Poverty, Reproduction, and Autonomy in the Welfare State: Some Thoughts on the Ethics of Social Policy Legislation POVERTY, REPRODUCTION, AND AUTONOMY IN THE WELFARE STATE: SOME THOUGHTS ON THE ETHICS OF SOCIAL POLICY LEGISLATION Joanna K. Weinberg* I. INTRODUCTION Autonomy for poor women is increasingly limited by a wide array of both state and federal legislative provisions governing social policy. Although the Supreme Court has forestalled the anticipated demise of Roe v. Wade' with its recent decision in Planned Parenthood v. Casey,' it is nevertheless appropriate at this time to focus the attention of pro-choice advocates on the role the state legislatures may play in the continuing struggle to establish reproductive freedom. In this paper I will suggest that reproductive autonomy must be understood within the larger context of the growing conflict between welfare state ideology and women's autonomy generally. This conflict impacts a broad range of issues that affect women's lives. I will address the way in which women's autonomy is threatened not only by overt attacks on abortion rights, but also by limitations on autonomy imposed upon beneficiaries of welfare entitlement programs, who are chiefly women.' I will also discuss the expansion of legislative restrictions on autonomy. * J.D. Harvard, LL.M. Columbia. Fellow, Institute for Health and Aging, University of California at San Francisco; Adjunct Professor, Hastings College of the Law, University of California. The author wishes to acknowledge the valuable research assistance of Anna Newman, Hastings College of the Law 1993. The author also wishes to thank David I. Levine, Martha A. Fineman, and participants in the Feminist Theory Workshop for their input and support. 1 410 U.S. 113 (1973). 2 112 S. Ct. 2791 (1992). 1 In many of the welfare entitlement programs, women are not themselves designated as beneficiaries; AFDC, for example, is technically an entitlement that inures to minor children. However, women are, if not the actual, the surrogate beneficiaries of such programs. The benefit "follows" the child and is paid to the person who has actual custody of the child. It is the custodial parent (usually the mother) whose income is measured for eligibility determinations, and whose lifestyle and living arrangements are scrutinized in determining the scope and amount of the benefit. For this reason, I characterize women as beneficiaries, even though according to statutory guidelines the children are the legal beneficiaries. Columbia Journal of Gender and Law While Roe has not been explicitly overruled, it remains in a persistent vegetative state. The judicial analysis of abortion as a fundamental right has evolved such that abortion now has the status of an "interest" that can be limited by states as long as the limitations are not "unduly burden- some." 4 Autonomy for poor women is increasingly limited by a wide array of both state and federal legislative provisions governing social policies: Therefore, before turning to state or federal legislatures to replicate the protections once afforded by Roe, it is important to examine some of the peculiar properties of the legislative arena. Utilizing the legislative and judicial forums may not be the best way to design policies that bring about improved social conditions for women. Traditionally, the women's movement has relied on the judiciary to correct the emblematic defects of legislation.5 However, in recent years, the judiciary has limited its protection, in part by giving legislation affecting women "intermediate scrutiny," and by cutting back on the scope of privacy protections in the reproductive area. Moreover, the collective voices of women have too often been minimized in the negotiating process that frames litigation and legislative issues.6 Through this process, legislation has often limited both autonomy and state benevolence in ways that severely disadvantage women.7 ' The undue burden standard was first articulated by Justice O'Connor in her dissenting opinion in City of Akron v. Akron Ctr. for Reprod. Health, Inc., 462 U.S. 416, 461-74 (1983). The standard was adopted by the majority in Casey, where the Court stated: To protect the central right recognized by Roe v. Wade, while at the same time accommodating the State's profound interest in potential life, we will employ the undue burden analysis. . . . An undue burden exists, and therefore a provision of law is invalid, if its purpose or effect is to place a substantial obstacle in the path of a woman seeking an abortion before the fetus attains viability. Casey, 112 S. Ct. at 2821. The Court went on to reject the "rigid trimester framework" of Roe. Id. I Some feminist legal theorists have suggested that too much attention has been focused on the rights paradigm and the courts for resolving conflicts about women's choices and women's lives. See, e.g., Elizabeth Schneider, The Dialectic of Rights and Politics: Perspectives from the Women's Movement, 61 N.Y.U. L. Rev. 589 (1986); Robin West, Jurisprudence and Gender, 55 U. Chi. L. Rev. 1 (1988). Others suggest that the appropriate construct for assessing such claims is in the context of accommodat- ing difference. See generally Martha Minow, Making All the Difference: Inclusion, Exclusion and American Law (1990). 6 See, e.g., Martha Albertson Fineman, Images of Mothers in Poverty Discourses, 1991 Duke L.J. 274, 289-93; see also Lisa Peattie & Martin Rein, Women's Claims: A Study in Political Economy 19, 21 (1983). The authors use the vocabulary of"claiming" to describe the "social negotiation of reality" and how women traditionally have been underrepresented in this process. Id. at vii. ' This has been best documented in the areas of family law, divorce, and child [Vol. 3:1 Poverty, Reproduction, and Autonomy Autonomy is a term that is used in a number of disciplines with differing meanings.' In the context of social science or psychology, autonomy refers to how people structure their social relationships. In the field of mental health, autonomy has come to mean an ability to conduct one's affairs without "unusual" assistance from others. Thus, those who are able to engage in "normal" social intercourse, without being overly dependent on others and in a way that can be readily comprehended by others, are deemed autonomous, competent adults. It is difficult to state a clear legal definition of autonomy. Autonomy is characterized as an intermarriage of legal independence or competence, and personal privacy. These two definitions may lead in different directions. Thus, within a legal framework, autonomy might mean that a person can act in a legally responsible way, managing her own financial and personal affairs. Or, it might refer to a sphere of individual behavior that is free from government regulation. In either case, it is a foreign concept to the social scientist, health professional, or philosopher. It is also becoming clear that there are important gender differences in the way in which both ethics and autonomy are defined and analyzed. For example, some feminists focus on autonomy as a construction of interdepen- dence rather than independence.9 Feminists looking at issues of ethics use a similar mode of analysis.' 0 In this paper, when I refer to the longstanding definition of autonomy (the one that incorporates potentially conflicting attributes), I use the term traditional autonomy. I use the term relational autonomy to signal my construction of the way autonomy ought to be viewed. The latter does not necessarily include physical independence or self-sufficiency, but incorpo- rates an individual's relationships with others. The concepts I view as central to relational autonomy include aspects of interdependence, the ethic custody. See, e.g., Martha Fineman, Dominant Discourse, Professional Language, and Legal Change in Child Custody Decisionmaking, 101 Harv. L. Rev. 727 (1988). 8 For psychological and sociological perspectives on autonomy, see, e.g., Lawrence Haworth, Autonomy: An Essay in Philosophical Psychology and Ethics (1986); Phillip Slater, The Pursuit of Loneliness (1970). For a view from the medical field, see George J. Agich, Reassessing Autonomy in Long-Term Care, Hastings Center Rep., Nov.-Dec. 1990, at 12; Onora O'Neill, Paternalism and Partial Autonomy, 10 J. Med. Ethics 173 (1984). For a summary of general theoretical perspectives, see Gerald Dworkin, The Theory and Practice of Autonomy 3-20 (1988). 9 This is the approach taken by Carol Gilligan, who points to the "ethic of care" as a manifestation of the way in which interdependence is characterized. See Carol Gilligan, In a Different Voice (1982). Martha Minow takes a similar approach, focusing on characterizations of community. See Martha Minow, The Supreme Court, 1986 Term-Foreword: Justice Engendered, 101 Harv. L. Rev. 10 (1987). "0 See, e.g., Karen Lebacqz, Feminism and Bioethics: An Overview, Second Opinion, Oct. 1991, at 10. 19921 Columbia Journal of Gender and Law of care, and mutuality of responsibility." Relational autonomy also incor- porates a kind of sliding-scale perspective on autonomy, a construction that recognizes that people are at times more, and at other times less, autono- mous or independent. Relational autonomy is a construction that is in the process of formation, and part of my inquiry here is to assess whether a concept of relational autonomy can serve as a means of effectively restructuring the legislative policy-making process. II. AUTONOMY AND SOCIAL POLICY LEGISLATION The Conflicting Goals of Legislation Legislation in the social policy arena often has dual goals: to provide a specific service-e.g., money, goods, counseling-and to deliver a message, often implicating a moral or political agenda. These goals are not always consistent. When they are inconsistent, the service or policy purposes may be thwarted. Moreover, goals of different pieces of legislation often conflict with one another, even when legislation is directed at the same population or addresses similar problems. While there is a comprehensive literature that assesses the impact of judicial decisions on autonomy, there is no corresponding literature with regard to legislative encroachments on women's autonomy. I explore the issue of legislative encroachments on women's autonomy for two reasons. First, in the arena of legislative policy making, which is more directly controlled by the forces of politics and public policy than the judiciary, it is important to identify both overt and hidden impacts on autonomy. Second, with the erosion of its underlying constitutional foundation, legislation regulating reproductive autonomy will begin to look more and more like social welfare legislation of the New Deal/Great Society eras. Legislation in the welfare and health areas relies on policy-oriented arguments rather than on constitutionally mandated individual rights protections."2 Thus, abortion rights activists should draw " See, e.g., Women's Caring: Feminist Perspectives on Social Welfare (Carol T. Baines et al. eds., 1991). For a discussion of constructs of autonomy in relation to medical caregiving issues, see Bart Collopy et al., The Ethics of Home Care: Autonomy and Accommodation, Hastings Center Rep. Spec. Supp., Mar.-Apr. 1990, at 8. 12 To date, the Supreme Court has not recognized a "rights" argument for the provision of welfare and health services, except for procedural due process. See generally William Simon's critique of the new property concept, in The Invention and Reinvention of Welfare Rights, 44 Md. L. Rev. 1, 2-3, 23-25 (1985); Goldberg v. Kelly, 397 U.S. 254 (1970) (holding that procedural due process requires a pre- [Vol. 3:1 Poverty, Reproduction, and Autonomy important lessons from welfare and health initiatives before promoting legislative action as a way of securing the rights once guaranteed by Roe. I will identify and discuss some of the ethical dilemmas created by legislative enactments that affect women's autonomy, particularly with regard to the diminished power of the constitutionally based privacy analysis to protect autonomy. By ethical dilemmas, I mean the adoption of programs or policies that provide a benefit or entitlement to a particular group or class (e.g., women with young children), but which also exact a "fee" from those beneficiaries, often in the form of restrictions on lifestyles, life choices, and physical autonomy. Most of these dilemmas arise in the implementation of entitlement programs, and raise questions in the minds of program beneficiaries, service providers, and legislators about the true motivations of those charged with implementing the programs. The failure of those who structure the programs and analyze their impact to recognize such dilemmas has seriously undermined the autonomy of poor women in their reproductive choices and other aspects of their lifestyles, and threatens to encroach upon the autonomy of all women. My intent in presenting this analysis is to make clear that conflicting goals give rise to ethical questions that should be addressed at the time that legislation is designed, rather than later, when these questions arise as incidental outcomes or unintended consequences of well-meaning legislative policy making. I will analyze three social policy areas in which legislation has played a major autonomy-shaping role during the latter half of the twentieth century: reproductive autonomy and health; welfare and income support; and the "ethic of care" and women's caregiving roles. Before doing this, however, I will explore the issue of autonomy in greater detail. The Meaning of Autonomy: Competing Paradigms In order to fully understand what autonomy means in the context of a welfare state, it is necessary to review some of the history of social policy legislation in the United States. An examination of social policy in the United States shows that two contradictory paradigms of liberal theory have been the hallmarks of social policy development in this country. The first is the traditional autonomy or individual rights paradigm, characterized by a strong adherence to an ethic of individual freedom from state interference, initially in a political or constitutional context, and ultimately in a social context. 3 The individual rights paradigm defines the termination evidentiary hearing be held when public assistance payments to welfare recipients are discontinued); Flemming v. Nestor, 363 U.S. 603, 608 (1960) (holding that social security benefits are not accrued property rights). 13 See, e.g., Ronald Dworkin, Taking Rights Seriously (1977); John Rawls, A 19921 Columbia Journal of Gender and Law autonomous individual as one who is free from all state regulation except that which is necessary for public safety and welfare. 4 However, while courts have approved regulation of individual behavior in order to protect the civil rights of minorities from any political majority, they have not accepted the argument that this principle should be used to permit redistribution of wealth through legislated social welfare policies.15 The second paradigm is the benevolent state paradigm, exemplified by a state obligation to care for those who, in the eyes of the government, cannot care for themselves or who are otherwise deserving of assistance.' 6 This paradigm expands or contracts according to its adherents' views, often culturally biased, as to whom is justifiably considered vulnerable in society. Children, the "infirm," and the elderly have traditionally been included in this category. Until the mid-twentieth century, women were also placed in this category,' 7 as evidenced by the broad spectrum of common law and legislation that limited women's freedom to make contracts, and inherit or own property.' 8 Theory of Justice (1971); see also Roberto Torricella, Comment, Babalu-Aye Is Not Pleased: Majoritarianism and the Erosion of Free Exercise, 45 U. Miami L. Rev. 1061 (1991) (exploring the transition of First Amendment theory from a political to a social context). (Babalu-Aye is god of the sick in the Santeria religion.) 14 For a critique of this Rawlsian concept of the autonomous individual, see R. Paul Wolff, Understanding Rawls (1977). With regard to legal decision making, these concepts are frequently elaborated in the context of formalism. See Roberto Unger, Law in Modern Society: Toward a Criticism of Social Theory (1976); Ernest Weinrib, Legal Formalism: On the Imminent Rationality of Law, 97 Yale L.J. 949 (1988). 15 See, e.g., William Simon, Rights and Redistribution in the Welfare System, 38 Stan. L. Rev. 1431, 1432-35 (1986); Fineman, supra note 6. For a critical perspective on social welfare theory, see Anne Orloff, The Political Origins of America's Belated Welfare State, in The Politics of Social Policy in the United States (Margaret Weir et al. eds., 1988); Theda Skocpol & John Ikenberry, The Political Formation of the American Welfare State in Historical and Comparative Perspective, 6 Comp. Soc. Res. 87 (1983). 16 The history and philosophy of state benevolence has been amply documented. See, e.g., Willard Gaylin et al., Doing Good: The Limits of Benevolence (1978); Michael B. Katz, In the Shadow of the Poorhouse: A Social History of Welfare in America (1986); Francis Fox Piven & Richard Cloward, Regulating the Poor: The Functions of Public Welfare (1971). 17 For a historical perspective, see, e.g., Grace Abbott, The Dependent and the Delinquent Child: The Child of Unmarried Parents (1938); see also Joel F. Handler, The Conditions of Discretion: Autonomy, Community, Bureaucracy (1986). For a general discussion of the origins of welfare policy in the United States, see Walter I. Trattner, From Poor Law to Welfare State: A History of Social Welfare in America (2d ed. 1979). S Richard Chused, Married Women's Property Law: 1800-1850, 71 Geo. L.J. 1359, 1361 (1983); Hendrik Hartog, Marital Exits and Marital Expectations in 19th Century America, 80 Geo. L.J. 95, 96-99 (1991). [Vol. 3:1 Poverty, Reproduction, and Autonomy I assert that legislation that limits women's autonomy is always a manifestation of the benevolent state paradigm, and, as such, has the effect of characterizing, and ultimately marginalizing, women as vulnerable or dependent. In the late twentieth century, such characterizations have developed an ambivalent nature. Much of the legislation examined herein justifies limitations on autonomy under the benevolent state paradigm, while at the same time using the language of the individual rights paradigm to support the limitation or termination of entitlements. 9 Both the individual rights and state benevolence paradigms originated in Western European and American political theory and were consequently incorporated into Anglo-American political structures.2' Both paradigms have limitations that at times belie the positive aspects of their philosophy. The individual rights paradigm is the paradigm of laissez-faire, a construct that allows the state to regulate only in limited areas, and demands that individuals be free to maintain their social and economic independence.2' The benevolent state paradigm is the paradigm that allows the state to exercise social control over the behavior and activities of its citizens.' This latter construct is often employed to accomplish governmental social control of persons whose behavioral norms or beliefs are at variance with those of the mainstream governmental authority. The possibility of cultural bias is increased where there are racial, ethnic, or gender differences between the service providers/policy makers and the beneficiaries/ recipients. In such instances, the distinction between benevolent state assistance and demeaning social control may be greatly diminished.' 19 For a discussion of legislation of this nature, see infra notes 48-104 and accompanying text. 20 For a discussion of the paradoxes of these paradigms in the context of autonomy theory, see Dworkin, supra note 8, at 12-20. The history of the parens patriae model as the model for "liberty as state benevolence" is discussed in Trattner, supra note 17, at 108 n.9; see also Developments in the Law-The Constitution and the Family, 93 Harv. L. Rev. 1156, 1198-1201, 1208-13 (1990) [hereinafter Developments-The Family]. 21 Ronald Dworkin, Liberalism, in Public and Private Morality 113 (Stuart Hampshire ed., 1978); see also James Schall, Human Rights as an Ideological Project, 1987 Am. J. Juris. 47 (discussing the symbolic value of rights); Stuart Scheingold, The Politics of Rights: Lawyers, Public Policy and Political Change (1974) (discussing the"myth" of rights theory and its instrumental manifestations). 22 See, e.g., William A. Galston, Pluralism and Social Unity, 99 Ethics 711 (1989); Piven & Cloward, supra note 16. 23 See The Mean Season: The Attack on the Welfare State (Fred Block et al. eds., 1987) [hereinafter The Mean Season]; Anthony Alfieri, The Antinomies of Poverty Law and a Theory of Dialogic Empowerment, 16 N.Y.U. Rev. L. & Soc. Change 659, 666-82 (1987-88). 19921 Columbia Journal of Gender and Law The Challenge of Social Policy Legislation and Implementation Analyzing the ethics of social policy legislation leads to a persistent dilemma: while the goals of legislation are described in highly theoretical terms, the impact of legislation can only be understood at the level of implementation. By focusing on the ethics of legislation, I am demanding that the impact of legislation be acknowledged and addressed, and that those who design and negotiate legislation also think through and take responsibil- ity for the consequential impact. There are three aspects to this dilemma. The first is a translation dilemma: the necessity of articulating what the theoretical concepts mean in the context of specific policy questions and particular social problems. The second is a goals dilemma: the necessity for precisely articulating the social policy outcome desired and a methodology for reaching that goal. The third is an implementation dilemma: the necessity of determining who will draft the policy and design the program, how it will be administered, and who will serve as the "gatekeepers" to see that the policies remain consistent with the goals. Inherent in this dilemma is the necessity of building into the process a mechanism to enable legislatures or courts to respond to changing social conditions that may rapidly transform any of these three decisional areas. It is best to analyze the dilemma from this three-tiered approach because, unlike judicial decisions (which, while presumed to be self-effectu- ating, often are not), legislation necessarily involves several levels of interpretation and implementation. There is often a gulf between legislation as it is enacted by the legislative body and the program as it is implement- ed. Moreover, as was stated earlier, legislative themes can be expressed explicitly, in the body of legislation, or implicitly, by virtue of the means used to implement the legislation. It is problematic for the women's movement that critical feminist theory has avoided recognizing the dilemmas of legislation within traditional constructions of autonomy.24 Those traditional constructions generally ignore some of the fundamental features of women's lives that are most affected by social policy legislation, features that commentators such as Carol Gilligan have gathered together under the "ethic of care" umbrel- la.' Feminist theory needs to articulate a way for the themes of feminism 24 See, e.g., Gerda Lerner, The Creation of Patriarchy (1986); Johanna Brenner, Towards a Feminist Perspective on Welfare Reform, 2 Yale J.L. & Feminism 99 (1989). ' Gilligan, supra note 9. [Vol. 3:1 Poverty, Reproduction, and Autonomy to "translated" into legislative mandates.26 In the discussion of autonomy that follows, I have attempted to expand upon and to incorporate the alternative construction of autonomy that I have presented, and to examine legislative and judicial policy making from the perspective of autonomy. III. REPRODUCTIVE AUTONOMY AND HEALTH Indirect Effects of Legislation on Reproductive Autonomy There are two ways in which legislative provisions regarding reproduc- tive autonomy can affect women. Much of the legislation in this area has a direct and intended impact on reproductive autonomy.27 However 26 The "ethic of care" and the roles of caregiving are examples of one "feminist theory-legislative policy" connection that might be made. See, e.g., Carol Gilligan, Remapping the Moral Domain: New Images of the Self in Relationship, in Recon- structing Individualism: Autonomy, Individuality and the Self in Western Thought 237 (Thomas C. Heller et al. eds., 1986). Another example is Martha Minow's critique of traditional family law models as implementing masculine values and ignoring the women-centered values of connection and caretaking, in Martha Minow, Forming Underneath Everything That Grows: Towards a History of Family Law, 1985 Wis. L. Rev. 819, 893-94. See also Katharine T. Bartlett, Re-Expressing Parenthood, 98 Yale L.J. 293, 296 (1988) (arguing that, with regard to parenthood, the balance should tilt more toward responsibilities than rights). See generally Nel Noddings, Caring: A Feminine Approach to Ethics & Moral Educ. 14 (1984). 27 Virtually all of the post-Roe generation of restrictive legislation has been of the overt/direct type. These provisions are intended to limit access to abortion directly by making abortions more "burdensome" for both the pregnant woman and the abortion provider. At the same time, the "burdens" must not be so great as to exceed constitutional bounds. Initially the concept of "burdensome" merely referred to provisions which had a chilling effect on a woman's right to abortion, or, in a practical sense, on the ability of providers to offer affordable abortion services. Justice O'Connor gave the term a constitutional significance in City of Akron v. Akron Center for Reproductive Health, Inc., when she stated that only provisions that "unduly burdened" a woman's ability to obtain an abortion would be considered unconstitution- al. 462 U.S. 416, 453 (1983) (O'Connor, J., dissenting), overruled in part by Planned Parenthood v. Casey, 112 S. Ct. 2791 (1992). While this characterization initially appeared to be her way of proffering a middle ground between those on the Court who would overturn Roe and those who were unwilling to do so, more recent opinions using this terminology suggest that the burdens involved must be quite heavy before they will be considered unconstitutional. See, e.g., Webster v. Reproductive Health Servs., 492 U.S. 490 (1989); Rust v. Sullivan, 111 S. Ct. 1759, 1777 (1991) (concluding that Title X prohibitions of abortion referral and information provision do not "impermissibly burden a woman's ... rights"). In Casey, the Supreme Court's most recent abortion-related decision, the Court upheld as not "unduly burdensome" most of a restrictive Pennsylvania statute that imposed limitations on abortion services. The statute mandated, among other things, that a woman wait 24 hours between her decision to seek an abortion and the abortion 19921 Columbia Journal of Gender and Law troubling overt restrictions on abortion are, I am chiefly interested in studying indirect restrictions on autonomy. This form of legislation is more complex, because the limitations on reproductive autonomy often come about as a side effect of other legislative policies. The "good news" about the Casey decision, and the news that has received the most attention, is that by a narrow margin, the Supreme Court refused to overturn Roe, and, in fact, upheld the right to abortion.' However, in other respects, the Casey decision follows a pattern established in the other recent Supreme Court abortion determinations regarding legislation that curtails women's reproductive autonomy, a pattern that leaves a right or entitlement intact, but curtails the autonomy of those who exercise the right or entitlement. For example, the stated purpose of the record-keeping requirements upheld in Casey is to promote public health; however, their more subtle effect is to make abortions more expensive, and to discourage abortion providers by making it clear that their services will be publicly scrutinized.29 Often itself, 18 Pa. Cons. Stat. Ann. § 3205(a) (Supp. 1992); that doctors give women detailed and graphic information about fetal development, id. § 3205(a)(1)-(2); and that abortion providers keep extensive records, id. § 3214(a). The only portion of the legislation to be struck down was a provision requiring that a father-husband be notified of an abortion. Id. § 3209. The Court held that the spousal notification provision of Pennsylvania's statute placed an "undue burden" on a woman's right to abortion and was therefore invalid. 112 S. Ct. at 2826-31. The plurality opinion stated that "[t]he husband's interest in the life of the child his wife is carrying does not permit the State to empower him with this troubling degree of authority over his wife." Id. at 2831. Ironically, many of the provisions upheld in Casey had been rejected by the Supreme Court in Thornburgh v. American College of Obstetricians and Gynecologists, 476 U.S. 747 (1986), overruled in part by Casey, 112 S. Ct. 2791. In Thornburgh, the Court reaffirmed Roe in recognizing a woman's right to an abortion prior to fetal viability without significant interference from the state, id. at 759, and, as a result, rejected informed consent provisions and reporting requirements. While the Court did agree that the state had some power to limit or restrict abortions where a fetus was viable, it did not address the question of when and how viability should be determined. The Thornburgh line of cases predates the "unduly burdensome" standard that now appears to apply to all abortions, pre- and post-viability. As formulated by Justice O'Connor, the "unduly burdensome" standard gained real constitutional significance in Webster, 492 U.S. at 522, 530 (O'Connor, J., concurring). Other legislative provisions, however, have an indirect but equally grave effect on physical and reproductive autonomy. One of the effects of the Supreme Court's adoption of the "unduly burdensome" standard for evaluating restrictions on the right to abortion is that this language has the effect of legitimizing the more indirect restrictions as not "unduly" burdening a woman's right to abortion. Webster and the post-Webster cases, particularly Rust and Casey, fall into this category. 2 The Court reaffirmed the "central right recognized by Roe v. Wade," a woman's right to pre-viability abortions. 112 S. Ct. at 2821. 29 18 Pa. Cons. Stat. Ann. § 3214(a) (Supp. 1992). The indirect restrictions in Casey require extensive keeping of records, which must be filed and available for public inspection. Id. §§ 3207, 3214 (1983 & Supp. 1992). These requirements do not say [Vol. 3:1 Poverty, Reproduction, and Autonomy legislative restrictions on autonomy are hidden in seemingly benevolent policy language and in implementation regulations.3 Indirect restrictions such as these provide an insidious means of restricting autonomy, for their true impact is often not understood until legislation is enacted and imple- menting regulations are approved. Legislation that indirectly limits reproductive autonomy inordinately affects poor women and young women, mainly because of the limitations of the health-care network and the dependence of poor women on state assistance. The most obvious legislation of this type regulates the behavior of abortion providers who receive public funds. In two recent and widely discussed Supreme Court cases, the Court reviewed legislation or regula- tions that included components which impacted indirectly upon women's autonomy. In Webster v. Reproductive Health Services," the Supreme Court upheld a Missouri statute drastically limiting the kinds of reproduc- tive health services that publicly funded hospitals, clinics, and doctors could provide. The opinion intentionally left vague the precise definition of "public facility," thus leaving open the question of whether institutions utilizing public electric or sewer lines were included in that definition; this question has not been addressed by the Missouri Supreme Court.32 More recently, in Rust v. Sullivan33 the Supreme Court upheld a federal statute that prohibited federally funded reproductive health facilities from informing pregnant women that abortion was one of the available options that they could consider.' The obvious impact is restricted anything directly about access to abortion, and are not, as the Supreme Court held, "unduly burdensome" to that right. But the additional paperwork required will have the effect of making abortions more expensive, and the public filing requirements may discourage providers from offering abortion services and may make patients more reluctant to seek abortions. The statute states that records are necessary "to promote maternal health and life by adding to the sum of medical and public health knowledge through this compilation of relevant data." But as the statute also declares, such record keeping also "promote[s] the Commonwealth's interest in protect[ing] the unborn child." Id. § 3214(a) (Supp. 1992). 30 This has been the case in much of the workfare legislation that will be discussed infra part IV. 31 492 U.S. 490 (1989). 32 The author reviewed post-Webster cases in Missouri on this statute and turned up no further elaboration of this point. 3 111 S. Ct. 1759 (1991). 3" The Supreme Court upheld the statute as not "unduly burdensome" to the right to abortion; however, both the First and Tenth Circuits had found it unconstitutional. See Massachusetts v. Secretary of H.H.S., 899 F.2d 53 (1st Cir. 1990), vacated sub nom, Sullivan v. Massachusetts, 111 S. Ct. 2252 (1991); Planned Parenthood v. Sullivan, 913 F.2d 1492 (10th Cir. 1990), vacated, 111 S. Ct. 2252 (1991). The challenged provisions are contained in Department of Health and Human Services 1992] Columbia Journal of Gender and Law dissemination of information to a particular category of people-those who seek counseling in federally funded centers, most of whom are young and poor. However, the indirect effect of the Rust restrictions may be a general chilling: medical professionals may simply choose not to provide abortion services, and potential recipients of such services may delay or forego altogether seeking birth control counseling.35 Other indirect forms of regulation are beginning to have a profound impact on the availability of abortion and reproductive health services. For example, most state Medicaid programs do not cover abortion.36 Many private insurance policies offer limited or no coverage for obstetrical and reproductive services; many other working- and middle-class women have no private health insurance at all.37 Moreover, in the last fifteen years, increasing numbers of public and private hospitals and physicians have limited or eliminated abortion services. 3 Furthermore, in a growing regulations adopted pursuant to Title X of the Public Health Service Act, §§ 1002, 1008, as amended, 42 U.S.C.i§§ 300a, 300a-6 (1988). New directives were adopted in 1992 that relaxed some of the restrictions upheld by the Supreme Court in Rust, for example allowing physicians (but not other health professionals) to provide information on abortion. Id. § 300a-6. In October 1992, when Congress attempted to overturn the Rust restrictions, President Bush vetoed the legislation. Family Planning Amendments, H.R. 5495 and S.R. 2899, 102nd Congress, 2nd Sess., Oct. 1992. The issue remains pressing since without the backing of the Constitution or even substantive provisions of law, any regulatory actions are highly vulnerable to political winds. But see National Family Planning & Reprod. Health Ass'n, Inc. v. Sullivan, 979 F.2d 227 (D.C. Cir. 1992) (new directives are not exempt from the notice and comment rulemaking requirement of the Administrative Procedure Act, 5 U.S.C. § 553 (1988)). 15 For a discussion of the First Amendment implications, see Alexandra A.E. Shapiro, Title X: The Abortion Debate, and the First Amendment, 90 Colum. L. Rev. 1737, 1747-67 (written prior to the Supreme Court's decision in Rust). 36 See, e.g., The Medicaid Cutoff and Abortion Services for the Poor, 16 Fam. Plan. Persp. 170, 171 (1984) (examining effects of restrictions on abortion funding for Medicaid recipients); see also Carole A. Corns, The Impact of Public Abortion Funding Decisions on Indigent Women: A Proposal To Reform State Statutory and Constitution- al Abortion Funding Provisions, 24 U. Mich. J.L. Reform 371 (1991). According to Corns, as of 1991, only eight states voluntarily used state funds to finance all "medically necessary" abortions; courts in three additional states used the state constitution to invalidate restrictions on using state funds to pay for "medically necessary" abortions; and only California provided Medicaid funds for all abortions. Id. at 381-82. See also Rachel Benson Gold, Abortion and Women's Health: A Turning Point for America? (Alan Guttmacher Institute, 1990). " See, e.g., Rachel Benson Gold et al., Blessed Events and the Bottom Line: Financing Maternity Care in the United States (Alan Guttmacher Institute, 1987). 38 The number of hospitals performing abortions dropped from 1,405 in 1982 to 1,191 in 1985. Eighty-two percent of the counties in the U.S. had no clinic or hospital that performed abortions in 1985. Gold, supra note 36, at 25. The result is that women who are seeking abortions often must travel long distances at great expense. Stanley K. Henshaw, The Accessibility of Abortion Services in the United States, 23 [Vol. 3:1 Poverty, Reproduction, and Autonomy number of medical schools, medical students and Ob-Gyn residents are permitted to opt out of abortion training.39 Federal legislation has also made it more difficult for women to obtain family planning information.' Thus, abortion and many other reproductive services, including advice on birth control, have become extremely limited, particularly for low-income and rural women.4 The Pervasive Attack on Pregnant Women and Motherhood As legislation and government regulation has made abortion and birth control less available to young women and the poor, more women from these vulnerable populations are becoming pregnant. This has created an escalating series of ethical dilemmas in reproductive policy. It has become increasingly common for medical professionals to weigh the perceived interests of fetuses against the expressed interests of pregnant women in ways that affect not only the availability of abortion, but other procedures related to, or impacting upon, pregnancy. These include high-technology Fain. Plan. Persp. 246, 248 (1991). 39 Imagine a medical school allowing students and residents to decline training in treating liver diseases because they do not want to treat alcoholics, or a state bar association permitting lawyers to choose not to take the torts section of the bar exam because they do not want to handle personal injury cases. Yet The New York Times reports that many Ob-Gyn programs have either made abortion an elective for residents, or have stopped offering any abortion training at all. Tamar Lewin, Hurdles Increase for Many Women Seeking Abortions, N.Y. Times, Mar. 15, 1992, at Al. Residents in 18% of all United States residency programs were allowed no access to training in abortion in 1985. While most residency programs offered abortion training (72%) in 1985, this was a 22% decline in the number of programs offering abortion training as compared to 1976. This is despite the fact that abortion is the most common surgical procedure that women undergo. Philip D. Darney et al., Abortion Training in U.S. Obstetrics and Gynecology Residency Programs, 19 Fam. Plan. Persp. 158, 160, 161 (1987). Moreover, access is limited by other factors: 9% of non-hospital abortion patients must travel more than 100 miles to find a facility, and 18% must travel 50 to 100 miles. Many facilities (27-37%) do not treat patients who are HIV-positive. Henshaw, supra note 38, at 246. o Legislation such as that upheld in Rust, 111 S. Ct. 1759, has set strict limits on what family planning clinics can tell patients seeking information. At the same time, research on new birth control techniques has also been discouraged. See George J. Annas, Regulating the New Reproductive Technologies, in Reproductive Laws for the 1990s 411 (Sherrill Cohen & Nadine Taub eds., 1989). "' See, e.g., Stanley K. Henshaw & Jennifer Van Vort, Abortion Services in the United States, 1987 and 1988, 22 Fam. Plan. Persp. 102 (1990); Kathryn Kolbert, Developing a Reproductive Rights Agenda for the 1990s, in From Abortion to Reproductive Freedom: Transforming a Movement 297, 301-06 (Marlene Gerber Fried ed., 1990). 19921 Columbia Journal of Gender and Law testing of mothers and fetuses, potentially risky surgical and quasi-surgical interventions, and treatment for non-pregnancy-related gynecological condi- tions such as breast and ovarian cancer and menopause. 2 Criminal prosecutions of pregnant drug-abusing women weigh the perceived interests of fetuses and newborns against their mother's autono- my. Although these prosecutions (usually for the offense of distributing a controlled substance to a minor via the umbilical cord) have had mixed success in the criminal courts, they have received substantial publicity in the media. These prosecutions have spurred some child welfare agencies to attempt to "remove" unborn children from their pregnant mothers' "custody" prior to the child's birth, either by jailing the mother, or placing restrictions on the mother's behavior and jailing her if she violates the restrictions." Regrettably, at the same time that increased resources are dedicated to punishing pregnant drug-using women, there continues to be an extreme shortage of support services for such women. Many pregnant women who seek treatment for substance abuse are turned away from treatment facilities, and those few treatment facilities that do accept pregnant women have long waiting lists.' These trends raise questions concerning how comprehensively the government can regulate any pregnancy. For example, while smoking cigarettes or consuming alcohol is not illegal, during pregnancy or otherwise, the medical profession considers this behavior to be as harmful to fetuses as drug use.45 The aforementioned interventions raise the specter of government regulation of behavior by pregnant women that violates a doctor's recommendation. For example, should the government be allowed to intervene if a pregnant woman disregards a doctor's recommendation regarding sexual intercourse, athletic activity, or working 42 See Susan S. Mattingly, The Maternal-Fetal Dyad: Exploring the Two-Patient Obstetric Model, Hastings Center Rep., Jan.-Feb. 1992, at 13; see also Janet Gallagher, Fetus as Patient, in Reproductive Laws for the 1990s, supra note 40, at 185. " See, e.g., Gallagher, supra note 42, at 214. In California, a state court ultimately dismissed charges brought by the state against a woman who had disregarded her doctor's instructions not to have intercourse during the later stages of her pregnancy, People v. Stewart, No. M508197, slip op. at 10-11 (San Diego Mun. Ct., Cal. Feb. 26, 1987). However, a Florida court convicted Jennifer Johnson for delivering drugs to a minor via her umbilical cord, Johnson v. Florida, 578 So. 2d 419 (Fla. Dist. Ct. App. 1991) (affirming convictions of lower court), certified question answered, quashed, 602 So. 2d 1288 (Fla. 1992). See also Shona B. Glink, The Prosecution of Maternal Fetal Abuse: Is This the Answer?, 1991 U. I11. L. Rev. 533, 536-37. " Glink, supra note 43, at 544-45. 45 See, e.g., Bonnie Steinbock, The Relevance of Illegality, Hastings Center Rep., Jan.-Feb. 1992, at 19, 20; Note, Maternal Rights and Fetal Wrongs: The Case Against the Criminalization of "Fetal Abuse," 101 Harv. L. Rev. 994 (1988). [Vol. 3:1 Poverty, Reproduction, and Autonomy hours?' Recently, welfare recipients have been disparaged by state legislatures facing fiscal constraints and by politicians seeking elective office. The "blame welfare moms" phenomenon ties together two issues that excite political conservatives and shows increasing appeal among traditional "liberals" as well. These issues are widespread myths about the sexual promiscuity of poor women in addition to exaggerated notions about the number of years the average recipient remains on the welfare rolls.' The result has been a spate of proposals for sweeping reform of the welfare system, in ways that will substantially diminish women's reproductive autonomy. This issue is addressed in the next section. IV. LIMITATIONS ON THE AUTONOMY OF WELFARE RECIPIENTS The provision of benefits and services to those in need has always had a profound impact on the autonomy of women, both young and old. In the twentieth century the design of social service programs affecting women has been reshaped in ways that reflect an ambivalent attitude toward women and work. In addition, the rhetoric and values of the women's movement have been assimilated by policy makers in a way that limits the autonomy of women who, because of age or economic status, were not able to take advantage of the opportunities brought about by the women's movement. 46 See Steinbock, supra note 45, at 19. Steinbock expresses concern that this might be an insidious way of regulating behavior in a manner which would not be permissible under current interpretations of child protection laws. Id. at 21. She also notes that while Congress has mandated that states spend up to 10% of their federal Alcohol, Drug Abuse and Mental Health Services block grant funds on services for pregnant women, most states have failed to comply with this mandate, suggesting that the public issue is one of rhetoric rather than reality. Id. at 22. ' The rhetoric of the 1992 Presidential campaign provides a concrete example of how these vague presumptions can potentially be converted into legislative policy. In speeches at the Republican National Convention, the "us versus them" rhetoric made the conservative position particularly mean-spirited. It was clear to everyone who heard the references to Murphy Brown that no one intended to target a fictional middle-class- professional-unwed mother; the real targets were all those "promiscuous welfare recipients." But many of the sentiments expressed by conservatives were also present, in a more modified way, in the calls for welfare reform by Governor Clinton and other Democrats in the Presidential campaign and in proposals under consideration for the 1993 Congressional Session. See, e.g., The Downey/Hyde Child Support Enforcement & Assurance Proposal (copy of proposed bill provided to author by committee staff). 1992] Columbia Journal of Gender and Law The Family Support Act One example of the overt and hidden policy implications of social policy legislation is the Family Support Act of 1988 (FSA) and the proposals for amendments and local policy variations that have emerged since its enactment.4" I have chosen to analyze this legislation for several reasons. First, the FSA represents the first substantial alteration in the philosophy of the national welfare program, Aid to Families with Depen- dent Children (AFDC), by imposing mandatory work requirements upon women with children over the age of three (the JOBS Program). 9 Second, the legislation is structured in a way that allows state and local governments broad discretion in developing programs to move welfare recipients into the workforce (and, not incidentally, to remove 'ecalcitrant welfare recipients from the rolls)." Third, as state and local variations have been implemented, the combination of increased local discretion and an anti-welfare political climate has fostered a revisionist interpretation of the purpose behind the FSA. This interpretation has had the effect of limiting the personal and reproductive autonomy of welfare recipients, in ways that thwart the stated purpose of the legislation. Thus, the FSA serves as a good example of how legislation which appears intended to enhance women's autonomy, in this case by assisting welfare recipients to move from welfare into the workforce, ultimately limits the autonomy of its recipients. Some commentators suggest that the FSA intensified revisionist thinking about welfare because it implicitly granted credence to traditional myths about welfare recipients, revealing a thinly-veiled racism.5 The founding premise of AFDC was that mothers of children (up to eighteen 48 Family Support Act of 1988, Pub. L. No. 100-485, 102 Stat. 2343 (1988) (codified in scattered sections of 42 U.S.C.). 41 Pub. L. No. 11-485, 102 Stat. 2343 (1988); 45 C.F.R. § 250.30(b)(9)(i) (1991). 10 Most of these provisions are contained in the section entitled Job Opportunities and Basic Skills Training Program, in Title IV-F, 42 U.S.C. §§ 681-87 and 45 C.F.R. Pt. 250 [hereinafter the JOBS program], which replaced the Work Incentive program (WIN) Title IV, Parts A and C of the Social Security Act of 1967, 42 U.S.C. § 602 (1967), which was never fully implemented. Other relevant provisions of the FSA are the Child Support Enforcement provisions, at Title IV-D, 42 U.S.C. §§ 651-59, and the provisions governing child care and other support services, 42 U.S.C. § 602(g) and 45 C.F.R. Pt. 255. 51 See Joel F. Handler & Yeheskel Hasenfeld, The Moral Construction of Poverty: Welfare Reform in America 26-29 (1991). The authors note that the term "underclass" is an umbrella label for many of the tragedies of the African-American urban ghettos, and that for most of the history of welfare programs in this country, African-Americans and other racial minorities have been considered among the most "undeserving" poor. Id. at 27. [Vol. 3:1 Poverty, Reproduction, and Autonomy years of age) should not be forced to work, but should receive a welfare benefit in lieu of work.5" Following the implementation of the "Great Society" programs in the 1960s, that premise gradually changed to incor- porate the notion that welfare recipients should be encouraged to move into the workforce, generally through voluntary participation programs. 3 With the passage of the FSA, the work requirement was made mandatory. The Act was passed by a coalition of liberals and conserva- tives. The liberals held out for training and education programs and transitional benefits, such as medical insurance and child care (the "car- rots"). The conservatives held out for punitive sanctions against non-partic- ipating recipients (the "stick"). 54 When the FSA was first passed, it seemed that there would be a relatively equal balance between these goals in the program's implementa- tion. However, as states and localities have structured their programs, the conservative philosophy has prevailed. This is due, in part, to funding concerns. Many of the women mandated to work by the FSA are unable 11 Mimi Abramovitz, Regulating the Lives of Women: Social Welfare Policy from Colonial Times to the Present 313-14 (1988). The average AFDC benefit level provides income substantially below the poverty level in all states, and is less than two- thirds of the poverty level in 45 states. Thus, AFDC is not a program that provides anything close to a sufficient income. See, e.g., Adele M. Blong & Timothy J. Casey, AFDC Program Rules for Advocates: An Overview, 25 Clearinghouse Rev. 874 (1991). While total welfare spending on AFDC and other benefits has grown, the average combined grant for food stamps and AFDC has dropped significantly in virtually every state over the last 20 years. In 1972, the average annual AFDC benefit in California was $11,752 (adjusted for 1991 dollars); in 1991, the average benefit was $9,906. The figures for New York were $13,882 in 1972 and $9,114 in 1991; in Wisconsin, $12,666 in 1972 and $8,419 in 1991. Even in Alabama, which has traditionally had one of the lowest levels of AFDC payments, the figures dropped from $6,899 in 1972 (in 1991 dollars) to $4,812 in 1991. Jason DeParle, Why Marginal Changes Don't Rescue the Welfare System, N.Y. Times, Mar. 1, 1992, at D3 (citing the Report of the House Ways and Means Committee). Clearly, one of the reasons that benefits are so high in states like New York and California is that the cost of living (COL) is significantly higher in those states. However, the COL difference does not account for the full differential. Additionally, while the percent of the total population receiving AFDC benefits has remained at four to five percent since 1970, the percentage of children living in poverty who actually receive welfare benefits has decreased, from more than 80% in 1973 to about 50% in 1991. DeParle, supra, at D3. The program was designed to shed women recipients as their children attained majority. As a result, most AFDC recipients are under age 35. However, there is a growing population of women who receive AFDC as custodians for their grandchildren. See Nancy A. Naples, Activist Mothering: Cross-Generational Continuity in the Community Work of Women from Low-Income Urban Neighborhoods, 6 Gender & Soc'y 441, 442-44 (1992) (discussing the concept of "othermothers"). 13 See Handler & Hasenfeld, supra note 51, at 132-37. 14 See id. at 201-09. 19921 Columbia Journal of Gender and Law to do so because child care and other necessary support services have not been fully funded." Child care has not been made available to the extent promised. 6 Transitional medical benefits are available to work program participants, but only for the first year of their employment. Since many of the jobs in the private sector do not provide health insurance, participants are often faced with a difficult choice-to work without insurance, or to remain on welfare. 7 Moreover, there is often a large gap between the work requirements themselves and the availability of jobs in the communi- ties in which the participants live. 8 Additionally the salary for entry-level positions is frequently less than the welfare stipend, particularly since many of the available jobs are part-time.59 Many restrictive measures proposed or enacted by states in 1991 and 1992 are premised upon myths often cited by proponents of the stick philosophy. One such myth, as noted above, is that most welfare recipients are long-term dependents who have additional children solely to increase their benefit levels. In fact, most AFDC recipients receive welfare for less than two years,' and the number of children per AFDC family has actually declined since 1972.61 Another myth is that high benefit levels discourage recipients from working and discourage marriage. If this were true, the states with the lowest benefit levels should have the smallest percentage of residents receiving welfare. However, Alabama, which pays a single mother with two children an average of $401 per month, has the 11 See, e.g., Judith M. Gueron & Edward Pauley, From Welfare to Work 64-67 (1991). 56 Gueron and Pauley suggest that the data from studies of welfare-work programs make clear that paid child care is a high priority for enabling women with children to participate. Id. at 232-33. This is especially true for women with pre-school age children. Anecdotal evidence of the early JOBS implementation also bears this out. In Illinois, for example, two women who were eligible for participation in the JOBS program have challenged the Illinois Department of Public Aid's policy of placing applicants and participants in the Illinois JOBS program, Project Chance, on waiting lists for the program if they need child care assistance in order to participate. 25 Clearinghouse Rev. 747 (1991). 17 See Handler & Hasenfeld, supra note 51, at 212-13. 58 See, e.g., id. at 195-96. When jobs or child care are not available, AFDC recipients are not penalized, as long as they remain willing to participate in the JOBS program. However, to the degree that they are not participating in JOBS programs, they represent a statistic that can be used to buttress the argument that the program is not effective. 9 See id. at 185-87. 0 Michael B. Katz, The Undeserving Poor: From the War on Poverty to the War on Welfare 220-21 (1989) (citation omitted); see also David Ellwood, Targeting "Long- Term" Recipients of AFDC (1986); Gueron & Pauley, supra note 55, at 53; Handler & Hasenfeld, supra note 51, at 208. 61 Katz, supra note 60, at 220; see also DeParle, supra note 52. [Vol. 3:1 Poverty, Reproduction, and Autonomy same percentage of long-term welfare recipients as California, which pays the same family an average of $826 per month. 62 The difference in the cost of living may account for some of the difference, but not all. Moreover, the highest percentage of children living in mother-headed families are in the lowest benefit status. Local Initiatives Under the Family Support Act While the purpose of giving broad discretion to the states under the FSA was ostensibly to encourage the design of innovative programs that provide training and work experience, 63 many states have recently begun to design and implement reforms that tie independence from welfare to performance of traditionally female jobs-particularly those in child care, food service, and clerical settings-under the guise of ending welfare dependency. 6' Some state and local initiatives adopted since 1991 also impose restrictions on the autonomy of welfare recipients by regulating their choices of when to marry and bear children.65 These statutes have been proposed and passed under the guise of "protecting" people from the stigma of multi-generational welfare dependency. 6' But the result of many of the restrictions will be a reduction in benefits even for families with just two or three children.67 Moreover, the proposals include education and job-training components that pay scant attention to the structure of the workplace, the availability of full-time work, and caregiving demands on the welfare recipients. In these proposals, autonomy is under attack, not just on the reproductive front, but in all aspects of women's lives. One area in which welfare reforms severely limit women's autonomy is child support enforcement. The FSA received wide support for addres- sing the difficulty state welfare departments face in coordinating their efforts with those of child support enforcement agencies. 68 Failure of 62 David T. Ellwood, Poor Support: Poverty in the American Family 61 (1991). 63 See Handler & Hasenfeld, supra note 51, at 228-29. 6 See, e.g., Joanna Weinberg, The Dilemma of Welfare Reform: "Workfare" Programs and Poor Women, 26 New Eng. L. Rev. 415, 440-42, 448-50 (1991). See generally Brenner,- supra note 24. 65 See infra notes 77-80 and accompanying text. 66 Cf. Handler & Hasenfeld, supra note 51, at 229-30. 67 See, e.g., Weinberg, supra note 64. An example of other state efforts is the "learnfare" provision in Wisconsin, which requires public school students of AFDC families to meet certain school attendance requirements at the risk of their families losing all or some AFDC benefits. Handler & Hasenfeld, supra note 51, at 225-26. 6 See Handler & Hasenfeld, supra note 51, at 221-22; see also Joel F. Handler, 19921 Columbia Journal of Gender and Law absent parents, usually fathers, to provide child support, is often given as a key cause of the increase in the numbers of welfare recipients over the last twenty-five years.69 The FSA increases cooperation between AFDC programs and state child support collection agencies,7" and requires implementation of a statewide computerized management system. 7' Under this system, money is paid directly to welfare departments, enabling those departments to adjust recipients' grants accordingly when child support is received. 72 A family is permitted to keep only the first fifty dollars of the child support payment before its grant will be reduced by the amount of the payment.73 Even in a family where more than one absent parent pays child support, the family may still keep only the first fifty dollars. 4 Perhaps most troubling, the FSA places added burdens on poor women seeking welfare by requiring them to reveal the name of any absent fathers and to permit the state to pursue that parent for payment,75 regardless of the potential impact on the family structure or the potential for physical or psychological abuse.76 Another way in which the FSA limits the autonomy of women receiving AFDC benefits is by capping benefits for additional children born while their mothers are receiving grants. President Bush endorsed such a plan in Wisconsin.' Under the plan, a woman will receive payments only for two children, even if she has more than two children when she begins receiving benefits. 7 A woman can receive additional payments, however, if she marries the father of her children.79 Teenage parents would be required to live with a parent or guardian in order to receive benefits, and The Transformation of Aid to Families with Dependent Children: The Family Support Act in Historical Context, 16 N.Y.U. Rev. L. & Soc. Change 457, 509-12 (1987-88) (examining the structural changes accompanying the FSA). 69 See Handler & Hasenfeld, supra note 51, at 221-22. 70 42 U.S.C. § 602(a)(11) (1991). 71 Id. § 654(16). ' Id. § 602(a)(28). 7' Amy Hirsch, Income Deeming in the AFDC Program, 16 N.Y.U. Rev. L. & Soc. Change 713, 721 (1987-88). 74 Id. " 42 U.S.C. § 602(a)(26)(B) (1991). 76 Hirsch, supra note 73, at 724-25. Exceptions to the paternity requirement are permitted on a showing of a specifically enumerated "good cause," 42 U.S.C. § 602(a)(25)(B) (1991), but the fact remains that this requirement, taken together with other restrictions on welfare recipients, threatens the autonomy of poor women. 77 Marsha Mercer, Race, Welfare and Politics in U.S., Atlanta J. & Const., May 16, 1992, at A19. 78 Id. 71 MacNeil/Lehrer NewsHour, Apr. 13, 1992. [Vol. 3:1 Poverty, Reproduction, and Autonomy these benefits could be cut by up to $200 each month if the teenage parent fails to attend school.' ° These proposals reflect a draconian attitude toward public assistance recipients. The New Jersey legislature has passed a similar set of restrictions, although its legislation has a more attractive "carrot" than most state initiatives: New Jersey permits JOBS participants with full-time jobs to retain a far greater portion of their income while continuing to receive full welfare benefits than the federal program requires,8 and allows recipients to keep some benefits even after marrying.u In California, more restric- tive provisions were proposed and defeated in a voter initiative entitled The Government Accountability and Taxpayer's Protection Act of 1992, Proposition 165 on the November 1992 ballot. The initiative would have cut benefits by twenty-five percent for recipients who stay on welfare for more than six months; 83 denied benefits for new children (once a person begins to receive welfare);' required teenage parents to live with a parent, legal guardian, or other adult relative;' and would pay teenage parents an extra fifty dollars per month if they remained in school.' Finally, for a period of one year, new stte residents would have been eligible for no more than the benefit level they received in their former state of resi- dence.' Other states have followed suit. A Maryland proposal would cut benefits if a parent fails to pay rent regularly, provide preventive health care for children, or keep children in school;88 and a Connecticut proposal I Jason DeParle, California Plan to Cut Welfare May Prompt Others to Follow, N.Y. Times, Dec. 18, 1991, at Al. 81 In calculating how much income a welfare recipient may keep without losing benefits, federal law requires that the following be disregarded: all the earned income of each dependent child who is a student, 42 U.S.C. § 602(a)(8)(A)(i) (1990); the first $90 of the monthly earnings of any other child or relative, up to an additional $175 for expenditures for dependents, up to $30 not otherwise disregarded, and one-third of the remainder, 42 U.S.C. § 602(a)(8)(A)(ii)-(iv) (1991); and the first $50 per month of child support income, 42 U.S.C. § 602 (a)(8)(A)(vi) (1991). 82 New Jersey Legislation Act of Jan. 21, 1992, ch. 526. Under the New Jersey legislation, teen-mother applicants for welfare will receive half of the additional incremental child benefit if they have a second child while on welfare and no additional income if they have a third child. 83 Government Accountability and Taxpayer Protection Act of 1992, (Proposition 165) § 7(a)(1). 94 Id. § 7(b)(2). 85 Id. § 6(a)(1). 86 Id. § 9. 87 Id. § 7(2)(c). 88 Isabel V. Sawhill, The New Paternalism: Earned Welfare, The Responsive Community, Spring 1991, at 26, 26. 19921 Columbia Journal of Gender and Law would drop from the welfare rolls drug abusers who refuse treatment. s9 These proposals are closely tied to issues of race, gender, and class that became popular political issues in the 1992 Presidential campaign. As The New York Times noted, "candidates of both parties have been emboldened by the recession to make welfare a seminal political issue."' "In some cases, the political message on welfare has been communicated subliminally, seemingly aimed at evoking common stereotypes."' For those concerned with issues of women and the workplace, the resulting spate of proposals within the state legislative arena is one of the most troubling aspects of the work-requirement provisions of the FSA. The work requirements have the potential to disadvantage women in several ways, including requiring women with young children to work without being able to receive adequate support services, such as child care and health benefits.' The Concept of Work and the Work Requirements of the FSA Work-requirement programs also have the effect of promoting a dual-class structure of both jobs and services. The traditional dichotomy in American economic structure, between the "haves," with jobs and income, and the "have-nots," who require income supports, is shifting to a different structure. In this new structure, the haves are divided into two classes: those who hold jobs in the private sector with income well above 89 DeParle, supra note 52, at D3. I Kevin Sack, The New, Volatile Politics of Welfare, N.Y. Times, Mar. 3, 1992, at A16. The article points to President Bush's comment in the 1992 State of the Union Address that welfare was never intended to become "a narcotic," and California Governor Pete Wilson's comments about the "imbalance between taxpayers and tax receivers." Id. 91 Id. 9 While programs such as JOBS and its state counterparts, such as GAIN in California, may be a low-cost means of moving welfare recipients from welfare to work, they do not necessarily provide sufficient income for women, in particular, to remain off welfare. Fineman, supra note 6, at 277 n.4. Poor, female-headed, single-parent families probably need a job that pays considerably more than the minimum wage in order to leave welfare and cover expenses such as child care and medical expenses. Id. at 277-78 nn.4 & 5 (citing Christopher Jencks & Kathryn Edin, The Real Welfare Problem, 1 Am. Prospect 31 (1990)); Mimi Abramovitz, Why Welfare Reform Is a Sham, The Nation, Sept. 26, 1988, at 24; see also Joanna K. Weinberg, Workfare-It Isn't at Work, It Isn't Fair, N.Y. Times, Aug. 19, 1988, at A27 (op ed) (criticizing workfare programs for creating few jobs, cycling people into low-paying jobs requiring few skills, and being generally less successful than they are given credit for). [Vol. 3:1 Poverty, Reproduction, and Autonomy the poverty line, and those who hold public-sector jobs or marginal, temporary employment with income at or below poverty levels. At the core of this issue is the nature of work and training requirements, and the way in which they mesh with the needs of the job market. These requirements fall generally into two categories: those which emphasize job searches or job placement, and those which emphasize supported work and skill training, known as human capital development. 9 The very idea of conditioning benefits upon a "social contract," whereby the state and the recipient agree upon mutually beneficial goals, harkens back to the Progressive Era programs, which provided support only to those who behaved in socially acceptable ways (by looking for work, participating in workhouses, or seeking child support from an absent parent).' In this respect, it is not just the nature of job training that disadvantages women, rather, the entire concept of mandated work preserves an entitlement system that itself is essentially paternalistic. The revisionist provisions also perpetuate the presumption that raising children is not to be valued as "work." Under the FSA, while women are required to work (or to participate in training or education programs), they are often steered to traditional women's work, such as work in child care centers, nursing homes, or cleaning services. 95 Not only are these jobs at the lowest end of the pay scale, but the pattern contains a certain irony. Women may be trained to care for other people's children, but not their own; to take care of other people's dependent relatives, but not their own; to organize other people's homes, but not their own. It should be noted that an ideology encouraging work does not itself disadvantage women. Such a program can be perceived as a way to enable women to become economically self-sufficient. However, the problem with the current structure and implementation of work programs is that they fail to recognize the considerable barriers to full-time employment many of these women face. Too often, single mothers are forced to rely on inadequate day care or to combine full-time work with other caretaking responsibilities, without receiving additional public benefits. Women with young children often cannot find full-time work that is compatible with the demands of caregiving.96 It is also important to note that while transition- 9 Gueron & Pauley, supra note 55, at 22. 9' See Handler & Hasenfeld, supra note 51, at 45-48, 229-30. 9' Weinberg, supra note 64, at 440-42, 448-50. 96 Cf. Brenner, supra note 24, at 128. The Department of Labor notes that one in four women work part-time, while only one in ten men work part-time. One in seven employed women indicate that they were working part-time because they could not find full-time work that allowed them to fulfill their other responsibilities. Bureau of Labor Statistics, U.S. Dept. of Labor, Employment in Perspective: Women in the 1992] Columbia Journal of Gender and Law al child-care benefits will be paid for up to one year after a JOBS program participant finds work,' it is not clear whether these benefits will be paid for part-time work. Indeed, some employers may deliberately limit the number of hours employees work so that they may avoid having to cover these workers under a pension plan.9" Furthermore, the FSA does not pay transitional caregiver benefits to participants who care for disabled or older family members." Under the FSA, three groups are targeted as recipients of the program benefits: the principal wage earner in every two-parent recipient family"° (under the FSA, all states are required to implement a welfare program for unemployed two-parent families as a way of keeping families together);' 0' young parents without a high school degree; 1 2 and women whose chil- dren are almost eighteen (when the youngest child turns eighteen, a recipient's eligibility ends). 0 3 Because the principal wage earner in a two-parent family is presumed to be the father, men thereby become favored recipients of job opportunity and skills training programs. The last category, the older welfare recipient, is the most threatened population, because these women will lose not only their welfare benefits, but also their qualification for participation in JOBS-related programs once their youngest child turns eighteen. It is therefore the category for which the most careful program design is necessary. It is still too early to see whether any state programs have been successful in reaching this population. The proliferation of autonomy-limiting provisions at the state and local levels raises concerns about the lessons to be learned from implementation of the FSA as a whole. Preliminary studies suggest that the current economic recession and persistent state funding shortfalls in the provision of transitional benefits are seriously undermining the program, causing the entire program to be scaled back."'4 Labor Force 1 (2d Quarter 1989). 9 42 U.S.C. § 606 (1991); see Handler & Hasenfeld, supra note 51, at 212. 98 Vicki Gottlich, The Tax Reform Act of 1986: Does It Go Far Enough to Achieve Pension Equity for Women?, 4 Wis. Women's L.J. 1, 15 (1988). 9 42 U.S.C. § 606 (1991). 'o Id. § 602(19)(D). Note, however, that this subsection allows a state, at its option, to require both parents to participate in a job skills and opportunities program if the state guarantees child care consistent with federal guidelines. 'o' Id. § 606(a)(1). 'o Id. § 682(d)(2). 103 Id. § 606(a)(2)(A); see also Gueron & Pauley, supra note 55, at 57-58. 1 Limited information available from the General Accounting Office suggests that more than two-thirds of the states reported that they were unable to supply child care and transportation to all JOBS participants who needed them. U.S. General Accounting [Vol. 3:1 Poverty, Reproduction, and Autonomy V. CAREGIVING: HIDDEN LIMITS ON AUTONOMY AND THE FEMINIZATION OF THE AGING POOR The Relationship Between Caregiving and Economic Dependence Advances in personal autonomy and employment equity during the last twenty-five years have particularly inured to middle-class and younger women. 5 These gains have not equally benefitted poorer and older women for several reasons. First, most of the new jobs created during the economic boom of the eighties, when so many women moved into the workforce, were at the lower end of the wage scale-minimum wage, often part-time, without benefits." During the 1980s, income disparities between low-, middle-, and upper-income families in the United States increased.0 7 The disparity affects women, particularly older women, far more than it does men, and is a result of the male-oriented access structures of pension and health-care systems. 08 Women who take time out of their work lives to raise children or care for frail relatives, or who work part- time, have limited job opportunities in the short run and, over a lifetime, make significantly lower wages than men. 109 Office, Welfare to Work: States Begin JOBS, but Fiscal and Other Problems May Impede Their Progress (Sept. 1991) (cited in Center on Social Welfare Policy and Law, 25 Clearinghouse Rev. 1283, 1291 (1991-92)). 105 Younger middle-class women have been the chief beneficiaries of the equal pay and equal opportunity legislation of the civil rights era and of the advances brought about by the women's movement. See Ruth Sidel, On Her Own: Growing Up in the Shadow of the American Dream (1990); see also Cynthia Fuchs Epstein, Deceptive Distinctions 136-64 (1988) (discussing the structure of work). "0 Note that the service economy, especially in the low-wage occupations, has increasingly become dependent on the supply of low-wage women workers, and this has generated pressure on welfare policy to regulate the flow of low-income women into those jobs. Handler and Hasenfeld note that workfare requirements are used in this manner to discourage women from leaving the labor force. Handler & Hasenfeld, supra note 51, at 136-37. However, it also goes without saying that these jobs are highly sensitive to variations in the economy, and that these women are also the first to lose their jobs in a recession. In recessionary times, workfare requirements have little impact on welfare recipients and are most effective as political coin. 107 See Steven P. Erie et al., Women and the Reagan Revolution: Thermidor for the Social Welfare Economy, in Families, Politics and Public Policy: A Feminist Dialogue on Women and the State 94-119 (Irene Diamond ed., 1983). 108 Jill Quadagno & Madonna Harrington Meyer, Gender & Public Policy, Generations, Summer 1990, at 64, 66. 109 Id. at 64; see also Beth B. Hess, Gender & Aging: The Demographic Parameters, Generations, Summer 1990, at 12, 14. 19921 Columbia Journal of Gender and Law The women's movement has often been criticized for ignoring the needs of poor women and women of color while attending to the interests of white, middle-class women." 0 However, responsibility for the plight of poor women and women of color may more accurately rest with government and its structuring of the welfare system. As the numbers of single mothers raising children with no help from husbands increase, there is greater demand upon the state for welfare support."' In the 1980s, there were substantial cuts in welfare and other social service programs, with some costs shifted from the federal government to state and local programs, and other programs eliminated altogether." 2 The toughening of eligibility standards for disability assistance and medical care caused many working families to fall through the "safety net" of the public benefit system, and the decreasing availability of low-rent housing forced many families into homeless shelters or welfare hotels." 3 Workfare programs also tend to ignore the needs of mid-life and older women, particularly those who have never worked or whose spouses were not employed in the mainstream workforce." 4 Generally, work-training programs are designed for entry-level workers. Services required by older women, such as preparation for retirement or day care for husbands or parents who may be under their care, are seldom addressed. Moreover, there is a double-edged dilemma inherent in the type of jobs in which participants are placed. Many of the jobs are in the realm of "community service": work in shelters, day-care centers, medical facilities, and similar "soft-money" services. While these services are essential, and could forge worthwhile alliances between entitlement recipients and the services they use, these jobs are low-paying and have limited advancement potential. Most detrimentally, these jobs for which women are trained fall within the realm of "women's work." A person's value to society has traditionally been assessed on the basis of his or her ability to perform paid work. The free market is the determinant of the worth of various types of work. The 110 See, e.g., Trina Grillo & Stephanie M. Wildman, Obscuring the Importance of Race: The Implication of Making Comparisons Between Racism and Sexism (Or Other -Isms), 1991 Duke L.J. 397. .. See Fred Block, Rethinking the Political Economy of the Welfare State, in The Mean Season, supra note 23, at 109 (criticizing the "realist" model of capitalist individualism). 112 Handler & Hasenfeld, supra note 51, at 170-71. "a See generally Mimi Abramovitz, The Reagan Legacy: Undoing Class, Race and Gender Accords, J. Soc. & Soc. Welfare, Mar. 1992, at 91, 102-03; Kevin Phillips, The Politics of Rich and Poor: Wealth and the American Electorate in the Reagan Aftermath 20, 184 (1990). "* Doris B. Hammond, Health Care for Older Women: Curing the Disease, in Women as Elders: The Feminist Politics of Aging (Marilyn J. Bell ed., 1986). [Vol. 3:1 Poverty, Reproduction, and Autonomy traditional work of women-caregiving, homemaking, child care-has not even been considered remunerative "work" throughout much of our history and in many contexts. Government policies tend to limit the ability of working women to perform these "traditional" tasks (in the context of their own families) and at the same time to participate meaningfully in the private work economy. President Bush's veto of the 1990 and 1992 Family and Medical Leave Acts are examples of this policy." 5 Older Women: Increased Burdens on Autonomy Another issue, more difficult to describe and document because it has so far evaded close scrutiny, is the gradual and pervasive constriction of autonomy for older women. The constriction is evident with regard to several interconnecting issues-health care, competency, income support, and workplace structure. Decreased autonomy is also closely linked with the role of women as caregivers throughout their life cycle and the failure of policy makers to acknowledge and adequately value this role. Many older women have minimal employment, pension, and health insurance resources. Years spent as caretakers often leave them with few resources to face old age and little training for entry into the workplace. Legislators have created and perpetuated a gender-biased policy of dealing with the problems of caregiving. Legislation providing benefits for older people has traditionally focused on the relationship of the individual's prior work life to his or her need for benefits. Social Security retirement income, for example, is entirely based upon a person's work history (or that of her or his spouse). Health Care and Older Women The crisis in health care illustrates the inadequacy of existing public benefit systems to cope with the caregiving needs of women, particularly older women. The rhetoric of the women's movement places a premium on physical autonomy and control over one's body. As has been shown, such autonomy is less and less possible in the areas of reproduction and economic self-sufficiency. Its limits are also becoming clear in the area of health care, particularly for the elderly." 6 "' President's Message to the Senate Returning Without Approval the Family and Medical Leave Act of 1992, 28 Weekly Comp. Pres. Doc. 1722 (Sept. 22, 1992); President's Message to the House of Representatives Returning Without Approval the Family and Medical Leave Act of 1990, 26 Weekly Comp. Pres. Doc. 1030 (June 29, 1990). 116 An illustrative and contributing factor is the near-total exclusion of women from 19921 Columbia Journal of Gender and Law "Advances" in health care have highlighted technology and profession- alization at the expense of basic care-comprehensive responses to even simple problems, like well-child health care, are compromised by costly solutions. Health care is increasingly governed by private insurance programs, which are employment-based." 7 People who are not employed do not qualify for private insurance (unless they have a spouse who is covered); this means that many people do not receive health care at all.' Studies of Medicaid suggest that in many parts of the country, medical assistance recipients receive a level of health care that is vastly inferior to that available in the private sector, and similar problems are beginning to show up in Medicare, the health-care system for the elderly." 9 Given the glaring lack of services and limitations on autonomy in the area of reproductive health for women in their childbearing years, it is not surprising that older women are an even more underserved community with regard to reproductive health care. All older women have reproductive health concerns, yet feminist theorists in the area of reproductive health many health and medication studies. See Rebecca Dresser, Wanted: Single, White Male for Medical Research, Hastings Center Rep., Jan.-Feb. 1992, at 24; see also Bernardine Healy, Women's Health, Public Welfare, 266 JAMA 566 (1991) (editorial). This is an even greater problem for older women because most pharmaceutical studies have used young men as their standard for setting optimum dosages. As a result, there is little knowledge about the effect of widely prescribed medications on older people. The General Accounting Office recently brought this problem to public attention. NIH Reauthorization and Protection of Health Facilities: Hearings on Reauthorization of the National Institutes of Health Before the Subcomm. on Health and the Environment of the House Comm. on Energy and Commerce, 101st Cong., 2d Sess. 210-12 (1991) (statement of Mark V. Nadel, Associate Director, Nat. & Pub. Health Issues, Human Resources Div., General Accounting Office). Mark Nadel testified that the NIH has made little progress in implementing its policy to encourage the inclusion of women in research study populations. Id. at 211. Although the original policy announcement encouraged researchers to analyze study results by gender, NIH officials have taken little action to implement this element of the policy. Id. at 212. 17 Paul Starr, The Social Transformation of American Medicine 333-34 (1982). 18 Medicaid provisions cover only some of the health expenses of the elderly poor; according to statistics, only about 42% of individuals with incomes below the poverty level receive Medicaid. The working poor, and increasingly, working middle-income families, tend not to have disposable income available for health care. As private insurance has become more expensive, many smaller employers have raised the costs of insurance to their employees, or abandoned health insurance altogether. Up to 75 % of the uninsured are working people or their dependents. Michele Medlen et al., Health Care Rights of the Poor: An Introduction, 25 Clearinghouse Rev. 896, 900 & n.49 (1991) (citing The Pepper Commission (U.S. Bipartisan Commission on Comprehensive Health Care)); American Medical Association, A Call for Action: Final Report, in Health Access America 59-60 (1990) (A.M.A., Chicago, IL). 119 Robert Wood Johnson Foundation, Special Report, Access to Health Care in the United States: Results of a 1986 Survey (1987); see also Quadagno & Meyer, supra note 108. [Vol. 3:1 Poverty, Reproduction, and Autonomy have failed to attend to the specific reproductive health concerns of older women. 20 Diseases of the reproductive system, the need for preventive care for breast cancer, ovarian cancer, and cervical cancer are, for the most part, discussed in relation to younger women. Concerns relating to the problems of menopause, and to the ability of the medical profession to address the issue in an adequate manner, also have not been addressed sufficiently.12 1 While the relative economic status of the elderly has improved over the last three decades, primarily because of increased social security benefits tied to the rising cost of living, and the institution of the Medicare program, women comprise a higher percentage of the elderly poor than men.122 Moreover, because of their relatively marginal workplace participation, and their participation in welfare programs, women are disproportionately represented in the ranks of those who have no medical insurance or inadequate medical insurance. Older women are also less likely to receive social security or private 10 For example, menopause is universally considered to be a form of "disease," rather than a normal stage of the reproductive process. Thus its manifestations become "symptoms," treated medically as "abnormal conditions"; little attention has been paid to the function of menopause as an integral aspect of the reproductive life cycle of women. See Hammond, supra note 114; see also Rosetta Reitz, Menopause: A Positive Approach (1977). See generally Ethel D. Kahn, The Women's Movement and Older Women's Health: Issues and Policy Implications, Women & Health, Winter 1984, at 87, 92. Menopause has recently become a "hot" issue for mid-life feminists. See, e.g., Germaine Greer, Change: Women, Aging and the Menopause (1992); Gail Sheehy, The Silent Passage: Menopause (1992). "' A recent Hastings Center Report details numerous instances of women being excluded from important medical studies-the relationship between low cholesterol and heart disease, many AIDS studies, and even a pilot project on the impact of obesity on breast and uterine cancer. The first 20 years of a major study on health and aging included only men. Dresser, supra note 116; see also Gena Corea, The Hidden Malpractice: How American Medicine Treats Women as Patients and Professionals 232-37 (1977); Myrna Lewis, Older Women and Health: An Overview, Women & Health, Summer-Fall 1985, at 1, 14. 122 In 1986, 29% of persons over 65 and 38% of persons over 75 had incomes below the poverty level (in contrast to 13 % of the population at large). Only children have a higher poverty rate. Moreover, an even higher proportion of the elderly are "near poor," living at or slightly above the poverty level. Kathleen S. Short & Mark S. Littman, U.S. Dept. of Commerce, Bureau of the Census, Current Population Reports, Household Economic Studies, Series P-70, No. 18, Transitions in Income and Poverty Status: 1985-1986, at 3-7 (1990). The large majority of this population are women: 88% of women over age 75 are widowed and living alone. Old, Alone & Poor: A Plan for Reducing Poverty Among Elderly People Living Alone, Common- wealth Fund, April 16, 1987 [hereinafter Old, Alone & Poor]. See also Chairman of House Select Committee on Aging, 100th Cong., 2d Sess., Report on The Quality of Life for Older Women: Older Women Living Alone (Comm. Print 1988) [hereinafter Quality of Life for Older Women]; Karen Davis et al., Alone & Poor: The Plight of Elderly Women, Generations, Summer 1990, at 43, 43-45. 1992] Columbia Journal of Gender and Law pensions than their male counterparts. 21 Women without a consistent work history are increasingly likely to fall through the cracks of a system that ties benefits for older people to past participation in the workforce. 24 The fact that more older women than older men are single means that more women live in poverty. Statistics show that the change in status from married to single does not significantly affect the economic position of men, but the same change in status for women reduces, on average, their income by nearly half."z Most of the time when health care is discussed, the focus is on the health aspect of the issue: the dilemmas of medical technology, health insurance, and treatment modalities. But the health-care dialogue has failed to address the care aspect of the health-care dilemma: the fact that a significant portion of a woman's life cycle may be spent as a caregiver and health-care policies fail to recognize this function as a compensable commodity. 26 Both private and public insurance programs see women as costless caregivers; the services provided by women as caregivers are rarely reimbursable or compensable. 27 The United States Senate Special Committee on Aging reported in 1988 that "elderly persons who have family members to assist them tend to enter nursing homes at a much 123 A 1985 study revealed that only 10% of female spouses had primary pension benefits from their husbands' pensions. Linda Drazga Maxfield & Virginia P. Reno, Distribution of Income Sources of Recent Retirees: Findings from the New Beneficiary Survey, 48 Soc. Security Bull. 7, 9 (1985). 124 Until recently, widows without a work history were required to meet a more stringent test for disability payments under SSA (the Social Security Act disability payment program) than workers. See, e.g., Eileen Sweeney, OBRA-90 Changes in Social Security and SSI Will Benefit Applicants and Beneficiaries, 24 Clearinghouse Rev. 1266 (1991). 121 Short & Littman, supra note 122, at 8. 126 The Commonwealth Fund projects that future poverty rates among elderly widows will increase. Old, Alone & Poor, supra note 122. This is partly because women live longer, but it is also because women are forced to "spend down" a greater proportion of their resources caring for an ill spouse or family member, leaving them with little for their own aging years. Moreover, women tend to have more intermittent work histories because of these other duties. 127 The benefit/work connection has an impact here, as well. The Bush Administra- tion's health insurance proposal was tied to employer contributions to private insurance systems, and many of the alternatives proposed by candidates in the 1992 elections used a similar tie-in to employment, with an expansion of a Medicaid-like program for the un- or under-insured. This does not assure any improvement in the quality of care for those insured by public programs, nor does it address the cost issues posed by private insurance programs. At the time of publication, the Clinton proposal is still vague, although there is increasing pressure to cover the unemployed. It is therefore left to speculation when universal coverage would occur, or whether people ineligible for Social Security benefits would qualify. [Vol. 3:1 Poverty, Reproduction, and Autonomy higher level of impairment than do those without such help.""z The United States relies heavily on informal unpaid caregivers, the majority of whom are either wives or adult daughters. 129 These unpaid caregivers, who are often middle-aged or elderly themselves, are an underserved community, whose problems have not been addressed. 130 The increase in the rate of divorce, and the increased participation of women in the paid labor force may limit the availability of in-home unpaid care in the years to come.' 3' Income Supports and Older Women The dilemma of caregiving suggests another issue that must be addressed in a restructured welfare state: the "graying" of those in need of state assistance. Today's young female AFDC recipients will no longer qualify for public assistance when their grown children leave home. If they do not have functional employment skills, they will again need welfare supports that the programs are not designed to provide. Women who are divorced are not automatically assured of a share in their former spouse's retirement or pension benefits (except for Social Security), and often do not consider their retirement needs at the time that a divorce settlement is negotiated. 3 2 Again, autonomy-limiting legislation does nothing to end this cycle.'33 Even women who have been consistently employed do not benefit as 128 Special Comm. on Aging, Developments in Aging, S. Rep. No. 291, 100th Cong., 2d Sess. (1988) at 18 (citation omitted). 129 Id. at 19. 13o See Hess, supra note 109, at 15; Emily K. Abel, Love Is Not Enough: Family Care of the Frail Elderly 10-12 (1987); Quality of Life for Older Women, supra note 122, at 3. '3" Id. at 25. 132 See Quality of Life for Older Women, supra note 122, at 2; Grace Ganz Blumberg, Adult Derivative Benefits in Social Security, 32 Stan. L. Rev. 233, 257-60 (1980). "' The FSA makes some effort to address this issue by targeting for service priority women whose youngest child is near the age of 18. It is still too early to tell whether this strategy will work, or whether it will provide these women with sufficient education or wage-earning capacity to attain self-sufficiency. The FSA mandates evaluation studies early in the administration of the Act. Pub. L. No. 100-485, 102 Stat. §§ 2343, 2399 (1988). Funding for these studies, however, has not been authorized. A preliminary assessment of the early years of the FSA JOBS program is somewhat pessimistic in this regard. It appears that higher-cost service may be needed to raise the earnings of the more dependent welfare recipients-those without a high school diploma who have had no recent earnings and have been on welfare for more than two years. Gueron & Pauley, supra note 55, at 211-12. 19921 Columbia Journal of Gender and Law equals with men because existing retirement provisions, health benefits, and social security policies were designed with male recipients in mind." Older women are often left without sufficient resources, particularly those women who survive their spouses, or have been part-time or marginal workers. In the future, the relationship between welfare and work that is the hallmark of the American version of the welfare state will fail to address the needs of the older population, The complex, interrelated structure of welfare and social security must also be understood in the context of family law in the United States. In some respects, both areas of law underscore the dependency and lack of autonomy of women, and their need for a "man in the house." While private family law governs the relationships between individuals and their families, and public welfare and social security law governs relationships between poor or elderly individu- als and the state, the two systems have become closely intertwined in recent years. Lawyers practicing in these areas readily admit the necessity for practitioners to be expert in both areas, in order adequately to advise their clients. 135 VI. TOWARD A NEW DEFINITION OF AUTONOMY One way to understand limitations on autonomy, and the magnitude of their impact on policies relating to women, is to understand that these issues 134 See, e.g., C. Arthur Williams, Jr., Social Security: Past, Present, and Future, in Search for a National Retirement Income Policy (Jack L. VanDerhei ed., 1987). There is also a large disparity in the provision of health care and benefits between older men and women. See Quality of Life for Older Women, supra note 122. 3 Amy Hirsch notes that low-income women are physically bounced back and forth between the public and private systems, sometimes being required to go to family court, in order to get permission to apply for welfare, resulting in welfare determinations based upon information disclosed in family court. Hirsch, supra note 73, at 734-35. Similarly, the mechanism for appointing a public guardian to manage the affairs of an elderly individual whose competency is in doubt requires a close coordination between the requirements of state guardianship statutes and regulations, implemented in probate court, and the federal and state protective services and placement structures, which are often determined by what kinds of services will be covered by Medicare or Medicaid. This becomes especially complex when there is a competent spouse who is unable to manage the affairs of the frail spouse him or herself, but who needs his or her own resources to live. See, e.g., Chairman of the Subcommittee on Health and Long-Term Care, House Select Committee on Aging, 100th Cong., 1st Sess., Report on Abuses in Guardianship of the Elderly and Infirm: A National Disgrace (Comm. Print 1987); Penelope A. Hommel, Trends in Guardianship Reform: Implications for the Medical and Legal Professions, in Law, Medicine & Health Care 213 (1990); Jan Ellen Rein, Preserving Dignity and Self-Determination of the Elderly in the Face of Competing Interests and Grim Alternatives: A Proposal for Statutory Refocus and Reform, 60 Geo. Wash. L. Rev. 1818, 1840-41 (1992). [Vol. 3:1 Poverty, Reproduction, and Autonomy are really a metaphor for the highly controversial issue of whether, and to what extent, American society under the guise of state regulation should regulate personal autonomy. '36 I believe that autonomy is socially useful, as long as it is an autonomy that incorporates an ethic of interdependence. It is no coincidence that these issues are coming to a head as we move into the 1990s. In light of the weakening of its constitutional underpin- nings, many aspects of personal autonomy have become susceptible to governmental and legislative regulation. While the reality of life for many women and children in this society is grim, some policy makers have adopted a "blame-the-victim" approach. There is a resurgence of a eugenics undertone in much of the current reproductive policy debate, a hint that perhaps some people-persons with AIDS, alcoholics, drug addicts-should not be permitted to have children;'37 and that others, for example welfare mothers, should not be permitted to keep having children, and should be required to work instead.'38 While almost no one would openly advocate forced sterilization, voluntary sterilization may be required to get a good job, because a company's unstated "fetal protection" policy requires it,"' or to receive public housing or even obstetrical care. The trend advocating sterilization or incarceration of substance-abusing mothers, mothers who abuse their children, or HIV-infected women who become pregnant, are examples of the neo-eugenic position. 4 Because advancing 136 See, e.g., Dworkin, supra note 21. In Robert N. Bellah et al., Habits of the Heart: Individualism and Commitment in American Life (1985), the authors characterize autonomy in American life as something close to what de Tocqueville described in his model of the "independent citizen." But they note a significant change brought about in the twentieth century, stating that we have moved from a society "in which economic and social relationships were visible and, however imperfectly, morally interpreted as parts of a larger common life." The twentieth-century society is one in which "the individual can only rarely and with difficulty understand himself and his activities as interrelated in morally meaningful ways with those of other, different Americans." Id. at 50. "3 See, e.g., Glink, supra note 43, at 538-41. 138 See, e.g., Corns, supra note 36; Weinberg, supra note 64; Lucy A. Williams, The Ideology of Division: Behavior Modification in Welfare Reform Proposals, 102 Yale L.J. 719, 741-46 (1992). "' The decision in UAW v. Johnson Controls, Inc., 111 S. Ct. 1196 (1991), would appear to prohibit this practice, at least with regard to an existing employment relationship. However, some commentators have expressed concern that, after the Johnson Controls decision, employers may be less likely to hire women of childbearing years in the first place, so that they do not run the risk of possible lawsuits down the road. See, e.g., Mary E. Becker, From Muller v. Oregon to Fetal Vulnerability Policies, 53 U. Chi. L. Rev. 1219 (1986); Frances Olsen, From False Paternalism to False Equality: Judicial Assaults on Feminist Community, Illinois 1869-1895, 84 Mich. L. Rev. 1518 (1986). o See Glink, supra note 43, at 538-41 (discussing the increase in criminal 19921 Columbia Journal of Gender and Law technology allows more control of reproduction and fertility, it has caused *the generation of new ideas and methods for regulating fertility. Society has begun to devise more efficient ways of taking children away from parents who harm them: stronger child abuse laws that reach children in the womb, or incarceration or sterilization of women for having chil- dren. 4 Finally, an issue that has been only minimally examined is the definition of autonomy in the context of care for the frail elderly. The structure of health-care benefits for the elderly have reified.technology and professionalization at the expense of basic care. 42 While everyone who receives any form of medical care is affected by this trend, women seem to bear the sharpest impact of these policies, for the reasons described earlier in this paper. Because women are seen as costless caregivers, and their services are rarely reimbursed by public or private health benefit programs, they become invisible statistics in the development and implementation of policies. 1" They are either too preoccupied with the demands of their caregiving roles to take part in policy-making decisions, or when they are finally free, they are so old and frail that their voices are still unheard. Women are also increasingly (and disproportionately to their numbers) losing personal autonomy in formal legal proceedings, such as public appointments of guardians or conservators.I" Elderly women face diminishing control over their assets and personal care decisions through the welfare state's newly developed protective "tools," which justify such encroachments as benevolent interventions. Encroachments on autonomous decision makers in the health-care context are troubling for several reasons. First, there is the threshold question of whether the doctrine of personal privacy under the United States Constitution, or similar state laws, extends to individuals' decisions about prosecutions against drug-dependent women). 14 See Maura A. Ryan, The Argument for Unlimited Procreative Liberty: A Feminist Critique, Hastings Center Rep., July-Aug. 1990, at 6. 142 See Collopy et al., supra note 11, at 2; Jane Aronson, Women's Sense of Responsibility for the Care of Old People: "But Who Else Is Going to Do It?," 6 Gender & Soc'y 8, 25-27 (1992); see also Daniel Callahan, Setting Limits: Medical Goals in an Aging Society (1987); Ken Dychtwald, The Aging of America: Overview, in Wellness and Health Promotion for the Elderly 1, 1, 16 (Ken Dychtwald & Judy MacLean eds., 1986). "I Helen I. Marieskind, Women in the Health System: Patients, Providers, and Professions 146, 319 (1980); Hammond, supra note 114; Ethel D. Kahn, The Women's Movement and Older Women's Health: Issues and Policy Implications, Women & Health, Winter 1984, at 87, 90; Lewis, supra note 121, at 9-10. 144 See generally Susan Miler & Sally Balch Hurme, Guardianship Monitoring: An Advocate's Role, 25 Clearinghouse Rev. 654 (1991); Lebacqz, supra note 10. [Vol. 3:1 Poverty, Reproduction, and Autonomy their physical care, including life or death choices (for themselves or others). The "demise" of Roe will make this question far more complex. A second related question involves the extent to which the state may assert interest in an individual's physical person, as an extension of the state's generic parens patriae interest in the public welfare. A third issue centers on the meaning of the term "best interests" as a legal justification for state involvement impacting personal decisions. How does the social construc- tion of responsibility for one's self or one's intimates fit within the legal definition of autonomy, and when may the state interfere with individual choices about self and family care? Finally, we must acknowledge the impact of the shifting legal and social framework upon professionals and administrators who make health-care policy and deliver health-care services. Historical Perspectives: Best Interests Versus Individual Rights What ties these issues together is a radically changing definition of the role of the state with regard to personal autonomy and privacy. Paradigms of American social thought regarding autonomy involve conflicting views of the power and duties of the state in relationship to people who need some degree of assistance, and the high premium placed on individual rights. The benevolent state paradigm described earlier evolved from the notion that certain people were unable to exercise individual self-determination, due to age or infirmity. 45 As the paradigm evolved, the state's equitable power of parens patriae became the justification for court or state action to protect the welfare of "infants, idiots and lunatics. "'u Early in the '4 The best interests standard initially found broad acceptance as a standard for decision making in child custody and foster care placement cases. Joseph Goldstein et al., Beyond the Best Interests of the Child (1973). It has become a well-established standard in case law in the area of custody adjudications. See Bennett v. Jeffreys, 356 N.E.2d 277, 280 (N.Y. 1976); Krohn v. Krohn, 47 N.W.2d 869 (Neb. 1984). See generally Robert H. Mnookin, Child, Family and State 447-754 (1978). The best interests concept has since been expanded as a mechanism for deciding other issues involving persons with diminished competence. See Judith Areen, Intervention Between Parent and Child: A Reappraisal of the State's Role in Child Neglect and Abuse Cases, 63 Geo. L.J. 887 (1975); David J. Rothman & Sheila M. Rothman, The Conflict over Children's Rights, Hastings Center Rep., June 1980, at 7; Michael S. Wald, State Intervention on Behalf of "Neglected" Children: Standards for Removal of Children from Their Homes, Monitoring the Status of Children in Foster Care, and Termination of Parental Rights, 28 Stan. L. Rev. 623, 634 (1976); Developments-The Family, supra note 20, at 1221-23. '" 3 William Blackstone, Commentaries 47 (1768). While some commentators have interpreted the standard to mean that the state should duplicate the decision that its ward would make if competent, this is not always possible. See Rawls, supra note 19921 Columbia Journal of Gender and Law development of legal doctrine in this area, it was recognized that the parens patriae power ended when a person (usually a child) was considered sufficiently mature and competent to make a reasoned decision.147 Furthermore, the ability of third parties to make decisions for dependent individuals was balanced against the notion that the individual's interests and welfare were paramount, and not to be sacrificed to the welfare of others. '41 The "best interests standard" provided a mechanism through which courts made substantive decisions under the parens patriae power and under which legislatures designed protective legislation. Because the state had an interest in preserving its notion of the child's (or incompetent's) welfare, that welfare, or "best interests," became the determinative factor. 49 In the United States, the parens patriae power was vested in the legislatures, which often delegated that authority to the courts. Gradually, the doctrine expanded to include provisions that allowed the state to remove children from their homes,"' ° and civil commitment statutes that authorized the involuntary commitment to mental institutions of persons who were considered mentally ill.I 1 In a less direct way, the best interests standard also provided justification for legislation that "protected" women and children in the area of employment and working conditions." 2 Broadsweeping parens patriae powers have led to accusations of abuse directed at judges and protective services agencies. Discretionary decision making by judges and elected officials has afforded them the opportunity to intervene in families on the basis of highly subjective standards and values. 13, at 248-49, noting that if knowledge of the preference of the individual is not available, decisions should be made in accordance with the "theory of primary goods," a slight variation of the best interests standard. In other words, even though the state should attempt to duplicate the decision that an individual would have made if competent, the need to ascertain each individual's value structure makes this process unworkable. 14' Developments-The Family, supra note 20, at 1223. '41 Developments in the Law-Civil Commitment of the Mentally I11, 87 Harv. L. Rev. 1190, 1208 (1974) [hereinafter Developments-Civil Commitment]. 141 Developments-The Family, supra note 20, at 1223. "I0 Id. at 1224; see also Ex parte Crouse, 4 Whart. 9 (Pa. 1839) (per curiam); Farnham v. Pierce, 6 N.E. 830 (Mass. 1886); Milwaukee Indus. Sch. v. Supervisors of Milwaukee County, 40 Wis. 328 (1876). '5 In Mormon Church v. United States, 136 U.S. 1, 56-57 (1889), the Supreme Court described the parens patriae power as rooted in the very nature of the state in modern society. It has therefore been viewed as a power which the members of the community have granted the state for the protection of their future well-being. See Developments-Civil Commitment, supra note 148, at 1208. ..2 See, e.g., Frances E. Olsen, The Family and the Market: A Study of Ideology and Legal Reform, 96 Harv. L. Rev. 1497, 1498-99 (1983). [Vol. 3:1 Poverty, Reproduction, and Autonomy The specific interests of children or incompetent persons have been sublimated in order to promote often vague standards of appropriate parenthood.'53 Thus, the "best interests" standard became a rather fluid concept which gave broad discretion to the decision maker and placed less emphasis upon precise determinations of what those individuals perceived to be their own interests. More recently, two relatively distinct threads of analysis about "best interests" standards have emerged. One thread attempts to construct or reconstruct a person's will, through testamentary- evidence or some other form of factual information." This is an appropriate mechanism where an individual has been competent in the past, but is no longer competent, as is the case with some elderly or disabled people. The other, more traditional, thread relies on the discretion of the decision maker to ascertain what course of action, in the decision maker's view, will best benefit the individual. 5 Lon Fuller has described decisions of the first type as person-oriented rather than action-oriented, because the focus of the decision-making process is on the nature and character of the person involved.'56 Under this mechanism, however, the decision maker is still '" Developments-The Family, supra note 20, at 1224. See generally Anthony M. Platt, The Child Savers: The Invention of Delinquency (2d ed. 1977); Ellen Ryerson, The Best-Laid Plans: America's Juvenile Court Experiment (1978); Judge Edward Schoen, The Field of the Juvenile Court, in Proceedings of the Conference on Juvenile Court Standards, 97 U.S. Dept. of Labor, Children's Bureau 32, 34 (1922) ("The State is bound to step in and give the child the same protective care and training that nonnal parents give to normal children in normal homes in civilized, socialized, conventional communities." (emphasis added)). 154 Another term for this thread of analysis is the doctrine of substituted judgment, which is occasionally considered a distinct and separate category. However, its objective is to set a standard for decision making that would effectuate the best interests of the person. See Scott D. Hughes, Comment, Civil Commitment: Guardianship, Substituted Judgment, and Right to Refuse Psychiatric Treatment, 20 Gonz. L. Rev. 479, 491 (1985); see also In re Custody of a Minor, 434 N.E.2d 601, 608-09 (Mass. 1982). 155 This is most commonly articulated in the child welfare area, although it has been applied with regard to mentally disabled adults, and increasingly to partially competent adults and the infirm elderly. Goldstein et al., supra note 145, frames this analysis in psycho-social terms, with regard to child custody decisions, arguing that custody decisions need to be formulated in such a way as to incorporate the particular psychological orientation of children. The standard has become widely accepted in the years since the book was written; however, courts do not always inquire into the wishes of a particular child when making decisions, but often rely upon their judgment or that of a social worker or psychologist in reaching this determination. See Arlene Skolnick, The Limits of Childhood: Conceptions of Child Development & Social Context, 39 Law & Contemp. Probs. 38 (1975). 56 Lon Fuller, Two Principles of Human Association, in Nomos XI 3, 17-19 (J. Roland Pennock & John W. Chapman eds., 1969). 1992] Columbia Journal of Gender and Law obligated to make a decision that is in the "best interests" of the individual, and must therefore attempt to make that determination. While it is possible to design judicial mechanisms that reference an individual's interests, the process breaks down in the legislative arena when used to design social policy, primarily because, by its nature, legislation must be broadly drawn. This makes "person-oriented" decisions difficult. Moreover, because the very purpose of some legislation is to limit autonomy as a condition of providing an entitlement, the essence of the "best interests" concept is undermined. Finally, where legislative categories are not sufficiently narrow, competent adults may be placed in the same category as incompetent adults and children. The inclusion of "incompetents" in the category justifies the use of the "best interests" standard in decision making for the entire category. Consequently, the autonomy of those competent adults may be seriously undermined. This has been the case with the generation of what I term "post-modern welfare state" legislation. At the other end of the spectrum, historically, is the concept of what I call autonomous self-determination, whereby the state is free to act only if, in its application of the "best interests" standard, it does not interfere with an individual's exercise of her right of self-determination.' 57 Under this concept, the state cannot use the parens patriae power in order to shield itself from the requirements of procedural or substantive due process, whereby all state actions must be reasonably related to a valid state goal and actions affecting fundamental liberties are justified only if they promote a compelling state interest."I This analysis supports the superimposition of a right to privacy or self-determination as an underlying baseline, inherently limiting the invasive power of the best interests standard. Constitutional Standards and Legislative Categories Much of the history of social policy, at least in the late nineteenth and twentieth centuries, is about finding a balance between the traditional individual rights paradigm and the benevolent state paradigm. However, the paradigms do not fully explain the design of modem social policy ' As the doctrine has been interpreted in the child welfare area, a state might characterize its role as acting for, rather than imposing its will on, a child as a substitute decision maker. See Parham v. J.R., 442 U.S. 584, 618-19 (1979). This implies a narrower reading of the best interests standard than most courts and commentators have given when adults are involved as plaintiffs. '58 Cf. Developments-Civil Commitment, supra note 148, at 1210. See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432 (1985); see also Roe v. Wade, 410 U.S. 113 (1973). [Vol. 3:1 Poverty, Reproduction, and Autonomy because there are some inherent flaws in the paradigmatic characterization. The two paradigms, like most theories and much legislation, are designed to apply to absolutes-absolute categories and absolute types of people. The individual rights paradigm is most appropriate when applied to people whose autonomy can be fully realized or actualized, people who are able, in the true sense of the classic autonomy theorists, to exercise complete intellectual and social independence.' 59 Similarly, the benevolent state paradigm is most successful conceptually when it is applied to people who are dependent in a classical sense, namely children and the mentally incompetent.I" Most people, however, fall somewhere in between these polar absolutes. Nevertheless, those who design social policies conceptualize the populace in these terms, with results that are unfair as well as inappropri- ate. To a certain extent, legislation is shaped by judicially defined characterizations of various groups within society, particularly with respect to constitutional issues.161 As a result of this characterization, legislation that has a disparate impact on people because of their identification with one of these groups is not subject to the highest level of judicial scrutiny, which is applied in instances of race-based impact. Legislation affecting women is reviewed under the intermediate scrutiny standard, requiring that the legislation relate to an important government purpose. 62 Legislation affecting the elderly, the poor, gays and lesbians, and the disabled is subject only to the lowest level of judicial scrutiny, requiring merely that legislation be rationally related to a legitimate government purpose.'63 19 Joel Handler describes the inherent conflicts in the construct of autonomy when this is not the case. Handler, supra note 17; see also Joel F. Handler, Social Movements and the Legal System: A Theory of Law Reform and Social Change (1978). " See, e.g., Goldstein et al., supra note 145. For a discussion of the uncertainty of benevolence theory in the context of welfare, see William H. Simon, Legality, Bureaucracy and Class in the Welfare System, 92 Yale L.J. 1198 (1983); Frank I. Michelman, The Supreme Court 1968 Term-Foreward: On Protecting the Poor Through the Fourteenth Amendment, 83 Harv. L. Rev. 7, 40-47 (1969). 161 See generally Susan Moller Okin, Justice, Gender and the Family (1989); Ruth Bader Ginsburg, Some Thoughts on Autonomy and Equality in Relation to Abortion, 63 N. Cal. L. Rev. 375 (1985); Sylvia A. Law, Rethinking Sex and the Constitution, 132 U. Pa. L. Rev. 955 (1984); Frances Olsen, Comment: Unraveling Compromise, 103 Harv. L. Rev. 105 (1989). 162 See, e.g., Frontiero v. Richardson, 411 U.S. 677 (1973); Reed v. Reed, 404 U.S. 71 (1971); see also Milner S. Ball, Judicial Protection of Powerless Minorities, 59 Iowa L. Rev. 1059 (1974). 163 See, e.g., Bowers v. Hardwick, 478 U.S. 186 (1986); City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432 (1985). The ultimate impact of the recently enacted Americans With Disabilities Act will be interpreted by the courts in subsequent years. 42 U.S.C. § 1211 et seq. See The Supreme Court: Leading Cases, 99 Harv. L. Rev. 120, 161-73 (1985); see also Mark F. Kohler, Note, History, Homosexuals 19921 Columbia Journal of Gender and Law While these standards are of judicial origin, they find their way into legislation, particularly in the social policy arena. Legislators anticipate constitutional challenges to laws limiting the autonomy of persons in these categories. Legislators also tend to apply the benevolent state paradigm broadly, overzealously limiting autonomy where an individual requires assistance in only one aspect of life. It is therefore appropriate at this time to address the issue of recasting these paradigms in a form more accommo- dating to current social structures and theory. The development of social welfare legislation in the twentieth century is intertwined with the formulation of a progressive-era notion of the welfare state, probably best characterized by David Rothman."6 The progressive model is of a state that has a particularized interest in the welfare of its citizens, but is (ideally) bounded at its parameters by a respect for and awareness of the autonomy of individuals. This construction has led, in turn, to a model of a welfare state that takes responsibility for those within the society who are deemed to be "dependent." In the first- half of the twentieth century, legislators embraced this notion. Beginning in the 1960s, however, the courts took on the role of protecting autonomy, and the ideology of autonomy was radically changed. Counterpoint to the progressive formulation of the welfare state's obligations was the other historical event of the twentieth century, the glorification of the liberal individual rights analysis, moving beyond the political sphere and applying it to social and personal rights. VII. CONCLUSION: CONNECTING THE THREADS- A FINAL WORD The conflict between the theoretical frameworks of self-determination and "best interests" is particularly relevant to the provision of social and medical services in the modern social welfare state. Like the liberal state, the modern social welfare state places a high value on personal autonomy and seeks societal protection against threats to autonomy. However, the ideology of liberalism is not necessarily appropriate for the construction of a modern welfare state, because that ideology does not always recognize the interdependent, rather than independent, nature of autonomy. This is one of the reasons why many who support some of the themes of liberalism find and Homophobia: The Judicial Intolerance of Bowers v. Hardwick, 19 Conn. L. Rev. 129 (1986). '" David J. Rothman, Conscience and Convenience: The Asylum and Its Alternatives in Progressive America (1980); David J. Rothman, The Discovery of the Asylum: Social Order and Disorder in the New Republic (1971). [Vol. 3:1 Poverty, Reproduction, and Autonomy that liberal ideology does not adequately fulfill the needs of a more complex society.'6 In any modern society, public social services are not designated to specific individuals, but are generally allocated on the basis of decisions about community or interest group needs. Distribution of these services occurs through a form of negotiation, often in the political process, but sometimes in the context of professional-client relationships, through a kind of claiming process."' 6 The public fisc pays for such services, usually through taxation mechanisms. Under present interpretations of constitution- al law, most social services are not guaranteed to individuals by right, but are considered to be entitlements proffered by the state. 67 Allocation of these goods and services is, however, subject to due process mecha- nisms. 16 Recently, however, the foundations of both the liberal-progressive welfare state and the liberal rights ideal have been substantially shaken, either by inherent fault or external pressure. However, there is increasingly a concern that the wealthy might choose to opt out of the public service claiming, distribution, and provision system, by choosing to pay individual- ly (or in smaller interest groups) for privately funded services, such as private schools, recreational facilities, and neighborhoods .169 When the 165 See Bruce A. Ackerman, Social Justice in the Liberal State 3 (1980); West, supra note 5. However, this paper is not intended to be a critique of liberalism, or as an argument that liberalism has fundamentally failed in its mission, but rather to suggest that even though the fundamental principles of liberal theory may be valid, those principles should be phrased so that they reflect the highly relational and interdependent nature of individual autonomy. See, e.g., Michelman, supra note 160; Michael J. Sandel, Liberalism and the Limits of Justice (1982); see also Amy Gutmann, Communitarian Critics of Liberalism, 14 Phil. & Pub. Aff. 308 (1985). In my use of the concept of liberalism, I attempt to incorporate an ethos of community within the liberal definition of substantive due process. See, e.g., Gilligan, supra note 9; Minow, supra note 9. 166 See Peattie & Rein, supra note 6, at vii, wherein the vocabulary of "claiming" is used to "develop a language which makes it possible to make the connection between our understanding of political and social processes at the micro level, in which there is a social negotiation of reality, and the more macro-social and political economy." The authors note: People who work in the micro tradition talk about the negotiation of reality between doctor and patient, citizen and bureaucrat, husband and wife; there is a tendency to leave this account at the interpersonal level. Those working at the macro level try to give accounts of broader institutional processes, which tend to be bereft of the personal content of role and experience. Id. Social conventions can therefore remove aspects of the distributional system from public discussion, and thereby delineate a false consciousness of rights. 167 See, e.g., Wyman v. James, 400 U.S. 309 (1971). 6 See, e.g., Goldberg v. Kelly, 397 U.S. 254 (1970). 169 This is the theme of two recent books by social commentators. See generally 19921 Columbia Journal of Gender and Law wealthy choose not to claim the benefits of the public system, they also tend to opt out of the funding mechanism, that is, to refuse through the political process to authorize taxation for such services.170 This has created a kind of disequilibrium in the structure of public and social services that cuts off fiscal authority for these services, just as the groups who now need them are able to claim more political power. It is important for feminists to understand the nature of this disequilibrium, and to find ways to overcome it, by looking more closely at the conflict between the communitarian ideology of the welfare state and the "liberal legal" ideology of autonomy. This might be done in several ways. First, feminists need to become more overtly involved in the design and implementation of social policy legislation, perhaps developing a set of concrete "litmus tests" that would subject such legislation to scrutiny of its hidden as well as its overt agendas. But it is also necessary to develop a new model for social policy legislation that incorporates a feminist view of autonomy. This model should incorporate several themes: the necessity for using interdependence in positive ways; an ideology of and context for care; and an equal as well as an equitable allocation of resources based upon a system of negotiated claims. Phillips, supra note 113; Robert B. Reich, The Work of Nations (1991). 170 Reich, supra note 169. It should also be noted that the trend toward rejecting increased taxes for social spending continues, as many bond issues as well as tax provisions were voted down in the November 1992 elections. See, e.g., Vlae Kershner, California Voters Reject 9 of 13 Ballot Measures, S.F. Chron., Nov. 5, 1992, at A15. [Vol. 3:1