrich feminist l egal s cholarship? One a rea t hat c ould b enefit i s feminist analysis of laws governing marriage and child-parent relations. Many feminists take the position that the institution of marriage is oppressive and harmful to women. They regard out-of-wedlock childbearing and single- parenthood as viable social forms that can improve women's lives. Because these assertions are empirically testable, they cry out for systematic assessment in light of facts. In evaluating these claims and recommendations, anecdotes should not substitute for a statistically sound investigation of the effects of institutions and practices on real people. Yet anecdote, rather than sound analysis, often rules the debate. Basic assumptions that serve as starting points for discussion often lack evidentiary support or fly in the face of existing data. The recommendations that emerge from such shaky foundations are unlikely to command the respect of critical thinkers or mainstream policymakers. The picture on the consequences of marriage for women is complex and decidedly mixed, with available social science evidence suggesting that there is both good news and bad news for women in marriage. On the one hand, married women do as well or better than single women on a variety of indicators. 13 Moreover, the data indicates that many women are eager to marry and stay married, which means that they regard themselves as better off married than single. Finally, although marriage rates have declined somewhat and age of first marriage has increased, the great majority of women marry at some time in their lives. Straightforward economic analysis and simple common sense point to the conclusion that men and women marry and stay married because they see marriage as improving their lives. On the other hand, there is evidence that men gain more from marriage than women. Women's priorities get less play within the marital relationship and married women's labor market prospects suffer relative to single w omen. In e conomic terms, men command t he l ion's s hare o f t he "surplus" utility or benefits generated from the marital partnership.14 The reasons for these patterns are complex and can best be understood by employing competitive market and bargaining models that take cognizance of differences in men's and women's preferences, reproductive behavior, and labor market prospects. 13 See, e.g., Linda J. Waite & Maggie Gallagher, The Case for Marriage: Why Married People Are Happier, Healthier, and Better Off Financially (2001). 14 For more on this issue, see Amy L. Wax, Bargaining in the Shadow of the Market: Is there a Future for Egalitarian Marriage?, 84 Va. L. Rev. 509 (1998). 2003] 553 Columbia Journal of Gender and Law The foregoing discussion shows that attention to evidence and economic modeling does not always make trouble for feminist intuitions. The analysis that indicates that men benefit more from marriage than women and that caretaking is undervalued should add to the credibility of feminist claims and command the attention of those who would otherwise be disinclined t o h eed feminist c omplaints. O n t he o ther h and, e vidence- based analysis can yield challenges for feminist positions and sometimes warrants reassessment of their conclusions. This point applies to the feminist view on single-parenthood, which is bound up with concern for poor m others and children. T he c ase for aid and s upport t o single-parent families and tolerance of out-of-wedlock childbearing is often coupled with assertions that (1) many poor women's failure to marry before bearing children is "not their fault," but rather can be traced to concrete obstacles, such as lack of access to contraception and a scarcity of marriageable men due to high unemployment and incarceration; and that (2) single-parent families can function as well as more traditional two-parent units. Unfortunately, the demographic evidence offers little support for these propositions. Consider, for example, the debate surrounding the reproductive b ehavior o f African American w omen, w ho h ave a n o ut-of- wedlock birth rate roughly three times as high as white women and a much lower marriage rate. Contrary to common assumption, these patterns cannot be explained by high rates of unemployment and incarceration of African American men. As Christopher Jencks and Orlando Patterson, among others, have shown, the marriage rate has declined dramatically even among African American men with jobs, and incarceration among African American men accounts for at most a fraction of low marriage rates in this community. 5 Moreover, any claim that single-parenthood among the poor is a product of the unavailability of contraceptives-which leaves women with no "choice" but to become pregnant--does not withstand scrutiny: ethnographic studies indicate that most young women who bear children out of wedlock do not consistently use contraception despite its availability.' 6 The picture that emerges from these data on marital behavior and out-of- wedlock childbearing suggests that these patterns are not the direct product of ineluctable obstacles or externally imposed conditions such as a simple absence of jobs, a disparity in the number of male and female partners, or 15 Christopher Jencks, Rethinking Social Policy: Race, Poverty, and the Underclass 130-36 (1992) (showing that increases in unemployment made only a "modest contribution" to decline in marriage rates among black males, id. at 133); Orlando Patterson, Rituals of Blood: Consequences of Slavery In Two American Centuries (1998) (ascribing low marriage rates among blacks primarily to cultural rather than economic factors); see also June Carbone, Has the Gender Divide Become Unbridgeable? 36-46, 53-56 (2003) (draft on file with author). 16 Kristin Luker, Dubious Conceptions: The Politics of Teenage Pregnancy 139-51 (1996) (discussing teenagers' use of contraception). 554 [Vol. 12:3 Converted or Unconverted lack of access to birth control. Rather, they are the product of choices and decisions made by individual men and women in light of evolving norms, mores, and cultural patterns. It is much harder .to convince society to support single parents on a model that points to a key role for individual values and initiative rather than demographic necessity and economic hardship. Yet feminists would do well to face up to these realities and to engage the full complexity of behavior in these areas. Failure to do so undermines feminists' ability to convince those who take issue with their views. A similar point can be made about the position that the law should refrain from favoring two-parent families and should no longer "privilege" and encourage stable heterosexual marriage. The claim is that single-parent families are potentially as stable and effective as two-parent families and that children need be no worse off for growing up with only one parent. To the extent t here a re disadvantages, a n adequate commitment o f resources can ameliorate them. Unfortunately, evidence painstakingly assembled by social science over many years casts doubt on these positions. Although the jury is still out on two-parent homosexual families, the evidence on heterosexual families is abundant and unequivocal. Children in blended, step-parent, divorced, and single-parent families have worse outcomes on many measures of social and individual well-being than those raised in stable, intact biological families with equivalent socioeconomic status and financial resources.1 7 This strongly suggests that money cannot wholly compensate for the drawbacks of being raised in s ingle-parent or broken homes, and that biological fathers within intact families enhance the well- being of children. That the feminist agenda in favor of single-parenthood runs up against these inconvenient facts does not spell the end of the program. As noted above, maintaining stable two-parent families within conventional heterosexual marriage is a project that entails some costs for women. Creating conditions that enhance women's personal freedom and foster their occupational success is perhaps worth some degree of sacrifice in children's well-being-but that is an argument for tradeoffs rather than for unalloyed good and admittedly a harder sell to the "unconverted" who have different ideas about the appropriate order of social priorities. Yet the evidence demands that these arguments be confronted if feminist positions are to persuade and prevail. Another area where attention to empirical observation and economic theory is essential to an assessment of feminists' claims and 17 See evidence summarized and collected by June Carbone, From Partners to Parents: The Second Revolution in Family Law (2000). See also Mary Parke, Are Married Parents Really Better for Children? What Research Says About the Effects of Family Structure on Child Well-Being, Center for Law and Social Policy, Brief #3 (May 2003), available at http://www.clasp.org. 2003] Columbia Journal of Gender and Law priorities and has the potential to provide powerful arguments in favor of the feminist perspective concerns the reconciliation of work and family life. Feminist legal scholars interested in improving women's employment status have advocated family friendly workplace reforms that make it easier to combine work with women's primary responsibility for children. The call for family friendly workplace reform runs up against the contention that workplace practices currently in place have been adopted because they are efficient. Therefore interventions that seek to rearrange work and compensation schedules will shrink the pie of resources available to all. Game-theoretic models of workplace interaction are potentially useful in revealing the flaws in this line of reasoning: competition among employees in the face of employers' difficulties in evaluating employee quality can lead to inflexible workplace patterns that are inefficient in light of employee preferences. This suggests that reforms that feminists seek- such as mandated family and medical leave, and greater flexibility in work schedules, demands, and conditions-may actually prove more efficient than the status quo and hold the potential to make everyone better off.18 Yet another example of the relevance of social science and economic models to feminist concerns pertains to the feminist hostility to the recent push, initiated by the Clinton administration and carried forward under President Bush, to reform the federal welfare system by imposing strict work requirements as a condition for receiving cash benefits. These measures have been condemned as punitive, destructive, and harmful to children and to the women who stayed home to care for them under pre- reform welfare programs. Available evidence simply does not support these claims of harm. A recent study of thousands of low-income families found no significant association between mothers' move from welfare to work and negative outcomes for children. Indeed, adolescents of employed mothers showed improvements in behavior and mental health.' 9 These results should cause those concerned with the well-being of poor women and children, including feminist legal scholars, to rethink their positions on work-based welfare reform. Finally, another area of inquiry that holds the potential to enrich feminist scholarship is evolutionary theory, particularly evolutionary psychology. Although a systematic treatment of the implications of evolutionary thought for feminist analysis is an extensive undertaking that is well beyond the scope of this paper, a few points are worth noting here. First, although feminists in general have l ong been chary o r even h ostile 18 See Amy Wax, The Family Friendly Workplace: Making the Case for Change (draft on file with author). 19 See P. Lindsay Chase Lansdale et al., Mothers' Transitions from Welfare to Work and the Well-Being of Preschoolers and Adolescents, 299 Sci. 1548 (2003). [Vol. 12:3 Converted or Unconverted towards evolutionary analysis, 20 a nuanced assessment of the best work in this field reveals that this stance is overdrawn. The evolutionary analysis of behavior-which attempts to understand individual psychology and human social interactions as an outgrowth of tendencies formed in response to evolutionary forces-is not the bete noir that many feminists fear. A sophisticated application of evolutionary theory does not yield the cardboard portrait of human (and female) possibility that feminists find so threatening. To be sure, feminist wariness is not wholly without foundation: as a conceptually and intellectually demanding field, evolutionary theory is vulnerable to distortion and misuse by both proponents and opponents alike. 2' Evolutionary science and scholarship that draws insights from it are replete with oversimplification, misunderstanding, and elementary analytical errors. Unfortunately, wholesale hostility to evolutionary approaches as a response to the misuse of its ideas implicitly indulges the same simplistic assumptions and errors that bedevil work drawing upon evolutionary insights. The proper response is not fear or avoidance, but debunking. One persistent error is the assumption that any behavioral pattern or tendency that is characterized as the product of evolutionary pressures must thereby be rigidly "determined" and impervious to environmental influence or modification. So, for example, the view that behavioral differences between men and women have an evolutionary source is taken to imply that neither culture nor socialization can do much to alter the behavioral patterns that men and women display. But evolutionary psychology, properly understood, suggests no such thing. First, the expression of genetically programmed tendencies creates a broad range of behaviors that depend critically on environmental influence. Second, many scholars in social science fields (including law) who draw on evolutionary paradigms downplay crucial aspects of evolutionary theory, such as sexual selection and group selection, that confound attempts to make straightforward predictions about what is behaviorally possible.2 2 It is now thought that evolutionary forces create competition among groups as well as individuals. These multilevel processes generate complex patterns of human conduct in which environmental influences and social conditions play a large role in shaping behavior. In particular, human beings have evolved to be amenable to influence by moral codes of conduct, cultural conventions and practices, and social mores. These tendencies are a key part of our "nature" and 20 See, e.g., Anne Fausto-Sterling, Myths of Gender: Biological Theories About Women and Men (1985); see also Baker, supra note 4, and references cited therein. 21 See, e.g., Janet Radcliffe Richards, Human Nature After Darwin: A Philosophical Introduction (2001); Pinker, supra note 2. 22 See Amy Wax, The End of Evolutionary Psychology: Group Selection and Sexual Selection in Law and Policy (forthcoming in Law and Philosophy 2004). 2003] Columbia Journal of Gender and Law account for much of the flexibility and adaptiveness that make our species uniquely successful. The vulnerability of individuals to cultural influence, even to the point of effectively suppressing individual, competitive self- interested behavior i n favor of group priorities, is as much a part of our 23evolved nature as pure self-interest. Predictions about competitive behaviors that seem to bear straightforwardly on reproductive success-to which some evolutionists point as the source of enduring sex differences- must give way t o the complex, unpredictable forces of social convention and higher cultural forms. These aspects of our nature make it difficult to speak with confidence about the limits and possibilities of human behavior. Although the position that evolution shaped human behavior has undeniable cogency, those who would use evolutionary theory to posit the outer limits of behavioral possibility must proceed with extreme caution. One example of the potential complexity that emerges from evolutionary analysis comes from the work of Sarah Blaffer Hrdy. Hrdy argues that the picture of the passive, nurturing, risk-averse, sexually timid mother found in the literature on the behavioral evolution of the sexes fails to account for the full complement of social and biological factors that bear on women's reproductive fate. A more complete analysis suggests that women's success as mothers depends on securing the survival and flourishing of their children within complex social structures and demands a nuanced juggling act that balances the mother's ability to wield cultural and economic influence with the performance of more elemental nurturing functions. What emerges is a picture of women as flexible and adaptable creatures equipped by "nature" to take on a multiplicity of social roles and to compete with men in a wide variety of spheres as circumstances demand and allow.2 4 The final and perhaps most important reason why feminists should not shrink from engaging work done in the field of evolutionary psychology is that it offers an opportunity to stand firm against fundamental analytical errors that undermine clear thinking about the role o f women in society. With respect to evolutionary or "biological" explanations of human behavior, one central mission must be to identify the naturalistic fallacy in all its persistent and pernicious incarnations and to refute it with rigor and clarity. Because this fallacy rears its ugly head relentlessly, this task is never-ending and calls for constant vigilance. Thus feminists and clear thinkers of all stripes must constantly be on guard against any attempt to move from positive descriptions of reality to normative conclusions. That something is true does not mean that it ought to be true. That a behavior can 23 For more discussion, see Amy L. Wax, Against Nature: On Robert Wright's The Moral Animal, 63 U. Chi. L. Rev. 307 (1996). See also Paul H. Rubin, Darwinian Politics: The Evolutionary Origin of Freedom (2002); Elliott Sober & David Sloan Wilson, Unto Others: The Evolution and Psychology of Unselfish Behavior (1999); Wax, supra note 22. 24 See Hrdy, supra note 3. [Vol. 12:3 Converted or Unconverted be characterized as "natural" makes it neither inevitable nor desirable. (Indeed, the category of "natural" as applied to human conduct is vacuously inclusive: anything that people do is in some sense "natural" in being the outgrowth of brain programs that make it possible.) In sum, the distinction between what is natural and unnatural cannot be the source of any normative justification, nor can it provide any basis for a positive or negative judgment. As stated, these propositions seem obvious. Yet variations on arguments from "is" to "ought" show up frequently in scholarship that draws on economics and evolutionary theory. The specter raised by the naturalistic fallacy holds another key lesson for feminists: although scholars must constantly avoid indulging these solecisms in their work, they must also refrain from condemning entire fields just because those are prone to misunderstanding. Indeed, the blanket rejection of evolutionary psychology based on instances of sloppy thinking only reinforces the naturalistic fallacy by implying that normative conclusions that are claimed to follow from scientific observations really are valid. The feminist response to erroneous naturalistic thinking in evolutionary theory or in any other field should not be hostility or cavalier dismissal, but rather active engagement and a response on the merits. An example of the naturalistic fallacy in action can be found in the work of Kingsley Browne, who has written a number of law review articles on topics of interest to feminists that make use of evolutionary paradigms. In one piece,25 Browne claims that women's failure to achieve high level leadership positions in the workplace is not due primarily to labor market discrimination but rather stems from choices that women make in light of their evolved preferences and desires. As Browne notes, evolutionary theory predicts that women will tend on average to be less competitive than men. Browne claims that this tendency finds expression in women's avoidance of career paths that require fierce competition for scarce leadership positions. This suggests that women's labor market status is most likely due to individuals' preference-driven "supply side" choices rather than to "demand side" discrimination. The trouble comes in the final part of the article in which Browne explores the implications for law and policy of his view that "much of what we see in the workplace" can be traced to a "biological basis for sex differences in temperament. 6 Browne rightly points out that, insofar as these insights suggest that labor market discrimination is not the main culprit impeding women's progress in the workplace, stepping up enforcement of existing anti-discrimination laws is unlikely to be effective 25 Browne, Sex and Temperament in Modem Society, supra note 3; see also Kingsley R. Browne, Biology at Work: Rethinking Sexual Equality (2002) (expanding on themes in Sex and Temperament in Modem Societ, supra). 26 Browne, Sex and Temperament in Modem Society, supra note 3, at 1101. 20031 Columbia Journal of Gender and Law in improving women's prospects. He notes also that much policy discourse in this area, including the call for more aggressive anti-discrimination enforcement, is based on the assumption he challenges: "that men and women are the same in all respects relevant to the workplace. 2 7 However, although he acknowledges that the facts about sex differences should not dictate values and choices, Browne makes clear that he does not favor aggressive social intervention to improve women's labor market status. He observes that arguments that something must be done to address women's stalled progress in the workplace are "based upon an erroneous premise of societal culpability '28 rather than on the more accurate understanding that existing patterns stem from choices driven by individual preferences. Because the justification usually offered for remedial action is grounded in a faulty view of social reality, policies to correct the situation "must rest on an alternative rationale., 29 That rationale, claims Browne, "has not yet been offered."3 ° The final p art o f h is argument i s naive a s a matter o f social a nd political analysis. Assuming that men's higher status, greater power, and stronger financial position are ultimately traceable to tastes that are "hard- wired" and that the relative positions of men and women are the consequence of choices made in light of their respective "natures," there are strong arguments for not leaving these disparities undisturbed. Indeed, the main thrust of a dominant strain of Anglo-American philosophical and political theory provides more support for an interventionist view than for the contrary. According to the position known as "luck egalitarianism," a social system that works to systematically disadvantage one group (e.g., women) should not necessarily be left intact if the status quo results from undeserved or unchosen differences in natural endowment rather than free choices exercised from equivalent starting points.3' Although there is room for dispute about what attributes should be considered "unchosen," and what types of choices are "free," Browne's insistence that the differences between men and women are "hard-wired," unalterable endowments of human nature would, on this view, argue in favor of social intervention or compensation to alter disparate results. In other words, if "natural" differences between men and women account for women's disadvantage, justice requires that society act to correct resulting disparities. 27 Id. 28 Id. at 1102. 29 id. 30 Id. 31 On luck egalitarianism, see Daniel Markovits, How Much Redistribution Should There Be?, 112 Yale L.J. 2291 (2003); Samuel Scheffler, What is Egalitarianism?, 31 Phil. & Pub. Aff. 5 (2003); Amy L. Wax, Something for Nothing: Liberal Justice and Welfare Work Requirements, 52 Emory L.J. 1 (Winter 2003). [Vol. 12:3 Converted or Unconverted To be sure, this critique of Browne's discussion is more a matter of emphasis than overt omission; although Browne nods in the direction of feminist views, he is too ready to dismiss them. His error is in arguing that if women's lower workplace status is due to innate attributes, it necessarily follows that no massive social corrective is warranted. Rather, things should be left as they are. He gives the luck egalitarian view, as such, no play at all. Yet as the vast and sophisticated literature on luck egalitarianism reveals, the laissez faire position requires an extensive defense, and the arguments in its favor are controversial and theory-laden. This does not mean that Browne's causal account is without interest or wholly false. Nor does it mean that there is no warrant for his complaint that feminists focus too narrowly on discrimination as the source of women's labor market status, or his assertion that their program may prove counterproductive. Rather, it is important never to lose sight of the question of what follows from Browne's causal story, even assuming it is true. It is all too easy to let ideology color the response to the question "Now what do we do?" That an able scholar like Browne equivocates on the relationship between his evolutionary account and a laissez faire agenda shows how easy it is to fall into naturalistic thinking. As already noted, however, Browne's article provides a cautionary lesson for feminists. That Browne is perhaps overly eager to use evolutionary principles to tilt against interventionist social policies-in disregard of decades of liberal political thought--does not justify demonizing his method or dismissing his speculation about the causes of observed labor market patterns as presumptively worthless. It does not follow from the fact that his conclusion may be infirm that his premise is without foundation. To indulge that reverse inference is to offer aid and comfort to the enemies of clear thinking everywhere. The final reason that feminists should educate themselves about evolutionary psychology is that it may have something useful to teach them about how best to achieve their desired goals.32 Although defective naturalistic reasoning should be trounced whenever it appears, arguments that exemplify the naturalistic fallacy are not always easy to distinguish from more cogent lines of reasoning that move from fact to value with the aid of suppressed premises. Empirical observations do sometimes carry normative implications in light of shared commitments or values. To label a behavior "natural" or "efficient" can serve as shorthand for predicting that individuals will make certain choices under common conditions, or that patterns of behavior will tend to emerge under competitive pressures, or that certain practices will endure against alternatives. In a society that values freedom, individual choice, the benefits of free markets, and the right to be let alone, such predictions have important implications. In our society, 32 Browne makes this point as well. See supra note 25 and accompanying text. 2003] Columbia Journal of Gender and Law acting on judgments that certain social outcomes are undesirable and should be altered by deliberate intervention may prove costly, entail problematic sacrifices or tradeoffs, or have perverse consequences.33 Although reliance on assumptions about "human nature" to justify complacency in the face of injustice and suffering must be viewed with suspicion, hard-headed concern about real world consequences of social interventions must be part of any humane program of social change. History is replete with examples of the untoward effects of social engineering projects built on a view of people as "blank slates" without common desires or needs.34 Meaningful reform of undesirable conditions, including women's subordination, proceeds best in light of evidence about most people's preferences and responses, if only because it is important to minimize the costs of change and maximize the chance that social policies will work. The real world tradeoffs that most reform programs necessarily entail are hard for some idealists to swallow. In particular, the tendency of people to adjust better to gradual rather than abrupt change makes many reformers impatient. Yet social science suggests that gradualism may sometimes make for the most enduring and least costly outcomes. Similar realism about real-world conditions and caution about confusing fact and value is also in order with respect to economic analysis. Economics cannot tell us what to care about. It cannot define our ultimate goals, tell us what normative commitments to make, nor dictate the preferences we ought to have. Indeed, most conventional economic analysis takes preferences and commitments as given. Once these are defined, however, economic models are very useful in helping sort out the costs and benefits of particular programs and policies. That the tradeoffs economics helps identify are only a starting point for argument does not mean they are irrelevant or can be ignored. A clear picture of the consequences of a course of action can help develop effective methods for making good on ideal commitments and aid refinement and rethinking of cherished goals. Moreover, as noted above, it is a wrong to assume economic analysis has no room for the full range of our values and commitments or necessarily posits a narrowly selfish, "rational" economic man. Economics can canvass values if the preferences included in the cost/benefit equation are defined broadly enough. Although principles of economics cannot, for example, lead us to care about future generations or the continuation of the race, they do not exclude consideration of those sorts of concerns. We may want to reward those who are responsible for children, whether we benefit directly and materially or not. We may decide to commit ourselves to the proper care of the most vulnerable, even at some personal sacrifice. We may harbor a desire that civilization continue after we die, although we receive no 33 See Wax, Against Nature, supra note 23. 31 See Pinker, supra note 2. [Vol. 12:3 Converted or Unconverted tangible benefit from this. To be sure, the amorphous nature of many of these commitments resists precise modeling and measurement. For that reason, "other-regarding" values or ideals are often left out of the equation, a limitation that accounts in part for feminist skepticism towards economic approaches. But that skepticism is less and less justified in the face of sophisticated developments in economic theory, including the recent explosion in experiments and models that take account of choices driven by emotions, social preferences, or utility.3 5 Nonetheless, because economic models can mislead or produce perverse conclusions if human motivation is too narrowly defined, it is essential to the feminist project to remain vigilant in describing the values and commitments that are minimized or omitted. Identifying these types of defects is hard work. It demands critical, case-by- case, sustained consideration of the application of economic principles to social problems, not wholesale and disdainful dismissal of established methods. Feminist legal scholars must strive to develop the necessary expertise to master basic economic analysis and statistical methods and to become conversant with evolutionary ideas. Understanding these fields is the only sound starting point for a productive dialogue between the "converted" and the "unconverted" that, it is hoped, will help win new adherents to feminist causes. Only when feminist legal scholarship earns its stripes of intellectual honesty will feminist law journals be worthy of attention from scholars of all stripes. Only then will legal scholars read and discuss their contents with interest and respect. 35 See, e.g., Colin F. Camerer & Ernst Fehr, Measuring Social Norms and Preferences Using Experimental Games: A Guide for Social Scientists, Jan. 2002, available at http://www.iew.unizh.ch/wp/iewwp097.pdf. 2003] AUTONOMY AND INTEGRATION Choosing Which Master to Serve