Vol. 96, No. 5

The Court’s Morality Play: The Punishment Lens, Sex, and Abortion

June Carbone* & Naomi Cahn†
Constitutional Law, Supreme Court
Articles

This Article uncovers the hidden framework for the Supreme Court’s approach to public values, a framework that has shaped—and will continue to shape—the abortion debate. The Court has historically used a “punishment lens” to allow the evolution of moral expression in the public square, without enmeshing the Court itself in the underlying values debate. The punishment lens allows a court to redirect attention by focusing on the penalty rather than the potentially inflammatory subject for which the penalty is being imposed, regardless of whether the subject is contraception, abortion, Medicaid expansion, or pretrial detention.

This Article is unique in discussing the circumstances in which the Court has simultaneously concluded that the state could regulate but could not punish, even if that means redefining a sanction as not punitive. By making visible this framework, we offer the Court and the states a potential off-ramp from the continuation of an ugly and litigious future on abortion access. If the Supreme Court seeks to deflect the outrage over Dobbs, the simplest way to do so would be to take seriously the statement that all it has to do is to return the issue to the states. In that case, the Court’s focus should be, as Justice Kavanaugh suggested in his concurrence, on the impermissibility of punishment that infringes on established rights, independent of a right to abortion, such as the right to travel, the First Amendment right to communicate accurate information about abortion availability, or doctors’ efforts to perform therapeutic abortions necessary to preserve a pregnant person’s health. The Court would not pass judgment on the permissibility of abortion, and it could affirm the propriety of state bans, but still strike down heavy-handed prosecutions and ill-defined prohibitions that impose undue penalties.

After Dobbs v. Jackson Women’s Health Organization, this Article is particularly important for three reasons. First, this Article examines the ways in which the Court has used considerations of punishment to deflect irreconcilable values clashes. Second, a focus on punishment often illuminates the “dark side” of government action, justifying limits on such actions. Third, a focus on “punishment” often illustrates the consequences of government actions, consequences that may be an indirect result of statutes or regulations but that have disproportionate effects on marginalized communities. Understanding how the Court has used this elusive concept in the past may thus help shape the response to Dobbs.

INTRODUCTION

The concept of punishment is central to the Supreme Court’s jurisprudence on abortion—and, beyond abortion, to the expression of moral values in the public square. In Dobbs v. Jackson Women’s Health Organization, Justice Alito found “an unbroken tradition of prohibiting abortion on pain of criminal punishment” throughout the common law until the Court’s decision in Roe v. Wade in 1973.¹ He noted that “the great common-law authorities—Bracton, Coke, Hale, and Blackstone—all wrote that a post-quickening abortion was a crime”² and he traced these developments from the thirteenth century forward.³

The Dobbs opinion, like most criminal law discussions, assumes that the power to prohibit includes the power to punish violations of those prohibitions. And, indeed, criminal law scholars have produced an extensive literature on the justifications for the imposition of criminal sanctions and the constitutional limits on that imposition.⁴

What neither that vast literature nor the Dobbs opinion addresses, however, is the role of punishment in the evolution of the jurisprudence addressing the expression of public values, separate and apart from the existence of the laws prohibiting conduct. As this Article shows, when the Supreme Court has focused on the state’s justification for punishment independently from the underlying policy, it has often used the nature of punishment as a justification for striking down legislation—even when the Court concedes that the state purpose is otherwise legitimate.⁵ And it sometimes uses the declaration that onerous provisions are not “penalt[ies]” to uphold coercive legislation that, as a practical matter, limits access to what the Court otherwise recognizes as important rights.⁶ As these cases show, outside of the narrow context of whether a criminal prohibition justifies the imposition of a particular sentence,⁷ punishment has an ill-defined life of its own in Supreme Court jurisprudence.

This Article is the first to detail how the Supreme Court has viewed the concept of “punishment” as a justification for upholding or invalidating government acts in the context of issues involving contested values.⁸ While an intense debate raged at mid-century over whether the state should regulate morality, that debate generally assumed that if the state could regulate, it could also punish.⁹ This Article is unique in discussing the circumstances in which the Court has simultaneously concluded that the state could regulate but could not punish. For example, the Court held that a state could discourage teen sex but not by encouraging pregnancy as the consequence¹⁰ or could adopt restrictive measures,¹¹ such as blanket refusals to fund medically necessary abortions, so long as the statute did not prohibit abortion or penalize those seeking one.¹²

This Article is particularly important following Dobbs for three reasons. First, it illustrates the ways in which the Court has used considerations of punishment to deflect irreconcilable values clashes. For those who would like to extricate the Court from the conflicts Dobbs has inflamed, limiting punishment, for example, of those exercising a constitutionally protected right to travel, offers a potential off ramp.

Second, a focus on punishment often illuminates the “dark side” of government action. The opinion in Griswold v. Connecticut placed great weight on the intrusiveness of policing the use of contraceptives in the marital bedroom.¹³ The ugliness of imposing punishment may similarly become a focal point for organization in response to the patchwork of state laws after Dobbs.¹⁴

Third, a focus on “punishment” is often used to illustrate the consequences of government actions, consequences that may be an indirect result of statutes or regulations but that have disproportionate effects on marginalized communities.¹⁵ Abortion bans may aggravate race and class-based differences,¹⁶ prompting greater recognition of the rights of the pregnant to obtain the medical care needed to safeguard their health.¹⁷ Understanding how the Court has used this elusive concept in the past can thus help shape the response to Dobbs.

The Supreme Court’s conception of “punishment” underlying these considerations is slippery, perhaps intentionally so.¹⁸ The Court uses the concept at both an expressive level, reinforcing public norms, and a practical level, specifying the consequences for the violation of government mandates, both civil and criminal. Most critically for this Article, it provides the Court with a way to shape emerging norms in the context of public unease.

After describing the multidisciplinary literature on punishment’s multiple roles, we examine the way that the Court has deployed punishment as a rationale for invalidating government action, particularly in the context of cases involving sexual morality. Eisenstadt v. Baird, which stuck down bans on the sale of contraceptives to single women, provides a classic case: the Court simultaneously “conceded” that “the State could . . . regard the problems of extramarital and premarital sexual relations as ‘(e)vils’ ”¹⁹ but still held that this could not be the purpose of the Massachusetts legislation because it “would be plainly unreasonable to assume that Massachusetts has prescribed pregnancy and the birth of an unwanted child as punishment for fornication . . . .”²⁰ The irrationality of the punishment, not the permissibility of unmarried sex, provided the basis for the decision—and implicitly for the limitation of state power to regulate sexual morality.

We then explore how the Court has used the determination of what is a punishment to affirm state decision-making power in a federal system. The question of when the state is inflicting a punishment as opposed to imposing a reasonable condition or proceeding on appropriate administrative grounds arises in contexts ranging from welfare “home visits” to detention to Medicaid expansion, with the Court using the punishment lens to sidestep the substantive bases for these decisions.²¹

Finally, this Article considers cases that directly engage the relationship between punishment and the underlying values debate. Lawrence v. Texas, which invalidated Texas’s same-sex sodomy statute, provides the most striking example. Justice Kennedy’s majority opinion did not just strike down the criminalization of the sexual conduct.²² It affirmed the dignity and worth of the expression of intimacy in the case. Justice Scalia’s dissent, by contrast, saw punishment as the point, with both the majority and dissent agreeing that the values debate was central to the discussion.²³

This Article observes that the “punishment lens” provides a powerful tool for shaping the evolution of public values without enmeshing the Court in the underlying values debate. We consider whether the punishment lens can be successful in two ways: guiding the evolution of public values without triggering a backlash that further entrenches polarized opposition²⁴ or, failing that, reaching decisions in controversial cases that do not undermine the Court’s own legitimacy and authority.²⁵ By this standard, Eisenstadt v. Baird, which used the punishment lens to avoid the underlying values questions while striking down barriers to contraceptive access, and Lawrence v. Texas, which instead of relying on the punishment lens directly engaged the values questions, both succeeded in resolving issues in ways that helped move public opinion and lock in legal conclusions that remain embedded in American law. Whether applying the punishment lens to abortion can enjoy similar success remains to be seen, but this Article concludes by outlining the possibilities a focus on punishment can offer.

I. PUNISHMENT AND THE RULE OF LAW

The role of state-administered punishment is much studied—and much contested. Existing literature addresses the questions of what might justify the ability of the state to inflict intentionally burdensome treatment on its citizens,²⁶ what purposes such punishment should serve,²⁷ and what constitutes appropriate punishment.²⁸ As this scholarship establishes, law enforcement—and punishment of egregious crimes—is essential to a state’s legitimacy. Without punishment of criminal acts, a state cannot govern—and command either the support of its constituents or deference from the international order.²⁹ This literature, however, in its most idealized form, tends to assume a straightforward relationship between crime and punishment: the state prohibits certain acts, imposes prescribed penalties for the violation of the law, and administers the penalties in accordance with principles of procedural and substantive justice, emphasizing due process rights for the accused and fairness defined in terms of proportionality between the crime and the punishment.³⁰

This Section goes beyond the conventional analysis of criminal punishment to explore the expressive role that punishment serves. It shows that the judicial oversight of punishment serves four roles that pose difficult challenges in the face of contested or changing values: establishing shared societal values, maintaining or dismantling social hierarchies, mediating disputes over the authority of governmental actors to impose punishment, and channeling the individual desire for vengeance into state-approved channels.

First, the administration of punishment defines and reinforces societal values, often in symbolic ways. For example, with the recognition that smoking caused cancer and other health risks,³¹ the perceived acceptability of smoking changed.³² In the United States, the state did not respond by prohibiting smoking. Instead, government entities gradually limited the places where smoking was permitted, first, creating “no smoking” areas and ultimately banning smoking in restaurants, offices, and other places.³³ Over time, enforcement of these rules—and the imposition of sanctions on violators—did not just shift norms of politeness; they expressed moral disapproval of smoking as undesirable and deviant.³⁴ In 1993, the Supreme Court of the United States embraced the shift in attitudes in a decision that held placing a nonsmoking prison inmate in a cell with a five-pack-a-day smoker could constitute constitutionally impermissible “cruel and unusual punishment.”³⁵ In so ruling, the Court did not limit the word “punishment” to the prescribed penalties for a criminal act.³⁶ Instead, its finding of “cruel and unusual punishment” reflected and reinforced the changed social meaning of smoking from an acceptable activity to one that violated evolving “standards of decency,”³⁷ and concluded that violating this new moral sensibility could constitute “punishment” within the meaning of the Constitution. The act of placing a nonsmoker with a smoker thus became punishment because of the changed moral status of smoking.

Second, legal scholars have argued that beyond merely maintaining order, much of the power of state-administered punishment comes from this expression of “moral condemnation”³⁸ and its role in establishing social hierarchies within a society. In accordance with this analysis, moral condemnation does not just declare particular conduct to be illegal; it establishes and reinforces social order and social standing in a society.³⁹ Criminal acts threaten to upend the social order, as the person committing the crime asserts the right to defy established law and norms. Imposing punishment that carries moral condemnation with it restores the moral order,⁴⁰ affirming the victim’s superior status to that of the violator.⁴¹ Punishment can thus signal that “the community values the victim”⁴² while the failure to punish can indicate indifference, or even disdain, toward the victim.⁴³ Accordingly, both imposing punishment and failing to punish send important messages about what a society values. State-administered punishment can thus establish and reinforce norms in ways that contribute to social cohesion,⁴⁴ cohesion operating at the group as well as the individual level.⁴⁵ Disturbingly, the imposition of punishment can breed cohesion even if there is no crime;⁴⁶ nonetheless, the state reaffirms its legitimacy and authority when it punishes in the name of a value or ideal, rather than simply because it can.⁴⁷

The role of punishment in establishing social hierarchies, particularly when it operates at a group-based level, contributes to the dark side of punishment.⁴⁸ Brain imaging studies⁴⁹ show that the act of punishing engages the part of the brain that produces feelings of reward—the same area of the brain involved in drug addiction.⁵⁰ Individuals may thus derive pleasure from imposing punishment on others even when imposing punishment makes the punisher worse off.⁵¹

This psychological dimension corresponds to some descriptions of the retributivist purpose of punishment. Nietzsche argues that cruelty—and the satisfaction some derive from it—is the point of punishment.⁵² Even Oliver Wendell Holmes agreed that at least in some cases, punishment “is inflicted for the very purpose of causing pain” and “one of its objects is to gratify the desire for vengeance.”⁵³ And the anger and moral outrage that fuels demand for punishment can be manipulated.⁵⁴ Research has tantalizingly suggested that the act of punishment itself reinforces perception of harm.⁵⁵ Cultural cognition studies further show that people associate behavior contrary to their moral norms with socially detrimental consequences.⁵⁶

Precisely because the administration of punishment reinforces social standing at both the individual and the group level, it has implications that go beyond the punishment administered to any particular individuals. The decisions about which punishments to implement (such as firing an employee who refuses to be vaccinated or imposing work requirements as a condition of eligibility for state subsidized health insurance benefits) can create group-based winners and losers, elevating the status of one group at the expense of another.⁵⁷ Yet, denying the legitimacy of such demands for punishment—or imposing them too harshly—can also undermine respect for law.⁵⁸

This leads to the third role of judicial oversight of punishment: mediating conflicts that involve the authority of different governmental actors to impose punishment. In the United States, for example, the Supreme Court has overseen evolving conflicts between the states and the federal government in the administration of family law. The U.S. Constitution has historically been viewed as entrusting family law to the states,⁵⁹ but the Supreme Court has selectively intervened, at times to enhance or restrain state authority to impose punishment. In Stanley v. Illinois,⁶⁰ for example, the Supreme Court held that Illinois could not treat Peter Stanley as an unfit parent⁶¹—and thus deprive him of standing to seek the custody of his children after their mother’s death—solely because he had not married the mother.⁶² The Court intervened to limit the power of the state to punish unmarried fathers, at a time when attitudes were changing toward unmarried relationships.

Finally, the courts have historically overseen punishment in order to channel vengeance into socially constructive venues. The failure to punish perceived wrongs may persuade wronged individuals or groups to “take the law into their own hands” or to impose punishments out of proportion to the wrongful act.⁶³ The courts, in contrast, are supposed to act “judiciously” in administering punishment in a neutral manner, not just on behalf of the wronged individual, but because the assertion of the moral values of the social order can contribute to a sense of social order and cohesion.⁶⁴

The challenge of serving these four roles increases as social norms change.⁶⁵ The tension between maintaining order and imposing destabilizing punishments is particularly difficult if some social groups reject the norms, while others respond to the increasing defiance of the first group by calling for greater punishment as violations increase.⁶⁶ The imposition of punishment thus involves an “ever-shifting relationship between a regime and a given population that makes up the most essential element in any political order.”⁶⁷

These four roles make the administration of punishment central to the rule of law. They are also evident as a longstanding aspect of Supreme Court jurisprudence. Yet, managing the tensions between these objectives can undermine as well as maintain social cohesion. Congress and various state legislatures, for example, have attempted to shift norms surrounding intimate relationships by changing the laws governing sexual assault to make date rape easier to prosecute.⁶⁸ Imposing more serious penalties, however, may make judges and juries more reluctant to convict—and failures to impose punishment can undermine, in turn, the efforts to shift norms and also lead victims to feel even more isolated and aggrieved.⁶⁹ Expressing moral condemnation while keeping punishments commensurate with the perceived seriousness of the offenses thus requires walking a tightrope, one that sways with changing public sensibilities. Abortion, perhaps as much as if not more than any other issue, involves “irreconcilable disagreement” that challenges the legitimacy of the judicial system itself. The issues of punishment in the abortion context will test whether the judiciary generally, and the Supreme Court in particular, retain any capacity for guiding the recreation of shared social values.⁷⁰

II. SEX AND PUNISHMENT: RECOGNIZING REPRODUCTIVE RIGHTS

At the time Roe v. Wade was decided in 1973, the Supreme Court was carefully navigating a revolution in sexual mores.⁷¹ Sexual morality presents a classic case for the expressive role of punishment,⁷² with punishment serving to reinforce what are seen as consensus-based moral values broadly shared by the public.⁷³ Enforcing such norms also involves, however, punishment of private consensual conduct.

This Part shows how the Supreme Court focused on the acceptability of punishment as a rationale for state action rather than on the changing norms themselves. It did so through a series of cases that addressed contraception, nonmarital children’s legitimacy status, welfare benefits, parentage—and ultimately abortion—though the lens of punishment for sexual conduct. Within this new jurisprudence, the Court carved out a right to privacy that did not address the propriety of intimate conduct, but rather evaluated the permissibility of state action designed to shape private conduct.

A. Contraception and the Propriety of Pregnancy as Punishment for Sex

Starting with Griswold v. Connecticut⁷⁴ in 1965, the Supreme Court began to strike down legislation that regulated sexuality in ways that the Court deemed needlessly punitive. In doing so, the Court never waged a frontal assault on the moral order that channeled sexuality into marriage.⁷⁵ Instead, the Court examined the rationales underlying the laws and the consequences of imposing punishment.

Griswold addressed the constitutionality of a law that forbade the use of contraception. Anthony Comstock had spearheaded prohibition of contraceptives in the nineteenth century, convinced that they “facilitate[d] immoral conduct” because they “reduce[d] the risk that individuals who engage[d] in premarital sex, extramarital sex, or prostitution [would] suffer the consequences of venereal disease or unwanted pregnancy.”⁷⁶ Comstock persuaded Congress to outlaw “print and pictorial erotica, contraceptives, abortifacients, information about contraception or abortion, sexual implements and toys, and advertisements” in 1873⁷⁷ and the states adopted their own “Little Comstock laws” thereafter.⁷⁸ Connecticut’s statute, adopted in 1879,⁷⁹ was one of the most restrictive, banning not just the advertising and sale of contraceptives, but also the use of contraception.⁸⁰

The Court framed the case as one against defendants who “gave information, instruction, and medical advice to married persons as to the means of preventing conception.”⁸¹ In resolving the matter, the Court conceptualized a right to privacy, a right that justified looking the other way at sexual conduct. The Court wrote that “[w]e deal with a right of privacy older than the Bill of Rights older than our political parties, older than our school system.” ⁸² The Court did not mention married couples’ efforts to limit the number of children they had directly, although it did refer to the marital relationship as “intimate to the degree of being sacred” and suggested that enforcing a ban on contraceptive use would have “a maximum destructive impact upon [the marital] relationship.”⁸³ By contrast, the Court acknowledged the validity of the state’s purported rationale for the regulation: “the discouraging of extra-marital relations.”⁸⁴ While the Court stated that this rationale “is admittedly a legitimate subject of state concern,”⁸⁵ banning contraceptive use by married couples was simply too far removed from the purported subject of the statute to pass constitutional muster.⁸⁶ The Court suggested that the state could regulate the manufacture or sale of contraceptives but not their use within marital unions.⁸⁷ In short, the Court focused on the ugliness of enforcement⁸⁸ rather than on the permissibility of the underlying conduct—the use of contraception.⁸⁹

Connecticut did not often enforce its ban on married couples’ contraceptive use, but the fact that the law was on the books effectively limited the ability to use contraception to those with access to doctors and pharmacists.⁹⁰ While the Griswold decision did not mention the issue,⁹¹ a major reason for challenging the ban on contraception was the unequal nature of contraceptive access.⁹² By striking down criminal penalties for contraceptive sales, the Court effectively allowed doctors and clinics to make contraception more broadly available. The implicit principle at the core of this decision was that, while the state could steer sexuality into marriage, it could no longer seek to ensure that pregnancy be the unavoidable consequence of sexual relationships.

Eisenstadt v. Baird,⁹³ decided in 1972, expanded the principle—that pregnancy was an unreasonable punishment—beyond marriage. Eisenstadt struck down a Massachusetts statute that prohibited supplying contraception to single, as opposed to married, individuals. As in its decision in Griswold, the Court “conceded” that “the State could, consistently with the Equal Protection Clause, regard the problems of extramarital and premarital sexual relations as ‘(e)vils.’ ”⁹⁴ Nonetheless, it concluded that this could not be the purpose of the Massachusetts legislation because it “would be plainly unreasonable to assume that Massachusetts has prescribed pregnancy and the birth of an unwanted child as punishment for fornication.”⁹⁵ The Court acknowledged, as it did in Griswold, that notwithstanding the law, contraceptives are widely available, and thus “the rationality of this justification is dubious.”⁹⁶

By 1977, the Supreme Court was willing to say that the state could not prescribe pregnancy as the punishment for sex even where the state had a clear interest in discouraging sex between minors. In striking down a state law that prohibited selling contraceptives to minors under the age of sixteen, the Court noted the state interest in regulating the “morality of minors” in its efforts to promote “the State’s policy against promiscuous sexual intercourse among the young.”⁹⁷ Again, however, the Court accepted the legitimacy of the state interest, but rejected the connection between such a state interest and the prohibition on sales of contraceptives to minors. The Court observed that, “with or without access to contraceptives, the incidence of sexual activity among minors is high, and the consequences of such activity are frequently devastating,” but observed that there was little evidence that banning contraception had much impact.⁹⁸ The Court thus concluded that the state could not promote an otherwise legitimate objective—discouraging “promiscuous sexual intercourse among the young”—by making pregnancy the punishment for sex and criminalizing efforts to avoid the consequences.⁹⁹ And it emphasized that the justification for banning contraceptive sales became that much weaker as the evidence mounted that the laws on the books did not have the desired effect. It thus concluded that the “punishment” (pregnancy) did not serve the interests of either deterrence (teens with still have sex) or an appropriate desert (a child) for a wrongful act.

B. Public Recognition and the Removal of the Scarlet Letter from Children

The Supreme Court relied on similar reasoning in dismantling the distinctions between “legitimate” and “illegitimate” children, with the Court ultimately concluding that the states could not seek to channel childbearing into marriage by punishing children for their parents’ conduct. In the “seminal” case of Levy v. Louisiana,¹⁰⁰ the Court considered a Louisiana law that restricted the ability to bring a tort action for the wrongful death of a parent to “legitimate children.”¹⁰¹ As a result, an unmarried mother’s five children, who lived with her, and whom she raised on her own earnings, had no right to sue for their mother’s allegedly wrongful death. The Court, in striking down the statute in a brief opinion, observed that the Court could imagine no reason “why, in terms of ‘equal protection,’ should the tortfeasors go free merely because the child is illegitimate?”¹⁰² The Court reasoned that the circumstances of the birth had “no relation to the nature of the wrong allegedly inflicted on the mother;” the children, “though illegitimate, were dependent on her.”¹⁰³ The Court even recounted how the mother in the Levy case supported her children by working as a domestic servant, “taking them to church every Sunday and enrolling them, at her own expense, in a parochial school.”¹⁰⁴ In this opinion, the Court identified no countervailing state interest; the children were deprived of the right to sue for the loss of their mother simply because of the circumstances of their birth.

The Supreme Court in Levy did not mention the issue of race, but amicus briefs filed in the case emphasized that, particularly in Louisiana, the distinctions between marital and nonmarital children had a significant racial impact. Indeed, an amicus brief filed by Illinois law professor Harry Krause (and others) argued explicitly that the statute “discriminates on the basis of race.”¹⁰⁵ The brief maintained that the discrimination stemmed partly from the fact that “disproportionately more Negro children than white children are born out of wedlock,” and, partly from the fact that “a high percentage (70%) of white illegitimate children are adopted . . . whereas very few (3-5%) Negro illegitimates find adoptive parents.”¹⁰⁶ As a result, “95.8 percent of all persons affected by discrimination against illegitimates under the statute are Negroes.”¹⁰⁷ The brief concluded, “the classification of illegitimacy . . . is a euphemism for discrimination against Negroes.”¹⁰⁸

Louisiana denied that it sought to punish the children for immorality in sexual behavior, but it nonetheless maintained that it sought to encourage marriage.¹⁰⁹ And the state asserted: “If the community grants almost as much respect for non-marriage as for marriage, illegitimacy increases” and that “illegitimate daughters tend to err in the manner of their illegitimate mothers, producing more illegitimate children.”¹¹⁰ In short, Louisiana did argue that it was necessary to punish the children to deter their parents, if not quite in so many words. And the children who would be punished as a result were overwhelmingly Black.¹¹¹ Louisiana’s efforts to punish nonmarital births thus reinforced a racial as well as moral line, though the majority opinion for the Court did not directly address the racial issue.

In subsequent cases, the Court made the role of punishment even more explicit. In 1972, the Court reaffirmed Levy in striking down a Louisiana statute that defined “child” so that only marital children were eligible for insurance benefits resulting from their father’s death.¹¹² Justice Powell’s majority opinion held that the “status of illegitimacy has expressed through the ages society’s condemnation of irresponsible liaisons beyond the bonds of marriage,” but still concluded that imposing “this condemnation on the head of an infant is illogical and unjust.”¹¹³ Powell concluded that the distinction between marital and nonmarital children was not justified by any state interest.¹¹⁴

In 1977, the Court revisited the issue of inheritance,¹¹⁵ invalidating an Illinois statute that permitted nonmarital children to inherit only from their mothers, not their fathers.¹¹⁶ In a 5-4 decision, Justice Powell reiterated that “visiting this condemnation on the head of an infant is illogical and unjust.”¹¹⁷ He emphasized that, while the parents’ behavior might have been immoral, that was not the fault—nor the responsibility—of the children. The opposition to the punishment of children commanded a majority of an even more conservative Court than the Warren Court that had initially struck down such classification.¹¹⁸

C. The Right to Abortion: Part I

The Supreme Court’s 1973 decision in Roe v. Wade situated the case within the punishment lens the Court had constructed to deal with reproductive rights more generally. The case never squarely fit there, however, because abortion did not just involve the regulation of sexual behavior between consenting partners; it also raised issues about the involvement of the medical profession and the status of the fetus. Nonetheless, the Court framed the decision as a right centered on the irrationality of the state prescription of childbirth as a way to prevent illicit sex and a jurisprudence conscious of the consequences, intended and unintended, of regulating sexual morality. It thus treated laws banning abortion as imposing punishment—on the pregnant for incurring an unwanted pregnancy, on doctors for exercising medical judgment in treating patients, and on those who felt compelled to seek illegal abortions in unsafe circumstances.

Among the telling aspects of this analysis is the way the Court articulated the state interests at stake. The Court identified the first such interest as one based on “a Victorian social concern to discourage illicit sexual conduct.”¹¹⁹ Curiously, though, the Court acknowledged that Texas did not articulate that justification in Roe, and it appeared that courts and commentators had not actually taken the argument seriously.¹²⁰ On the other hand, however, the Comstock laws, which banned abortifacients along with pornography and contraception, treated the regulation of sexual morality as of a piece with abortion.¹²¹ The Court thought the connection between an abortion ban and the regulation of morality sufficiently important to mention—and dismiss.

Second, the Court acknowledged that forcing a woman to carry an unwanted pregnancy to term is cruel.¹²² It referred to the burdens of pregnancy and childbirth, including the possibility that childbirth “may force upon the woman a distressful life and future,” her “[m]ental and physical health may be taxed by child care,” and the unwanted child may cause “distress, for all concerned.”¹²³ The opinion acknowledged the hardship involved in bringing a child into a family that could not care for the child, and the potential for stigmatizing a nonmarital mother.¹²⁴ The Court accordingly echoed earlier cases treating avoidable pregnancy and childbirth an inappropriate way to advance state purposes because of the burden imposed.

Third, the Court was aware that the states often brought criminal actions against doctors.¹²⁵ One of the parties in Roe, Dr. James Hubert Hallford, allegedly had faced prosecutions for violations of the Texas abortion statutes.¹²⁶ Hallford maintained that the applicable statutes were unconstitutionally vague because he could not determine whether his patients’ situations would qualify as exceptions to the abortion ban,¹²⁷ so he faced punishment for exercising a good faith medical judgment about his patients’ therapeutic needs. Justice Blackmun’s initial draft proposed striking down Texas’s anti-abortion law as unconstitutional only on the grounds that it was void for vagueness.¹²⁸ The punishment that doctors faced in making delicate judgements was clearly a factor in the subsequent Roe decision and in its declaration that abortion decisions should be left to “the woman and her responsible physician.”¹²⁹

Fourth, the Court dismissed state assertions that banning abortion was necessary to protect women’s health, observing that mortality rates during the first trimester of pregnancy “appear to be as low as or lower than the rates for normal childbirth” in contrast with the “prevalence of high mortality rates at illegal ‘abortion mills.’ ”¹³⁰ While less explicit than the Court’s acknowledgment of the burdens of pregnancy, the Court recognized that resort to unsafe abortions was a punitive consequence of the prohibition of legal abortions.

In the background of the case, states’ law on abortion had begun to change, with some states repealing their anti-abortion statutes entirely and others reforming their law to expand the availability of therapeutic abortions.¹³¹ A practical consequence was that, as with contraception, the availability of abortion, particularly safe abortion, differed significantly by race, location, and class.¹³² Partly as a result, women of color were substantially more likely—by some estimates twelve times more likely¹³³—to die from illegal abortion than white women.¹³⁴

In limiting the state ability to restrict abortion, the Court treated these restrictions as imposing impermissible penalties on those seeking abortion. The penalties were not so much the criminal sanctions themselves; these were rarely imposed on the individuals who secured abortions.¹³⁵ Instead, states banning abortion were making childbirth the consequence of unprotected sex—and the risk of death the price of seeking an illegal abortion. The Court found that unacceptable. And while the Court recognized the state interest in protecting fetal life, it balanced that interest against the woman’s interest in deciding whether to give birth. Fetal life, as an interest separated from the sexuality (and women’s bodies) that produced it, would become more prominent as an issue only after Roe was decided.¹³⁶

In these cases, the Supreme Court helped oversee a shift in sexual mores during a period where nonmarital sexuality was becoming more common and accepted. In focusing on the acceptability of the punishment, the Court did not endorse the changes directly; instead, it addressed the rationality of widely violated restrictions that imposed serious, arbitrary and discriminatory harms. The Court’s use of the term “punishment” was not, however, consistent or the subject of a coherent jurisprudence. Sometimes, it referred to the state rationales (deterring sex by limiting access to contraception, making pregnancy the “punishment” for fornication), sometimes it referred to the intrusive nature of criminal enforcement (searching the marital bedroom) rather than the imposition of criminal sanctions, and sometimes it considered the collateral consequences of government action (the stigma and limitations associated with nonmarital births). In the process, however, the Court used the punishment lens to oversee a wholesale effort to strike down what it saw as the outdated remnants of “Victorian” sexual mores without disavowing the legitimacy of state efforts to channel sexuality into marriage.

III. PUNISHING PARENTS

The era that produced Roe involved overlapping interests reducing the support for a punitive approach to sexual morality: a change in sexual norms, a remaking of women’s roles, and more urgent calls for racial equality.¹³⁷ In addition, the parties were less ideologically polarized, with greater elite consensus.¹³⁸

Nonetheless, by the mid-seventies, another jurisprudential revolution was taking place: one embedding a neoliberal¹³⁹ view of the state into Supreme Court jurisprudence. The Warren Court had been sympathetic to calls not just for racial equality, but also for greater economic rights.¹⁴⁰ These claims often took the form of calls to treat government benefits as entitlements, with more equal access to the benefits and more obstacles to denying eligibility. The neoliberal era taking hold by the late seventies rejected these claims. The Court embedded this perspective in the same way it had overseen the change in sexual mores: by using the punishment lens to resolve issues that involved farther reaching clashes in values. The Court did so by denying the very fact of punishment. It concluded that if a given regulation did not penalize the individuals subject to it for protected activity, no constitutional issue arose at all. In the process, the Court upheld regulations that supervised poor women’s sexuality and denied access to abortion funding.

This Section focuses on how the punishment lens applies in more varied civil settings, tracing the evolution of the Court’s treatment of government benefits. The first part of this Section describes how the Court deemed public benefit requirements non-punitive in order to uphold limitations on government benefits under the Aid to Families with Dependent Children (AFDC) and Medicaid programs; the second part of the Section shows how the punishment lens applies outside of the sexual-morality context, analyzing how it has been used to limit access to benefits under the Affordable Care Act.

A. Welfare Benefits and the Rejection of Positive Rights

In the 1960s, the Supreme Court addressed the relationship between sexuality and eligibility for government benefits during a period in which the Court was enhancing access to government benefits more generally.¹⁴¹ The original Aid for Dependent Children (ADC) program was adopted in the 1930s as part of the New Deal’s far-reaching social legislation. The United States, unlike many European nations, had never adopted a universal system of family allowances to support childrearing but instead had a variety of state programs designed to provide widows’ pensions to support children who would otherwise land in orphanages because their mothers could not support them.¹⁴² In the 1930s, Congress nationalized these efforts, providing federal funding for a state-run system to compensate for the loss of a male breadwinner.¹⁴³ Congress limited aid to children who had “been deprived of parental support or care by reasons of the death, continued absence from the home, or physical or mental incapacity of a parent” and allowed the states to impose additional eligibility standards, such as “moral character” requirements that excluded the children of unmarried parents from the program.¹⁴⁴

As early as the 1940s, critics argued that the moral requirements “were habitually used to disguise systematic racial discrimination; and that they senselessly punished impoverished children on the basis of their mothers’ behavior.”¹⁴⁵ The federal government sought to discourage the moral requirements.¹⁴⁶ By the late 1960s, the states had shifted from outright prohibition of benefits to “man in the house rules” that deemed the income of a man who cohabited with a welfare recipient to be available to the family, thereby affecting the family’s qualification for public welfare.¹⁴⁷ These regulations were understood to serve the dual purpose of punishing African Americans and privatizing dependency by withholding public benefits from nonmarital families.¹⁴⁸

In King v. Smith, the Supreme Court examined the punitive nature of these requirements. The Court sidestepped the constitutional issues in the case, striking down the Alabama regulation at issue on statutory grounds, noting that federal law precluded states from denying public welfare to children because “of their mothers’ alleged immorality or to discourage illegitimate births.”¹⁴⁹ The Court concluded that “Congress has determined that immorality and illegitimacy should be dealt with through rehabilitative measures rather than measures that punish dependent children, and that protection of such children is the paramount goal of AFDC.”¹⁵⁰ Justice Douglas’s concurrence, however, would have reached the constitutional issue. He saw Alabama officials as discriminating against children on the basis of illegitimacy and therefore acting at odds with the ruling in Levy v. Louisiana, decided during the same term.¹⁵¹ He wrote that “the Alabama regulation is aimed at punishing mothers who have nonmarital sexual relations.”¹⁵² In administering the provisions, the “economic need of the children, their age, their other means of support, are all irrelevant. The standard is the so-called immorality of the mother.”¹⁵³ He viewed that standard—and the attendant punishment—inflicted on the mother to be constitutionally impermissible.¹⁵⁴

By the time the Supreme Court decided the case in 1968, the nature of the AFDC program had changed. While 43% of the ADC caseload in 1937 consisted of widows, only 7% were in 1961.¹⁵⁵ And as documented in an amicus brief in Levy v. Louisiana, decided the same term, the statute was both “overt discrimination on the basis of the criterion of illegitimacy,” and “covertly discriminate[d] on the basis of race.”¹⁵⁶ The Court almost certainly saw the two cases as linked, although only Justice Douglas’s concurrence in King made the connection directly.¹⁵⁷

In deciding King v. Smith, the majority opinion, however, dealt with these issues only obliquely. Instead, it focused on the irrationality of the punishment imposed—the denial of benefits in a program intended to help children that would disproportionately disadvantage the very children the program was intended to help. The Court did not endorse a right to nonmarital sexuality.¹⁵⁸ It did not discuss the discriminatory motive and effect underlying the regulations.¹⁵⁹ It did not recognize an affirmative “right” to federal benefits¹⁶⁰ nor a right to privacy for benefit recipients.¹⁶¹ Instead, it focused solely on the legitimacy of the punishment, concluding that children could not be deprived of benefits in an effort to change their mothers’ conduct. It treated the man-in-the-house rules not as a rational effort to determine the resources available to the family, but as a subterfuge to continue morals regulation in the face of federal disapproval.¹⁶² The case was thus of a piece with the contraception and legitimacy cases in challenging irrational punishments: punishments that were irrational because once they failed to deter nonmarital sexuality in an era of changing mores, their application became arbitrary and discriminatory.

In subsequent cases, however, the Supreme Court upheld provisions that burdened the poor and their children by deeming such provisions non-punitive. Thus, in Wyman v. James, the Court found constitutional a New York statute mandating home visits,¹⁶³ that were in line with federal law’s requirements that aid be provided only after consideration of the family’s resources and only to children who were not being neglected.¹⁶⁴ The “visits” could prove embarrassing in front of children and guests, and could serve to police sexual relationships.¹⁶⁵ The Court refused to find that mandated visits were a penalty at all, terming them instead a condition of benefit eligibility¹⁶⁶ and not a substantive, much less punitive, standard tying loss of benefits to impermissible or arbitrary considerations.¹⁶⁷

The dissent objected on the grounds that welfare rights should be seen as entitlements.¹⁶⁸ While both the majority and the dissent focused on the status of welfare benefits, Justice Blackmun’s majority opinion used the conclusion that the “conditions” on receipt of benefits were not penalties to lock in a neoliberal view of government action: because there is no right to benefits, the state could impose whatever standards it chooses as preconditions for eligibility, and those conditions never become punishment subject to constitutional scrutiny.¹⁶⁹

B. Punishing Sex

In subsequent cases, the Court’s characterization of a particular government action as non-punitive allowed it to uphold conditions that were challenged as discriminatory, cruel, or unjust. The results were particularly striking when the issue turned to abortion. Legislators who opposed abortion and who could not overturn Roe v. Wade directly sought to express their disapproval of abortion by prohibiting the use of public funds to pay for abortions, while permitting those funds to be used for pregnancy and childbirth.¹⁷⁰ Were these bans penalization of a constitutionally protected right—the right to elect abortion to terminate a pregnancy—or were they simply the exercise of legislative policy preferences to allocate public funds to support some activities and not others? The Supreme Court used the punishment lens to resolve the issue. Since individuals enjoyed no positive right to health care—or to abortion funding—the denial of funding could not constitute a penalty and thus had no constitutional implications.¹⁷¹

An initial case upheld Connecticut regulations limiting public funding of abortions to medically necessary abortions during the first three months of pregnancy.¹⁷² Justice Powell wrote for the 6-3 majority that the Constitution did not impose any obligation on the states to pay pregnancy-related medical expenses of low-income women or any other medical expense.¹⁷³ He noted that the Court had not found in previous cases that wealth was a suspect class¹⁷⁴ and that Connecticut was accordingly free to subsidize childbirth and not abortion as an expression of state policy designed to encourage the former.¹⁷⁵

By 1980, Congress had gone further, adopting the Hyde Amendment, a prohibition on the use of federal funds to reimburse the cost of abortions under the Medicaid program, including abortions that were the result of rape or incest or medically indicated.¹⁷⁶ In a 5-4 opinion later that year, the Supreme Court upheld the constitutionality of the Amendment. The majority opinion treated the issue as a classic one of negative liberty, explaining that the freedom to choose to have an abortion, even a medically necessary one, does not carry with it a government obligation to fund the abortion.¹⁷⁷ It then explained that a woman’s poverty was “the product not of governmental restrictions on access to abortions, but rather of her indigency.”¹⁷⁸ Accordingly, the Court concluded that the failure to pay for abortions was not punishment and thus not subject to constitutional review.

The four dissenters viewed the Hyde Amendment as punitive and cruel. Justice Blackmun made the point that the legislators championing the Hyde Amendment cynically sought to express their own views on the morality of abortion by imposing those views “only upon that segment of our society which, because of its position of political powerlessness, is least able to defend its privacy rights from the encroachments of state-mandated morality.”¹⁷⁹ He would have accordingly subjected the legislation to more exacting judicial review.¹⁸⁰ Justice Stevens emphasized that “the Court expressly approves the exclusion of benefits in ‘instances where severe and long-lasting physical health damage to the mother’ is the predictable consequence of carrying the pregnancy to term” and, indeed, “even if abortion were the only lifesaving medical procedure available.”¹⁸¹ He concluded that the result “is tantamount to severe punishment” for wanting an abortion.¹⁸² Justice Marshall emphasized the racial impact of denying abortion funding and also noted that the Hyde Amendment resulted in “excess deaths.”¹⁸³

In Harris, the Court upheld the validity of an extraordinarily cynical statute. Congress, in effect, limited poor women’s abortion access because it could—it could allow expression of the anti-abortion sentiments of members of Congress at the expense of a relatively powerless group.¹⁸⁴ By declaring that forced birth due to the failure to secure funding for an abortion was not a punishment, the Court avoided addressing the question of whether it burdened a constitutional right.

In Wyman and Harris, neither the majority nor the dissenting opinions treated these cases as imposing punishment for sex, and the majority opinions rejected even the premise that the aid recipients had been punished for the exercise of constitutional rights (privacy in Wyman, abortion in Harris). The reasoning in the cases backtracked on the entitlement language that had been building in the welfare rights era, leading to the conclusion that if the benefits at issue were not entitlements, the failure to provide them could not be seen as punishment—effectively ending the discussion of whether the provisions at issue were unduly cruel or whether they reinforced class- or race-based social hierarchies.¹⁸⁵

C. The Punishment Lens Beyond Sex

The litigation over the Affordable Care Act (ACA)¹⁸⁶ involves the clash of values we have described in this Article and the use of the punishment lens to resolve some of the challenges. The ACA was the largest expansion of public largesse in a half century and therefore a direct challenge to neoliberal values. The legislation’s principle of universal health insurance coverage clashed with those who wished to limit government benefits altogether or to withhold them from those deemed unworthy, such as those who were not working, reinforcing class and racial hierarchies. In addition, by treating contraception as an integral part of women’s health care, the ACA conflicted with the views of some Christian employers who opposed contraception. The legislation thus involved, on a much larger scale, the clash of values underlying the characterization of government benefits in Wyman and Harris.

In the cases discussed in this Section, the Supreme Court returns to the issue of punishment, though without any more precise a definition of the concept. Instead, the Court repeatedly faced the question of whether the ACA provisions operated as a tax or a penalty, a condition or a penalty, and a provision of alternative means of compliance or a penalty, and used the characterization of the actions as penalty or not to resolve the cases.¹⁸⁷ The net effect for the ACA was a compromise: the ACA endured but on somewhat more neoliberal terms than the Obama Administration and the Congress that enacted the legislation might have intended.¹⁸⁸

The ACA, in attempting to provide universal health care access,¹⁸⁹ included a series of alternatives that were designed to balance the principles of expanded access, adequate funding, and reasonable private choice.¹⁹⁰ In National Federation of Independent Business v. Sebelius, the most prominent of the ACA cases,¹⁹¹ the Court addressed two issues that turned on the concept of a penalty. The first involved the “individual mandate,” which required an individual who did not otherwise receive health insurance through their employers or other state provisions, to purchase health insurance on state exchanges or pay what the legislation described as a “penalty” collected by the Internal Revenue Service with the filing of individual tax returns. The Court rejected the government’s claim that the commerce clause authorized the mandate,¹⁹² but upheld it instead as a “tax.”

The Court reasoned that under the ACA, “if an individual does not maintain health insurance, the only consequence is that he must make an additional payment to the IRS when he pays his taxes.”¹⁹³ The Government accordingly argued that the mandate could “be regarded as establishing a condition—not owning health insurance—that triggers a tax—the required payment to the IRS.”¹⁹⁴ Under this theory, the legislation does not establish “a legal command to buy insurance,” just a trigger for owing taxes, like “buying gasoline or earning income.”¹⁹⁵ Therefore, the Court concluded that the ACA was within the Congressional tax power.

Critical to the Court’s reasoning was its decoupling of a requirement to buy insurance, which the Court concluded that Congress could not do, and a requirement to pay an amount, deemed by the Court a “tax,” intended to finance the program. In reaching this conclusion, the Court explained that “[i]n distinguishing penalties from taxes, this Court has explained that ‘if the concept of penalty means anything, it means punishment for an unlawful act or omission.’ ”¹⁹⁶ The ACA mandate was not a penalty (or punishment) because while the mandate sought to incentivize health insurance purchases, it did not make the failure to do so “unlawful.”¹⁹⁷ The fact that Congress sought to influence individual behavior did not matter, just as Congress’s efforts to encourage childbirth rather than abortion did not matter in Harris v. McRae; so long as the federal government did not outlaw the failure to buy insurance, the individual mandate was a tax, not a penalty (and not punishment for the failure to buy insurance). It was therefore constitutional.¹⁹⁸

The second issue the Court addressed was Medicaid expansion, which the Court again decided in terms of the acceptability of the Act’s “penalties.” Congress revised the existing Medicaid program, which is a federal-state partnership, to cover individuals within 138% of the poverty line, and to bring Medicaid coverage in line with the coverage health insurance policies offered on the exchanges.¹⁹⁹ Congress then gave the states a choice: accept federal funding in accordance with the new expanded Medicaid program or forego federal Medicaid funding.²⁰⁰ The majority in Sebelius objected that the “choice” was too coercive, effectively mandating state participation in the program.²⁰¹ It reasoned that while Congress could condition state eligibility for federal funding under a new program, it could not “penalize States that choose not to participate in that new program by taking away their existing Medicaid funding,”²⁰² describing the “inducement” in the Act as “a gun to the head.”²⁰³ Justice Ginsburg’s dissent objected that Congress was, as it had done in the past, just requiring states to comply with “conditions” imposed by Congress to receive Medicaid funding.²⁰⁴

The parallels between Sebelius and Wyman v. James are striking. The requirement that the states adopt Medicaid expansion in order to participate in the Medicaid program could have been described, as Justice Ginsburg wrote, as a condition for participation in a federally funded program. The Sebelius Court disagreed, finding that it penalized the states for the failure to agree to the program’s terms.²⁰⁵ The Court effectively treated the state’s existing funds as an entitlement the federal government could not threaten to take away in order to obtain the performance it sought. In Wyman, because welfare was not an entitlement, a welfare recipient’s failure to consent to intrusive home visits was not considered a penalty at all; it was labelled as “a condition of eligibility” to the continued receipt of benefits.²⁰⁶ The label—condition or penalty—resolved each case without engaging the substantive issue of whether the conditions themselves were reasonable or justified.

In Sebelius, the result cloaks the real issues underlying Medicaid expansion—skepticism about whether the poor merit medical benefits²⁰⁷ and opposition to the state role in meeting such needs.²⁰⁸ Indeed, the federal government picked up 100% of the initial costs associated with implementing the program, and 90% thereafter so that the financial burden on the states was relatively minimal²⁰⁹—and less than the state share of the pre-ACA Medicaid program²¹⁰ and arguably much less of a burden on the states than asking a welfare recipient to consent to frequent, unannounced, and intrusive home visits (or the uninsured to go without health care).²¹¹ What Sebelius did not address is why states opposed Medicaid expansion, given the substantial financial incentives in the ACA for the states to do so. Most commentators attribute the opposition to the states’ ideological opposition to government provision of health insurance, if not outright hostility to the poor people in their states.²¹² Some states continue to resist Medicaid expansion, despite widespread public support for it.²¹³ In effect, the Court, in the name of federalism, authorized the states to act with impunity in frustrating Congressional efforts to ensure accessible health insurance at the expense of people in their states who qualified for the benefits.

In a later ACA case,²¹⁴ the Supreme Court also used the concept of punishment to address the employer mandate, which gave businesses the choice of providing health insurance that met federal standards for their employees or contributing to the exchanges so that employees could purchase their own insurance.²¹⁵ Hobby Lobby, a closely held, for-profit corporation, provided health insurance for its employees, but refused to comply with federal requirements to cover certain forms of contraception, including the morning after pill, because, according to the company, they acted as an abortifacient.²¹⁶ In a 5-4 decision, the Court held that requiring a company to cover certain mandated health care benefits, such as the pills in question, violated the Religious Freedom Reformation Act.²¹⁷ The Court gave little regard to women’s loss of access to the contraceptives, holding that the federal government, if it chose, could provide them through “less restrictive means.”²¹⁸ In short, the Court held that it would be an unjustifiable penalty to compel corporate owners to comply with the terms of a neutral government program that benefitted their employees, if those terms conflicted with the owners’ religious beliefs.²¹⁹

The employer mandate was essentially the same as the individual mandate—it gave those affected, whether individuals or employers, a choice: meet the ACA requirements (individuals by purchasing insurance that met federal standards or employers by providing such insurance) or pay the mandated sums to the federal treasury, in each case less than the cost of the insurance. With respect to the individual mandate, the Court concluded that the payment was a tax on those without insurance and not a penalty because the federal government had not (and could not) compel the purchase of insurance. In the case of the employer mandate, the Court concluded that the required payments were, in effect, a penalty for Hobby Lobby’s desire to act on its religious beliefs, rather than a condition for participation in a program providing federal subsidies.

To be sure, the two cases do not arise under identical bodies of law. Sebelius addressed two distinct legal issues: Congressional power to enact the individual mandate under the Commerce Clause and the taxing power, and the limits of Congressional power under a federal system to incentivize state participation in a federal program. Hobby Lobby was decided in accordance with a third body of law, determining the religious rights of for-profit corporations. Yet, in each case, the Court’s framing of the law as punishment or not—that is, whether the intricate provisions of the ACA acted as sanctions designed to compel specific behavior—determined the outcome. And, in each of these cases, the Court upheld moral hierarchies: protection of religious employers at the expense of employees denied access to federal contraception benefits, protection of states disapproving of health care subsidies at the expense of their citizens who would benefit from such subsidies, and limits on the power of the federal government vis-à-vis other actors, including the states and privately-held businesses. ²²⁰

IV. RETURNING MORALITY TO THE PUBLIC SQUARE

In focusing on punishment, the Supreme Court oversaw a revolution in sexual mores without directly engaging the issue of what values should govern in the public square. The Court has also strengthened a neoliberal regime by simultaneously holding that imposing conditions on program beneficiaries does not constitute punishment while imposing conditions that require coverage constitutes a constitutionally unacceptable “penalty.” In relatively few of these cases did the Court, particularly in its majority opinions, directly engage the underlying values clash. The exception has come in the discussion of LGBT rights—and increasingly in the Court’s opinions on abortion. These exchanges pull back the curtain on the role of punishment in Supreme Court jurisprudence. In these cases, the argument for the losing parties, embraced by the dissents, maintain that punishment is the point—the necessary component to affirming the “right values” in the public square. In response, the Court, in a way it did not do so in the earlier cases, directly addresses the relationship between the status of those affected by punishment and the values they express by engaging in the prohibited activity.

A. LGBTQ+ Rights and the “Homosexual Agenda”

One of the clearest clash of values prior to Dobbs occurred in the Supreme Court’s decision in Lawrence v. Texas.²²¹ In Bowers v. Hardwick,²²² the Court had considered whether there was a fundamental right to engage in same-sex sodomy, a formulation that the Court repeated in Dobbs. In both cases, the Court referred to the long history of criminalizing the conduct at issue,²²³ with those arguing for the constitutionality of such criminal penalties maintaining that the history of punishment reflected disapproval of the underlying conduct and provided evidence of the continuing legitimacy of such sanctions.²²⁴

Lawrence, which involved a criminal prosecution for same-sex sodomy, directly involved the issue of punishment. The two men in the case were arrested in a private residence when the police arrived to investigate a purported weapons disturbance.²²⁵ In his opinion for the majority, Justice Kennedy’s opinion had two levels of analysis. Like the Griswold line of cases, it affirmed a right to privacy, observing that “[t]he statutes do seek to control a personal relationship that, whether or not entitled to formal recognition in the law, is within the liberty of persons to choose without being punished as criminals.”²²⁶ The majority opinion then emphasized that the Texas statute being enforced in the case was not just about prohibiting a “particular sexual act”;²²⁷ it involved intimate conduct as part of “a personal bond that is more enduring.”²²⁸ The opinion thus concluded that such punishment was not just constitutionally impermissible²²⁹ but that the behavior at issue had societal value.²³⁰

Justice O’Connor, in her concurrence in Lawrence, did not go as far as the majority. Instead, in a manner reminiscent of the earlier cases on contraception, she limited her analysis to a punishment lens, finding that Texas could not claim a legitimate interest. She thus rejected out of hand the asserted state interest in the case, which she described as nothing more than the “moral disapproval of an excluded group.”²³¹ For O’Connor, the impermissibility of the punishment—and its discriminatory character—were enough to strike down the statute without necessarily requiring an affirmation of the value of same-sex intimacy.

Writing in dissent, Justice Scalia made clear that he thought that moral disapproval of same-sex sexuality was exactly what the case should have been about. He denounced what he called the “homosexual agenda,” which he defined as “the agenda promoted by some homosexual activists directed at eliminating the moral opprobrium that has traditionally attached to homosexual conduct.”²³² He cast his dissent explicitly in terms of maintaining a moral hierarchy based on that opprobrium.²³³

The opinions in Lawrence thus frame, perhaps better than any of the other cases, the permissibility of punishment and the Court’s use of a punishment lens. They involve a clash between the ability to affirm moral values in the public square versus the preservation of private homes from the intrusion of the state. They also involve the use of the declaration of values to define those to be “protected,” in Scalia’s words, from those to be “excluded,” in O’Connor’s terms, thus reaffirming societal hierarchies between the groups. And they involve the permissibility of the imposition of criminal sanction to reinforce moral opprobrium, even when the behavior at issue is consensual conduct between two adults. The Lawrence Court’s 6-3 majority unequivocally rejected the propriety of punishment used to harden the lines between the protected and the excluded—and in the majority opinion, if not O’Connor’s concurrence,²³⁴ embraced an alternative view of the purpose of sexual conduct as an expression of commitment to a partner, not just as a means to procreation.

In Obergefell v. Hodges,²³⁵ the case upholding the right to marriage equality, the majority went even further in embracing same-sex relationships as an expression of family values²³⁶ while the dissents reaffirmed the need to channel sexuality into marriage—and to punish those who fell outside of such precepts. ²³⁷ Kennedy wrote that there “is dignity in the bond between two men or two women who seek to marry and in their autonomy to make such profound choices.”²³⁸

The majority opinion added that the right to marry is not just about the couples’ relationship to each other, but also about their children. “Without the recognition, stability, and predictability marriage offers,” Kennedy wrote, “their children suffer the stigma of knowing their families are somehow lesser . . . . The marriage laws at issue here thus harm and humiliate the children of same-sex couples.”²³⁹ The opinion thus saw denial of the ability to marry as a punishment imposed not only on the couple but on their children. ²⁴⁰ It accordingly equated the limitation of marriage to different-sex couples with imposition of a stigma on those raising families outside the institution.²⁴¹

The Obergefell majority did take sides in the culture wars—in recognizing the dignity and moral worth of same-sex relationships. In basing the decision on the changed nature of marriage, the Supreme Court acknowledged that marriage reflected a new moral sensibility: one that made autonomous choice, not religious or societal duty, the foundation of the marital relationship.²⁴² The Court accordingly went beyond the rejection of the punishment (while noting “the harm and humiliation” involved in the refusal to recognize same-sex families) to confer public recognition and moral worth on LGBT families.

The four justices who dissented rejected both the premise that marriage had changed and that the Supreme Court should acknowledge that change. Chief Justice Roberts’s dissent explained that “for the good of children and society, sexual relations that can lead to procreation should occur only between a man and a woman committed to a lasting bond.”²⁴³

This reasoning is the same as the reasoning that justified the vilification of nonmarital sexuality a half century ago. In accordance with this reasoning, heterosexual sex, not just procreation, needs to be channeled into marriage and marriage needs to be about a moral command to avoid nonmarital sexuality.²⁴⁴ Punishment, whether material or symbolic, is the necessary complement to this reasoning.²⁴⁵

Alito’s dissent made explicit his objection to overturning traditional moral hierarchies. He wrote: “I assume that those who cling to old beliefs will be able to whisper their thoughts in the recesses of their homes, but if they repeat those views in public, they will risk being labeled as bigots and treated as such by governments, employers, and schools.”²⁴⁶ In short, Alito’s concern lay directly with the ability to uphold the preferred values in the public square and fear that those who did so would now be the ones receiving punishment. And while he acknowledged that family understandings and behavior could change over time, he simply treated data such as the 40% nonmarital birth rate as further reason states could chose to double-down on traditional moral understandings²⁴⁷—drawing clear distinctions between preferred groups and those subject to moral condemnation even when a substantial or even majority of the public did not share such views.

Alito’s opinion accepted the right of moral traditionalists to insist on the primacy of heterosexual marriage and to punish those who create families or engage in sexual intimacy outside of marriage. He saw the majority, in contrast, as embracing same-sex families as entitled to equal moral worth and such views as necessarily punishing those who disagree as bigots. Moreover, he treated evidence of changing norms, such as the increase in nonmarital births, as evidence of a threat to the traditional moral order and therefore as additional reason for punishment. Framed in such terms, the legal question becomes one of power and authority to uphold the preferred views and, in Alito’s terms, punishment cannot be separated from the underlying values.

B. Abortion Revisited

With respect to abortion, however, neither the Court’s efforts to sidestep the morality of the underlying conduct nor its efforts to address the issues directly have yet succeeded. In the years after Roe, abortion became a political marker in part because the issue offers little opportunity for compromise.²⁴⁸ While the Court largely succeeded in making contraception more available without directly embracing the sexual revolution, the Court’s efforts to sidestep the moral issues underlying the abortion issue satisfied no one. Roe satisfied neither those who saw reproductive rights as essential for gender equality nor those who believe the status of the fetus is not an issue that could be “bracketed.” ²⁴⁹ These divisions, unlike those underlying recognition of LGBT relationships, have increased over time.

In Planned Parenthood v. Casey,²⁵⁰ the Court nonetheless tried to tamp down the divisions by directly engaging the values conflicts. Decided in the early 1990s, Casey²⁵¹ had been widely expected to reverse Roe outright.²⁵² Instead Casey preserved the core of the right to abortion, while permitting the states to impose new restrictions, such as waiting periods and parental consent provisions.²⁵³ Justice O’Connor’s plurality opinion was the only significant abortion decision for the Court written by a woman. She observed that the earlier decisions in Griswold, Eisenstadt, and Carey “support the reasoning in Roe relating to the woman’s liberty because they involve personal decisions concerning not only the meaning of procreation but also human responsibility and respect for it.”²⁵⁴ Casey, alone in the Supreme Court’s reproductive rights decisions, made women’s relationship to the growing fetus central to the decision.²⁵⁵ It succeeded, however, only in delaying the day of reckoning over Roe itself.

Dobbs v. Jackson Women’s Health Organization²⁵⁶ is radically at odds with previous decisions that have used the concept of punishment to distract attention from inflammatory subjects. It is also at odds with the conception of judicial statesmanship, through which courts legitimate the judicial system while also recognizing social change and creating community in the midst of conflicting values’ clashes.²⁵⁷ Although Justice Alito claimed otherwise, the decision is designed to inflame and, in doing so, it is likely to empower state officials who wish to exercise their authority to punish—in order to affirm the moral superiority of their position, to reaffirm their values in the public square, to impose dominance over outgroups, and to restore a sense of hierarchical order that validates their position in society. The opinion itself invites such a response.

First, it goes out of its way to say not just that opposing views, but Roe itself were never legitimate. Alito’s majority opinion declares that “Roe was egregiously wrong from the start. Its reasoning was exceptionally weak, and the decision has had damaging consequences.”²⁵⁸

Second, it dismisses women’s interest in their bodily integrity as of no consequence, suggesting that those interests are amply protected through existing laws.²⁵⁹

Third, while the opinion claims not to base the decision on recognition of a fetus as a human being from the moment of conception forward, it clearly views state actions based on such views as a legitimate basis for legislative action and declares that the fact that abortion serves to “destroy a ‘potential life’ ” justifies the Court’s treatment of Roe as precedent entitled to less deference than other Supreme Court precedents.²⁶⁰

Fourth, unlike other Supreme Court decisions announcing a major change in governing law (with all deliberate speed),²⁶¹ the Court provides no guidance for the states and no timelines for implementation. It simply overturns Roe and leaves the states—and the pregnant—on their own in the face of a rapidly shifting and still uncertain legal landscape.

The majority opinion thus has the hallmarks of an act of vengeance righting a wrong, rather than serving to provide judicial guidance in the face of contentious issues. It seeks to restore the moral hierarchy associated with the forces that see abortion as necessarily impermissible. It affirms states’ right to ban abortion without addressing the impact on the rights of states who wish to ensure its continuing availability. And in not only issuing the Dobbs’ decision, but in failing to restrain the states’ earlier vigilante laws,²⁶² the Court’s current stance suggests that the states will be free to treat abortion as murder and punish those who provide abortions, those who seek abortions, and those who aid and abet those involved with abortions in any way.

V. THE FUTURE OF ABORTION PUNISHMENT

Abortion has become a flash point for political division because it falls on the fault lines of cultural polarization and political realignment.²⁶³ After Dobbs, the factors that drive political divisions are likely to overlap with the factors driving calls to punish those seeking and providing abortions.

In analyzing and moving forward on these issues, it is first critical to understand the sources of the call for punitive measures and then to consider whether a focus on punishment can also provide a strategy for defusing the conflict.²⁶⁴ Without such a strategy, this Article concludes, the likely result is a replication of the conditions that preceded Roe: pregnancy as the punishment for sex, aggravating the existing class and regional bifurcation in unintended births; a high-profile fight between elite actors on the boundaries of post-Dobbs public morality; and selective enforcement that disproportionately penalizes poor and minority women. As an alternative, this Article proposes that using the punishment lens analysis can serve as a means to de-escalate the coming legal wars over abortion.

A. Values Polarization and Abortion Punishment

The analysis of the factors underlying the calls for punishment start with the factors driving political polarization. Political theorists link partisan polarization to a sorting between the parties based on cultural values. They describe those with conservative values orientations as favoring in-group unity and strong leadership, and having “a desire for clear, unbending moral and behavioral codes,” that include an emphasis on the importance of punishing anyone who strays from the code, “a fondness for systematization,” as well as “a willingness to tolerate inequality (opposition to redistributive policies).”²⁶⁵

Those with a liberal values orientation, in contrast, tend to be more tolerant to outsiders, to consider context rather strict rules adherence when it comes to determining appropriate behavior.²⁶⁶ They also demonstrate more empathy and less interest in strict punishment for violations of moral and behavioral rules and greater intolerance of inequality.²⁶⁷

Attitudes toward abortion both reflect and contribute to the partisan polarization.²⁶⁸ Abortion attitudes have become more partisan over time, and psychologist Drew Westen describes this outcome as a matter of intentional political strategy.²⁶⁹ Such a strategy was designed to attract people who see abortion in rigid moral terms to the Republican party in the 1990s,²⁷⁰ and as that happened, self-identified Republicans became more opposed to abortion.²⁷¹ Stances on abortion accordingly became a political marker.

Public opinion polls today confirm the high degree of partisan polarization on abortion. While 61% of all Americans believe that abortion should be legal in all or most cases, 60% of Republicans—and 72% of those who identify as “conservative Republicans”—believe that abortion should be illegal in all or most cases.²⁷² In contrast, 80% of Democrats and 90% of “liberal Democrats” believe that abortions should be legal in all or most cases.²⁷³ Public opinion polls indicate that support for the imposition of criminal sanctions closely tracks abortion views generally.²⁷⁴

These attitudes correspond to the purposes and pitfalls of punishment. All groups seek affirmation of their values, but the values to be expressed are not parallel in their relationship to the imposition of punishment. Abortion rights advocates seek to preserve a right to privacy free from government intrusion through the democratic process, including referenda as well as litigation. To the extent they wish to exact punishment for taking away abortion rights, they have suggested defeating anti-choice politicians at the ballot box,²⁷⁵ impeaching Supreme Court justices for perjury about their willingness to follow precedent,²⁷⁶ and requesting ethics investigations.²⁷⁷ We could also imagine more aggressive efforts to counter the efforts of anti-abortion activists who attempt to interfere with abortion in states where abortion remains legal.²⁷⁸ Some of the most important actions pro-choice states have taken, however, is greater support to assist those coming from out-of-state, protecting their own health care workers, and ensuring access to medication abortion. ²⁷⁹ The symbolism involves a greater and more visible state embrace of a right of abortion access.

The punishment desired by those opposed to abortion, by contrast, has two components. The first involves the expressive function of law and the declaration that abortion is wrong.²⁸⁰ The declaration reaffirms the moral hierarchy that elevates those who oppose abortion entirely; empirical studies indicate that when abortion is perceived as a “moral wrong” that produces outrage in those who oppose it; they dehumanize the women (and their partners) who seek abortions.²⁸¹ Expressing this moral opposition even has a “shaming effect” on those who require abortions because of significant health issues.²⁸² It also justifies subjecting those who seek therapeutic abortions to intrusive review of their doctor’s medical determinations or requiring those experiencing rape or incest to face onerous proof requirements, retraumatizing victims of sexual assault. Yet, the symbolic effect can occur with limited punishment, prosecuting only occasional cases that involve public defiance of the new abortion bans.

This dehumanization and shame, in turn, empowers those who would pursue the second component: waging a war to root out the practice. The National Right to Life Committee has proposed sweeping measures, for example, that would not only criminalize abortion itself, but treat it as a “criminal enterprise” that needs to be eliminated using “RICO-style laws” that would reach anyone providing any type of support to someone seeking an abortion. ²⁸³ These provisions target not only medical personnel but those providing abortion information.²⁸⁴ Others propose empowering not only state prosecutors but individual citizens to conduct surveillance on those visiting out-of-state abortion clinics, accessing internet websites providing abortion information, or even monitoring the pregnant (and their friends and family) more generally.²⁸⁵ These activities, particularly when carried out by private “vigilantes,” combine opposition to abortion with a moral crusade. While some laws immunize the pregnant from prosecution, existing laws in many states have already been used to prosecute women experiencing miscarriages for “feticide”²⁸⁶ and more draconian laws have been proposed that provide for prosecution for crimes based on an abortion.²⁸⁷ Even without new laws, the Attorney General of Alabama, for example, threatened to prosecute those crossing state lines to terminate their pregnancies or using abortion pills as child chemical endangerment, even if the patients legally obtain the pills within Alabama.²⁸⁸

Finally, prosecutions, particularly if they are brought against those who seek abortions, are likely to enforce gender, race and class hierarchies. As anti-abortion fervor has mounted, some states over the last decade have increased criminal investigations of various types of pregnancy loss, including not just self-induced abortions but also miscarriages, stillbirths, and any form of infanticide.²⁸⁹ These cases overwhelmingly target “pregnant people who are poor, young, have substance abuse issues or live in areas with limited health services.”²⁹⁰ Advocates fear the reversal of Roe will fuel more such cases and particularly harm women of color, already disproportionately overpoliced and prosecuted on pregnancy-related issues.²⁹¹ Farah Diaz-Tello, an attorney who works on reproductive health rights commented, “It’s this vicious cycle where lack of access, . . . increased scrutiny and stigma around abortion, as it becomes further restricted or criminalized, leads to more criminalization.”²⁹² And the fact that the individuals are poor, minority group members, substance abusers, or otherwise lack full control of their lives contributes to the willingness of others to impose moral condemnation on their behavior.

Dobbs will only make this worse.

B. Punishment in the Courts

Striking down Roe invited the states to adopt abortion bans that, in criminalizing abortion, also prescribe punishment. The courts have historically policed the limits of criminal punishment, requiring, for example, that criminal laws provide clear notice as to what acts are proscribed, that those accused enjoy appropriate procedural protections, and punishments are proportionate to the offence. This Article has gone beyond these traditional concerns to address how the Supreme Court uses a punishment lens to accomplish broader objectives, particularly in the face of irreconcilable and intrinsically divisive issues, and issues that may threaten judicial legitimacy. Abortion certainly qualifies as divisive, and Dobbs has already raised serious concerns about judicial legitimacy.

Indeed, in the years since Roe, anti-abortion activists have made the fetus the issue—with the impact on the person forced to give birth disappearing from view.²⁹³ When the fetus becomes the subject of concern, consensual sex—with no victims other than public mortality—is beside the point. When prosecutors act to prosecute abortions, they are passing moral judgment on the permissibility of the abortion itself and often imposing significant penalties.²⁹⁴

Two arenas in particular, however, offer the Court an opportunity to tamp down the Dobbs-inspired conflicts.

First, if the Supreme Court seeks to deflect the outrage over Dobbs, the simplest way would be to take seriously its own statement that all it has to do is to return the issue to the states.²⁹⁵ Taking that seriously requires protecting the rights of states that wish to secure access to abortion—and protecting, as Justice Kavanaugh suggested in his concurrence, the constitutional right to travel.²⁹⁶ The most basic question involving the right to travel is whether citizens of one state can travel to another state, return to their home state, and be punished for their out-of-state conduct. Existing precedent from the Roe era suggests that such conduct is constitutionally protected and other limits on state jurisdiction ordinarily preclude punishment for out-of-state acts.²⁹⁷ Affirming the constitutional right to travel should also mean that states cannot burden exercise of the right to travel, by punishing, for example, those within the state who assist the traveler in leaving the state or acts that a pregnant person takes within the home state, such as researching out of state options, packing one’s bags, or driving to the state line for the purpose of accessing abortion in another state, just as the Court concluded in Hobby Lobby that forcing an employee to choose between an ACA compliant health plan or a monetary contribution to ACA funding constitution a burden on religious freedom.²⁹⁸ The Court should also strike down punishment that creates obstacles to First Amendment rights of expression, ²⁹⁹ such as penalizing websites or advice to individuals that contain accurate information about abortion and out-of-state availability.³⁰⁰ The Court could also recognize that states encouraging private citizens to track those accessing out-of-state abortion clinics, websites, menstrual periods or other personal information either serves no legitimate state purpose to the extent it is intended to penalize the right to travel or, like searching the marital bedroom for contraceptives, is so intrusive as to be constitutionally suspect. ³⁰¹ Striking down punishment that burdens the right to travel could simultaneously affirm state abortion bans and still protect its availability in the states that permit it.³⁰²

The second arena where a punishment lens could be effective in defusing abortion controversies involves women’s right to medical treatment to protect their health. Statutes banning abortion pose a dilemma for doctors; they report that they fear retaliation for performing abortion-like procedures—even when the fetus is dead or the health threat to the patient is significant.³⁰³ In these cases, the risks are asymmetrical: the doctor faces punishment for “doing the right thing” and little in the way of negative consequences for not acting, even if the patient dies as a result.³⁰⁴ Uncertainty itself thus imposes punishment—and serves the purposes of those who would root out abortion (with inevitable spillover effects to abortion-like procedures). Yet, criminal prosecutions of the doctor in these cases, while risky and expensive for the doctor personally, could bring the criminal justice system into disrepute. For those seeking to ensure abortion access, the question therefore should be how to bring the issue of punishing doctors—and the corresponding ability of the pregnant to receive abortions necessary to protect their health—into public focus. Test cases on enforceability of abortion bans in circumstances threatening the life of the mother might bring greater clarity.³⁰⁵ Such suits could also focus attention on the health threat that punishment poses to pregnant patients. Heavy-handed interventions into newborn care, in which governors sought to prolong the lives of children born with substantial birth defects, helped to discredit the interventions.³⁰⁶ The same approach might work in the context of pregnancy care. Justice Blackmun’s initial draft opinion in Roe sought to focus on the issue of professional judgment.³⁰⁷ Partisan differences on abortion are smaller (and overall support for punishment is substantially less) when the mother’s health is at risk.³⁰⁸ Striking down abortion laws that do not clearly immunize doctor’s decisions about medically therapeutic abortions is a first step; recognizing that the pregnant have a right to abortions necessary to protect their health is an important second step.

In cases of rape and incest, the effort ought to go further to highlight the callous treatment of such victims. Governor Greg Abbott declared, in response to questions about precluding abortion for the victims of involuntary sexual activity, that “Texas will work tirelessly to eliminate all rapists from the streets of Texas . . . .”³⁰⁹ In short, the Governor tried to deflect claims of punishment of one type (forcing the victims of rape to carry the rapist’s child to term) by talking about another type of punishment—that imposed on rapists. The veracity of the claim is not the issue, particularly because Texas has one of the highest rape rates in the country and Abbott had done little to combat it.³¹⁰ As with abortions necessary to protect the lives of the pregnant, partisan differences narrow considerably on cases of rape and incest and the failure to provide such exceptions underscores the punitive nature of the restrictions. ³¹¹

Finally, cases in which patients are prosecuted ought to be used to highlight the cruelty associated with abortion restrictions in the United States.³¹² Restricting access to abortion is in fact just one more form of punishment of the marginalized, with the same groups that support abortion restrictions also opposing more generous provisions to the poor.³¹³ White evangelical Protestants, for example, the religious group most opposed to abortion,³¹⁴ is also one of the groups most likely to respond that aid to the poor does more harm than good.³¹⁵ And the same groups have become more likely to oppose immigration and efforts to promote racial equality³¹⁶ and to favor imposition of preferred values through authoritarian means.³¹⁷ The cruelty of abortion bans is a large part of what motivated the decision in Roe. With abortion opponents calling for draconian enforcement measures,³¹⁸ it should be a factor in mobilizing the opposition to post-Dobbs enforcement of abortion restrictions.³¹⁹

CONCLUSION

Focusing on punishment will not resolve intractable values disputes; it simply changes the subject. Changing the subject, however, does offer a tactic for diffusing intractable disputes—or a long-term strategy for reframing what is at stake. In either case, it makes visible the consequences of public actions, such as abortion bans, on those affected by them in ways that can serve to underscore their cruelty. The public wants its core values expressed and respected in the public square; in cohesive societies the values are consensus based, and punishment reinforces them. The urge to punish, when embedded in group conflict, inflames divisions (threatening violence or civil war); channeling it effectively is central to the rule of law. Understanding this dynamic gives the Court tools (and a motive) to construct an offramp: it also allows states to decide their own approaches to abortion while protecting the pathways out of the states that ban it, and ensures that doctors can save the lives of their patients.

96 S. Cal. L. Rev. 1101

Download

June Carbone*

* Robina Chair in Law, Science and Technology, University of Minnesota Law School.

Naomi Cahn†

† Justice Anthony M. Kennedy Distinguished Professor of Law, Nancy L. Buc ’69 Research Professor in Democracy and Equity, University of Virginia School of Law. Thanks to workshop participants at the University of Minnesota Law School Squaretable for comments and to Sam Turco for research assistance, and to Katherine Bake, Mary Anne Case and John Q. Barrett for comments on an early draft.