Because sex does not dictate the capacity to provide care in the home or work in the market, sex-equality law combats harmful sex ste­reotypes by eliminating statutes and regulations that assign these roles on the basis of sex. When it comes to pregnancy, though, courts and commentators alike chart a very different course. They assume that pregnancy is a biological event that is almost exclusively for women. Thus, equal protection jurisprudence accepts the legal assignment of carework during pregnancy to women, and a range of laws regulating pregnancy carework—from prenatal leave under the Family and Medical Leave Act to health benefits under the Affordable Care Act to employment protections under the Pregnancy Discrimination Act—apply only or mostly to women. Even though the sexed law of preg­nancy stands in stark contrast to the unsexed law of parenting, the sexed preg­nancy has avoided challenge and largely escaped notice.

This Article makes visible the law of the sexed pregnancy, identifies and evaluates the core tension it generates in the law of sex equality, and considers how to unravel this tension. Of course, typically only women can physically carry a child, and therefore some pregnancy reg­ulations are appropriately sex specific. But the nine months of preg­nancy encompass a range of carework, much of which has little or nothing to do with the physical fact of pregnancy. Expectant fathers can, for example, buy a carseat, quit smoking, take a childcare class, and choose a pediatrician or daycare center for the child. Given the ability to disaggregate sex from much of the carework of pregnancy, the law’s failure to do so marks women for caregiving and men for breadwinning in the same problematic way that sex-equality law has tried to combat after a child is born. And while pregnancy implicates real concerns about a woman’s constitutional right to bodily autonomy, this concern alone cannot justify the failure to scrutinize all sex-based pregnancy regulations, because much prebirth carework does not involve the woman’s body at all.  After surfacing the law’s anomalous sexed treatment of pregnancy, this Article considers how to harmonize the law of sex equality.  This effort can advance not only the goal of equality between the sexes, but also equality for lesbian, gay, and transgender parents, while at the same time enhancing women’s autonomy.


    1. Why We Unsex Parenting
    2. How We Unsex Parenting
    1. Why We Sex Pregnancy
    2. How We Sex Pregnancy
      1. Constitutional Law
      2. Statutes and Regulations
    1. Pregnancy as Parenting
    2. Pregnancy and Autonomy
    1. Scrutinizing Pregnancy
      1. Constructing Scrutiny
      2. Deploying Scrutiny
    2. Protecting Autonomy Rights



The hallmark of sex-equality law is the aggressive policing of laws that classify individuals on the basis of sex and that are grounded in mere sex stereotypes. Nowhere have these unsexing efforts been more substan­tial than in the context of the carework involved in parenting. Because sex rarely if ever dictates one’s ability to parent, carework can be and has been gradually “disaggregated” from sex. 1 We borrow this language from Professor Mary Anne Case’s seminal article on the law of sex equality. See Mary Anne C. Case, Disaggregating Gender from Sex and Sexual Orientation: The Effeminate Man in the Law and Feminist Jurisprudence, 105 Yale L.J. 1, 2 (1995) [hereinafter Case, Disaggregating]. Unsexing the law of caregiving has been crucial to dismantling the separate sex-based spheres that pigeonhole women as caregivers and men as breadwinners and that have been so harmful to sex equality. Equal protection doctrine therefore closely scrutinizes sex classifications related to parenting carework, 2 The list of cases engaging in such efforts is long, and these cases are discussed in greater detail in section I.A. These efforts have been so persuasive that in 2017 a unanimous Supreme Court (with Justice Gorsuch recused) invalidated a statute in part because of its assumptions that women are caregivers and men are not. See Sessions v. Morales-Santana, 137 S. Ct. 1678, 1686 (2017) (striking down a sex-differentiated citizenship rule). and  laws  like  Title  VII  and  the  Family  and  Medical  Leave  Act  (FMLA)  take  aim  at  stereotypes  of  sex-based  caregiving  roles. 3    See infra notes 88–92. Yet despite this progress, one significant sphere of caregiving has remained immune to these efforts: pregnancy.

Sex-equality law’s pervasive efforts to disaggregate sex from caregiv­ing after birth are in stark contrast to its failure to do so before birth. While typically only women can bear children, 4 There is an exception to this deserving greater attention: the still rare but significant reality of the pregnant man. See Mathilde Cohen, The Lactating Man, in Making Milk: The Past, Present, and Future of Our Primary Food 141, 152–53 (Mathilde Cohen & Yoriko Otomo eds., 2017); Carma Hassan & Dakin Andone, ‘My Body Is Awesome’: Trans Man Expecting First Child, CNN, http://www.cnn.com/2017/06/08/health/trans-man-pregnant-trnd/index.html [https://perma.cc/G3NT-SU4X] (last updated June 8, 2017); infra notes 178–181 and accompanying text. an emerging consensus across a variety of scholarly fields recognizes the nine months of preg­nancy as much more than a physical fact. 5 See generally Anna R. Brandon et al., A History of the Theory of Prenatal Attachment, 23 J. Prenatal & Perinatal Psychol. & Health 201, 202–11 (2009) (discussing the relevant literature on prenatal attachment from both maternal and paternal perspectives). Earlier studies focused only on how behaviors during the pregnancy shaped later carework for women. See, e.g., Reva Rubin, Maternal Tasks in Pregnancy, 1 J. Advanced Nursing 367, 367–69 (1976) (focusing on how female caregiving affects post-delivery mothering). More recently, scholars have focused on how behaviors by both women and men during the pregnancy shape later caregiving by both. See John T. Condon, The Parental–Foetal Relationship—a Comparison of Male and Female Expectant Parents, 4 J. Psychosomatic Obstetrics & Gynecology 271, 280–83 (1985) (summarizing the relevant mechanisms and empirical findings). Rather, pregnancy involves a wide range of carework—such as quitting smoking, taking a childcare class, and choosing a pediatrician—that has more in common with child­rearing than childbearing.

Despite this, the Supreme Court has decided that pregnancy is an event almost exclusively for women and has therefore assumed that caregiving during pregnancy is almost exclusively for women too. 6 See infra notes 129–134 and accompanying text. Perhaps the most widely known—and reviled—of the many cases conflating pregnancy with physical gestation is Geduldig v. Aiello, in which the Court denied heightened scrutiny to a pregnancy classification because “pregnancy is an objectively identifiable physical condition with unique characteristics.” 417 U.S. 484, 496 n.20 (1974); see also Nguyen v. INS, 533 U.S. 53, 65 (2001) (deciding that mothers are more likely “to develop a real, meaningful relationship” with a newborn because of the “event of birth”); Lehr v. Robertson, 463 U.S. 248, 260–62 & n.16 (1983) (deciding that since “[t]he mother carries and bears the child[,] . . . her parental relationship is clear,” while the father must “grasp[ ] [the] opportunity” to become a parent through postbirth actions (internal quotation marks omitted) (quoting Caban v. Mohammed, 441 U.S. 380, 397 (1979) (Stewart, J., dissenting))); cf. Caban, 441 U.S. at 389 (“Even if unwed mothers as a class were closer than unwed fathers to their newborn infants, this generalization . . . would become less acceptable as a basis for legislative distinctions as the age of the child increased.”). The result is a wide swath of laws regulating pregnancy—including prenatal leave under the FMLA, essential health benefits under the Affordable Care Act, and employment protections under the Pregnancy Discrimination Act—that apply only or mostly to women. As  one  of  many  notable  exam­ples,   the  FMLA  provides  sex-neutral  leave  for  postbirth  caregiving  in  order  to  combat  sex  stereotypes, 7 See Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 736 (2003) (noting that Congress enacted the FMLA to address the problem that employers “often denied men similar accommodations [to women] or discouraged them from taking leave” because they “regard[ed] the family as the woman’s domain”). but it provides leave for prenatal care­giving only to women. 8 See infra section II.B.2.a (discussing the relevant statutory language and resulting regulations).

Scholars have likewise argued for pregnancy as a woman’s domain. 9 See, e.g., Katharine T. Bartlett, Pregnancy and the Constitution: The Uniqueness Trap, 62 Calif. L. Rev. 1532, 1532 (1974) (“Only women may experience pregnancy . . . .”); Lucinda M. Finley, Transcending Equality Theory: A Way Out of the Maternity and the Workplace Debate, 86 Colum. L. Rev. 1118, 1139 (1986) (“[P]regnancy is unique—[in] that it affects only women . . . .”); Sylvia A. Law, Rethinking Sex and the Constitution, 132 U. Pa. L. Rev. 955, 1007 (1984) (“[P]regnancy, abortion, reproduction, and creation of another human being are special—very special. Women have these experiences. Men do not.” (emphasis omitted)); Reva B. Siegel, Note, Employment Equality Under the Pregnancy Discrimination Act of 1978, 94 Yale L.J. 929, 942 (1985) [hereinafter Siegel, Employment Equality] (stating that pregnancy is a “biological difference central to the definition of gender roles” affecting women). While  pregnancy has  been  central  to  debates  over  sex-equality  law  for  decades,   all  sides  assume  that  pregnancy  represents  a  biological  sex  difference,   and scholars disagree only on whether women should be afforded special treatment given this physical difference. 10 Compare Herma Hill Kay, Equality and Difference: The Case of Pregnancy, 1 Berkeley Women’s L.J. 1, 1–2 (1985) (advocating for the special-treatment approach and noting that pregnancy is “a condition that only women experience”), with Wendy W. Williams, Equality’s Riddle: Pregnancy and the Equal Treatment/Special Treatment Debate, 13 N.Y.U. Rev. L. & Soc. Change 325, 327–28 (1984–1985) (advocating for the equal-treatment approach and noting that pregnancy is a “characteristic[ ] unique to one sex”). Questioning the sexed preg­nancy—as this Article does—challenges the very premise of these debates.

This Article renders the law of the sexed pregnancy visible, surfaces a central tension in the law of sex equality that it generates, and consid­ers how to unravel this doctrinal tension. Disaggregating sex from care­work at the beginning is important because sex-based caregiving stereotypes—and the sex-discriminatory laws that enforce them—are at the root of so much sex inequality. 11 See, e.g., Hibbs, 538 U.S. at 738 (stating that “the faultline between work and family” is “precisely where sex-based overgeneralization has been and remains strongest”); Ruth Bader Ginsburg, Speaking in a Judicial Voice, 67 N.Y.U. L. Rev. 1185, 1204 n.124 (1992) (explaining the argument that sex classifications “enshrining and promoting the woman’s ‘natural’ role as selfless homemaker, and correspondingly emphasizing the man’s role as provider, . . . impede[ ] both men and women from pursuit of the opportunities . . . that could enable them to break away from familiar stereotypes”). Dismantling these sex stereotypes after birth is too little because it is too late. Sticky behaviors marking women as caregivers and men as providers emerge during the pregnancy and are difficult to reverse after birth. We will never fully unsex parenting as long as pregnancy is sexed.

The sexed pregnancy is a key roadblock in the path not only to equality between men and women but also to equality for gay, lesbian, and transgender expectant parents. A legal system that enforces sex-based caregiving roles can exclude gay, lesbian, and transgender expectant parents from its protections simply because of their sex and also reinforce sex stereotypes that are especially difficult for nontradi­tional families. This Article emphasizes the interlocking sex stereotypes of women’s and men’s respective roles in the family and at work that fuel gendered distributions of caregiving. But, in doing so, the Article also highlights the damaging consequences of the sexed pregnancy for other family configurations. For this reason, when relevant, the Article refers to the nonpregnant expectant parent in sex-neutral terms, rather than to the expectant father.

Constitutional concerns related to women’s bodily autonomy that arise uniquely during pregnancy are critical considerations, but they can­not justify the law’s wholly distinct treatment of pre- and postbirth care­work. Precedents like Planned Parenthood of Southeastern Pennsylvania v. Casey do recognize a problem of constitutional magnitude with paternal involvement in pregnancy when it amounts to undue intrusion on the mother’s body. 12 See 505 U.S. 833, 898 (1992) (invalidating a spousal notification requirement because it gave the husband an “enforceable right to require a wife to advise him before she exercises her personal choices”). But the very circumstance that permits sex to be disaggregated from carework during pregnancy—that much carework during pregnancy is not tied to the physical fact of gestation—also per­mits paternal involvement in pregnancy in ways that do not involve the mother’s body and thus avoid triggering the important red flags that Casey raised. Once we recognize that fathers can do prebirth carework without involving the mother’s body, the objection from reproductive rights dissipates—although admittedly it does not disappear.

The roots of unraveling this key tension in the law of sex equality lie in the very cases that have constructed this tension in the first place. Sev­eral Supreme Court cases that serve as the foundation of the sexed preg­nancy need not be overruled but simply revisited. The Supreme Court, in its landmark Geduldig v. Aiello decision, determined that the pregnancy classification in that case was not sex discrimination subject to height­ened scrutiny under the Fourteenth Amendment. 13 417 U.S. 484, 494–97, 496 n.20 (1974). The Court also sug­gested, however, that under the right circumstances, pregnancy dis­crimination could be the type of sex discrimination that the Equal Protection Clause scrutinizes and rejects. 14 See id. at 496 n.20 (“Absent a showing that distinctions involving pregnancy are mere pretexts designed to effect an invidious discrimination against the members of one sex or the other, lawmakers are constitutionally free to include or exclude pregnancy from the coverage of legislation . . . .”); see also Gen. Elec. Co. v. Gilbert, 429 U.S. 125, 135–36 (1976) (stating in an analogous Title VII context that a “distinction which on its face is not sex related might nonetheless violate the Equal Protection Clause if it were in fact a subterfuge to accomplish a forbidden discrimination”); id. at 149 (Brennan, J., dissenting) (“Geduldig’s outcome was qualified by the explicit reservation of a case where it could be demonstrated that a pregnancy-centered differentiation is used as a ‘mere pretext . . . designed to effect an invidious discrimination against the members of one sex . . . .’” (alterations in original) (quoting Geduldig, 417 U.S. at 496 n.20)). This Article argues that many of the sexed pregnancy regulations identified here are precisely the type of pregnancy regulations that warrant heightened scrutiny and that could wither under its exacting gaze.

Although courts may strike down sex-based pregnancy regulations, this Article considers whether the remedy should be to “level up” by extending some existing laws to nonpregnant expectant parents when physical sex differences do not justify their exclusion. 15 See, e.g., Sessions v. Morales-Santana, 137 S. Ct. 1678, 1698 (2017) (“[W]hen a statute benefits one class . . . and excludes another[,] . . . ‘[a] court may either . . . order that its benefits not extend to the class that the legislature intended to benefit, or it may extend the coverage of the statute . . . .’” (quoting Califano v. Westcott, 443 U.S. 76, 89 (1979))). Many pregnancy protections and benefits can be extended to and utilized by those other than the pregnant woman and without meaningfully involving her. 16 See infra section II.B.2. Extending these protec­tions and benefits not only bolsters sex equality but can also enhance—rather than infringe on—women’s constitutionally guaranteed autonomy. As feminist scholars have articulated, autonomy for caregivers can come from support rather than separateness—and this is precisely what extend­ing pregnancy protections and benefits can help to achieve. 17 See infra notes 340–342 and accompanying text.

This Article proceeds in four parts. Parts I and II demonstrate the distinct ways that sex-equality law treats carework after birth as compared with carework before birth. Part III establishes that this is a distinction without a meaningful difference. Sex-based regulations of carework in the prebirth period, like those in the postbirth period, generate and rein­force gendered caregiving distributions that are damaging to the law’s sex-equality goals. While pregnancy raises autonomy concerns that are absent in the postbirth period, these alone do not justify the unique treatment of prebirth caregiving, given that many pregnancy regulations do not implicate autonomy concerns. Part IV considers ways to alleviate this key tension in the law of sex equality by showing how this tension is not doctrinally inevitable. Supreme Court precedents can be revisited and reread to relieve the tension, and Part IV fills in the details about how courts could do so.

I. Unsexing Parenting

Sex-equality law in the United States has focused on what this Article calls unsexing. The idea is a simple one: If it is not necessary that sex determine one’s capacity because of a real difference between the sexes, then sex should not be made to determine one’s capacity by the force of law. 18 See United States v. Virginia, 518 U.S. 515, 533 (1996) (deciding that physical sex differences may still justify sex classifications under the Equal Protection Clause); Geduldig, 417 U.S. at 496 n.20 (1974) (noting that physical differences between the sexes can be the basis for permissible classifications that are effectively sex based). In such circumstances, sex can and should be “disaggregated” from capacity. 19 See Case, Disaggregating, supra note 1, at 2 (arguing that sex can be disaggregated from the performance of certain gender roles). Otherwise, the law creates a self-reinforcing set of sex roles based on stereotypes, and sex will unnecessarily and unfairly limit one’s station in life. 20 See Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 736 (2003) (“These mutu­ally reinforcing stereotypes created a self-fulfilling cycle of discrimination that forced women to continue to assume the role of primary family caregiver, and fostered employers’ stereotypical views about women’s commitment to work and their value as employees.”).

One of the most important realms the law has unsexed is carework within the family. Prescribing sex roles in the family generates sex inequality that extends far beyond it, especially into the workplace. 21 See, e.g., id. at 738 (“[T]he faultline between work and family[ ] [is] precisely where sex-based overgeneralization has been and remains strongest . . . .”). The Supreme Court has recognized the important role the law has played in allocating “family duties” on the basis of sex. 22 Id. at 730. Applying heightened scru­tiny under the Fourteenth Amendment, the Supreme Court has held both (1) that there is no necessary link between the ability to perform these carework duties and the sex of the parent, and (2) that laws relying on such a link reinforce sex stereotypes that limit both women’s roles at work and men’s roles in the family. 23 See infra note 73 (collecting cases). Unsexing carework has entailed not only removing sex-based barriers created by legislation through the applica­tion of heightened scrutiny but also enacting legislation to create a world where these sex-based limits would no longer apply. 24 See infra notes 88–93 and accompanying text.

This Part begins by discussing why we unsex carework (that is, the theory of unsexing carework) and then turns to explaining how we unsex postbirth carework (that is, the doctrine of unsexing carework).

A. Why We Unsex Parenting

The law did not always unsex parenting. For most of American his­tory, the life and law of the sexes were of separate spheres. 25 See Ruth Bader Ginsburg, Remarks on Women Becoming Part of the Constitution, 6 Law & Ineq. 17, 19 (1988) [hereinafter Ginsburg, Remarks]; see also Sessions v. Morales-Santana, 137 S. Ct. 1678, 1683 (2017) (referencing “an era when the Nation’s lawbooks were rife with overbroad generalizations about the way men and women are”). Carework was presumed to be both aggregated with and distributed by sex. Under this view, as Justice Ginsburg has explained, “It was man’s lot, because of his nature, to be breadwinner, head of household, representative of the family outside the home; and it was woman’s lot, because of her nature, not only to bear, but also to raise children, and keep the home in order.” 26 Ginsburg, Remarks, supra note 25, at 19. This breadwinner–homemaker dichotomy served as the founda­tion for many Supreme Court decisions upholding laws that distinguished roles—especially work and family roles—on the basis of sex. 27 See, e.g., Hoyt v. Florida, 368 U.S. 57, 58–62 (1961) (upholding a state law providing that women would not be called for jury service unless they specifically volunteered, because women’s place was at “the center of home and family life”); Goesaert v. Cleary, 335 U.S. 464, 465–67 (1948) (upholding a state law providing that no women, except wives and daughters of male tavern owners, could tend bar); Muller v. Oregon, 208 U.S. 412, 416–17, 423 (1908) (upholding a state law limiting the hours women could work); cf. Bradwell v. Illinois, 83 U.S. (16 Wall.) 130, 141 (1873) (Bradley, J., concurring in the judgment) (“The constitution of the family organization, which is founded in the divine ordinance, as well as in the nature of things, indicates the domestic sphere as that which properly belongs to the domain and functions of womanhood.”). Because such laws were subject only to rational basis review, there was no need to show that men and women were actually different to justify treating them differently. The Court upheld such laws on the basis of “opinion” 28 Goesaert, 335 U.S. at 466–67 (quoting Tigner v. Texas, 310 U.S. 141, 147 (1940)) (“Since the line [Michigan legislators] have drawn is not without a basis in reason, we cannot give ear to the suggestion that the real impulse behind this legislation was an unchivalrous desire of male bartenders to try to monopolize the calling.”). or “widespread belief” 29 Muller, 208 U.S. at 420 (opining that the “widespread belief that woman’s physical structure, and the functions she performs in consequence thereof,” justifies legislation limiting women’s work hours). The Court did extensively discuss the evidence of sex differences put forward in the Brandeis brief—his first—to distinguish the case from Lochner, which struck down a similar restriction that applied equally to men and women. See id. at 419–23; Lochner v. New York, 198 U.S. 45, 52, 64–65 (1905). See generally Melvin I. Urofsky, Louis D. Brandeis: A Life 216–19 (2009) (describing the contents of, and the Court’s treatment of, Brandeis’s Muller brief). that there is “a wide difference in the respective spheres and destinies of man and woman.” 30 Bradwell, 83 U.S. (16 Wall.) at 141 (Bradley, J., concurring in the judgment) (“Man is, or should be, woman’s protector and defender. The natural and proper timidity and delicacy which belongs to the female sex evidently unfits it for many of the occupations of civil life.”).

The Court’s rejection of separate spheres through the application of heightened scrutiny was founded on the core equality principle of the Equal Protection Clause: Treat likes alike. 31 See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439–42 (1985) (recognizing that the Fourteenth Amendment requires that “all persons similarly situated should be treated alike”). The big leap forward was appreciating that men and women were more alike than had previously been thought when it comes to the carework involved in parenting. 32 See Ginsburg, Remarks, supra note 25, at 21 (noting how women’s increasing presence in the workforce in the 1970s made it more difficult for judges to accept the traditional breadwinner and homemaker stereotypes). This progress turned on the judicial recognition of two related princi­ples: (1) that much of the work of parenting was not biologically or otherwise necessarily sexed, and thus (2) that the legal assignment of sex roles in the family was harmful to the cause of sex equality. 33 See id. at 20 (“To turn in a new direction [of sex-equality law], the Court first had to comprehend that legislation apparently designed to benefit or protect women could often, perversely, have the opposite effect.”).

As  for  the  first  principle,   constitutional  law  has  long  recognized  all  of  the  work  that  goes  into  parenting. Domestic  responsibilities  were  seen  to  include  two  aspects — “home  [life]”  and  “family life” 34 Hoyt v. Florida, 368 U.S. 57, 62 (1961); see also Muller, 208 U.S. at 419–23 & n.1 (upholding a state law limiting the hours women could work and citing as justification, among other reasons, “the rearing . . . of the children,” and “the maintenance of the home” (quoting Brief for Defendant in Error at 97, Muller, 208 U.S. 412 (No. 107), 1908 WL 27605)). —with “women’s place at ‘the center’” of both. 35 Ginsburg, Remarks, supra note 25, at 19 (quoting Hoyt, 368 U.S. at 62) (explaining the dual aspects of women’s domestic responsibilities: “rais[ing] children[ ] and keep[ing] the home in order”). The Court has recognized this postbirth care­­­work to encompass a range of responsibilities, 36 Other family caregiving, especially for elderly parents, has been recognized as substantial and substantially distributed by sex, although this caregiving falls outside the scope of this Article. See H.R. Rep. No. 103-8, pt. 1, at 24 (1993) (recognizing within the legislative history of the Family and Medical Leave Act that “[t]wo-thirds of the nonprofessional caregivers for older, chronically ill, or disabled persons are working women”). from basic physical main­tenance and supervision, 37 See Weinberger v. Wiesenfeld, 420 U.S. 636, 652 (1975) (discussing the importance of “care” and “custody” of children (quoting Stanley v. Illinois, 405 U.S. 645, 651 (1972))); cf. 29 U.S.C. § 2612(a)(1)(A) (2012) (recognizing the need for leave from work “[b]ecause of the birth of a son or daughter of the employee and in order to care for such son or daughter”). to moral education, 38 See Muller, 208 U.S. at 419 n.1 (including within women’s family responsibilities “education of the children” (quoting Brief for Defendant in Error at 97, Muller, 208 U.S. 412 (No. 107), 1908 WL 27605)). to emo­tional bonding with the child. 39 See Weinberger, 420 U.S. at 652 (discussing the “companionship” of children (internal quotation marks omitted) (quoting Stanley, 405 U.S. at 651)); id. at 655 (Rehnquist, J., concurring in the result) (highlighting the importance of “the personal care and attention” of a parent); cf. 29 C.F.R. § 825.120(2) (2018) (“Both parents are entitled to FMLA leave to be with the healthy newborn child (i.e., bonding time) during the 12-month period beginning on the date of birth.” (emphases added)). In addition, the Court has suggested a mana­gerial role associated with carework, which includes its administra­tive elements, such as planning meals, making appointments, and keep­ing the family calendar. 40 See Weinberger, 420 U.S. at 652 (referencing the “management” aspects of parenting (internal quotation marks omitted) (quoting Stanley, 405 U.S. at 651)); Hoyt, 368 U.S. at 62 (proclaiming women to be the “center of home and family life” and suggesting that they are responsible for running the domestic sphere); Bradwell v. Illinois, 83 U.S. (16 Wall.) 130, 141 (1872) (Bradley, J., concurring in the judgment) (referencing the “noble and benign offices of wife and mother” and suggesting the administrative aspects of these roles (emphasis added)). See generally Elizabeth F. Emens, Admin, 103 Geo. L.J. 1409, 1412–17 (2015) (describing the phenomenon of “admin,” which is the “the office-type work that people do to manage their lives,” and its gendered distribution).

All of this carework that went into maintaining “the domestic sphere” was thought to “properly belong[ ] to the domain and functions of womanhood.” 41 Bradwell, 83 U.S. (16 Wall.) at 141 (Bradley, J., concurring in the judgment). There was the additional implication that men could work while fulfilling their family responsibilities, whereas women could not. See id. (“The harmony, not to say identity, of interests and views which belong, or should belong, to the family institution is repugnant to the idea of a woman adopting a distinct and independent career from that of her husband.”). The Court only later came to appreciate that men and women could do this work equally. Weinberger v. Wiesenfeld, described by Justice Ginsburg as the “most critical” of the sex discrimination cases decided by the Court in the 1970s, 42 Ruth Bader Ginsburg, Interpretations of the Equal Protection Clause, 9 Harv. J.L. & Pub. Pol’y 41, 43 (1986). made clear that carework should be unsexed because there was no reason that only women could fulfill the domestic tasks just described. 43 See 420 U.S. at 652. Weinberger involved a provision of the Social Security Act that granted survivors’ benefits upon the death of a husband to his minor children and his surviving wife, but granted the same benefits upon the death of a wife only to her minor children and not her surviving husband. 44 Id. at 637–38.

In a case decided a few years before Weinberger, the Court had explained the need for heightened scrutiny of sex classifications by disaggregating sex from social role: “[T]he sex characteristic frequently bears no relation to ability to perform or contribute to society.” 45 Frontiero v. Richardson, 411 U.S. 677, 686 (1973). In a unanimous decision in Weinberger, the Court applied this reasoning to carework. 46 See Weinberger, 420 U.S. at 651–53. Justice Douglas did not participate. See id. at 653. The purpose of the rule under question was “to permit women to elect not to work and to devote themselves to the care of children” following the death of their husbands. 47 Id. at 648. The Court recognized that men and women were simi­larly situated in their ability to care for their children and that the longstanding presumption otherwise was not based on any necessity. 48 The decision also acknowledged that men who assumed caregiving responsibilities would face the same challenges in balancing work and family as women. See id. at 652 (“[T]o the extent that women who work when they have sole responsibility for children encounter special problems, it would seem that men with sole responsibility for children will encounter the same child-care related problems.”). Notably, it did so in the context of a mother who had died in childbirth, 49 Id. at 639. therefore recognizing that from the moment of birth fathers are just as capable parents as mothers.

As  for  the  second  principle,    constitutional  law  came  to  recognize  that  assigning  caregiving  roles  based  on  sex  is  not  only  unnecessary  but  actually  harmful.  The  harms of  treating  similarly  situated  mothers  and  fathers differently, as the Social Security Act provision discussed in Weinberger did, flow from “a much broader pattern of sex-role enforcement that associate[s] men with the marketplace and women with the home.” 50 Cary Franklin, The Anti-Stereotyping Principle in Constitutional Sex Discrimination Law, 85 N.Y.U. L. Rev. 83, 124 (2010); see also Frontiero, 411 U.S. at 684 (noting that the “practical effect” of laws based on sex stereotypes was to “put women, not on a pedestal, but in a cage”). Legislating on the basis of sex stereotypes is so problematic because it renders these stereotypes self-reinforcing, and nowhere is this truer than in the family. As Justice Ginsburg has explained, sex classifications “enshrining and promoting the woman’s ‘natural’ role as homemaker, and correspondingly empha­sizing the man’s role as pro­vider, . . . impeded both men and women from pursuit of the very opportunities that would have enabled them to break away from familiar stereotypes.” 51 Ginsburg, Remarks, supra note 25, at 21; see also Califano v. Westcott, 443 U.S. 76, 89 (1979) (stating that sex-based classifications reflect and reinforce stereotypes that women’s role is at “the center of home and family life” (internal quotation marks omitted) (quoting Taylor v. Louisiana, 419 U.S. 522, 534 n.15 (1975))).

In  Nevada  Department  of  Human  Resources  v.  Hibbs,   for example,   the  Court  identified  many  laws  featuring  sex-based  classifications  premised  on  the  breadwinner–homemaker  dichotomy  that  had  pushed  women  away  from  work  and into  the  home. 52 See 538 U.S. 721, 729 (2003). State laws, for instance, had prohib­ited women from working as lawyers 53 See Bradwell v. Illinois, 83 U.S. (16 Wall.) 130, 131 (1873). or bartenders, 54 See Goesaert v. Cleary, 335 U.S. 464, 465 (1948). or from working more than a certain number of hours per week. 55 See Muller v. Oregon, 208 U.S. 412, 416–17 (1908). By barring women from entering certain professions or limiting the terms on which they could work, these laws not only reflected and reinforced stereotypes that women’s role was at home but also made it far more likely that this was so. Therefore, “the primary responsibility for family caretaking often falls on women, and such responsibility affects the working lives of women more than it affects the working lives of men.” 56 Hibbs, 538 U.S. at 728 n.2 (internal quotation marks omitted) (quoting 29 U.S.C. § 2601(a)(5) (2000)). In fact, Congress has traced women’s limited work opportunities directly to “the pervasive presumption that women are mothers first, and workers second,” which has led to “discrimination against women when they are mothers or mothers-to-be” 57 Id. at 736 (internal quotation marks omitted) (quoting The Parental and Medical Leave Act of 1986: Joint Hearing on H.R. 4300 Before the Subcomm. on Labor–Mgmt. Relations and the Subcomm. on Labor Standards of the H. Comm. on Educ. & Labor, 99th Cong. 100 (1986) (statement of Women’s Legal Defense Fund)); see also Frontiero v. Richardson, 411 U.S. 677, 686–87 (1973) (“[S]tatutory distinctions between the sexes often have the effect of invidiously relegating the entire class of females to inferior legal status without regard to the actual capabilities of its individual members.”). and left women with lower pay, less job security, shorter job tenure, and fewer accumulated benefits. 58 See Califano v. Goldfarb, 430 U.S. 199, 212 (1977).

Circumscribing men’s roles through overbroad sex classifications also causes harm. 59 As Professor Cary Franklin has documented, this recognition was harder to achieve. See Franklin, supra note 50, at 106–14 (describing resistance to the recognition that sex stereotypes held men back as well). By “presuming a lack of domestic responsibilities for men,” these types of sex classifica­tions not only pigeonhole women into domestic roles but also push men away from those roles and into the workplace. 60 See Hibbs, 538 U.S. at 736 (“Because employers continued to regard the family as the woman’s domain, they often denied men similar accommodations or discouraged them from taking leave.”). As the Supreme Court recently stated in Sessions v. Morales-Santana, “[S]uch laws may disserve men who exercise responsibility for raising their children.” 61 137 S. Ct. 1678, 1693 (2017). Stereotyp­ing men as breadwinners harms women, too, by pushing them still fur­ther into family roles and away from work 62 See Hibbs, 538 U.S. at 736 (recognizing that employers’ denial of caregiving leave to men “forced women to continue to assume the role of primary family caregiver”). and by making sure that these family roles remain undervalued. 63 See Case, Disaggregating, supra note 1, at 3 (“So long as stereotypically feminine behavior, from wearing dresses and jewelry to speaking softly or in a high-pitched voice, to nurturing or raising children, is forced into a female ghetto, it may continue to be devalued.”).

Finally, overbroad sex classifications premised on separate spheres “creat[e] a self-fulfilling cycle of discrimination” 64 Morales-Santana, 137 S. Ct. at 1693 (alteration in original) (internal quotation marks omitted) (quoting Hibbs, 538 U.S. at 736). not only by their impact on the behavior of women and men but also by their impact on third parties. For example, employment statutes relying on overbroad sex classifications can influence employer attitudes and behavior. Congress enacted the FMLA because laws that provided family leave only to women reflected and reinforced “employers’ stereotypical views about women’s commitment to work and their value as employees” 65 Hibbs, 538 U.S. at 736. and gener­ated “serious potential for encouraging employers to discriminate against [female] employees and applicants.” 66 29 U.S.C. § 2601(a)(6) (2012).

B. How We Unsex Parenting

Once the Court identified why equal protection requires dis­aggregating sex from carework, it started down the path to achieve this goal. The black letter law of unsexing is intermediate scrutiny, meaning that when a law classifies on the basis of sex, the government must show that it is attempting to further an important government interest by means that are substantially related to that interest. 67 See Craig v. Boren, 429 U.S. 190, 197 (1976). Intermediate scrutiny has sometimes been stated as requiring an “exceedingly persuasive” justification for the classification. United States v. Virginia, 518 U.S. 515, 533 (1996) (internal quotation marks omitted) (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 724 (1982)). Scholars agree, though, that when it comes to sex discrimination cases, the ball game is about stereotypes, not scrutiny. 68 See Mary Anne Case, “The Very Stereotype the Law Condemns”: Constitutional Sex Discrimination Law as a Quest for Perfect Proxies, 85 Cornell L. Rev. 1447, 1449 (2000) [hereinafter Case, Very Stereotype] (arguing that “the components of the intermediate scrutiny standard . . . have rarely been the moving parts in a Supreme Court sex discrimination decision” and that “the bulk of the work in these decisions . . . [is] the proposition that there are constitutional objections to ‘gross, stereotyped distinctions between the sexes’” (footnote omitted) (quoting Frontiero v. Richardson, 411 U.S. 677, 685 (1973))); Franklin, supra note 50, at 138 n.296 (explaining that “[t]he anti-stereotyping principle . . . shap[es] what constitutes an important interest and what means qualify as sufficiently narrowly tailored to serve this interest,” and noting that since the doctrine was first introduced, “the Court has never upheld a sex classification after determining that it reflects or reinforces sex stereotypes”). In 2017, the Supreme Court explained without dissent that “[o]verbroad generalizations” about the sexes are problematic, even if these generalizations might be true in “many” situa­tions. 69 Morales-Santana, 137 S. Ct. at 1692–93 (“Overbroad generalizations of that order, the Court has come to comprehend, have a constraining impact, descriptive though they may be of the way many people still order their lives.”); see also J.E.B. v. Alabama ex rel. T.B., 511 U.S. 127, 139 n.11 (1994) (“[G]ender classifications that rest on impermissible stereotypes violate the Equal Protection Clause, even when some statistical support can be conjured up for the generalization.”); Craig, 429 U.S. at 201–02 (invalidating a sex classification even though the evidence supporting it was “not trivial in a statistical sense” and noting that “prior cases have consistently rejected the use of sex as a decisionmaking factor even though the statutes in question certainly rested on far more predictive empirical relationships than this”); Case, Very Stereotype, supra note 68, at 1450 (“[V]irtually every sex-respecting rule struck down by the Court in the last quarter century embodied a proxy that was overwhelmingly, though not perfectly, accurate.”). As Professor Mary Anne Case has helpfully put it: “[T]he assump­tion at the root of the sex-respecting rule must be true of either all women or no women or all men or no men; there must be a zero or a hundred on one side of the sex equation or the other.” 70 Case, Very Stereotype, supra note 68, at 1449–50. In effect, this can make the doctrine stricter than strict scrutiny. 71 See id. at 1453 (“The perfect proxy test has always had the capacity to be more strict even than strict scrutiny.”).

Under this analysis, “[p]hysical differences between men and women” are the primary differences that can justify sex-based classifica­tion. 72 Virginia, 518 U.S. at 533. In a handful of cases from decades ago, the Court recognized two other circumstances when sex classifications are justified even if not based in physical differences: (1) when they are “used to compensate women ‘for particular economic disabilities [they have] suffered,’” and (2) when another sex-respecting classification not before the Court creates a perfect proxy between sex and the challenged classification. Case, Very Stereotype, supra note 68, at 1457–58 (quoting Virginia, 518 U.S. at 533). Sex classifications unjustified by physical differences are imper­missible, because there is then no necessary connection between sex and the classification, and thus the classification is an overbroad stereotype. 73 See, e.g., Morales-Santana, 137 S. Ct. at 1700–01 (striking down a sex-differentiated citizenship rule); Wengler v. Druggists Mut. Ins. Co., 446 U.S. 142, 152 (1980) (striking down a law that granted automatic workers’ compensation benefits to widows but not to widowers); Califano v. Westcott, 443 U.S. 76, 89 (1979) (striking down a law that provided benefits to children of unemployed fathers but not unemployed mothers); Caban v. Mohammed, 441 U.S. 380, 394 (1979) (striking down a law that allowed unwed mothers, but not unwed fathers, to block the adoption of a child by refusing consent); Orr v. Orr, 440 U.S. 268, 283 (1979) (striking down a law that provided alimony upon divorce for women but not for men); Califano v. Goldfarb, 430 U.S. 199, 216–17 (1977) (striking down a law that provided Social Security benefits to widows automatically but required more to provide benefits to widowers); Stanton v. Stanton, 421 U.S. 7, 15–17 (1975) (striking down a law providing different ages of majority for boys and girls and thereby affording boys more years of parental support on the assumption that girls would marry rather than continue their education); Weinberger v. Wiesenfeld, 420 U.S. 636, 642–45, 653 (1975) (striking down a categorical ban on Social Security survivors’ benefits for widowers, but not widows, with minor children); Frontiero, 411 U.S. at 688–91 (striking down a presumption that wives, but not husbands, of servicemembers were dependent on their spouses); Stanley v. Illinois, 405 U.S. 645, 658 (1972) (striking down a presumption that unwed fathers, but not mothers, were inadequate caregivers for their children). Importantly, though, the Equal Protection Clause scrutinizes sex classifications even when physical differences between the sexes matter. In United States v. Virginia, the Court considered a challenge to the male-only admissions policy at the Virginia Military Institute (VMI), a state-run military academy requiring “[p]hysical rigor” that men on average could more readily achieve than women on average. 74 518 U.S. at 522 (alteration in original) (internal quotation marks omitted) (quoting United States v. Virginia, 766 F. Supp. 1407, 1421 (W.D. Va. 1991)). The Supreme Court recognized that physical sex differences could in theory justify limiting admissions to men but would not do so in reality unless no women were capable of transcending them. 75 See id. at 533 (“Physical differences between men and women . . . are enduring . . . .”). Because  some  women  could  pass  VMI’s  admissions  bar,   only  one  Justice  believed  the  VMI  process  to  be  constitutional. 76 See id. at 566–67 (Scalia, J., dissenting). Chief Justice Rehnquist concurred with the majority in a separate opinion. See id. at 558 (Rehnquist, C.J., concurring in the judgment). Justice Thomas recused himself because his son was attending VMI. See id.

Under this doctrine of constitutional sex-equality law, parenting has been recognized as perhaps the most critical area to scrutinize because “the faultline between work and family” is “precisely where sex-based overgeneralization has been and remains strongest.” 77 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 738 (2003). The Court has compared mothers and fathers to assess whether they are similarly situ­ated in performing “family duties” 78 Id. at 730, 736. and balancing those duties with work responsibilities. 79 See Weinberger v. Wiesenfeld, 420 U.S. 636, 652 (1975) (“[T]o the extent that women who work when they have sole responsibility for children encounter special problems, it would seem that men with sole responsibility for children will encounter the same child-care related problems.”). The Court has repeatedly struck down sex classifi­cations in these circumstances as not grounded in any physical differ­ences but reflective of separate-spheres thinking. 80 See Ginsburg, Remarks, supra note 25, at 23–24 (“The framework evolving at the time of the Wiesenfeld case persists to this day. It has enabled the Supreme Court effectively to break the hold of the breadwinner–homemaker dichotomy . . . .”); supra note 73 (listing cases in which the Court struck down statutes based on overbroad stereotypes).

By 2003, Republican-appointed Chief Justice Rehnquist—no stalwart of women’s rights 81 See Reva B. Siegel, You’ve Come a Long Way, Baby: Rehnquist’s New Approach to Pregnancy Discrimination in Hibbs, 58 Stan. L. Rev. 1871, 1872 (2006) [hereinafter Siegel, Long Way] (noting that Rehnquist had been an Equal Rights Amendment opponent and “a vocal critic of the Court’s sex discrimination jurisprudence”); Linda Greenhouse, The Supreme Court: States’ Rights; Justices, 6-3, Rule Workers Can Sue States over Leave, N.Y. Times (May 28, 2003), https://www.nytimes.com/2003/05/28/us/supreme-court-states-rights-justices-6-3-rule-workers-can-sue-states-over-leave.html (on file with the Columbia Law Review) (noting the surprising Rehnquist vote and the shift it represented among the conservative Justices). —included broad language about the importance of unsexing parenting in a key sex discrimination precedent, Nevada Department of Human Resources v. Hibbs. 82 See 538 U.S. at 730–31 (accepting Congress’s finding that the “pervasive sex-role stereotype that caring for family members is women’s work” reinforced gendered binaries). The question in Hibbs was whether the FMLA’s abrogation of state sovereignty was valid under Section 5 of the Fourteenth Amendment, thereby allowing a state employee to sue a state employer under the statute. Id. at 726. This question required the Court to determine the contours of the Amendment’s first Section, as Congress’s Section 5 enforcement power applies only to laws remedying or deterring conduct that violates Section 1. See id. at 727–30 (“[T]he persistence of . . . unconstitutional discrimination by the States justifies Congress’ passage of prophylactic § 5 legislation.”). The Court decided that there are no differences between fathers and mothers that would justify grant­ing family leave after birth to mothers and not fathers beyond the “period of physical disability due to pregnancy and childbirth.” 83 Id. at 731 & n.4 (citing evidence suggesting that the typical medical recovery period for pregnancy lasts four to eight weeks). Any more extended maternity leave was “not attributable to any differential physical needs of men and women, but rather to the pervasive sex-role stereotype that caring for family members is women’s work.” 84 Id. at 731. So even after birth when there are physical sex differences, equal protection man­dates unsexing of any carework unrelated to these differences.

A critical recent pillar in the law’s unsexing of postbirth parenting has been recognizing the right to same-sex marriage. 85 See Obergefell v. Hodges, 135 S. Ct. 2584, 2604–05 (2015). Even while the Constitution policed laws enforcing sex roles in the family, sex-based par­enting roles remained a basis for rejecting same-sex marriage. 86 Discrimination on the basis of sexual orientation has been rooted in the challenges homosexuality poses to traditional sex roles more broadly. See Case, Very Stereotype, supra note 68, at 1488 (“[P]rohibitions on homosexuality rely on stereotypes in the sense that they are based on ‘fixed notions concerning the roles and abilities of men and women.’” (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 725 (1982))). The recognition of a federal right to same-sex marriage rejected the notion that there was anything necessary about having one parent of each sex. 87 See Obergefell, 135 S. Ct. at 2600 (“[M]any same-sex couples provide loving and nurturing homes to their children . . . . Most states have allowed gays and lesbians to adopt . . . . This provides powerful confirmation from the law itself that gays and lesbians can create loving, supportive families.”). The Supreme Court’s recognition that two men or two women can be just as good parents as a man and a woman means that sex has been still further decoupled from parenting as a matter of constitutional law.

Statutes that regulate at the “faultline between work and family” are now aimed at disaggregating sex from carework. 88 Hibbs, 538 U.S. at 738. Title VII 89 See Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e-2(a) (2012). and the Equal Pay Act 90 See Equal Pay Act of 1963, 29 U.S.C. § 206(d) (2012). together prohibit many forms of employment discrimination on the basis of sex, including when sex stereotypes about family responsibilities lead to discrimination, against either women or men. 91 See EEOC, Enforcement Guidance: Unlawful Disparate Treatment of Workers with Caregiving Responsibilities, 2 EEOC Compl. Man. (BNA) § 615 (2007), https://www.eeoc.gov/policy/docs/caregiving.pdf [https://perma.cc/5EKK-4U54] (noting that Title VII outlaws both “sex-based disparate treatment of female caregivers, focusing on sex-based stereotypes,” and “sex-based disparate treatment of male caregivers, such as the denial of childcare leave that is available to female workers”). And when formal equality under these statutes proved too anemic to unsex at the intersection of work and family, Congress acted by taking steps—albeit small ones—in passing the FMLA. 92 See Hibbs, 538 U.S. at 737–38 (explaining that the affirmative efforts of the FMLA were necessary because formal-equality statutes like Title VII were insufficient to address the inequality that women faced at home and at work). As  the  Supreme  Court  explained,   the  FMLA,  “[b]y  setting  a  minimum  standard  of  family  leave  for  all  eligible  employees,   irrespective  of  gender,  . . .  attacks the formerly state-sanctioned stereotype that only women are responsible for family caregiving, thereby reducing employers’ incentives to engage in dis­crimination by basing hiring and promotion decisions on stereotypes.” 93 Id. at 737.

II. Sexing Pregnancy

Courts and commentators across the jurisprudential spectrum have failed to recognize carework during pregnancy because they view preg­nancy as simply a biological event and thus impossible to disaggregate from sex. Cases like Geduldig v. Aiello and statutes like the FMLA view pregnancy as a purely physical experience, or a social experience deriva­tive of that physical experience, and scholars largely agree. 94 See 417 U.S. 484, 496 n.20 (1974) (“Normal pregnancy is an objectively identifiable physical condition with unique characteristics.”); 29 C.F.R. § 825.120(a)(4) (2018) (“An expectant mother may take FMLA leave before the birth of the child for prenatal care or if her condition makes her unable to work.”); Siegel, Employment Equality, supra note 9, at 942 (stating that pregnancy “is a biological difference central to the definition of gender roles, one traditionally believed to render women unfit for employment” (emphasis added)). Constitu­tional law protecting a woman’s right to choose likewise frames pregnancy as almost exclusively a biological event. 95 See, e.g., Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 869 (1992) (identifying “the urgent claims of the woman to retain the ultimate control over her destiny and her body, claims implicit in the meaning of liberty”); id. at 896 (“It is an inescapable biological fact that state regulation with respect to the child a woman is carrying will have a far greater impact on the mother’s liberty than on the father’s.”); id. (“The effect of state regulation on a woman’s protected liberty is doubly deserving of scrutiny . . . , as the State has touched . . . upon . . . the very bodily integrity of the pregnant woman.” (emphasis added)). Under this view, pregnancy loses its uniqueness only when one understands that men experience conditions that are comparably disabling in the physical limitations they produce and the social reactions they generate from others—but men experience these conditions only outside of pregnancy. 96 See Williams, supra note 10, at 326–27. The result is a wide swath of facially sex-discriminatory laws regulating “the faultline between work and family” 97 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 738 (2003). during the pregnancy that have never been noticed or scrutinized, let alone identified as in tension with the doctrine discussed in Part I, which requires disaggregating sex from carework.

This Part first explains why we sex pregnancy, unpacking the relevant pregnancy-discrimination jurisprudence and how it sits in tension with the sex-equality jurisprudence of parenting. Pregnancy-discrimination jurisprudence typically does not see the carework that transpires during the pregnancy. 98 See infra notes 129–133 and accompanying text. Even in those instances when it is identified, the care­work of pregnancy is seen as necessarily aggregated with and then assigned by sex, and therefore it necessarily reinforces a domestic role for women. 99 See infra notes 134, 139–140 and accompanying text. The assumption that pregnancy is only for the pregnant woman not only excludes expectant fathers from pregnancy but some­times also excludes nonpregnant expectant mothers, including lesbian partners. 100 See infra note 257 and accompanying text. For gay male couples who engage a surrogate, both expectant parents are excluded from the pregnancy by reason of their sex. 101 See infra note 255 and accompanying text.

This Part then sets forth how we sex pregnancy, cataloguing the range of legal rules that rely on sex classifications during the pregnancy and highlighting how the very same rules are sex neutral after birth. This catalogue does not aspire to be exhaustive but rather highlights some of the most significant rules that generate the law of the sexed pregnancy.

A. Why We Sex Pregnancy

After birth, the constitutional basis for disaggregating sex from parenting is the recognition of various “family duties” that need not be and thus should not be allocated on the basis of sex. 102 Hibbs, 538 U.S. at 730. During the nine months of pregnancy, there are substantial analogous “family duties” that have nothing to do with gestation and thus can also be disaggregated from sex. 103 See infra notes 110–128 and accompanying text. But unlike the postbirth period, this prebirth carework has gone unnoticed by law. This section first provides a typology of the substantial nonbiological carework that goes on before birth that is analogous to the postbirth carework that equal protection doctrine has recognized and required to be unsexed. It then explains the invisibility of this pregnancy carework as a result of the conception of pregnancy as a biological event whose social consequences flow only from this biological phenomenon.

Carework during pregnancy is substantial: Expectant parents of either sex can engage in significant investments before birth, similar to those they engage in after birth, to increase the chances of producing a happy and healthy child. 104 See Dara E. Purvis, The Origin of Parental Rights: Labor, Intent, and Fathers, 41 Fla. St. U. L. Rev. 645, 681 (2014) (listing examples of “prebirth parental labor” that can be separated from the gestational labor of pregnancy). And they do. Every year Americans spend billions of dollars and many hours preparing for the birth of their child. 105 See Elisabeth Rosenthal, American Way of Birth, Costliest in the World, N.Y. Times (June 30, 2013), https://www.nytimes.com/2013/07/01/health/american-way-of-birth-costliest-in-the-world.html (on file with the Columbia Law Review). Private firms offer highly valuable employees “parenting coaches” to manage these investments. 106 See Tara Siegel Bernard, Why Companies Have Started to Coach New Parents, N.Y. Times (July 22, 2016), https://www.nytimes.com/2016/07/23/your-money/why-companies-have-started-to-coach-new-parents.html (on file with the Columbia Law Review) (describing parenting programs that begin during pregnancy and “are providing employees with coaching sessions” that are “often available to new fathers”). Books like Getting Ready for Baby have sold many copies because of their lists of “things to remember [to do] throughout your pregnancy and in the first months after your baby’s birth.” 107 Hélène Tragos Stelian, Getting Ready for Baby: The Practical Parent’s Organizer 19 (2014). Pregnancy carework that can be done by either sex falls into three broad categories: physical capital, human capital, and social capital, which are discussed in turn. 108 While this Article focuses on the various ways that pregnancy carework entails acquiring capital, sometimes expectant parents will also simply have to do administrative work to prepare for the child, such as making arrangements for health insurance cover­age, and either parent can invest in this nonphysical capital. See Ester Bloom, The Unconscionable Difficulty of Getting Health Insurance for a Newborn, Atlantic (June 20, 2016), https://www.theatlantic.com/business/archive/2016/06/difficulty-of-getting-health-
insurance-for-a-newborn/484568 [https://perma.cc/P7X3-XNHG] (discussing the diffi­culty of planning to get health insurance coverage for a newborn).

First, expectant parents invest in physical capital in ways that can be unsexed. Just like being a parent requires acquiring many goods that are needed to care for the child, preparing to parent requires acquiring many goods that are needed either during the pregnancy or after birth. 109 Many parents borrow, rent, or are given these products rather than purchase them. Even the process of identifying used products requires an investment of resources that can be unsexed. Some of these products will be used exclusively by the pregnant woman during the pregnancy (for example, back pillow 110 See Jess Miller, Best Pregnancy Pillow: An Expert Buyers Guide to Maternity Pillows, Parent.Guide, http://parent.guide/pregnancy-pillows [https://perma.cc/NE7P-6NRR] (last updated Oct. 31, 2018) (“As your pregnancy progresses, you will notice your that [sic] tummy will continue to expand . . . . Your new size will bring with it a whole new set of challenges, one of which is sleeping comfortably. A pregnancy pillow is just the product you need to get a super comfortable sleep . . . .”). ) or after the pregnancy (for example, breastpump). Other products can be used by either expectant parent during the pregnancy, such as the Getting Ready for Baby book mentioned above. And still other products can be used by either expectant parent after the child is born 111 Even products that seem inherently sexed may not be. For example, a breastfeeding pillow can be used not only by the woman while breastfeeding but also by either parent when giving the baby a bottle. See How to Use a Breast Feeding Pillow, wikiHow, https://www.wikihow.com/Use-a-Breast-Feeding-Pillow [https://perma.cc/Y4B2-B4LN] (last visited Oct. 9, 2018). but are substantially more useful if purchased during the pregnancy. State laws require that expectant parents must have a carseat before they depart from a hospital in a vehicle with their newborn. 112 See Child Passenger Safety, Governors Highway Safety Ass’n, http://www.ghsa.org/html/stateinfo/laws/childsafety_laws.html [https://perma.cc/836D-DJQ5] (last visited Oct. 9, 2018) (explaining that “[a]ll states and territories require child safety seats for infants and children fitting specific criteria” based on age, weight, and height). The early days of a newborn involve hundreds of diapers and little time to purchase them. 113 See Newborn Checklist: Everything You Need Before Your Baby Arrives, Today’s Parent (Apr. 1, 2018), http://www.todaysparent.com/checklists/newborn-checklist [https://
Only about one in four parents “always” or “almost always” sleeps with their newborn, meaning that expectant parents usually need a bassinet or crib. 114 See William Sears et al., The Baby Sleep Book 107 (2005) (citing a survey by the National Center for Health Statistics). Regardless of which parent will use these products, any expectant parent can research the products and acquire them.

Second, expectant parents invest in human capital in ways that can be unsexed. Parents learn how to assemble toys, respond to the unique emo­tional composition of a child, and complete tasks more efficiently due to the time constraints of parenthood. Likewise, the pregnancy is a nine-month period to acquire the motivation and knowledge that will improve one’s performance as a parent. 115 Evolutionary biologists have theorized that the nine months of pregnancy were necessary precisely for expectant parents to learn the many skills needed to become productive parents. See Alison Gopnik, The Gardener and the Carpenter: What the New Science of Child Development Tells Us About the Relationship Between Parents and Children 52–60 (2016) (summarizing the literature). The Supreme Court has noted the importance of the nine-month period for generating one’s identity as a future parent, 116 See Lehr v. Robertson, 463 U.S. 248, 260 n.16 (1983) (“The mother carries and bears the child, and in this sense her parental relationship is clear.” (internal quotation marks omitted) (quoting Caban v. Mohammed, 441 U.S. 380, 397 (1979) (Stewart, J., dissenting))). and research has found that developing this capacity to care can occur regardless of sex. 117 See Brandon et al., supra note 5, at 202–11.

Consider, also, the example of attending a newborn-care class. Expectant parents learn basic physical maintenance, such as changing diapers; health and safety measures, such as ensuring that newborns do not sleep on their backs to reduce the risks of Sudden Infant Death Syndrome; and even infant first aid and CPR. 118 See Parent Education Classes & Support Groups, NYU Langone Health, https://
nyulangone.org/patient-family-support/for-new-expecting-parents/parent-education-classes-support-groups [https://perma.cc/Y88R-9WLM] (last visited Nov. 1, 2018) (providing an overview of various classes for new parents).
Expectant parents can also learn more generally about different approaches to parenting in ways that can positively inform their parenting forever.

Investments in human capital can also take the form of behavior modifications. Some behavioral changes, such as reducing caffeine con­sumption or avoiding sushi, are for the pregnant person. 119 The expectant father might choose to similarly restrict his diet or increase his caloric intake out of solidarity with the pregnant partner. See Lisa Belkin, Men Gain Weight During Pregnancy, N.Y. Times: Motherlode (June 2, 2009), https://parenting.blogs.nytimes.com/2009/06/02/men-who-swell-with-pregnancy [https://perma.cc/4URK-BCXH] (citing research “that the average weight gained during pregnancy is 14 pounds—and that’s by the babies’ fathers”). Other behav­ioral changes, such as quitting smoking, are important for all expectant parents so long as the expectant parent will be present during the preg­nancy. 120 There are dozens of studies with these empirical findings. For a recent and notable example, see Julian Laubenthal et al., Cigarette Smoke-Induced Transgenerational Alterations in Genome Stability in Cord Blood of Human F1 Offspring, 26 FASEB J. 3946, 3953–55 (2012). Quitting smoking during pregnancy is important not only because of the health risks of prenatal smoking but also because it prepares expectant parents to remain smoke-free after birth. 121 Again, studies on the harms of postnatal parental smoking are too numerous to list, but some good examples include C.M. Blackburn et al., Parental Smoking and Passive Smoking in Infants: Fathers Matter Too, 20 Health Educ. Res. 185, 190–93 (2005); Katherine King et al., Family Composition and Children’s Exposure to Adult Smokers in Their Homes, 123 Pediatrics 559, 562–63 (2009). Other behavioral changes during the pregnancy are specifically targeted at preparing to parent. For example, expectant parents—either mothers or fathers—may begin a workout regimen that will prepare them to lift an infant regularly without injury. Or expectant parents might try to adopt a host of behaviors to improve themselves—eating healthier, for example, or communicating better—to serve as good role models to their chil­dren. It is important to begin such behavioral changes during pregnancy because it takes time for new habits to form, 122 See Lisa Buckingham, How to Make Exercise a Habit, Guardian (Feb. 8, 2015), https://www.theguardian.com/lifeandstyle/2015/feb/08/get-habit-exercise-getting-fitter-new-year-resolutions [https://perma.cc/J4V6-CUP8] (citing a study concluding that it takes sixty-six days to form a new habit). and few parents of an infant have the time and energy necessary to form new habits.

Third, expectant parents invest in social capital in ways that can be unsexed. Parents rely on relationships in their communities and their workplaces to support their efforts. 123 See generally Laura Rosenbury, Between Home and School, 155 U. Pa. L. Rev. 833 (2007) (identifying sites of childrearing beyond the home and the school and exploring how law might recognize these sites). Through the course of the nine months, expectant parents likewise develop relationships with persons who help to care for the pregnancy or the resulting child: the obstetri­cian or midwife, the doula, and the pediatrician. 124 See Naomi Schoenbaum, The Law of Intimate Work, 90 Wash. L. Rev. 1167, 1174–85 (2015) (describing the close relationships that arise between consumers and providers of “intimate work,” including home health aides and medical workers). Expectant parents also find it useful to develop relationships with others who are going through the experience of expecting a child. These relationships may grow out of human capital investments—you meet people at a childbirth class. These relationships may also be in service of acquiring knowledge and skills—you turn to these persons for advice on matters of expecting a child, for instance. 125 See Rosario Ceballo & Vonnie C. McLoyd, Social Support and Parenting in Poor, Dangerous Neighborhoods, 73 Child Dev. 1310, 1311 (2002) (“[S]ocial networks and support may prevent or alleviate a number of potential familial problems.”); Carolyn Webster-Stratton, From Parent Training to Community Building, 78 Families Soc’y 156, 158–60 (1997) (discussing the need for more community networks and social support for good parenting outcomes, especially in low-income families). All expectant parents can form these relationships, regardless of sex. Even when it comes to relationships that are thought to run primar­ily to the pregnant woman—such as the obstetrician, midwife, and doula—any expectant parent can play a key role in forming, maintaining, and deepening these relationships. 126 See Carrie Murphy, What Dads Need to Know About Doulas, Parents.com, https://www.parents.com/pregnancy/giving-birth/doula/what-dads-need-to-know-about-doulas [https://perma.cc/EVH3-U8G4] (last visited Oct. 10, 2018) (explaining that “doulas aren’t there only for moms-to-be” and that “[t]hey also play a key role in helping their partners, offering them invaluable emotional and practical support during the overwhelming experience of childbirth”).

The presence of carework during the pregnancy that is analogous to carework after the pregnancy has escaped notice because sex-equality law has viewed pregnancy solely as a biological event. If pregnancy is only a matter of gestation, then there is no other carework to identify to ensure that it is not impermissibly allocated on the basis of sex. Take the Supreme Court’s 1974 decision in Geduldig v. Aiello, which rejected a constitutional sex discrimination challenge to a California disability insurance law that exempted certain pregnancy-related disabilities from its coverage. 127 417 U.S. 484, 496–97 (1974). The law originally excluded all disabilities “caused by or arising in connection with pregnancy” but was revised to exclude only disability associated with normal pregnancy. Id. at 490 n.15 (quoting the revised law); Aiello v. Hansen, 359 F. Supp. 792, 793 n.1 (N.D. Cal. 1973) (quoting the original law). The women challenging the law under the Fourteenth Amendment focused on pregnancy as a “physical” experience that is therefore “unique to one sex.” 128 See Brief for Appellees at 28–47, Geduldig, 417 U.S. 484 (No. 73-640), 1974 WL 185752. The State of California accepted this characterization of pregnancy, arguing that men could be compared to pregnant women only “for other physical or mental reasons.” 129 See Brief for Appellant at 19, Geduldig, 417 U.S. 484 (No. 73-640), 1974 WL 185750 (emphasis added). The par­ties disagreed about whether it is unconstitutional sex discrimination to single out a physical experience that is necessarily sexed—pregnancy—for unique treatment. But their disagreement was grounded in a critical premise on which both sides agreed: Pregnancy is a necessarily sexed experience unique to women. The Court in Geduldig embraced this fram­ing of pregnancy as necessarily sexed, describing pregnancy as “an objec­tively identifiable physical condition with unique characteristics.” 130 Geduldig, 417 U.S. at 496 n.20. To the extent that the Court suggested otherwise in a notorious footnote that distinguished “pregnant women” from “nonpregnant persons,” id., that suggestion has been almost universally reviled. See, e.g., Joanna L. Grossman, Pregnancy, Work, and the Promise of Equal Citizenship, 98 Geo. L.J. 567, 599 n.171 (2010) (“[Geduldig] has been rightly lampooned for its reasoning, which is essentially the same as holding that a law that discriminates against bachelors does not discriminate on the basis of sex.”); Kenneth L. Karst, Foreword: Equal Citizenship Under the Fourteenth Amendment, 91 Harv. L. Rev. 1, 54 n.304 (1977) (“[T]he constitutional sport of [Geduldig] . . . with [its] Alice-in-Wonderland view of pregnancy as a sex-neutral phenomenon, [is a] good candidate[ ] for early retirement.”); Law, supra note 9, at 983 (“Criticizing Geduldig has since become a cottage industry. Over two dozen law review articles have condemned both the Court’s approach and the result.”). Given that the Court treated pregnancy as a purely physical condition unique to women, its later disentanglement of pregnancy from sex went entirely unjustified. So neither the parties nor the Court recognized the nonbiological carework that pregnancy generates. 131 Two years later, when the Court decided in General Electric Co. v. Gilbert that pregnancy discrimination was not sex discrimination for purposes of Title VII, carework was likewise nowhere mentioned. See 429 U.S. 125, 133–37 (1976) (considering the sex-equality arguments related to pregnancy and referencing arguments by counsel for both sides).

Even when the Court has recognized the social impact of pregnancy, it has still viewed this social impact as flowing only from the physical experi­ence of being pregnant and thus not involving the type of carework that can be disaggregated from sex. For example, across many areas of law, the Supreme Court has referred to emotional bonding that transpires during the pregnancy, but it has assumed that this emotional bonding happens only as a result of the experience of gestation and is thus experienced by the mother alone. 132 See supra note 6; see also Michael M. v. Superior Court, 450 U.S. 464, 467–72 (1981) (plurality opinion) (Rehnquist, J.) (upholding a statutory rape law applying only to female victims in part because “[o]nly women may become pregnant, and they suffer disproportionately the profound physical, emotional, and psychological consequences of sexual activity”). The Court has never recognized any forms of carework done during the pregnancy that would produce emo­tional bonding by either the mother or father.

Even scholars who acknowledge certain social aspects of pregnancy have likewise missed the pregnancy carework that can be disaggregated from sex. 133 But see Purvis, supra note 106, at 681 (recognizing pregnancy carework that can be disaggregated from sex for the purpose of establishing parental rights). Scholars such as Professor Reva Siegel and then-Judge Ginsburg have argued that the focus on pregnancy as a physi­cal experi­ence is excessive. 134 See Ruth Bader Ginsburg, Some Thoughts on Autonomy and Equality in Relation to Roe v. Wade, 63 N.C. L. Rev. 375, 386 (1985) (“Overall, the Court’s Roe position is weakened, I believe, by the opinion’s concentration on a medically approved autonomy idea, to the exclusion of a constitutionally based sex-equality perspective.”); Reva Siegel, Reasoning from the Body: A Historical Perspective on Abortion Regulation and Questions of Equal Protection, 44 Stan. L. Rev. 261, 274 (1992) [hereinafter Siegel, Reasoning] (“Because Roe and its progeny treat pregnancy as a physiological problem, they obscure the extent to which . . . [it] impose[s] material deprivations and dignitary injuries on those who perform its work.”). To these scholars, pregnancy is also a moment defin­ing the social roles of expectant mothers—but only expectant mothers. Because pregnancy “is a biological difference” 135 Siegel, Employment Equality, supra note 9, at 942 (stating that pregnancy is “a biological difference central to the definition of gender roles, one traditionally believed to render women unfit for employment”). that “women[ ] distinc­tively” 136 Robin West, Jurisprudence and Gender, 55 U. Chi. L. Rev. 1, 2–3 (1988). experience, it defines social roles for women and not for men. The reason is that the social impact of pregnancy flows only from physical fact of gestation, not from the carework that preg­nancy generates. 137 See Siegel, Reasoning, supra note 136, at 267 (recognizing that “[s]ocial forces play a powerful part in shaping the process of reproduction” but nonetheless limiting the recognition of these social forces to how they act on the pregnant women, such as “[s]ocial forces determin[ing] the quality of health care available to a woman during pregnancy”). For  these scholars,   recognizing  the  social  impact  of  pregnancy  expands  not  the  class  of  relevant  stakeholders  in  the  preg­nancy  but  the  class  of  relevant  stakes  that  expectant  mothers — and expectant mothers alone — face during the pregnancy. So the only role that men play is as compa­rators in discrimination claims: Men experi­ence conditions that are comparably disabling in their physical limita­tions and the social reactions they generate from others as pregnancy—but men experience these conditions only outside of pregnancy. 138 See Williams, supra note 10, at 327–28, 359 n.130 (explaining how nonpregnant but similarly disabled men can serve as comparators to support pregnancy-discrimination claims but rejecting the idea that a “man w[ould] ever need a pregnancy-related leave” as implausible (alteration in original) (quoting Linda J. Krieger & Patricia N. Cooney, The Miller-Wohl Controversy: Equal Treatment, Positive Action and the Meaning of Women’s Equality, 13 Golden Gate U. L. Rev. 513, 532 (1983))).

B. How We Sex Pregnancy

Because the law treats pregnancy as a physical experience primarily for women, a law of pregnancy has developed that is sexed across a range of doctrines. Constitutional law sexes pregnancy by failing to treat women and men who can engage in the same carework during preg­nancy as similarly situated. Instead, decisions presume a pregnancy in which women do all of the work—whether biological or not. Robust scholarly debates around these decisions have missed this and have thus only reinforced an inherently sexed conception of pregnancy. And even if equal protection scrutiny were otherwise recognized to apply to preg­nancy, constitutional law might still prevent this under a reading of repro­ductive rights jurisprudence that raises concerns about any male involve­ment in pregnancy. In the wake of this constitutional treatment, a host of facially sex-discriminatory laws, including those assigning workplace rights, remain in the context of pregnancy, even though analogous laws have long been eradicated after birth.

1. Constitutional Law. — Sex-equality law—like so much of anti­discrimination law—is defined by comparisons. 139 See Suzanne B. Goldberg, Discrimination by Comparison, 120 Yale L.J. 728, 731, 748 (2011) (“Comparators become relevant to the analysis . . . because they help expose . . . that ‘likes’ have been treated in an ‘unlike’ fashion and give rise to the inference [of] discrimination . . . .”). After birth, equal protection cases examine whether federal and state laws “presume[ ] . . . [that] the mother is the ‘center of home and family life’” 140 Califano v. Westcott, 443 U.S. 76, 89 (1979) (quoting Taylor v. Louisiana, 419 U.S. 522, 534 n.15 (1975)). by comparing what care- and market-work the law expects of mothers and fathers. 141 See supra note 48. Before birth, though, equal protection cases reflect and reinforce the breadwinner–homemaker divide by not comparing the capacity of expectant mothers and expectant fathers to engage in care- and market-work. Courts and commentators across the jurisprudential spectrum agree that if there is an equal protection claim in the context of preg­nancy, it is because of a comparison between a pregnant woman and a man facing a similar physical complication, not because of a comparison between a pregnant woman and an expectant father. 142 See supra note 140 and accompanying text (describing similar scholarly views).

Consider, again, the Court’s decision in Geduldig v. Aiello. 143 417 U.S. 484 (1974). Counsel for both sides—as well as the majority and dissenting Justices—agreed that the relevant comparison for constitutional purposes would be bet­ween pregnant women and similarly situated physically limited employees (both male and female). 144 See id. at 496–97 (“There is no risk from which men are protected and women are not. Likewise, there is no risk from which women are protected and men are not.”). The plaintiffs suggested that the only male workers similarly situated to the pregnant plaintiffs were those at risk of heart attacks or facing other physically disabling conditions. 145 See Brief for Appellees, supra note 130, at 58–71 (arguing that disabilities caused by pregnancy are “similar in all relevant respects to disabilities now compensated” by California’s disability-benefits program).

Justice Ginsburg has been critical of Geduldig. 146 See Coleman v. Court of Appeals, 566 U.S. 30, 54 (2012) (Ginsburg, J., dissenting) (“[T]his case is a fit occasion to revisit that conclusion.”). Yet she has argued that the problem with Geduldig is that it has authorized “dis­crimination against women,” without recognizing that this is in part because of how pregnancy regulations stereotype both women and men. 147 Id. at 56 (Ginsburg, J., dissenting) (emphasis added); see also id. (“[C]hildbearing is not only a biological function unique to women. It is also inextricably intertwined with employers’ ‘stereotypical views about women’s commitment to work and their value as employees.’” (emphasis added) (quoting Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 736 (2003))). As Ginsburg herself argued as an advocate, “Fair and equal treat­ment for women means fair and equal treatment for members of both sexes.” 148 Brief for Petitioner-Appellant at 20, Moritz v. Comm’r, 469 F.2d 466 (10th Cir. 1971) (No. 71-1127) (on file with the Library of Congress, Manuscript Division, Ruth Bader Ginsburg Papers, Container 5, Folder: Moritz v. Comm’r (1971)). Even though Geduldig dealt with healthcare benefits, it is telling nonetheless that women’s rights advocates presented pregnancy in these purely biological terms. 149 This is not because they were seeking special legal treatment for pregnancy. In fact, in later litigation, the same advocate who represented the plaintiffs in Geduldig, Wendy Williams, was firmly opposed to any unique benefits being afforded to pregnancy out of concern that this would negatively stereotype women as facing a unique physical disability. See Brief Amici Curiae for the National Organization for Women et al. in Support of Neither Party at *5 n.10, Cal. Fed. Sav. & Loan Ass’n v. Guerra, 479 U.S. 272 (1987) (No. 85-494), 1986 WL 728368 [hereinafter Williams Amicus Brief, California Federal] (comparing pregnancy to other situations involving “extended physical health problems”).

Most notable is that women’s rights advocates, like Ginsburg, who were adamant about challenging laws based on stereotyped thinking about postbirth carework 150 See Franklin, supra note 50, at 89–90 (noting how Ginsburg and others pressed the arguments “that laws that steer men out of traditionally female [caregiving] roles effectively require women to assume those roles” and persuaded the Court to interpret the Equal Protection Clause “as a bar to such ‘role-typing’” (quoting Stanton v. Stanton, 421 U.S. 7, 15 (1975))). did not raise these same arguments in the prebirth period. Consider California Federal Savings & Loan Ass’n v. Guerra, a case deciding whether Title VII’s ban on pregnancy discrimina­tion preempted a state law granting unique workplace benefits to preg­nant women. 151 479 U.S. at 274–75. The plaintiff’s lawyer in Geduldig, Wendy Williams, sub­mitted an amicus brief in California Federal arguing against special treat­ment of pregnant women because it would stereotype women as workers facing a unique physical disability, even though men faced analo­gous physical disabilities. 152 Williams Amicus Brief, California Federal, supra note 151, at *14–17 (endorsing Congress’s view that “viewing pregnancy as sui generis has historically resulted in widespread and serious harm to pregnant workers” and noting that special treatment of pregnant women may lead to employer backlash). In contrast to their litigation strategy in the postbirth period, antistereotyping advocates never mentioned how the law would enforce stereotypical sex roles by presuming that pregnancy had no carework impact on expectant men. 153 See generally id. (mentioning men in several contexts but failing to acknowledge the carework of expectant men).

Even laws that classify on the basis of sex but relate to pregnancy are treated more deferentially than laws that make other sex-based classifica­tions because of the assumption that pregnancy is necessarily sexed. In contrast to the postbirth period, the Court unquestionably accepts a vari­ety of sex-specific consequences beyond the physical that are seen to flow inevitably from the biological difference of pregnancy. For example, in Michael M. v. Superior Court, the Court upheld a statutory rape law that only men could violate, on the basis of gendered physical difference. 154 See 450 U.S. 464, 472–73 (1981). The Court justified the law “by the immutable physiological fact that it is the female exclusively who can become pregnant.” 155 Id. at 467 (internal quotation marks omitted) (quoting Michael M. v. Superior Court, 601 P.2d 572, 574 (Cal. 1979)). These physical differ­ences also generated social consequences that applied uniquely to women: “Only women may become pregnant, and they suffer dispropor­tionately the profound physical, emotional, and psychological conse­quences of sexual activity.” 156 Id. at 471.

Constitutional law sexes pregnancy not just by failing to scrutinize sexed law under the Equal Protection Clause but also by potentially requiring that pregnancy be sexed because of constitutional reproductive rights. Typically, formal equality under the Equal Protection Clause presents no independent constitutional problem. If a university admissions policy is invalidated for discriminating on the basis of race, there is no constitu­tional problem with a race-neutral remedy (unless there is some remedial obligation to have a race-conscious program). 157 See, e.g., City of Richmond v. J.A. Croson Co., 488 U.S. 469, 519 (1989) (“[T]he rule against race-conscious remedies is already less than an absolute one, for that relief may be the only adequate remedy after a judicial determination that a State or its instrumentality has violated the Equal Protection Clause.”). If men and women are treated similarly during pregnancy, though, this could be seen to under­mine the woman’s right to choose whether to carry the pregnancy to term. For now it is worth noting that the sexed pregnancy is overlooked because of both the absence of scrutiny and the perceived demands of a constitutional right. This Article explains in later Parts why these inter­pretations—a complete absence of heightened scrutiny for pregnancy regulations and a constitutional mandate for a universally sexed preg­nancy—are misguided.

2. Statutes and Regulations

a. Workplace Leave. — Postbirth-leave protections for providing care are almost entirely sex neutral. 158 The Equal Protection Clause would presumably allow a leave policy for the period of physical recovery from childbirth to be granted only to women who have given birth. See supra notes 72, 82–84 and accompanying text (discussing how the doctrine permits distinctions based on these types of physical sex differences). Women and men are equally able to take time away from work to attend important appointments with the pediatrician after birth. 159 See, e.g., S. Rep. No. 103-3, at 11–12 (1993) (listing this example as part of what the FMLA would cover). Regulations implementing the FMLA make this same promise—a promise unfulfilled in the prenatal context. See 29 C.F.R. § 825.112(b) (2018) (“The right to take leave under FMLA applies equally to male and female employees. A father, as well as a mother, can take family leave for the birth, placement for adoption, or foster care of a child.”). Women and men are equally able to use leaves to find childcare once they return to work. 160 See 29 C.F.R. § 825.126(b)(3).

By contrast, prebirth-leave protections are chock-full of sex classifica­tions that protect pregnant women’s carework but not analogous care­work by expectant fathers. When it comes to prenatal leave, the FMLA is facially sex discriminatory. FMLA regulations provide only “mother[s]” with “FMLA leave . . . for prenatal care.” 161 29 C.F.R. § 825.120(a)(4). The  FMLA  draft­ers  men­tioned  “attending  appointments”  and  “obtaining  essential  knowledge  about  how  to  care  for  a  newborn”  as  categories  of  “prenatal  care”  that  could  qualify  for  job-protected leave  for  only “expectant moth­ers” under the FMLA. 162 Id. (“The expectant mother is entitled to FMLA leave . . . for prenatal care.”). If an expectant father wants job-protected leave to attend a prenatal appointment to view the ultrasound and bond with the child, this is not covered by the FMLA. 163 See id.

Expectant fathers can take prenatal FMLA leave, but only in a secon­dary capacity, to “care for” the expectant mother when she is “incapaci­tated” or if such spousal support is “needed to care for her during her prenatal care.” 164 Compare id. § 825.120(a)(5) (“A spouse is entitled to FMLA leave if needed to care for a pregnant spouse who is incapacitated or if needed to care for her during her prenatal care . . . .” (emphasis added)), with id. § 825.202(b)(1) (“A pregnant employee may take leave intermittently for prenatal examinations or for her own condition, such as for periods of severe morning sickness.” (emphasis added)). After birth, the FMLA does not condition the father’s ability to pick up a prescription for his child on the condition of the mother. 165 See id. § 825.112(a)(1), (3) (granting leave for either a mother or father “to care for [a] newborn child” or “[t]o care for” a “son” or “daughter . . . with a serious health condition”). Before birth, the expectant father can pick up a prescription for the expectant mother only if she is “incapacitated” and cannot travel herself. 166 Id. § 825.120(a)(5); see also Family and Medical Leave Act of 1993, 73 Fed. Reg. 67,934, 67,950 (Nov. 17, 2008) (explaining that “a husband is entitled to FMLA-protected leave if he is needed to care for his spouse who is incapacitated due to her pregnancy (e.g., if the pregnant spouse is unable to transport herself to a doctor’s appointment)”).

Otherwise, expectant fathers are entitled to FMLA leave only to the same extent as any other eligible employee. Typically, the FMLA requires employees to have a “serious health condition” to take self-care leave. 167 29 U.S.C. § 2612(a)(1)(D) (2012) (providing leave “[b]ecause of a serious health condition that makes the employee unable to perform the functions of the position of such employee”). This would require an expectant father to have “an illness, injury, impair­ment, or physical or mental condition that involves . . . inpatient care in a hospital, hospice, or residential medical care facility,” or “continuing treatment by a health care provider.” 168 Id. § 2611(11). By contrast, the FMLA lowers the bar for pregnant women, who can take leave for “[a]ny period of incapacity due to pregnancy,” regardless of whether her condition is “serious.” 169 29 C.F.R. § 825.102 (allowing leave for “[a]ny period of incapacity due to pregnancy”); id. § 825.120(a)(4) (“The expectant mother is entitled to leave for incapacity due to pregnancy even though she does not receive treatment from a health care provider during the absence, and even if the absence does not last for more than three consecutive calendar days.”). So a pregnant woman who is experiencing antenatal depres­sion due to worries about being a mother is entitled to FMLA leave if her depression crosses the threshold of “incapacity.” An expectant father who is suffering from antenatal depression—which researchers estimate happens to approximately ten percent of expectant fathers 170 See James F. Paulson & Sharnail D. Bazemore, Prenatal and Postnatal Depression in Fathers and Its Association with Maternal Depression: A Meta-Analysis, 303 JAMA 1961, 1967 fig.2 (2010). There is a broader condition called “couvade” that describes “[w]hen pregnancy symptoms such as nausea, weight gain, mood swings and bloating occur in men.” See Katherine E. Wynne-Edwards, Why Do Some Men Experience Pregnancy Symptoms Such as Vomiting and Nausea when Their Wives Are Pregnant?, Sci. Am., https://www.scientificamerican.com/article/why-do-some-men-experienc [https://perma.cc/7ZKT-5E7Q] (last visited Oct. 10, 2018) (“[E]stimates of the frequency [of couvade] in modern Western populations range from under 20 percent to more than 80 percent of expectant fathers.”); see also Hillary Klein, Couvade Syndrome: Male Counterpart to Pregnancy, 21 Int’l J. Psychiatry Med. 57, 57 (1991). —is not entitled to leave unless his depression reaches a higher bar. Likewise, a father is entitled to take leave, including to “provid[e] psychological comfort and reassurance” to the pregnant mother if she is incapacitated, 171 See 29 C.F.R. § 825.120(a)(5); Family and Medical Leave Act of 1993, 73 Fed. Reg. at 67,950. but she can take leave to care for him only when he has a serious health condition. 172 See 29 U.S.C. § 2612(a)(1)(C) (providing leave “to care for the spouse . . . if such spouse . . . has a serious health condition”).

Still further, the FMLA provides pregnancy-related benefits to an expectant father only when necessary “to care for a pregnant spouse.” 173 29 C.F.R. § 825.120(a)(5). This means that it is not actually the expectant father who is entitled to any pregnancy-related leave under the FMLA but rather the pregnant woman’s husband, regardless of whether that person is the expectant parent. When the father of the child is not married to the mother, he is entitled to no preg­nancy-specific FMLA leave even though the pregnant woman is. The restriction of leave to spouses is significant not only for denigrating the father’s role in the pregnancy but also for its practical consequences. Recent data suggest that forty percent of births are to unmarried moth­ers. 174 Unmarried Childbearing, CDC, http://www.cdc.gov/nchs/fastats/unmarried-childbearing.htm [https://perma.cc/98RR-ZFWX] (last updated Mar. 31, 2017). In those situations, mothers receive their FMLA benefits, and sin­gle fathers do not receive even the minimal benefits that married fathers receive.

Because these laws generally refer to “expectant mothers,” all expectant fathers, including gay men who have engaged a surrogate, will be excluded, meaning that no one in their families will be eligible for any prenatal FMLA benefits. On the flip side, precisely because the laws refer to “expectant mothers” 175 29 C.F.R. § 825.120(a)(4). rather than “pregnant women,” they might be read to cover nonpregnant expectant mothers, including lesbian partners. The FMLA’s limitation of prenatal leave to “expectant mothers” also has implications for transgender men. There is the rare but real phe­nomenon of the pregnant man. 176 The pregnant transgender man troubles the perfect identification of sex with the physical effects of pregnancy that some have assumed and relied on to argue for heightened scrutiny of pregnancy regulations. See Bray v. Alexandria Women’s Health Clinic, 506 U.S. 263, 323 n.20, 327 (1993) (Stevens, J., dissenting) (“As the capacity to become pregnant is a characteristic necessarily associated with one sex, a classification based on the capacity to become pregnant is a classification based on sex.”); Cass R. Sunstein, Neutrality in Constitutional Law (with Special Reference to Pornography, Abortion, and Surrogacy), 92 Colum. L. Rev. 1, 32–33 (1992) (arguing that abortion restrictions should be seen as sex discrimination because pregnancy is a “defining characteristic or biological correlate of being female”). Female-to-male transgender persons can become pregnant. 177 See supra note 4 and accompanying text. In some states, a pregnant man could be legally male, rendering him ineligible for sex-based pregnancy benefits like those provided by the FMLA. 178 See Changing Birth Certificate Sex Designations: State-by-State Guidelines, Lambda Legal, https://www.lambdalegal.org/know-your-rights/article/trans-changing-birth-certificate-sex-designations [https://perma.cc/P8VD-KPSN] (last updated Sept. 17, 2018) (summarizing states’ procedures for amending a sex designation on a birth certificate). This requires a transgender man to choose between his identified gender and legal benefits. 179 This concern will only grow as America’s transgender population continues to increase. See Andrew R. Flores et al., Williams Inst., How Many Adults Identify as Transgender in the United States? 2–3 (2016), https://williamsinstitute.law.ucla.edu/wp-content/uploads/How-Many-Adults-Identify-as-Transgender-in-the-United-States.pdf [https://perma.cc/38WB-DGV2] (noting that America’s transgender population has reached 1.4 million people).

b. Workplace Rights. — Workplace law also grants pregnant women accommodations and protections against discrimination for carework they perform during the pregnancy while excluding expectant fathers in similar caregiving situations. Many states now mandate reasonable work­place accommodations on the basis of pregnancy. 180 See, e.g., Cal. Gov’t Code § 12945(a)(3)(A) (2018) (making it unlawful “[f]or an employer to refuse to provide reasonable accommodation for an employee for a condition related to pregnancy”); Conn. Gen. Stat. § 46a-60(b)(7)(G) (2018) (requiring that employers create “reasonable” workplace accommodations for “an employee or person seeking employment due to her pregnancy, unless the employer can demonstrate” undue hardship). See generally Nat’l P’ship for Women & Families, Reasonable Accommodations for Pregnant Workers: State and Local Laws (2018) [hereinafter Nat’l P’ship, Reasonable Accommodations], http://www.nationalpartnership.org/research-library/workplace-fairness/pregnancy-discrimination/reasonable-accommodations-for-pregnant-workers-state-laws.pdf [https://perma.cc/EY6M-XQU2] (collecting the pregnancy accommodation laws of twenty-three states, four cities, and the District of Columbia). These state laws, though, echo the FMLA’s prebirth exclusion of fathers by providing workplace accommodations only to pregnant women. For example, New Jersey requires that employers “of an employee who is a woman affected by pregnancy shall make available to the employee reasonable accom­modation in the workplace . . . for needs related to the pregnancy when the employee, based on the advice of her physician, requests the accom­modation.” 181 N.J. Stat. Ann. § 10:5-12(s) (West 2018) (emphasis added). This means that a pregnant woman who needs a workplace accommodation for antenatal depression or to attend a prenatal appoint­ment is entitled to one under these laws, while a father who needs the same accommodation for antenatal depression or to attend a prenatal appointment is not.

Expectant fathers can also be punished and even terminated from employment for pregnancy-related caregiving behaviors. The FMLA prohibits punishment or termination of both mothers and fathers for parenting behavior after birth. 182 See 29 C.F.R. § 825.112(a)(6) (2018). By contrast, the Pregnancy Discrimination Act (PDA), amending Title VII to protect against preg­nancy discrimination, by its own terms covers only “women.” 183 42 U.S.C. § 2000e(k) (2012) (“The terms ‘because of sex’ or ‘on the basis of sex’ include . . . because of or on the basis of pregnancy . . . ; and women affected by pregnancy . . . shall be treated the same for all employment-related purposes . . . as other persons not so affected but similar in their ability . . . to work . . . .” (emphasis added)). This assumes that only women face adverse consequences at work due to preg­nancy and excludes fathers from protection. A man who attends a prenatal obstetrician appointment and is fired by his employer for being seen as more committed to family than work has no cause of action under the PDA, even though such a termination is premised on the very type of sex stereotype that Title VII was meant to eradicate. If a woman is terminated for attending such an appointment, this may give rise to an inference of impermissible pregnancy discrimination. 184 See, e.g., Peralta v. Roros 940, Inc., 72 F. Supp. 3d 385, 394 (E.D.N.Y. 2014) (finding an inference of pregnancy discrimination when the defendant manager openly disapproved of the plaintiff mother’s attendance at prenatal appointments). The PDA has become an increasingly litigated area of federal law, with more judicial resources being funneled into developing a body of law that protects only expectant mothers. 185 See Joanna L. Grossman & Gillian L. Thomas, Making Pregnancy Work: Overcoming the Pregnancy Discrimination Act’s Capacity-Based Model, 21 Yale J.L. & Feminism 15, 17 n.3 (2009) (“The EEOC reports a forty percent increase in charges related to pregnancy between 1997 and 2007 . . . .”).

If an employer generally allowed women but not men to attend prenatal appointments, a man might have a claim for sex discrimination under Title VII. But the right that the PDA grants—to be treated the same as other nonpregnant but similarly situated employees—would not protect him. 186 See supra note 185 and accompanying text. And even a sex discrimination claim might be doomed by employers defending on the ground that they granted the pregnant woman leave only because they were legally mandated to do so under either the FMLA or the PDA, neither of which extends its protections to expectant fathers.

Note also the intersection between the facial sex classification of the PDA and sexual orientation and transgender discrimination. The sex-discriminatory language of the PDA may permit discrimination between expectant fathers and lesbian expectant (but nonpregnant) mothers, who arguably could be covered by the term “women affected by preg­nancy.” 187 42 U.S.C. § 2000e(k). And the PDA, which by its terms applies to pregnant “women,” would, like the FMLA, deny protection to the pregnant transgender man. 188 Id.; see also supra notes 177–181 and accompanying text.

c. Workplace Subsidies. — Federal law requires employers to provide equal health benefits to mothers and fathers after birth. 189 For a general discussion of the sex-neutral approach to health benefits after birth reflected in the Patient Protection and Affordable Care Act, see Danielle Garrett et al., Nat’l Women’s Law Ctr., Turning to Fairness: Insurance Discrimination Against Women Today and the Affordable Care Act 3–4 (2012), https://nwlc.org/wp-content/uploads/2015/08/nwlc_2012_turningtofairness_report.pdf [https://perma.cc/GQ5Z-KR9N]. Before birth, by contrast, federal law permits employers to provide bene­fits only to women. For example, the Medicare Prescription Drug, Improvement, and Modernization Act (also called the Medicare Modernization Act) created modern health savings accounts (HSAs). 190 For the relevant history, see Alyssa A. DiRusso, Charity at Work: Proposing a Charitable Flexible Spending Account, 2014 Utah L. Rev. 281, 307–14 & n.164. HSAs, like Flexible Spending Arrangements (FSAs), allow many employees to set aside pretax wages in a separate account created by their employer to pay for eligible healthcare expenses. 191 See id. at 307–09; Louise Norris, What Is the Difference Between a Medical FSA and an HSA?, healthinsurance.org (May 8, 2018), https://www.healthinsurance.org/faqs/what-is-the-difference-between-a-medical-fsa-and-an-hsa/ [https://perma.cc/S8A9-974H]. Which healthcare expenses qualify for this special treatment is a matter of federal law. 192 The most relevant rules derive from the Internal Revenue Code as well as regulations issued by the Department of Treasury. For the various terms, see I.R.C. §§ 67(b)(5), 68(c)(1), 213(a) (2012); Treas. Reg. § 1.213-1(a), (e)(1) (as amended in 1979).

Federal law allows mothers and fathers to use FSAs on equal terms for the expenses of postbirth carework (for example, to attend a parent­ing class or purchase diapers). 193 See DiRusso, supra note 193, at 312–13 (noting the Code’s authorization of dependent-care assistance, which extends care benefits to family members, including newborns). Before birth, only expenses that relate to the woman’s physical experience of being pregnant or being a mother, such as those for pregnancy-support clothing or learning nursing tech­niques, are covered. 194 See ADP, Spending Account Eligible Expense Guide 9 (2013), https://www.nafhealthplans.com/files/1614/7282/0926/FSA-AF_Spending_Account_Eligible_Expense_Guide.pdf [https://perma.cc/ZTQ8-CJT8] (noting that maternity girdles, special support hose, and lactation consultants are covered but that “[n]ew parents, newborn childcare classes, or sibling classes are not eligible”). Note how Cigna, a major insurance company, summarized the situation:
Childbirth classes—Expenses for childbirth classes are reimbursable, but are limited to expenses incurred by the mother-to-be. Expenses incurred by a “coach”—even if that is the father-to-be—are not reimbursable. To qualify as medical care, the classes must address specific medical issues, such as labor, delivery procedures, breathing techniques and nursing.
Which Expenses Are Eligible for HSA, FSA and HRA Reimbursement?, Cigna, http://www.cigna.com/qualified-health-care-expenses [https://perma.cc/4CR2-SAF6] (last visited Oct. 10, 2018).
Other expenses that are not related to the physi­cal experience of pregnancy but are related to analogous prebirth care­work, like strengthening one’s core muscles to prepare for carrying a baby or learning bottle-feeding techniques, 195 There are even classes for partners to learn how to support breastfeeding. See Classes for Expectant Parents, Breastfeeding Ctr., http://breastfeedingcenter.org/classes/
classes-for-expectant-parents [https://perma.cc/ZX8F-DJPP] (last visited Oct. 10, 2018) (“Breastfeeding for Fathers and Partners[:] . . . This much loved class reviews the basics of breastfeeding and discusses how to support the breastfeeding mother. Questions are encouraged! No mothers allowed! This class is intended for fathers, partners and other support persons of the breastfeeding mother.”).
are not covered. 196 ADP, supra note 197, at 9. So the only way an expectant father can take advantage of these types of benefits is to pay for the pregnant woman’s eligible expenses. 197 See id.

The Patient Protection and Affordable Care Act (ACA) generally requires health plans to offer benefits to new mothers and fathers on equal footing. 198 See Garrett et al., supra note 192, at 4 (noting the ACA’s prohibition on sex discrimination in insurance company–provided health plans). By contrast, in many contexts the ACA requires covered employers to provide coverage for prenatal benefits only to pregnant women. 199 Patient Protection and Affordable Care Act § 2713(a), 42 U.S.C. § 300gg-13 (2012). While the specific required prenatal benefits were to be deter­mined later by an expert panel, 200 See id. § 300gg-13(a)(4) (requiring that insurance plans provide “such additional preventive care and screenings . . . as provided for in comprehensive guidelines supported by the Health Resources and Services Administration for purposes of this paragraph”). the  ACA  itself  suggests  that  the  man­date  was  to  increase  pregnancy-related  benefits  only  “with  respect  to  women.” 201 Id. Eventual regulations mandated that covered employers offer insurance plans providing a host of pregnancy-related benefits, such as breastfeeding pumps, counseling for tobacco users, and prenatal educa­tion interventions, but only for pregnant women. 202 See id. § 1396d(bb)(1) (requiring coverage only “for cessation of tobacco use by pregnant women” (emphasis added)); id. § 1396o(a)(2)(B) (requiring coverage and barring cost sharing for, among other things, “services furnished to pregnant women, if such services relate to the pregnancy or to any other medical condition which may complicate the pregnancy, and counseling and pharmacotherapy for cessation of tobacco use by pregnant women” (emphases added)); The Henry J. Kaiser Family Found., Preventive Services Covered by Private Health Plans Under the Affordable Care Act tbl.1 (2015), http://files.kff.org/attachment/preventive-services-covered-by-private-health-plans-
under-the-affordable-care-act-fact-sheet [https://perma.cc/4X8V-2BL5].
Employer-offered insurance plans need not provide comparable benefits to expectant fathers.

While some of these benefits are properly sex specific, others are not. Take the counseling for tobacco users. Fathers who smoke during a pregnancy and are around the pregnant woman (“passive smoking”) can generate negative outcomes comparable in frequency and magnitude to expectant mothers who themselves smoke. 203 See Esther M. John et al., Prenatal Exposure to Parents’ Smoking and Childhood Cancer, 133 Am. J. Epidemiology 123, 127–29 (1991); David H. Rubin et al., Effect of Passive Smoking on Birth-Weight, 328 Lancet 415, 417 (1986). Yet the ACA requires insur­ance plans to provide support only for pregnant women to quit smok­ing. 204 See supra note 205. Still further, the limitation of these benefits to “pregnant women” again excludes transgender pregnant men, as well as nonpregnant expectant mothers.

III. Sexing Parenting by Sexing Pregnancy

The unsexed law of postbirth carework and the sexed law of prebirth carework present a deep tension in sex-equality law. Of course, not all of pregnancy can or should be unsexed. However, like the postbirth period, the prebirth period involves carework that can be disaggregated from sex.

This Part makes the case that carework in the pre- and postbirth con­texts are ripe for comparison and should be treated consistently. It first explains how the two circumstances are worthy of comparison because they both shape a common and core concern of sex-equality law: the sexed distributions of care- and market-work. Both the pregnancy and the period immediately after birth are foundational moments that shape actual investments at home and in the market, as well as employers’ expectations of these investments. The failure to unsex pregnancy there­fore undermines legal efforts to unsex parenting. And it does so by its effects not only on heterosexual couples but also on other parenting con­figurations. The sexed pregnancy privileges different-sex over same-sex couples by denying many benefits when both expectant parents are male and reinforces notions of sexed caregiving by how it distributes benefits to gay, lesbian, and transgender expectant parents.

This Part then discusses how the concern about intrusion on women’s bodily autonomy does not justify wholly distinct treatment of sexed carework in the prebirth period. Carework during pregnancy—whether done by the woman or the man—often does not involve the woman’s body, and in these circumstances it can be appropriately com­pared to carework after birth.

A. Pregnancy as Parenting

Doctrinal differences within sex-equality law before and after birth generate tension because these two areas of doctrine regulate similar social realities. These two periods define how sex operates at the “fault­line between work and family.” 205 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 738 (2003). Laws regulating pregnancy that pre­sume that the “mother is the ‘“center of home and family life”’” 206 Califano v. Westcott, 443 U.S. 76, 89 (1979) (quoting Taylor v. Louisiana, 419 U.S. 522, 534 n.15 (1975)). start in motion “a . . . cycle of discrimination that force[s] women to continue to assume the role of primary family caregiver, and foster[s] employers’ stereotypical views about women’s commitment to work and their value as employees.” 207 Hibbs, 538 U.S. at 736.

Before proceeding further, it is important to note that the compari­son between the doctrine of pregnancy and the doctrine of parenting is anything but perfect. Due to meaningful differences between pre- and postbirth carework, the law need not treat “the allocation of family duties” 208 Id. at 730. identically before and after birth. The Supreme Court has said that “[p]hysical differences between men and women” are “enduring” and that these differences can justify sex-based classifica­tion. 209 United States v. Virginia, 518 U.S. 515, 533 (1996); see also Hibbs, 538 U.S. at 731 (suggesting that the “differential physical needs of men and women” could justify differential treatment); Geduldig v. Aiello, 417 U.S. 484, 494–97 (1974) (noting that physical differences between the sexes can be the basis for permissible sex-based classifications). There are more physical differences based on sex before birth than after birth, justifying more sex-based distinctions during the pregnancy.

It is simply impossible to unsex many physical aspects of pregnancy. Only the pregnant woman carries the fetus. Only she constantly feels the fetus and experiences the emotional connection this generates. 210 The studies about the expectant mother feeling the fetus suggest that fetal movement can create higher maternal–fetal attachment. See Susan M. Heidrich & Mecca S. Cranley, Effect of Fetal Movement, Ultrasound Scans, and Amniocentesis on Maternal–Fetal Attachment, 38 Nursing Res. 81, 83 (1989); see also Nguyen v. INS, 533 U.S. 53, 65 (2001) (“The mother knows that the child is in being and is hers and has an initial point of contact with him. There is at least an opportunity for mother and child to develop a real, meaningful relationship.”); Caban v. Mohammed, 441 U.S. 380, 405 & n.10 (1979) (Stevens, J., dissenting) (“During that [pregnancy] period, the mother and child are together . . . . The father, on the other hand, may or may not be present . . . .”). Only she experiences many of the physical manifestations of pregnancy, including suffering morning sickness and the other health risks associated with pregnancy. 211 See, e.g., Dormeyer v. Comerica Bank–Ill., 223 F.3d 579, 583 (7th Cir. 2000) (discussing physical complications of pregnancy, including morning sickness); Troupe v. May Dep’t Stores Co., 20 F.3d 734, 735 (7th Cir. 1994) (same); Haywood L. Brown, Physical Changes During Pregnancy, Merck Manual, https://www.merckmanuals.com/home/women-s-health-issues/normal-pregnancy/physical-changes-during-pregnancy [https://perma.cc/4TTP-TK24] (last updated Nov. 2016) (discussing both normal and more troubling physical consequences of pregnancy, including chronic fatigue, headaches, bleeding, shortness of breath, and constipation); Feras H. Khan, Hyperemesis Gravidarum in Emergency Medicine, Medscape, https://emedicine.medscape.com/article/796564-overview [https://perma.cc/P5TK-Z27K] (last updated Jan. 8, 2016) (discussing morning sickness, which can create constant and durable nausea and vomiting). But see supra note 172 (discussing couvade, a condition in which men experience physical symptoms of a partner’s pregnancy). Only she endures the stares from others. 212 See Siegel, Reasoning, supra note 136, at 374–75 (“A woman may find that pregnancy comes to embody her social identity to others, who may treat her with love and respect or, alternatively, abuse her as a burden, scorn her as unwed, or judge her as unfit for employment.” (footnotes omitted)). Only she will have the many col­leagues, friends, and strangers approach her with desired and undesired pregnancy advice and with a request to touch her burgeoning belly. 213 See id. (discussing the social aspects of pregnancy). Only she decides whether to continue the pregnancy. 214 See, e.g., Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 896–98 (1992) (invalidating a spousal notification requirement because it gave the husband an “enforceable right to require a wife to advise him before she exercises her personal choices”). This dimension of the pregnant woman’s experience runs in the other direction too: Only her behaviors affect the fetus on a constant basis. 215 For the best overview of the scientific literature on this point, see generally Annie Murphy Paul, Origins: How the Nine Months Before Birth Shape the Rest of Our Lives (2010) (exploring the burgeoning field of “fetal origins,” which studies how influences outside the womb during pregnancy can shape the physical, mental, and emotional well-being of the developing fetus for the rest of its life).

While these physical differences mean that laws regulating preg­nancy can be more sexed than those regulating parenting after birth, it does not render these areas of law entirely distinct. The goal of sex-equality law is to—as the Court put it in Hibbs—eliminate “the pervasive sex-role stereotype that caring for family members is women’s work.” 216 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 731 (2003); see also id. at 734 (“Congress could reasonably conclude that such discretionary family-leave programs would do little to combat the stereotypes about the roles of male and female employees that Congress sought to eliminate.”). The law of the sexed pregnancy encourages pregnant women to make investments in caregiving during the pregnancy and discourages expectant fathers from doing so in a way that persists after birth, regard­less of efforts to unsex after birth. The law of the sexed pregnancy also encourages employers to stereotype women as committed to care and men as committed to career in a way that persists after birth, regardless of any efforts to unsex after birth.

The Court in Hibbs recognized the foundational importance of the moments after birth for developing caregiving identities and attach­ments. 217 See supra notes 81–84 and accompanying text. The moments before birth are equally important. An entire field of “prenatal attachment studies” has shown that it is during the pregnancy that expectant parents first start to love their future child “both as an extension of self and as an independent object.” 218 See Brandon et al., supra note 5, at 202–03. The attachments generated—or not generated—during the pregnancy pre­dict later parental identity and involvement. If an expectant parent approaches the pregnancy without any change in self-conception or social role, this predicts lesser involvement after birth. 219 See Condon, supra note 5, at 280–82 (speculating that an expectant parent’s noninvolvement with the fetus could lead to deliberate avoidance of active parenting in the postnatal period). If,   on  the  other  hand,   an  expectant  parent  begins  to  think  of  herself  or  himself  as  a  par­ent  during  the  pregnancy,    that  predicts  greater  involvement  after  birth. 220 See id.

Engaging in prebirth carework is especially important for this iden­tity shift in expectant fathers precisely because they do not experience the physical aspects of pregnancy that help generate this identity shift in pregnant women. 221 See Nguyen v. INS, 533 U.S. 53, 65 (2001) (stating that mothers are more likely “to develop a real, meaningful relationship” with a newborn because of their physical experience of being pregnant). A range of prebirth carework that can involve the expectant father, such as attending the first ultrasound, selecting a crib, or finding a childcare center, may be significant motivating experiences. 222 Cf. Brandon et al., supra note 5, at 206 (summarizing a study finding that lack of interest on the part of husbands can delay expectant mothers’ feelings of attachment for their unborn child).

While cases like Hibbs identified the importance of unsexing care­work to combat stereotypes after birth, these efforts are undermined by a sexed law of pregnancy that provides pregnant women with greater incentives than expectant fathers to invest in carework during the preg­nancy. 223 See supra section II.B.2. Federal and state laws protect pregnant women but not expectant fathers from risking their job or other work penalties for tend­ing to prenatal matters. 224 See supra sections II.B.2.a–.b. Federal law reduces the cost of engaging in prebirth carework for pregnant women but not expectant fathers. 225 See supra section II.B.2.c.

The absence of any meaningful legal recognition of the idea that the expectant father should engage in the pregnancy has an even deeper and more pervasive cost. The idea that the expectant father should participate substantially in the pregnancy is considered “off-the-wall.” 226 See generally Jack M. Balkin, How Social Movements Change (or Fail to Change) the Constitution: The Case of the New Departure, 39 Suffolk U. L. Rev. 27, 52 (2005) (noting how certain legal claims are considered so outside of the jurisprudential mainstream as to be considered “off-the-wall”). Because such an idea is nowhere in law, it is (almost) nowhere in social life either. 227 One can find stray discussions, see, e.g., Renee Bacher, Prenatal Visits: Should Dads Go Too?, Parenting.com, https://www.parenting.com/article/prenatal-visits-should-dads-go-too [https://perma.cc/AMW3-2MHR] (last visited Oct. 10, 2018), but not the type of major public focus that paternal involvement in postbirth caregiving has received. This means that an expectant father who does participate in the pregnancy can risk severe stigma. 228 See Joan C. Williams et al., Cultural Schemas, Social Class, and the Flexibility Stigma, 69 J. Soc. Issues 209, 220–21 (2013) (discussing the stigma men face from their employers when they are involved in postbirth caregiving and have acted counter to gender-role expectations). So not only does the law fail to disrupt this stigma, but it might be held responsible for it.

When laws do try to encourage father involvement during the preg­nancy, they do so in a way that reflects and reinforces the separate-spheres mentality. The FMLA, for instance, justifies limiting most prena­tal leave to the expectant mother by stating that it is the expectant father’s role to “care for” the pregnant woman. 229 29 C.F.R. § 825.120(a)(5) (2018) (“A spouse is entitled to FMLA leave if needed to care for a pregnant spouse who is incapacitated or if needed to care for her during her prenatal care, or if needed to care for her following the birth of a child if she has a serious health condition.”). This is reflective of outdated sex stereotypes envisioning a pregnant woman dependent on her male partner, and the male partner himself without need for sup­port. 230 See Weinberger v. Wiesenfeld, 420 U.S. 636, 643 (1975) (striking down a sex classification for perpetuating the harmful stereotype that wives were dependent on husbands but husbands were not dependent on wives); Schlesinger v. Ballard, 419 U.S. 498, 507 (1975) (“In Frontiero, the assumption . . . was that female spouses of servicemen would normally be dependent upon their husbands, while male spouses of servicewomen would not.”). The  FMLA  does  include  “psychological  comfort  and  reassur­ance”  within  the  support  men  can  provide. 231 29 C.F.R. § 825.124(a). But,   critically,   the  expectant  father  himself  is  not  identified  as  needing  this  same  comfort  and  reassurance in return. 232 See supra notes 173–174 and accompanying text. The expectant father envisioned by the FMLA is the old-fashioned masculine man that Professor Case identified in her seminal article: the man who does not need emotional support from other people. 233 See Case, Disaggregating, supra note 1, at 12 (discussing the adjectives that are conventionally associated with masculinity, including “independent,” “individualistic,” and “self-sufficient”). While this view of the father may seem outdated, the law of the sexed pregnancy helps ensure that it will live on. 234 See supra notes 50–66 and accompanying text (discussing the self-reinforcing nature of sex-stereotypical laws).

The sexed law of pregnancy also undermines sex equality by sending powerful messages to third parties. The law generates the impression in employers that women will invest relatively more in the home and rela­tively less in work than men, fueling statistical discrimination against women at work. 235 See Gillian Lester, A Defense of Paid Family Leave, 28 Harv. J.L. & Gender 1, 24 (2005) (“The fact that women in this age group are more likely than their male counterparts to interrupt work for caregiving purposes may serve as a proxy for an actual finding of a weak work commitment of a particular worker.”). Even before a woman is visibly pregnant, she will often face employer discrimination because she is of an age or in a life situa­tion (for example, recently married) suggesting that she will imminently become pregnant. 236 See Michael Selmi, Family Leave and the Gender Wage Gap, 78 N.C. L. Rev. 707, 735 (2000) (“[T]he patterns [of exiting the labor force] for women differ significantly from those of men, and employers may be taking these differences into account in their employ­ment decisions.”). For this reason, women may conceal their pregnancies or their marital status during job interviews. See Lysak v. Seiler Corp., 614 N.E.2d 991, 992–93 (Mass. 1993) (holding that an employer’s termination of an employee for lying about her pregnancy during a job interview did not violate a state pregnancy-discrimination law); Alissa Quart, Opinion, Why Women Hide Their Pregnancies, N.Y. Times (Oct. 6, 2012), https://
www.nytimes.com/2012/10/07/opinion/sunday/why-women-hide-their-pregnancies.html (on file with the Columbia Law Review).
Once an employer knows that a woman is pregnant, the employer will presume that she has substantial obligations that will reduce her investment in work. In addition to the physical fact of pregnancy, this discrimination is based in the belief that prebirth carework—visiting childcare centers, attending childbirth classes, and the like—is the responsibility of the mother. 237 Cf. Jane A. Halpert et al., Pregnancy as a Source of Bias in Performance Appraisals, 14 J. Organizational Behav. 649, 655 (1993) (finding that employers discriminate against pregnant women not only because they view them as “physically limited,” “overly emotional,” and “often irrational” but also because they view them as “less than committed to their jobs” and not “dependable employees”). Laws such as the FMLA that facially discriminate during the pregnancy by giving women more benefits than men only make women more expensive to hire 238 See Christine Jolls, Accommodation Mandates, 53 Stan. L. Rev. 223, 264 (2000) (“[T]he Family and Medical Leave Act (FMLA) increases the average cost of employing disabled and female employees . . . .”); Naomi Schoenbaum, The Case for Symmetry in Antidiscrimination Law, 2017 Wis. L. Rev. 69, 91–98 [hereinafter Schoenbaum, Case for Symmetry] (arguing that this perverse statistical-discrimination effect is one of the reasons that symmetrical laws are so important). —exactly the opposite of the FMLA’s goal. 239 See 29 U.S.C. § 2601(a)(6) (2012) (finding that sex-specific family leave policies generated “serious potential for encouraging employers to discriminate against [women] employees and applicants”); Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 736 (2003) (explaining that Congress enacted the FMLA because laws that provided family leave only to women reflected and reinforced “employers’ stereotypical views about women’s commitment to work and their value as employees”).

On the flip side, employers view pregnancy as imposing no burdens on an expectant father’s work. Indeed, there is a “fatherhood premium” that expectant fathers start to enjoy during the pregnancy. 240 See Jane Waldfogel, Understanding the “Family Gap” in Pay for Women with Children, 12 J. Econ. Persp. 137, 143 (1998) (explaining that while women with children earn less than other women in the United States, “[t]here is no such family penalty for men” and in fact “married men, most of whom have children, earn more than other men”). Employers may actually prefer expectant fathers as compared to men not expecting children because they assume that expectant fathers will intensify their breadwinning efforts 241 See Shelly Lundberg & Elaina Rose, Parenthood and the Earnings of Married Men and Women, 7 Lab. Econ. 689, 705–06 (2000) (“Fatherhood leads to a 9% increase in wages for men . . . .”). —but only if they conform to this expected sex role. If the expectant father takes on too much carework, he is likely to experience a “flexibility stigma.” 242 See, e.g., Jennifer J. Berdahl & Sue H. Moon, Workplace Mistreatment of Middle Class Workers Based on Sex, Parenthood, and Caregiving, 69 J. Soc. Issues 341, 345–48 (2013) (examining how employers punish “conspicuous” fathers); Melissa J. Hodges & Michelle J. Budig, Who Gets the Daddy Bonus? Organizational Hegemonic Masculinity and the Impact of Fatherhood on Earnings, 24 Gender & Soc. 717, 740–42 (2010) (measuring who gets the fatherhood premium). Researchers have found that expectant fathers are less likely to take parental leave when they work in male-dominated workplaces or workplaces in which other fathers have tended not to take parental leave. 243 See Magnus Bygren & Ann-Zofie Duvander, Parents’ Workplace Situation and Fathers’ Parental Leave Use, 68 J. Marriage & Fam. 363, 368–70 (2006) (reporting data on how features of a father’s workplace may affect the length of his parental leave). The sexed law of pregnancy plays a key role in this. The expectant father who asks for prenatal accommo­dations is asking for something excessive under current law, which does not require such accommodation and renders even discretionary accom­modations for expectant fathers off the wall.

State interventions to remedy sex inequalities after birth will inevita­bly be too little because they are too late. Patterns of gendered behavior develop during the pregnancy that are hard to reverse. This is because individuals struggle to learn new skills when they are exhausted and overwhelmed. 244 See Lee Anne Fennell, Willpower Taxes, 99 Geo. L.J. 1371, 1390 (2011) (citing studies that conclude that “willpower works like a muscle that can become fatigued with use”). Add to that the likelihood that the pregnant woman pro­bably already has some of these skills, 245 Expectant mothers are more likely to have caregiving experience than expectant fathers. See Olga Khazan, Even in Babysitting, Men Make More than Women, Atlantic (Mar. 14, 2014), https://www.theatlantic.com/business/archive/2014/03/even-in-babysitting-men-make-more-than-women/284426 [https://perma.cc/GL54-2U22]. as  well  as  the  substantial  tran­saction  costs  involved  in  teaching  these  skills  to  the  expectant  father,  246 For the seminal treatment, see generally R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1, 15–19 (1960) (discussing the importance of transaction costs in studying legal systems). and it is clear that asking a new father to reconceive his domestic role during the first few months with a newborn is a difficult proposition.

These  mechanisms  of  sex  inequality  during  pregnancy  have  conse­quences  for  market  work. Because   pregnant  women  invest  relatively more  in  caregiving  than  expectant  fathers  from  the  beginning  of  the  preg­nancy,   they invest relatively less in market work during this time. It is not surprising, then, that the motherhood penalty women face at work starts to accumulate during the pregnancy, rather than after the birth of the child. 247 See Charles L. Baum, The Effect of Work Interruptions on Women’s Wages, 16 Labour 1, 25–30 (2002) (describing how the expectant mother might switch firms and sacrifice her firm-specific human capital); Lester, supra note 238, at 22–25 (identifying mechanisms generating inequality, including “foregone work experience, as well as depreciation of accumulated skills during the periods of [work] interruption”); Jennifer Bennett Shinall, The Pregnancy Penalty, 103 Minn. L. Rev. 749, 787–803 (2018) (presenting empirical evidence that pregnant women suffer a greater penalty on the labor market than mothers do). See generally Joni Hersch & Leslie S. Stratton, Housework, Wages, and the Division of Housework Time for Employed Spouses, 84 Am. Econ. Rev. 120, 124 (1994) (noting the “vicious cycle” by which women’s increased investment in domestic work increases the earnings differential between the sexes). The contrast with men is significant. Men start to work longer hours during the pregnancy. 248 See Lundberg & Rose, supra note 244, at 705–06. By the time the child is born, then, women have already lowered their market wages relative to men. As a result, fathers face a higher opportunity cost for spending time with their children than mothers do. 249 See Christine A. Littleton, Does It Still Make Sense to Talk About “Women”?, 1 UCLA Women’s L.J. 15, 36 (1991) (“Because men’s wages continue to be significantly higher than women’s, it would be economically rational for a married woman to take the leave alone, as she would forgo a smaller income than her husband would, and his larger income would continue.”).

The sexed pregnancy undermines legal efforts to unsex parent­ing not only by its effects on heterosexual couples but also by its effects on gay, lesbian, and transgender expectant parents. In Obergefell v. Hodges, the Supreme Court decided that it was unconstitutional to privi­lege different-sex over same-sex couples when it comes to many of the most important parts of parenting. 250 135 S. Ct. 2584, 2600–01 (2015); see also Pavan v. Smith, 137 S. Ct. 2075, 2077–79 (2017) (invalidating a state law treating birth certificates differently for different-sex as compared with same-sex married couples). For instance,   the  Court  listed  “birth . . . certificates[ ] . . . and  child  custody,   support,   and  visitation”  as  “aspects  of  marital  status”  that  needed  to  be  open  equally  to  different- and same-sex  parents. 251 Obergefell, 135 S. Ct. at 2601; see also Brooke S.B. v. Elizabeth A.C.C., 61 N.E.3d 488, 504 (N.Y. 2016) (Pigott, J., concurring) (“Same-sex couples are now afforded the same legal rights as heterosexual couples and are no longer barred from establishing the types of legal parent–child relationships that the law had previously disallowed.”). The constitutional mandate to equalize differ­ent- and same-sex parents in the period after birth presumably extends in substantial part to the period before birth as well.

The sexed pregnancy’s focus on the pregnant woman as the holder of benefits and protections violates this mandate. Sexed pregnancy regulations exclude all gay men expecting children because of their sex. So whereas the law of the sexed pregnancy grants protections and bene­fits to one expectant parent in heterosexual couples, it affords these same privileges to no one in gay-male couples. Under the sexed preg­nancy, gay-male couples are inadequate parents, not only because of their sexual orientation, but also because their family is missing an appropriate caregiver—a woman. This reinforces the constitutionally suspect stereotype that caring is women’s work, while also reinforcing the constitutionally suspect second-class status of same-sex families. 252 See United States v. Windsor, 133 S. Ct. 2675, 2694 (2013) (noting the importance of ensuring recognition for same-sex parents because their families must “understand the integrity and closeness of their own family and its concord with other families in their community”). And by excluding the transgender pregnant man from its benefits and protec­tions, the sexed pregnancy rejects the idea of men as caregivers, even when they are situated in precisely the same circumstance as women. 253 See generally City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439 (1985) (noting that equal protection requires “that all persons similarly situated should be treated alike”).

As for lesbian women, some of the laws discussed apply only to the pregnant woman herself, excluding nonpregnant expectant women just like they exclude expectant fathers. Other laws, according to their lan­guage, could be read to apply to nonpregnant expectant women. 254 See, e.g., supra note 190 and accompanying text (noting ambiguity about whether laws that grant rights to “expectant mothers” or “women affected by pregnancy” would apply in these situations). To the extent that any of these laws cover nonpregnant expectant mothers but not nonpregnant expectant fathers, they treat persons differently solely on the basis of sex, even when they are similarly situated. In so doing, they reinforce the constitutionally infirm notion that caring is a woman’s domain.

B. Pregnancy and Autonomy

Another fundamental consideration is present before birth that is not present after birth: the sex-equality implications of reproductive rights. 255 See Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 895–96 (1992) (“If these cases concerned a State’s ability to require the mother to notify the father before taking some action with respect to a living child . . . , it would be reasonable to conclude . . . that the father’s interest in the welfare of the child and the mother’s interest are equal.”). While this Article’s earlier discussions focused on how sex equal­ity can be undermined by too little parental involvement during the preg­nancy, reproductive rights jurisprudence has focused on how sex equality can be undermined by too much involvement. 256 See id. at 869 (identifying “the urgent claims of the woman to retain the ultimate control over her destiny and her body, claims implicit in the meaning of liberty”). Men’s involvement in pregnancy so starkly raises concerns about interference with women’s autonomy, and thus equality, due to one of the basic facts of pregnancy: It occurs in the woman’s body. 257 See supra note 97 and accompanying text.

Do the two different regulatory regimes identified in the first two Parts present a doctrinal distinction with a difference because of this? This section argues that carework before and after birth can be com­pared because sex can be disaggregated from pregnancy (like sex can be disaggregated from parenting). Disaggregating sex from pregnancy involves recognizing the range of caregiving tasks that expectant fathers can perform that do not compel the expectant mother to do anything at all, let alone anything related to her body.

Reproductive rights jurisprudence can certainly be understood to mean that deciding whether to carry a pregnancy to term is the defining sex-equality consideration during pregnancy. Because giving birth involves a woman’s body, and because it defines her for the rest of her life, 258 See Casey, 505 U.S. at 869 (noting how reproductive rights give a pregnant woman control “over her destiny”). it could raise sex-equality considerations different from those after birth. The Supreme Court has indicated that reproductive rights are particularly important rights for gender equality. 259 See id. at 856 (“The ability of women to participate equally in the economic and social life of the Nation has been facilitated by their ability to control their reproductive lives.”). If the woman’s right to choose is a trump card during the pregnancy period, then any sex-equality benefits of encouraging paternal involvement in the pregnancy must give way to the woman’s overriding reproductive rights.

Concededly, this is a plausible reading of the doctrine; it is not, how­ever, the best reading. Since Planned Parenthood of Southeastern Pennsylvania v. Casey, the test for protecting women’s reproductive auton­omy has been whether the government’s actions in purpose or effect place an “undue burden” on the woman’s right to choose whether to carry a pregnancy to term. 260 See id. at 878 (“Unnecessary health regulations that have the purpose or effect of presenting a substantial obstacle to a woman seeking an abortion impose an undue burden on the right.”). This is true regardless of the law’s goal. 261 See Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2309 (2016) (“The rule announced in Casey . . . requires that courts consider the burdens a law imposes on abortion access together with the benefits those laws confer.”); see also Linda Greenhouse & Reva B. Siegel, Casey and the Clinic Closings: When “Protecting Health” Obstructs Choice, 125 Yale L.J. 1428, 1432–33 (2016) (“We argue that Casey requires scrutiny of health-justified restrictions to ensure that they actually and effectively advance health-related ends and do not protect potential life in a manner the Constitution prohibits. . . . The undue burden framework is the gateway for making these determinations.”). The “purpose” 262 See Casey, 505 U.S. at 878 (noting that part of the doctrinal test involves an examination of the “purpose” of the law). of unsexing the social experience of pregnancy is to benefit pregnant women by creating a more equal distribution of care­work. The next question under Casey is whether the “effect” of the policy is substantially burdensome on the expectant mother. 263 Id. But the primary reason for—and likely effect of—unsexing the social experience of pregnancy is to reduce the burden on pregnant women. 264            See supra notes 247–252 and accompanying text (describing the mechanisms by which the sexed pregnancy contributes to sex inequality).

In Casey and other cases, the Court has also considered—even more directly—when third-party involvement in pregnancy interferes with reproductive rights. 265 See, e.g., Bellotti v. Baird, 443 U.S. 622, 642–43 (1979) (holding that if a state requires parental consent for minors seeking abortion, it must also provide an alternative procedure for authorizing an abortion); Planned Parenthood of Cent. Mo. v. Danforth, 428 U.S. 52, 74 (1976) (striking down a blanket ban on abortion for unmarried minors during the first twelve weeks of pregnancy absent parental consent). In Casey, the Court considered the constitutional­ity of a Pennsylvania statute providing that “except in cases of medical emergency, . . . no physician shall perform an abortion on a married woman without receiving a signed statement from the woman that she has notified her spouse that she is about to undergo an abortion.” 266 Casey, 505 U.S. at 887. The Court invalidated spousal notification because it gave the expectant father “dominion over his wife.” 267 Id. at 898. Note that after Obergefell this concern and the constitutional right that protects against such spousal control could eventually extend to all spouses, regardless of sex. 268 See Obergefell v. Hodges, 135 S. Ct. 2584, 2604–05 (2015). Admittedly, there are features of Casey suggesting that fathers pose unique risks to women’s autonomy because of the prevalence of domestic violence and because of the problematic history of husbands’ authority over wives. See Casey, 505 U.S. at 891 (noting the risks posed by “male partners”); id. at 898 (noting the “troubling degree of authority over his wife” that husbands enjoyed at common law).

Casey might be read to distinguish paternal involvement in the pre- and postbirth period because of the concern about bodily autonomy. The Court states that if it were considering a law that

require[d] the mother to notify the father before taking some action with respect to a living child raised by both, . . . it would be reasonable to conclude as a general matter that the father’s interest in the welfare of the child and the mother’s interest are equal.

Before birth, however, the issue takes on a very different cast. It is an inescapable biological fact that state regulation with respect to the child a woman is carrying will have a far greater impact on the mother’s liberty than on the father’s. 269 Casey, 505 U.S. at 895–96.

A closer reading reveals that the concern the Court had about paternal involvement in pregnancy is not about paternal involvement in pregnancy per se. Indeed, the Court suggests the importance of paternal involvement even during the pregnancy. 270 See id. at 895 (recognizing “that a husband has a ‘deep and proper concern and interest . . . in his wife’s pregnancy and in the growth and development of the fetus she is carrying’” (quoting Danforth, 428 U.S. at 69)). Not all paternal involvement raises the same concerns as the spousal-notification provision in Casey. The statute there gave the expectant father a right to compel the expectant mother to do something—notify the husband of her situation 271 See id. at 898 (noting that the problem is that the statute gives the husband an “enforceable right to require a wife to advise him before she exercises her personal choices”). —and to do something related to her body. 272 See id. at 896 (indicating concern when paternal involvement implicates “the very bodily integrity of the pregnant woman”); id. at 898 (arguing that if the law were upheld, it could also permit the state to “require a married woman to notify her husband before she uses a postfertilization contraceptive” or “before drinking alcohol or smoking”); see also Danforth, 428 U.S. at 71 (“[W]hen the wife and the husband disagree on [a pregnancy] decision, . . . only one . . . can prevail. Inasmuch as it is the woman who physically bears the child and who is the more directly and immediately affected by the pregnancy, as between the two, the balance weighs in her favor.” (emphasis added)). Disaggregating sex from pregnancy involves recognizing that the expectant father can participate in the pregnancy independently of the pregnant woman in a variety of ways, including buying a carseat, selecting a pediatrician, and visiting a childcare center. Once we disaggregate sex from pregnancy, we can recognize the host of ways that paternal involvement in pregnancy has nothing to do with the woman’s body, thus rendering carework in the pre- and postbirth periods ripe for compari­son. To the extent that paternal involvement during pregnancy does impli­cate the woman’s body, such as attending a prenatal appointment, the doctrine appropriately circumscribes his involvement such that it does not trouble Casey or the right to autonomy, as discussed in the next Part.

IV. Unsexing Pregnancy

The first three Parts of this Article identified a substantial tension in sex-equality law. Courts and commentators have focused significant atten­tion on disaggregating sex from parenting because no physical sex differ­ences dictate sex roles in parenting. But they have failed to do so during pregnancy, even though there are many facets of getting ready to parent during pregnancy that have nothing to do with physical sex differ­ences. This Part considers what a doctrine unraveling this tension would look like. It also considers how sex-equality law could unsex pregnancy to make the law of pregnancy consistent with the law of parenting. This Part focuses on doctrinal reconstruction, considering how existing sex-equal­ity law already relies on unsexing arguments and how these jurispruden­tial structures can construct a sex-equality law that unsexes pregnancy.

This Part first considers how sex-equality law can extend heightened scrutiny for sex classifications to the pregnancy period. Cases like Geduldig v. Aiello 273 417 U.S. 484 (1974). do not close the doctrinal door on this; rather, the argument is that cases like Geduldig opened the door to closer scrutiny of precisely the sort of invidious discrimination involved in sexing preg­nancy. This Part then considers how unsexing pregnancy could enhance sex equality and autonomy rights—consistent with reproductive rights jurisprudence—rather than force doctrine to choose between the two.

A. Scrutinizing Pregnancy

The doctrinal tension this Article has identified in the law of sex equality is a lack of comparable scrutiny of sex classifications regulating parenting in the pre- and postbirth periods. Sex-equality law could inva­lidate regulations in the pregnancy period that are based on “invidious” sex stereotypes related to parenting, like the one struck down in Weinberger v. Wiesenfeld. 274 See 420 U.S. 636, 651–53 (1975) (invalidating a provision of the Social Security Act permitting widows but not widowers to collect certain benefits). An even-handed application of equal protection in the pre- and postbirth periods would achieve the goals of unsexing parenting by freeing both women and men from the “very stereotype the law condemns.” 275 J.E.B. v. Alabama ex rel. T.B., 511 U.S. 127, 138 (1991) (internal quotation marks omitted) (quoting Powers v. Ohio, 499 U.S. 400, 410 (1991)). Cases like Hibbs have already started to move equal protection doctrine in this direction after birth. Cases like Geduldig should be read to mean not that pregnancy discrimination cannot be sex discrimination but that pregnancy discrimination consti­tutes sex discrimination only when founded on sex stereotypes.

1. Constructing Scrutiny. — Scrutinizing sex classifications has resulted in substantial progress in undoing separate-spheres thinking regarding women’s and men’s roles in the family and the market after a child is born by ridding the law of sex classifications that are not based in physical sex differences. Applying heightened scrutiny to sex classifica­tions in the pregnancy period could achieve the same success. Extending scrutiny to the context of pregnancy would require courts to distinguish between when sex classifications constitutionally regulate on the basis of physical sex differences and when they unconstitutionally regu­late on the basis of sex stereotypes. The law would consider who is simi­larly situated with regard to pregnancy in light of the goals of the regulation, regardless of sex. 276 See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439 (1985) (explaining that the Equal Protection Clause “is essentially a direction that all persons similarly situated should be treated alike”). The question would essentially be one of means–ends fit. The classification would have to be “substantially related” to the physical difference. 277 United States v. Virginia, 518 U.S. 515, 524 (1996) (stating that sex classifications must “serve[ ] important governmental objectives” and “the discriminatory means employed [must be] substantially related to the achievement of those objectives” (internal quotation marks omitted) (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 724 (1982))). If the classification were extended beyond what is needed to address the physical difference, it would be “irrational” and therefore an unconstitutional sex stereotype. 278 See Nguyen v. INS, 533 U.S. 53, 68 (2001) (“[T]he difference does not result from some stereotype, defined as a frame of mind resulting from irrational or uncritical analysis.”).

This doctrinal approach is consistent with postbirth jurisprudential trends that scrutinize sex discrimination even in the face of physical differences. The unsexing project made a big leap forward with the Court’s recognition in United States v. Virginia that sex classifications can be unconstitutional even when based on valid physical differences. 279 See Franklin, supra note 50, at 145–46 (explaining that Virginia’s “brand of analysis was familiar” but that “the Court’s treatment of the issue of ‘real’ differences marked a new departure for constitutional sex discrimination doctrine”). Before this, the equal protection treatment of sex classi­fications had been either–or: If there were a physical difference related to the law, it would justify the classification; there was no need to resort to a stereotyping analysis. 280 Id. In Virginia, even though physical differences between the sexes were relevant to the training afforded by the Virginia Military Institute, this did not end the matter. Instead, the Justices still assessed whether the sex classification was based in an “overbroad gener­alization[ ].” 281 Virginia, 518 U.S. at 533. So whereas “‘real’ differences [once] served as a check on the reach of anti-stereotyping doctrine,” after Virginia, the “anti-stereotyping doctrine serves as a check on the state’s regulation of ‘real’ differences.” 282 Franklin, supra note 50, at 145–46.

Hibbs extended this doctrinal innovation to the context of a physical sex difference that arises out of women’s capacity to become pregnant and bear children. 283 See id. at 149–54. Even though there are physical differences between the sexes when it comes to recovery from childbirth, the Court still care­fully scrutinized parental leave policies to ensure that any sex classifica­tion was fully supported by this difference and not sex stereotypes. 284 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 730, 733 n.6 (2003) (“States continue to rely on invalid gender stereotypes in the employment context, specifically in the administration of leave benefits. Reliance on such stereotypes cannot justify the States’ gender discrimination in this area.”). Unsexing pregnancy at the level of constitutional doctrine would require extending this recognition to the period of pregnancy itself: Laws that classify on the basis of sex might actually be based in stereotypes rather than physical differences between the sexes.

Although Hibbs marks progress toward recognizing heightened scru­tiny for sex stereotypes in the presence of physical sex differences, it did not go as far as it should—that is, to apply such scrutiny during the preg­nancy itself. Some have argued to the contrary—that Hibbs marks the first time the Court has recognized that pregnancy regulations can violate the Equal Protection Clause. 285 See Franklin, supra note 50, at 154 (“Hibbs teaches that pregnancy discrimination can constitute sex discrimination in instances in which it reflects and reinforces traditional conceptions of women’s sex and family roles.”); Siegel, Long Way, supra note 81, at 1873 (proclaiming Hibbs “the first Supreme Court opinion to recognize that laws regulating pregnant women can enforce unconstitutional sex stereotypes”). For Professor Siegel, the Court’s recog­nition in Hibbs that laws granting maternity leave beyond the period of physical disability after childbirth were constitutionally suspect was an instance of the Court treating a pregnancy regulation as unconstitutional sex discrimination. 286 See Siegel, Long Way, supra note 81, at 1891–92 (arguing that Hibbs makes clear that pregnancy regulations could amount to an unconstitutional sex classification under certain circumstances). But  Hibbs  did  not  address  sex  roles  during  preg­nancy. In  assessing  the  constitutional  status  of  sex-differentiated  postbirth  leave  policies  (that is,   maternity  leave  and  paternity  leave),   Hibbs  applied  the  already-established  principle  that  laws  may  not prescribe sex roles after a child is born when these roles are not justified by physical sex differ­ences. 287 The fact that it applied this scrutiny from the moment of birth is no advance over Weinberger. See supra note 49 and accompanying text. And even when it came to the period immediately following birth, the Court stopped short of its standard treatment of sex classifications. The Court suggested that medical evidence showing that women on average need four to eight weeks to recover from childbirth would justify granting this amount of leave to women as a class, even though some women in the class would not need this much leave. See Hibbs, 538 U.S. at 731 & n.4. This violates the perfect proxy rule since only some but not all women need even four weeks to recover from pregnancy. See supra notes 68–71 and accompanying text; see also Case, Very Stereotype, supra note 68, at 1450 (“[V]irtually every sex-respecting rule struck down by the Court in the last quarter century embodied a proxy that was overwhelmingly, though not perfectly, accurate.”). Given that administrative convenience cannot justify sex discrimination, a policy granting all women postbirth leave based on the average amount of recovery time women need should fail under heightened scrutiny. See, e.g., Frontiero v. Richardson, 411 U.S. 677, 690–91 (1973) (collecting cases on this point and holding that a law treating men and women differently “for the sole purpose of achieving administrative convenience” must fail). This narrower interpretation of Hibbs is only strengthened by the fact that, several years after Hibbs, the Court affirmed a federal statute featuring sex-role stereotypes related to pregnancy. 288 See Nguyen v. INS, 533 U.S. 53, 68 (2001). There, the Court upheld a law that treated mothers and fathers differently with regard to conferring citizenship on their children born abroad because “recognition that at the moment of birth . . . the mother’s knowledge of the child and the fact of parenthood have been established in a way not guaranteed in the case of the unwed father . . . is not a stereotype.” See id.

Importantly, too, approaching the Equal Protection Clause in the manner advocated here is not foreclosed by Geduldig and is even sug­gested by it. Courts and scholars disagree about the holding of Geduldig. Some have read the case to categorically exclude pregnancy classifica­tions from ever being sex classifications that would be subject to height­ened scrutiny under the Fourteenth Amendment. 289 See Neil S. Siegel & Reva B. Siegel, Struck by Stereotype: Ruth Bader Ginsburg on Pregnancy Discrimination as Sex Discrimination, in The Legacy of Ruth Bader Ginsburg (Scott Dodson ed., 2015) (“Geduldig is commonly read as holding that discrimination against pregnant women can never be sex discrimination.”). Others have read the case more narrowly, as leaving open the possibility that pregnancy regulations could be sex regulations that would receive heightened scrutiny. 290 See, e.g., Bray v. Alexandria Women’s Health Clinic, 506 U.S. 263, 327 (1993) (Stevens, J., dissenting) (“Nor should Geduldig be understood as holding that, as a matter of law, pregnancy-based classifications never violate the Equal Protection Clause.”).

The text of the opinion itself makes quite clear that Geduldig held only that the pregnancy regulation under consideration there was not an impermissible sex classification, not that a pregnancy regulation could never be an impermissible sex classification. 291 See Geduldig v. Aiello, 417 U.S. 484, 496 n.20 (1974) (“[N]ot . . . every legislative classification concerning pregnancy is a sex-based classification . . . .); see also Bray, 506 U.S. at 327 (“In fact, as the language of the opinion makes clear, what Geduldig held was that not every legislative classification based on pregnancy was equivalent, for equal protection purposes, to the explicitly gender-based distinctions struck down in Frontiero v. Richardson and Reed v. Reed.” (citation omitted)). Moreover, Geduldig states that under certain circumstances pregnancy classifications could be pre­texts to effect invidious sex discrimination, and that in such circum­stances pregnancy classifications could be invalidated. 292 Geduldig, 417 U.S. at 496 n.20 (“Absent a showing that distinctions involving pregnancy are mere pretexts designed to effect an invidious discrimination against the members of one sex or the other, lawmakers are constitutionally free to include or exclude pregnancy from the coverage of legislation . . . .”). The Court affirmed this limited reading of Geduldig two years later in General Electric Co. v. Gilbert, which considered whether pregnancy discrimination could be impermissible sex discrimination under Title VII. 293 See 429 U.S. 125, 135–36 (1976) (“[A] distinction which on its face is not sex related might nonetheless violate the Equal Protection Clause if it were in fact a subterfuge to accomplish a forbidden discrimination.”); id. at 149 (Brennan, J., dissenting) (“Geduldig’s outcome was qualified by the explicit reservation of a case where it could be demonstrated that a pregnancy-centered differentiation is used as a ‘mere pretext . . . designed to effect an invidious discrimination against the members of one sex . . . .’” (quoting Geduldig, 417 U.S. at 496 n.20)).

Even those commentators who have read Geduldig to leave open the possibility of invalidating pregnancy regulations as impermissible sex-based classifications have struggled to provide examples of such regula­tions. In holding that the pregnancy classification at issue there did not amount to unconstitutional sex discrimination, Geduldig distinguished Frontiero v. Richardson, which invalidated a law that granted residential and medical benefits to wives of male servicemembers automatically but granted these same benefits to husbands of female servicemembers only upon a showing of economic dependence. 294 See 411 U.S. 677, 678–79, 688 (1973) (explicitly applying heightened scrutiny to a sex classification for the first time). Frontiero was explicit about extinguishing sex stereotypes that make up the homemaker–breadwin­ner dichotomy, because “the sex characteristic frequently bears no rela­tion to ability to perform or contribute to society.” 295 Id. at 686. Therefore, when Geduldig refers to “invidious discrimination” during the pregnancy that would be struck down like the “sex-based classification . . . considered in . . . Frontiero,” 296 Geduldig, 417 U.S. at 496 n.20. it should be read to mean laws that enforce sex-stereotypical roles in the family without any necessary connection to phy­sical sex differences.

2. Deploying Scrutiny. — Applying heightened scrutiny to pregnancy would allow courts to classify pregnancy regulations into one of three types: (1) those regulations that do not implicate physical sex differ­ences, (2) those regulations that implicate physical sex differences and are justified by them, and (3) those regulations that implicate physical sex differences but are not justified by them. The remainder of this subsection discusses these three types of pregnancy regulations in turn, as well as the implications for sex equality and for equality for gay and lesbian parents—a critical part of unsexing parenting under Obergefell. 297 See supra notes 253–255 and accompanying text.

The first type of regulation—those not based in any physical differ­ences—would be impermissible. For example, laws that provide support only for pregnant women to attend newborn care classes covering diaper­ing, sleep, and the like would be automatically suspect. 298 See supra section II.B.2 (discussing how federal law grants mothers but not fathers the right to use pretax dollars to pay for some types of prenatal education classes). Unsexing preg­nancy requires a doctrine far more skeptical of sex classifications based in assumptions that expectant fathers cannot or will not prepare for or bond with the expected child during the course of the pregnancy. 299 See, e.g., Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 736 (2003) (“Stereo­types about women’s domestic roles are reinforced by parallel stereotypes presuming a lack of domestic responsibilities for men.”); Weinberger v. Wiesenfeld, 420 U.S. 636, 652 (1975) (“It is no less important for a child to be cared for by its . . . parent when that parent is male rather than female.”).

The second and third types of cases—those based in physical differ­ences—are harder to distinguish. As in Virginia and Hibbs, a physical difference would not itself resolve the constitutionality of a sex-based pregnancy regulation. 300 See supra notes 284–289 and accompanying text. Courts would instead consider whether men and women are “similarly situated” with regard to the regulation, regard­less of any physical sex differences. 301 See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439 (1985) (explaining that the Equal Protection Clause “is essentially a direction that all persons similarly situated should be treated alike”); Frontiero v. Richardson, 411 U.S. 677, 690 (1973) (explaining that heightened scrutiny of sex classifications under the Fourteenth Amendment is meant to ensure that “similarly situated” men and women are treated the same (internal quotation marks omitted) (quoting Reed v. Reed, 404 U.S. 71, 77 (1971))) (quoting Reed v. Reed, 404 U.S. 71, 77 (1971)))r exclude pregnancy from the coverage of legislation . . . .”ent parental consen. This would require courts to con­sider the purpose of the sex classification (whether it is sufficiently “important”) and also whether the sex classification actually serves that purpose (whether it is “substantially related” to achieving the pur­pose). 302 United States v. Virginia, 518 U.S. 515, 524 (1996) (stating that the sex “classi­fication [must] serve[ ] important governmental objectives and that the discriminatory means employed [must be] substantially related to the achievement of those objectives” (internal quotation marks omitted) (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 724 (1982))). In practice, sex must serve as a “perfect proxy” for the law’s objective. 303 See supra notes 70–71 and accompanying text. If the sex classification extends beyond what is needed to achieve the law’s aim, it will be an unconstitutional “overbroad gener­alization” premised on sex stereotypes. 304 Virginia, 518 U.S. at 533–34 (striking down the male-only admissions policy to a military institute as an “overbroad generalization” based in and furthering harmful sex stereotypes, despite the relevance of physical sex differences, because some women could meet the admissions standard).

Applying heightened scrutiny to pregnancy in this way would of course still leave certain sex classifications intact. There are some prod­ucts that could be provided only to pregnant women without raising constitutional concerns. 305 The state might also decide what pregnancy-related products health insurers have to cover and for whom. See supra notes 201–205 and accompanying text for a discussion of how the ACA does this. A back-support pillow designed for pregnancy is one such product. 306 See supra note 112 and accompanying text. The benefits of such a product are derived only based on the physical difference of carrying the pregnancy, and a non­pregnant person would derive no benefit from the product. 307 But cf. supra note 113 (noting that seemingly sexed products like a breastfeeding pillow might be used by expectant parents of either sex).

Even in these instances, the law should classify on the basis of preg­nant as compared with nonpregnant persons, rather than on the basis of sex, to account for pregnant transgender men. 308 See supra notes 178–181 and accompanying text. In a rare set of cases, it might be sex rather than pregnancy that dictates the need for pregnancy-related support. Women can take drugs to stimulate lactation even in the absence of a prior pregnancy. Elizabeth LaFleur, Infant and Toddler Health, Mayo Clinic (Jan. 20, 2016), http://www.mayoclinic.org/healthy-lifestyle/infant-and-toddler-health/expert-answers/induced-lactation/faq-20058403 [https://perma.cc/B7ZF-K883]. Therefore, the provision of support to learn how to breastfeed and devices aimed at supporting breastfeeding might be needed even by some subset of women who did not give birth. So a regulation grant­ing support that would benefit only those physically carrying a child should still grant such support in a sex-neutral way, to the pregnant per­son, not to the “wom[a]n affected by pregnancy” 309 42 U.S.C. § 2000e(k) (2012) (protecting from employment discrimination only “women affected by pregnancy, childbirth, or related medical conditions”). or “expectant mother,” 310 29 C.F.R. § 825.120(a)(4) (2018) (“An expectant mother may take FMLA leave before the birth of the child for prenatal care or if her condition makes her unable to work.”). as current law sometimes does.

But even when pregnancy regulations relate to physical differences between the sexes, these differences may not always be sufficiently related to achieving the purpose of the law to justify a sex classification. Take laws that aim to promote the health of the expected child by promoting a healthier fetal environment. The Affordable Care Act, for example, requires insurers to provide support for pregnant women to quit smoking and engage in other behaviors to advance fetal health. 311 See 42 U.S.C. § 1396d(bb). Another similar example would be programs aimed at reducing stress in the environment. Despite  physical  differences,    such  laws  should  be  constitutionally  suspect. Some  studies  have  found  that  fetal  exposure  to  secondhand  smoke  has nega­tive  consequences  for  the  fetus,    even  in  the  absence  of  smoking  by  the  pregnant  woman. 312 See John et al., supra note 206, at 128 (describing the results of a study suggesting a correlation between childhood cancer and exposure to a father’s secondhand smoke); Rubin et al., supra note 206, at 415 (“Exposure to smoking by the mother was found to reduce birth-weight, and indirect or passive exposure to smoking by the father had nearly as large . . . an effect.”). Therefore, even if a physical sex difference means that secondhand smoking by expectant fathers harms fetuses through a different physical mechanism than smoking by pregnant women, the exclusion of men can­not be justified. 313 See Sessions v. Morales-Santana, 137 S. Ct. 1678, 1696–97 (2017) (“It will not do to ‘hypothesiz[e] or inven[t]’ governmental purposes for gender classifications ‘post hoc in response to litigation.’” (quoting United States v. Virginia, 518 U.S. 515, 533 (1996))); Frontiero v. Richardson, 411 U.S. 677, 689 (1973) (requiring the government to put forward evidence that the sex classification serves its claimed objective). Some studies show that paternal smoking has a substantial fetal impact but a lesser impact than maternal smoking. 314 See, e.g., Rubin et al., supra note 206, at 417 (documenting one such study). Even then, the complete exclusion of expectant fathers may not be justi­fied. While “the Equal Protection Clause does not require that a State must choose between attacking every aspect of a problem or not attack­ing the problem at all,” 315 Dandridge v. Williams, 397 U.S. 471, 486–87 (1970). the lesser impact of paternal smoking—and not stereotypical views of the relative importance of pregnant women and expectant fathers—must have been the actual reason for the law’s reliance on sex. 316 See Morales-Santana, 137 S. Ct. at 1696–98 (holding a sex classification unconstitu­tional because it was premised on a stereotypical view of the mother’s role as a caretaker).

Biological sex differences during the pregnancy become still less persuasive justifications for sex classifications given that pregnancy regulations may be aimed not only at changing the fetal environment but also at setting up parental habits for after the child is born, when physical sex differences are irrelevant. In the smoking cessation example, encouraging a pregnant woman or an expectant father to quit smoking during the pregnancy matters both for its impact during pregnancy and for its impact on the environment in which the child is raised. An equal protection doctrine that disaggregates sex from pregnancy would recognize that pregnancy regulations matter for how they set up path-dependent behaviors for the postbirth period and thereby undermine sex equality not only during the pregnancy period but far beyond it. 317 See supra section III.A. Once this is recognized, the exclusion of expectant fathers will often become harder to justify, regardless of physical differences.

Applying heightened scrutiny to disaggregate sex from pregnancy in this way has important implications not only for heterosexual couples but also for gay, lesbian, and transgender expectant parents. By disaggregat­ing sex from pregnancy, the law not only decouples sex from the role of the pregnant partner but also decouples sex from the role of the nonpregnant partner. In other words, it is not only that expectant fathers can serve in the capacity that has been traditionally reserved for pregnant women but also that women can serve in the capacity that has been tradi­tionally reserved for expectant fathers. Lesbian couples benefit from the recognition that the partner of the pregnant woman could be a woman. 318 See Douglas NeJaime, The Nature of Parenthood, 126 Yale L.J. 2260, 2268–69 (2017) (“The law has traditionally connected women to motherhood as biological destiny, and thus crediting the social aspects of motherhood is necessary to value the parenting work of women who break from conventional roles.”). Recognizing that women can be either the pregnant partner or the nonpregnant expectant partner furthers the goals of unsexing parenting by combatting sex stereotypes about who plays which roles in the family from both sides. 319 See Schoenbaum, Case for Symmetry, supra note 241, at 98–102 (explaining the importance of combatting sex stereotypes from both sides).

B. Protecting Autonomy Rights

If applying heightened scrutiny under our rereading of Geduldig ren­ders a sex classification during pregnancy constitutionally infirm, what is the proper remedy? The answer could have important implications for women’s autonomy, as we might worry that some sex-neutral remedies could result in excessive intrusion into the pregnancy by the nonpregnant partner, with ill effects on reproductive freedom. Disaggre­gating sex from pregnancy means separating out those parts of the pregnancy that necessarily involve the woman’s body and therefore her autonomy rights from those that involve carework that can be unsexed. For those parts of the pregnancy that can be disaggregated, this section argues that extending legal benefits to those to whom they were formerly denied will tend to be the better legal remedy for most of these equal protection violations. Involving nonpregnant partners in the carework of pregnancy can actually promote rather than undermine both women’s equality and autonomy.

When a statute that “benefits one class . . . and excludes another from the benefit” is found to violate the Equal Protection Clause, “[a] court may either declare [the statute] a nullity and order that its benefits not extend to the class that the legislature intended to benefit, or it may extend the coverage of the statute to include those who are aggrieved by exclusion.” 320 Morales-Santana, 137 S. Ct. at 1698 (internal quotation marks omitted) (quoting Califano v. Westcott, 443 U.S. 76, 89 (1979)); see also Heckler v. Mathews, 465 U.S. 728, 740 (1984) (“[W]hen the ‘right invoked is that to equal treatment,’ the appropriate remedy is a mandate of equal treatment, a result that can be accomplished by withdrawal of benefits from the favored class as well as by extension of benefits to the excluded class.” (quoting Iowa–Des Moines Nat’l Bank v. Bennett, 284 U.S. 239, 247 (1931))). We can refer to the first of these alternatives as “leveling down” and the second of these alternatives as “leveling up.” Although leg­islative intent is the guidepost for deciding between these alternatives, 321 See Morales-Santana, 137 S. Ct. at 1699 (“The choice between these outcomes is governed by the legislature’s intent, as revealed by the statute at hand.”). ordinarily “extension, rather than nullification, is the proper course.” 322 Id. (internal quotation marks omitted) (quoting Westcott, 443 U.S. at 89). This has been true in the sex discrimination context. 323 See id. (stating this generality in the context of a sex discrimination case and citing several examples in that context); Califano v. Goldfarb, 430 U.S. 199, 202–04, 213–17 (1977) (plurality opinion) (Brennan, J.) (extending survivors’ benefits after striking down a sex classification); Frontiero v. Richardson, 411 U.S. 677, 678–679, 679 n.2, 691 & n.25 (1973) (plurality opinion) (Brennan, J.) (extending military spousal benefits after striking down a sex classification). But see Morales-Santana, 137 S. Ct. at 1700 (“Although extension of benefits is customary in federal benefit cases . . . all indicators in this case point in the opposite direction.”).

Leveling up is particularly important for achieving sex equality when it comes to caregiving responsibilities. The Supreme Court has explained, in the context of the FMLA, why it is critical for sex equality to grant not only a right of nondiscrimination on the basis of sex but also gender-neutral affirmative caregiving benefits. In a world where women engage in far more caregiving than men, and where employer “practices continue to reinforce the stereotype of women as caregivers,” a rule of formal equality that allows for no caregiving benefits at all “would exclude far more women than men from the workplace.” 324 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 738 (2003). Sex-neutral affirmative caregiving benefits “ensure that family-care leave w[ill] no longer be stigmatized as an inordinate drain on the workplace caused by female employees, and that employers [cannot] evade leave obligations simply by hiring men.” 325 Id. at 737. Sex-neutral affirmative caregiving benefits are also essential for combatting the “stereotype that only women are respon­sible for family caregiving, thereby reducing employers’ incentives to engage in discrimination by basing hiring and promotion decisions on stereotypes.” 326 Id.; see also 29 U.S.C. § 2601(b)(4) (2012) (explaining that an affirmative caregiving benefit as provided by the FMLA “minimizes the potential for employment discrimination on the basis of sex by ensuring generally that leave is available for eligible medical reasons (including maternity-related disability) and for compelling family reasons, on a gender-neutral basis”).

But will leveling up present a problem for women’s autonomy? This Article argues that the proper limits in place under Planned Parenthood of Southeastern Pennsylvania v. Casey 327 505 U.S. 833 (1992). will mitigate any such concerns. As discussed in section III.B, Casey has sometimes been wrongly read to sug­gest a universal harm of male involvement in the pregnancy writ large. Casey was about being able to choose to terminate a pregnancy free from excessive control by spouses. 328 Id. at 895–98 (focusing on the problematic “dominion” and “authority” that the statute there gave the husband over his wife). Within this paradigm, Casey rightly recog­nizes that some forms of involvement in the pregnancy amount to excessive coercion. In Casey, the Court struck down a spousal notification requirement because “[t]he husband’s interest in the life of the child his wife is carrying does not permit the State to empower him with this trou­bling degree of authority over his wife.” 329 Id. at 898.

Leveling  up  pregnancy  benefits  need  not  present  this  concern  because  pregnancy  can  be  disaggregated  from  sex. Once  we  recognize  the  different  ways  in  which  the  expectant  father  can  participate  in  the  pregnancy,   it  becomes  clearer  that  these  different  ways  can  have  different  relationships  to  the  mother’s  body  and  thus  to  her  autonomy. This  section  distinguishes  between  those  ways  in  which  participation  would  require  the  involvement  of  the  mother’s  body  and  those  in  which  it  would  not. Expectant fathers’ joint participation in the pregnancy, such as when attending a prenatal appointment, poses the greatest risk for infringing autonomy. In such circumstances, the pregnant woman’s con­sent should be required before the expectant father can play any role. The expectant father would have no constitutional 330 See Plotnick v. DeLuccia, 85 A.3d 1039, 1052–54 (N.J. Super. Ct. Ch. Div. 2013) (rejecting claims by an expectant father that the expectant mother must permit his attendance at birth because “modern constitutional jurisprudence has clearly confirmed that any interest a father has before the child’s birth is subordinate to the mother’s interests”). or other legal right to attend. 331 The expectant mother is the “patient” that the obstetrician treats during the pregnancy. See Denise Grady, Gynecologists Run Afoul of Panel when Patient Is Male, N.Y. Times (Nov. 22, 2013), https://www.nytimes.com/2013/11/23/health/gynecologists-run-afoul-of-panel-when-patient-is-male.html (on file with the Columbia Law Review) (“[T]he American Board of Obstetrics and Gynecology insisted that its members treat only women, with few exceptions . . . .”). Except in unusual situations, individuals who are not patients can only be present during appointments with the consent of the patient. See 45 C.F.R. § 164.510 (2017) (“A covered entity may use or disclose protected health information, provided that the individual is informed in advance of the use or disclosure and has the opportunity to agree to or prohibit or restrict the use or disclosure, in accordance with the applicable requirements of this section.”); see also Health Insurance Portability and Accountability Act of 1996, Pub. L. No. 104-191, sec. 201, § 1128C(a)(3)(B), 110 Stat. 1936 (1996) (codified as amended at 42 U.S.C. § 1320a-7c(a)(3)(B)) (requiring the creation of “guidelines relating to the furnishing of information by health plans [and] providers,” including “procedures to assure that such information is provided and utilized in a manner that appropriately protects the confidentiality of the information and the privacy of individuals receiving health care services and items”). Prenatal benefits exercised after the pregnant woman’s con­sent can be distinguished from the father-involvement laws that courts have invalidated. The invalidated laws required mothers to involve fathers in the pregnancy, whereas the paternal involvement proposed here may be exercised only if the mother agrees. 332 Casey, 505 U.S. at 898 (focusing on the problematic aspect of male “authority” over his pregnant partner). With this limit, sex-neutral preg­nancy benefits do not present the type of excessive coercion that con­cerned the Court in Casey. This holds true regardless of whether the nonpregnant expectant parent is male or female and whether the preg­nant person is female or male.

Even if leveling up benefits in the pregnancy context does not pre­sent a constitutional problem, we might be concerned about other significant expressive or material harms to women’s autonomy that could come from leveling up. As for the expressive harm, one might be con­cerned that any legal movement toward recognizing fetal interests—espe­cially as independent from the pregnant woman—will tend to limit women’s reproductive rights. 333 See Greenhouse & Siegel, supra note 266, at 1445–49 (discussing this concern in the context of fetal-protective policies); Grossman & Thomas, supra note 188, at 19 (discussing this concern in the context of the PDA). But disaggregating sex from pregnancy by providing sex-neutral pregnancy benefits is mostly not about the fetus—and certainly not about providing any rights vis-à-vis the fetus—but about preparing to parent the child after birth. Expectant parents invest in physical capital buying a carseat not for a fetus but for a child. Expectant parents invest in social capital finding a childcare center not for a fetus but for a child. Expectant parents invest in human capital by quitting smoking not only for the fetus but at least as much for a smoke-free environment for the child. This should mitigate—albeit not elimi­nate—concerns that disaggregating sex from pregnancy would communi­cate messages harmful to women’s autonomy. 334 See generally Robert C. Post & Reva B. Siegel, Legislative Constitutionalism and Section Five Power: Policentric Interpretation of the Family and Medical Leave Act, 112 Yale L.J. 1943, 2005–20 (2003) (discussing messages sent by laws at the intersection of work and family, such as the FMLA).

Still, curing equal protection violations by leveling up pregnancy benefits could affect bargaining dynamics between pregnant women and expectant fathers. An expectant father who begins to cultivate his domes­tic productivity is more likely to want to have a say over pregnancy-related matters. Even if she can veto attendance at a prenatal appointment, a pregnant woman may still feel that her autonomy is intruded upon if a one-night stand or an ex-lover or even her spouse has an opinion on the carseat or pediatrician or childcare center—not because he has any right to dictate these matters but simply because he is involved with them at all. These concerns may arise for transgender pregnant men and for pregnant women with female partners as well.

This harm is a real concern, and real enough that concerns about bargaining dynamics during pregnancy can overwhelm any attempt to compare pregnancy to parenting. But these possible harms can take on a different cast if we reframe the concept of autonomy in the way that feminists have long urged. Femi­nist legal theorists have critiqued liberal theory for viewing people as atom­ized individuals rather than in critical relationships with each other, and for failing to account for these relationships in considering what makes individuals free. 335 See, e.g., Susan Moller Okin, Justice, Gender, and the Family 25–40 (1989) (arguing that understanding what constitutes a just society requires considering the connections we have with others, especially in the family); West, supra note 138, at 1–2 (arguing that law and legal theory are pervasively marked by a view of people as separate rather than connected); see also Michael J. Sandel, Liberalism and the Limits of Justice 133 (1982) (stating that most contemporary political theory is grounded in the view that “what separates us is in some important sense prior to what connects us,” and thus “[w]e are distinct individuals first, and then we form relationships”). In her seminal book, The Autonomy Myth, law professor Martha Fineman emphasized that the human condition is marked far more by connection than separation. 336 See Martha Albertson Fineman, The Autonomy Myth: A Theory of Dependency (2004). She argued that those with caregiving responsibil­ities—disproportionately women—do not achieve freedom by being left alone but are far more free—and thus far more equal—when they are supported. 337 See id. at xix (“If we want our families to shoulder responsibility for dependency then we must look directly at that task and build policy to foster and facilitate caretaking.”). Because Casey was concerned with how a spousal notification requirement could interfere with women’s autonomy, the Court had little opportunity to consider the positive ways in which connections can foster autonomy. See Casey, 505 U.S. at 896 (“The Constitution protects individuals, men and women alike, from unjustified state interference, even when that interference is enacted into law for the benefit of their spouses.” (emphasis added)). Other courts have shown glimmers of recognition of the fact that autonomy turns on connection. One recent decision considered a Texas abortion regulation that would force the closure of a significant number of abortion clinics in the state, requiring some women to travel a substantial distance to obtain an abortion. See Whole Woman’s Health v. Lakey, 46 F. Supp. 3d 673, 682–83 (W.D. Tex. 2014), aff’d sub nom. Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292 (2016). The restriction was struck down in part based on the recognition that exercising the right to an abortion might rely on support from others, such as those who might provide “childcare” or “transportation,” and that “poor, rural, or disadvantaged women” would not be able to pay for this support in the absence of connections to provide it. Id. at 683; see also June Med. Servs. v. Kliebert, 250 F. Supp. 3d 27, 83 (M.D. La. 2017) (“Women who cannot afford to pay the costs associated with travel, childcare, and time off from work may have to . . . rely on predatory lenders, or borrow money from family members or abusive partners or ex-partners, sacrificing their financial and personal security.”).

Under this view, freedom for caregivers inevitably turns on connec­tion, not separation—on support, not solace. Rather than necessarily interfering with autonomy, participation in pregnancy to support women in performing the caregiving tasks that they tend to bear alone can be a means toward greater freedom and thus greater equality. The law should not ignore this but should instead try to ensure that critical relationships are constructive relationships. Rather than acting only as a barrier to expectant fathers unduly interfering in pregnancy in a way that under­mines sex equality, as in Casey, law can play a role in encouraging expectant fathers (and nonpregnant expectant mothers) to participate in pregnancy in a way that promotes sex equality.

Disaggregating sex from pregnancy might be viewed as part of a long line of laws that have garnered the support of liberal feminists despite the fact that they likely play into a woman’s calculus about whether to carry a pregnancy to term. It would seem strange to think of laws like the PDA, the FMLA, or child-support requirements 338 Indeed, one of the state law provisions upheld in Casey required women to be provided with information about child support from the father. See Casey, 505 U.S. at 881. as autonomy-reducing measures simply because they affect women’s reproductive choices, as they have been shown to do. 339 See Jocelyn Crowley et al., The Effect of Child Support Enforcement on Abortion in the United States, 93 Soc. Sci. Q. 152, 165–69 (2012) (finding a relationship between child support policies and childbearing decisions); Robert Plotnick et al., The Impact of Child Support Enforcement Policy on Nonmarital Childbearing, 26 J. Pol’y Analysis & Mgmt. 79, 96 (2007); see also Shari Motro, The Price of Pleasure, 104 Nw. U. L. Rev. 917, 971 (2010) (considering and defending legal rules that aspire to improve the conditions of pregnancy and therefore affect abortion rates). This  is  because,   as  with  laws  that  would  unsex  pregnancy,   such  laws  reduce  abortion  by  changing  the  social  cir­cumstances  of  pregnancy  in  ways  that  support  the  pregnant  woman  and  make  pregnancy  less  costly. 340 See Crowley, supra note 344, at 169–70 (finding that the rate of abortions falls as the expectation that men will pay child support rises because women feel more confident about the behavior of the expectant father). Indeed, such prenatal supports can be seen as autonomy enhancing, in that they open up choices for women that they might not have had before.

This doesn’t mean that paternal involvement in the pregnancy will always be easy for pregnant women. As in the postbirth period, women may feel territorial over this domain that they have typically occupied alone, leading to maternal gatekeeping. 341 See Naomi R. Cahn, Gendered Identities: Women and Household Work, 44 Vill. L. Rev. 525, 536–41 (1999) (discussing how gendered norms may lead mothers to engage in “gatekeeping” vis-à-vis fathers to retain the primary caretaking role). And gendered bargaining dynamics may play out such that women do not always put their own interests first. 342 See generally Amartya Sen, Gender and Cooperative Conflicts, in Persistent Inequalities: Women and World Development 123 (Irene Tinker ed., 1990) (discussing how gender influences intrafamily bargaining and how women can place family interests ahead of their own). But, despite these difficulties, research shows how con­structive paternal participation in the pregnancy advances sex equality. Expectant fathers whom pregnant women think will be more responsible generally empower rather than constrain these women. 343 See Shari Motro, Preglimony, 63 Stan. L. Rev. 647, 657 (2011) (“Thus, increasing support for pregnant women regardless of the pregnancy’s outcome is likely, over time, to change abortion from being used as a form of birth control that lets men off the hook into a last resort that both parties are invested in preventing.”).

Successful comparative examples confirm that paternal involvement in the pregnancy—within appropriate limits—and repro­ductive autonomy are not only consistent but mutually reinforcing. The United Kingdom has created a sex-neutral paid prenatal leave program 344 See, e.g., Press Release, Jo Swinson, Dep’t for Bus., Innovation & Skills, New Right for Fathers and Partners to Attend Antenatal Appointments (Oct. 2, 2014), https://www.gov.uk/government/news/new-right-for-fathers-and-partners-to-attend-antenatal-appointments [https://perma.cc/HY4U-DEN8] (granting fathers and same-sex partners the right to leave work to attend several prenatal appointments) [hereinafter U.K., New Right]. alongside vigorous protections of the right to choose. 345 United Nations, Dep’t of Soc. & Econ. Affairs, Population Div., World Abortion Policies 2013 (2013), https://www.un.org/en/development/desa/population/publications/pdf/policy/WorldAbortionPolicies2013/WorldAbortionPolicies2013_WallChart.pdf (on file with the Columbia Law Review). Consistent with a shift toward sex neutrality in caregiving, such laws provide benefits not only to expectant fathers but to nonpregnant expectant mothers (for example, lesbian partners) as well. 346 See U.K., New Right, supra note 349.


The Supreme Court’s opinion in Obergefell v. Hodges recognizing a constitutional right to same-sex marriage is replete with “end of history” rhetoric. 347 See Obergefell v. Hodges, 135 S. Ct. 2584, 2604 (2015) (“Especially against a long history of disapproval of their relationships, this denial to same-sex couples of the right to marry works a grave and continuing harm.”); NeJaime, supra note 323, at 2260 (“In the wake of Obergefell v. Hodges, courts and legislatures claim in principle to have repudiated the privileging of different-sex over same-sex couples and men over women in the legal regulation of the family.”). Law is one long march in which “new insights and societal understandings can reveal unjustified inequality within our most funda­mental institutions that once passed unnoticed and unchallenged.” 348 Obergefell, 135 S. Ct. at 2603. The law has gradually moved from constraining the sexes to liberating them. This is particularly true when law interjects itself at the “faultline between work and family.” 349 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 738 (2003) (explaining that this is “where sex-based overgeneralization has been and remains strongest”). Women can be workers and parents, and so can men.

Except when it comes to pregnancy. Federal and state governments spend billions of dollars each year subsidizing women devoting them­selves to carework during the pregnancy, and encouraging men not to do so. 350 See Nat’l P’ship, Reasonable Accommodations, supra note 182. However far we want to go forward toward sex equality in the future, we have to start further back at the pregnancy itself in consider­ing when sex inequality begins. This Article moves us backward in pre­cisely this way to move the law forward in achieving its goal of sex equality.