The Yale Law Journal


Sex Equality’s Inner Frontier: The Case of Same-Sex Marriage

21 Feb 2013
This Essay was adapted from remarks delivered at Equality’s Frontiers, a panel discussion celebrating Justice Ginsburg’s gender-equality jurisprudence and analyzing its relationship with new developments in the law of equality. The discussion preceded Justice Ginsburg’s Gruber Distinguished Lecture in Women’s Rights, held on October 19, 2012, at Yale University.

I am honored to be here. I thank the Gruber Foundation, my fellow panelists, the organizers, and, of course, the Justice. In my remarks, I will address some of the achievements and limitations of constitutional sex-equality jurisprudence. I will then consider the implications of that jurisprudence for the case study of same-sex marriage.

It is now familiar history that modern sex-equality jurisprudence under the Equal Protection Clause begins with the 1971 case of Reed v. Reed.1 Reed involved an Idaho statute that favored men over women as executors of estates. The Supreme Court invalidated this statute without formally raising the level of review under the Equal Protection Clause. Especially at the time, that application of rational basis review was anomalous, as such review is famously deferential. A colleague of mine has quipped that such review only requires legislation to be framed in grammatically complete sentences. I suspect the Justice knows of several cases in which the Court has waived even that grammar requirement.

The Court had to make sense of what it had done in Reed. Two years later, in Frontiero v. Richardson,2 a plurality of the Court contended that gender-based classifications merited strict scrutinythe most rigorous form of scrutiny drawn by race and national-origin classifications. However, Justice Brennan could not gather a majority of the Court for this proposition. It took three more years before a majority of the Court settled on a middle ground. In the 1976 case Craig v. Boren,3 the Court held that sex-based classifications “must serve important governmental objectives and must be substantially related to achievement of those objectives.”4 We still live under that intermediate scrutiny standard today.

That standard has had incalculable effects on the lives of men and women in this nation. Consider the 1982 case of Mississippi University for Women v. Hogan,5 in which an all-female, state-sponsored nursing school declined Joe Hogan admission. The Court applied intermediate scrutiny and struck down the discriminatory admissions policy, stating that it rested on the “stereotyped view of nursing as an exclusively woman’s job.”6 The mirror image of Hogan was the 1996 case of United States v. Virginia.7 By then, Justice Ginsburg had taken her seat on the Court. She wrote the epochal majority opinion in that case, which held that the Virginia Military Institute—a public academy dedicated to training citizen-soldiers—could not constitutionally deny women admission.

I want to tarry over the Virginia case because many have argued that this case increased the stringency of intermediate scrutiny. It was undisputed that intermediate scrutiny required the classification to be “substantially related” to “important governmental objectives.”8 Yet the majority opinion also stated that the classification had to have an “exceedingly persuasive justification.”9 In his concurrence, then-Chief Justice Rehnquist stated that the Court had taken a phrase hitherto used to describe the rigor of the intermediate scrutiny test and potentially transformed it into an additional element of that test.10 I do not wish to travel too far into this debate because each phrase—“exceedingly persuasive justification,” “substantially related,” and “important governmental objectives”—is so abstract that it is hard to ascertain what it requires.

The stronger evidence that the Virginia case ratcheted up the intermediate scrutiny standard lies elsewhere. Writing for the majority, Justice Ginsburg noted: “[G]eneralizations about ‘the way women are,’ estimates of what is appropriate for most women, no longer justify denying opportunities to women whose talent and capacity place them outside the average description.”11 Notice the radical nature of this statement. This logic requires that if any woman can avail herself of an opportunity, no woman can be denied it by the state.

When I teach this case, I ask my students what daylight remains between intermediate and strict scrutiny. My students respond that Justice Ginsburg herself provides a distinction in Virginia. After observing that “[s]upposed ‘inherent differences’ are no longer accepted as a ground for race or national origin classifications,” she observes that “[p]hysical differences between men and women, however, are enduring.”12 Real biological differences might provide a ground on which the state could make legitimate distinctions between the sexes.

Unfortunately, however, the Court has taken an overly deferential posture toward such biological differences, thereby imposing a serious constraint on the principle of sex equality. In 2001, the Court decided Nguyen v. INS.13 Nguyen concerned a congressional statute that explicitly favored mothers over fathers. If a citizen mother had a child with an alien father out of the country and out of wedlock, she passed her citizenship to her child at birth. However, if a citizen father had a child with an alien mother in otherwise identical circumstances, the child did not receive automatic citizenship.

Writing for the majority, Justice Kennedy applied intermediate scrutiny, but upheld the statue. He found that the sex-based classification was substantially related to the important governmental interests of assuring the existence of a biological parent-child relationship and of creating a link between the child and the United States through the citizen parent.14 He rejected the notion that equality required a sex-neutral rule by underscoring “that the mother is always present at [the] birth” of the child, while “the father need not be.”15

This reasoning problematically ignores the Court’s prior analysis. To take the logic of Virginia seriously is to observe that if even one man were capable of meeting the standards of conferring automatic citizenship (i.e., knowing and bonding with his child), then no man should be denied the opportunity to do so. This would be the case even if women were more likely on average to develop such bonds. The Court thus permitted the government’s enunciation of a real biological difference to become a Trojan horse through which cultural assumptions—including stereotypes—were imported.

To be clear, I am not gainsaying the existence of biological differences between men and women. What I am saying is that the line between biology and culture is drawn in culture. Because the location of that line is a matter of such contestation, the Supreme Court must scrutinize where the state has drawn it. The Nguyen majority refused to do so.

So what we have now is a great achievement and a limitation. We have an enhanced intermediate scrutiny standard from United States v. Virginia. Yet we have a limitation on that standard flowing from the real biological differences argument articulated in Nguyen v. INS.

I now consider the counterintuitive implications of this jurisprudence in the context of same-sex marriage. The same-sex marriage cases used to rely much more heavily on the principle of sex equality. In the 1993 case of Baehr v. Lewin,16 the Hawaii Supreme Court held that the definition of marriage as a union of one man and one woman had to be subjected to the strict scrutiny triggered by sex-based classifications under the state constitution. This analysis is cogent: these marriage statutes mention men and women, not gays and straights. They formally distinguish on the basis of sex, not sexual orientation.

Since then, a gradual shift has occurred toward analyzing marriage statutes as a form of discrimination based on sexual orientation. This shift reflects the success of the gay-rights movement, which has sought to have sexual-orientation discrimination recognized in its own right. Recently, the Second Circuit issued a landmark ruling that accorded classifications based on sexual orientation heightened scrutiny.17 The only problem with this shift toward the orientation-equality claim is that it sometimes leaves the sex-equality claim behind, erroneously treating the two claims as mutually exclusive.

I wish to return attention to the sex-equality claim at both the “wholesale” level and the “retail” level. The wholesale argument contends that laws defining marriage as a union between a man and a woman should be struck down because they constitute sex discrimination. I will bracket that argument here because it has already been ably set forth elsewhere.18 I focus instead on the retail argument. The retail argument does not seek to assail a marriage law tout court, but rather to force the government to retire certain justifications for the law.

To give an example, opponents of same-sex marriage often contend that only opposite-sex couples can create the optimal child-rearing environment. According to these litigants, the optimal child-rearing environment is a married man and woman who are both genetically related to the child. This environment has two components. One component is the “genetic tie”—the idea that we treat individuals better when they are blood relations. The other is a “gender-differentiated parenting” argument, which asserts that children need both a male and a female role model in the family. These are distinct arguments. In the hypothetical situation where two women could merge their genetic material and create a child, the first argument would evanesce while the second would endure.

Based on our current sex-discrimination jurisprudence, we might expect opponents of same-sex marriage to lead with the “genetic tie” argument. This argument rests primarily on biological predicates, while the “gender-differentiated parenting” argument rests primarily on cultural ones. However, in practice, the opposite appears to be the case.

Litigants may be relatively reluctant to press the “genetic tie” argument because it is politically unpalatable and empirically questionable. It is politically unpalatable because it charges all adoptive parents—including heterosexual parents—with providing a suboptimal child-rearing environment. It is empirically questionable because adoptive parents are so carefully screened that they may in fact perform better than the average genetic parents. As one of the witnesses for the defendant-intervenors testified in the Proposition 8 trial, adoptive parents “actually on some outcomes outstrip biological parents in terms of providing protective care for their children.”19

Instead, opponents of same-sex marriage insistently press the argument about “gender-differentiated parenting.” The proponents of Proposition 8 used such arguments liberally in their campaign. The district judge in the federal case challenging Proposition 8 quoted one such argument in his opinion:

  When moms are in the park taking care of their kids they always know where those kids are. They have like a, like a radar around them. They know where those kids are and there’s just a, there’s a bond between a mom and a kid different from a dad. I’m not saying dads don’t have that bond but they don’t.20  

“I’m not saying dads don’t have that bond but they don’t”—that is as succinct a description of that individual’s ambivalence as you could hope to find. It limns the distance between an ideal of sex equality and the felt reality of sex inequality.

The widespread acceptability of such statements is mysterious. In an era where men can go to nursing school, surely men can also perform “maternal” child-rearing functions. In an era where women can become citizen-soldiers, surely women can perform “paternal” child-rearing functions. We must ask why these constitutional norms—now so clearly established with regard to state action in the public sphere—have yet to be applied to state action in the private sphere.

The most intuitive answer is probably the perceived effect that true enforcement of such sex-equality norms would have on children. If a patient has a male nurse, few would think that the patient’s gender identity is going to change as a result. In contrast, if a child is raised by parents of the same sex, many more deleterious consequences are imagined.

The title of this panel is Equality’s Frontiers. Generally, a frontier is a border that gets pushed outward. Today, however, equality has an inner frontier. Well-established principles of sex discrimination need to be pressed into the public regulation of the family.

Finally, I would like to address a few closing words to Justice Ginsburg, who has been my role model since my law school days. I regard her as a founding father of sex-equality jurisprudence. I leave it to her whether she would prefer to be referred to as a founding mother, founding parent, or simply the founder. But I will not negotiate about returning to her the words of Chief Justice Marshall about some of the original Founding Fathers: “No tribute can be paid to [you] which exceeds [your] merit.”21

* * *

Justice Ginsburg offered the following response:

  Thank you. Kenji, I am overwhelmed by your generous comment. But if truth be told, I was very lucky to be born when I was—to become a lawyer when I did—because there were women and men saying the same things that we said in the 1970s, in the 1920s, the 1930s, the 1940s, and 1950s. But society was not yet ready to listen. I was there when society, even the “conservative” Burger Court, would listen to arguments that escaped the comprehension of the “liberal” Warren Court. I have heroines, Pauli Murray for one, who said earlier exactly what I said later. Society had changed by the 1970s, and that made it possible for my arguments to be heard.  

* * *

Kenji Yoshino is the Chief Justice Earl Warren Professor of Constitutional Law at the New York University School of Law.

Preferred citation: Kenji Yoshino, Sex Equality’s Inner Frontier: The Case of Same-Sex Marriage, 122 Yale L.J. Online 275 (2013),