Fourteenth Amendment, Section 1 — Equal Protection: Gender-Based Classifications

Amdt14.S1.8.8.1 Doctrine on Gender Classifications from 1870s to 1960s

Shortly after ratification of the Fourteenth Amendment, the refusal of Illinois to license a woman to practice law was challenged before the Supreme Court, and the Court rejected the challenge in tones that prevailed well into the twentieth century. For example, the Court stated in 1873 that “[t]he civil law, as well as nature itself, has always recognized a wide difference in the respective spheres and destinies of man and woman. 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 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.”¹ On the same premise, a statute restricting the franchise to men was sustained.²

The greater number of cases involved legislation aimed to protect women from oppressive working conditions, as by prescribing maximum hours³ or minimum wages⁴ or by restricting some of the things women could be required to do.⁵ A 1961 decision upheld a state law that required jury service of men but that gave women the option of serving or not. “We cannot say that it is constitutionally impermissible for a State acting in pursuit of the general welfare, to conclude that a woman should be relieved from the civic duty of jury service unless she herself determines that such service is consistent with her own special responsibilities.”⁶ Another type of protective legislation for women that was sustained by the Court is that premised on protection of morals, as by forbidding the sale of liquor to women.⁷ In a highly controversial ruling, the Court sustained a state law that forbade the licensing of any female bartender, except for the wives or daughters of male owners. The Court purported to view the law as one for the protection of the health and morals of women generally, with the exception being justified by the consideration that such women would be under the eyes of a protective male.⁸

A wide variety of sex discrimination by governmental and private parties, including sex discrimination in employment and even the protective labor legislation previously sustained, is now proscribed by federal law. In addition, federal law requires equal pay for equal work.⁹ Some states have followed suit.¹⁰ While the proposed Equal Rights Amendment was before the states and ultimately failed to be ratified,¹¹ the Supreme Court undertook a major evaluation of sex classification doctrine, first applying a “heightened” traditional standard of review (with bite) to void a discrimination and then, after coming within a vote of making sex a suspect classification, settling upon an intermediate standard. These standards continue, with some uncertainties of application and some tendencies among the Justices both to lessen and to increase the burden of governmental justification of sex classifications.

Amdt14.S1.8.8.2 Doctrine on Gender Classifications During the 1970s

In Reed v. Reed,¹ the Court held invalid a state probate law that gave males preference over females when both were equally entitled to administer an estate. Because the statute “provides that different treatment be accorded to the applicants on the basis of their sex,” Chief Justice Burger wrote, “it thus establishes a classification subject to scrutiny under the Equal Protection Clause.” The Court proceeded to hold that under traditional equal protection standards—requiring a classification to be reasonable and not arbitrarily related to a lawful objective—the classification made was an arbitrary way to achieve the objective the state advanced in defense of the law, that is, to reduce the area of controversy between otherwise equally qualified applicants for administration. Thus, the Court used traditional analysis but the holding seems to go somewhat further to say that not all lawful interests of a state may be advanced by a classification based solely on sex.²

Amdt14.S1.8.8.3 General Approach to Gender Classifications

It is now established that sex classifications, in order to withstand equal protection scrutiny, “must serve important governmental objectives and must be substantially related to achievement of those objectives.”¹ Thus, after several years in which sex distinctions were more often voided than sustained without a clear statement of the standard of review,² a majority of the Court has arrived at the intermediate standard that many had thought it was applying in any event.³ The Court first examines the statutory or administrative scheme to determine if the purpose or objective is permissible and, if it is, whether it is important. Then, having ascertained the actual motivation of the classification, the Court engages in a balancing test to determine how well the classification serves the end and whether a less discriminatory one would serve that end without substantial loss to the government.⁴

Some sex distinctions were seen to be based solely upon “old notions,” no longer valid if ever they were, about the respective roles of the sexes in society, and those distinctions failed to survive even traditional scrutiny. Thus, a state law defining the age of majority as eighteen for females and twenty-one for males, entitling the male child to support by his divorced father for three years longer than the female child, was deemed merely irrational, grounded as it was in the assumption of the male as the breadwinner, needing longer to prepare, and the female as suited for wife and mother.⁵ Similarly, a state jury system that in effect excluded almost all women was deemed to be based upon an overbroad generalization about the role of women as a class in society, and the administrative convenience served could not justify it.⁶

Even when the negative “stereotype” that is evoked is that of a stereotypical male, the Court has evaluated this as potential gender discrimination. In J. E. B. v. Alabama ex rel. T. B.,⁷ the Court addressed a paternity suit where men had been intentionally excluded from a jury through peremptory strikes. The Court rejected as unfounded the argument that men, as a class, would be more sympathetic to the defendant, the putative father. The Court also determined that gender-based exclusion of jurors would undermine the litigants’ interest by tainting the proceedings, and in addition would harm the wrongfully excluded juror.

Assumptions about the relative positions of the sexes, however, are not without some basis in fact, and sex may sometimes be a reliable proxy for the characteristic, such as need, with which it is the legislature’s actual intention to deal. But heightened scrutiny requires evidence of the existence of the distinguishing fact and its close correspondence with the condition for which sex stands as proxy. Thus, in the case that first expressly announced the intermediate scrutiny standard, the Court struck down a state statute that prohibited the sale of “non-intoxicating” 3.2 beer to males under twenty-one and to females under eighteen.⁸ Accepting the argument that traffic safety was an important governmental objective, the Court emphasized that sex is an often inaccurate proxy for other, more germane classifications. Taking the statistics offered by the state as of value, while cautioning that statistical analysis is a “dubious” business that is in tension with the “normative philosophy that underlies the Equal Protection Clause,” the Court thought the correlation between males and females arrested for drunk driving showed an unduly tenuous fit to allow the use of sex as a distinction.⁹

Invalidating an Alabama law imposing alimony obligations upon males but not upon females, the Court in Orr v. Orr acknowledged that assisting needy spouses was a legitimate and important governmental objective. Ordinarily, therefore, the Court would have considered whether sex was a sufficiently accurate proxy for dependency, and, if it found that it was, then it would have concluded that the classification based on sex had “a fair and substantial relation to the object of the legislation.”¹⁰ However, the Court observed that the state already conducted individualized hearings with respect to the need of the wife, so that with little if any additional burden needy males could be identified and helped. The use of the sex standard as a proxy, therefore, was not justified because it needlessly burdened needy men and advantaged financially secure women whose husbands were in need.¹¹

Various forms of discrimination between unwed mothers and unwed fathers received different treatments based on the Court’s perception of the justifications and presumptions underlying each. A New York law permitted the unwed mother but not the unwed father of a child born out of wedlock to block his adoption by withholding consent. Acting in the instance of one who acknowledged his parenthood and who had maintained a close relationship with his child over the years, the Court could discern no substantial relationship between the classification and some important state interest. Promotion of adoption of children born out of wedlock and their consequent “legitimation” was important, but the assumption that all unwed fathers either stood in a different relationship to their children than did the unwed mother or that the difficulty of finding the fathers would unreasonably burden the adoption process was overbroad, as the facts of the case revealed. No barrier existed to the state dispensing with consent when the father or his location is unknown, but disqualification of all unwed fathers may not be used as a shorthand for that step.¹²

On the other hand, the Court sustained a Georgia statute that permitted the mother of a child born out of wedlock to sue for the wrongful death of the child but that allowed the father to sue only if he had “legitimated” the child and there is no mother.¹³ Similarly, the Court let stand, under the Fifth Amendment, a federal statute that required that, in order for a child born out of wedlock overseas to gain citizenship, a citizen father, unlike a citizen mother, must acknowledge or “legitimate” the child before the child’s eighteenth birthday.¹⁴ The Court emphasized the ready availability of proof of a child’s maternity as opposed to paternity, but the dissent questioned whether such a distinction was truly justified under strict scrutiny considering the ability of modern techniques of DNA paternity testing to settle concerns about parentage.

The issue of sex qualifications for the receipt of governmental financial benefits has divided the Court and occasioned close distinctions. A statutory scheme under which a serviceman could claim his spouse as a “dependent” for allowances while a servicewoman’s spouse was not considered a “dependent” unless he was shown in fact to be dependent upon her for more than one half of his support was held an invalid dissimilar treatment of similarly situated men and women, not justified by the administrative convenience rationale.¹⁵ In Weinberger v. Wiesenfeld,¹⁶ the Court struck down a Social Security provision that gave survivor’s benefits based on the insured’s earnings to the widow and minor children but gave such benefits only to the children and not to the widower of a deceased woman worker. Focusing not only upon the discrimination against the widower but primarily upon the discrimination visited upon the woman worker whose earnings did not provide the same support for her family that a male worker’s did, the Court saw the basis for the distinction resting upon the generalization that a woman would stay home and take care of the children while a man would not. Because the Court perceived the purpose of the provision to be to enable the surviving parent to choose to remain at home to care for minor children, the sex classification ill-fitted the end and was invidiously discriminatory.

But, when, in Califano v. Goldfarb,¹⁷ the Court was confronted with a Social Security provision structured much as the benefit sections struck down in Frontiero and Wiesenfeld, even in the light of an express heightened scrutiny, no majority of the Court could be obtained for the reason for striking down the statute. The section provided that a widow was entitled to receive survivors’ benefits based on the earnings of her deceased husband, regardless of dependency, but payments were to go to the widower of a deceased wife only upon proof that he had been receiving at least half of his support from her. The plurality opinion treated the discrimination as consisting of disparate treatment of women wage-earners whose tax payments did not earn the same family protection as male wage earners’ taxes. Looking to the purpose of the benefits provision, the plurality perceived it to be protection of the familial unit rather than of the individual widow or widower and to be keyed to dependency rather than need. The sex classification was thus found to be based on an assumption of female dependency that ill-served the purpose of the statute and was an ill-chosen proxy for the underlying qualification. Administrative convenience could not justify use of such a questionable proxy.¹⁸ Justice John Paul Stevens, concurring, accepted most of the analysis of the dissent but nonetheless came to the conclusion of invalidity. His argument was essentially that while either administrative convenience or a desire to remedy discrimination against female spouses could justify use of a sex classification, neither purpose was served by the sex classification actually used in this statute.¹⁹

Again, the Court divided closely when it sustained two instances of classifications claimed to constitute sex discrimination. In Rostker v. Goldberg,²⁰ rejecting presidential recommendations, Congress provided for registration only of males for a possible future military draft, excluding women altogether. The Court discussed but did not explicitly choose among proffered equal protection standards, but it apparently applied the intermediate test of Craig v. Boren. However, it did so in the context of its often-stated preference for extreme deference to military decisions and to congressional resolution of military decisions. Evaluating the congressional determination, the Court found that it has not been “unthinking” or “reflexively” based upon traditional notions of the differences between men and women; rather, Congress had extensively deliberated over its decision. It had found, the Court asserted, that the purpose of registration was the creation of a pool from which to draw combat troops when needed, an important and indeed compelling governmental interest, and the exclusion of women was not only “sufficiently but closely” related to that purpose because they were ill-suited for combat, could be excluded from combat, and registering them would be too burdensome to the military system.²¹

In Michael M. v. Superior Court,²² the Court expressly adopted the Craig v. Boren intermediate standard, but its application of the test appeared to represent a departure in several respects from prior cases in which it had struck down sex classifications. Michael M. involved the constitutionality of a statute that punished males, but not females, for having sexual intercourse with a nonspousal person under eighteen years of age. The plurality and the concurrence generally agreed, but with some difference of emphasis, that, although the law was founded on a clear sex distinction, it was justified because it served an important governmental interest—the prevention of teenage pregnancies. Inasmuch as women may become pregnant and men may not, women would be better deterred by that biological fact, and men needed the additional legal deterrence of a criminal penalty. Thus, the law recognized that, for purposes of this classification, men and women were not similarly situated, and the statute did not deny equal protection.²³

In a 1996 case, the Court required that a state demonstrate “exceedingly persuasive justification” for gender discrimination. When a female applicant challenged the exclusion of women from the historically male-only Virginia Military Institute (VMI), the State of Virginia defended the exclusion of females as essential to the nature of training at the military school.²⁴ The state argued that the VMI program, which included rigorous physical training, deprivation of personal privacy, and an “adversative model” that featured minute regulation of behavior, would need to be unacceptably modified to facilitate the admission of women. While recognizing that women’s admission would require accommodation such as different housing assignments and physical training programs, the Court found that the reasons set forth by the state were not “exceedingly persuasive,” and thus the state did not meet its burden of justification. The Court also rejected the argument that a parallel program established by the state at a private women’s college served as an adequate substitute, finding that the program lacked the military-style structure found at VMI, and that it did not equal VMI in faculty, facilities, prestige, or alumni network.

The Court in Sessions v. Morales-Santana applied the “exceedingly persuasive justification” test to strike down a gender-based classification found in a statute that allowed for the acquisition of U.S. citizenship by a child born abroad to an unwed couple if one of the parents was a U.S. citizen.²⁵ The law at issue in Morales-Santana, which had been enacted many decades earlier, conditioned the grant of citizenship on the U.S. citizen parent’s physical presence in the United States prior to the child’s birth, providing a shorter presence requirement for an unwed U.S. citizen mother relative to the unwed U.S. citizen father.²⁶ According to the majority, such a classification “must substantially serve an important government interest today,”²⁷ and the law in question was based on “two once habitual, but now untenable, assumptions”: (1) that marriage presupposes that the husband is dominant and the wife is subordinate; (2) an unwed mother is the natural and sole guardian of a non-marital child.²⁸ Having found that the law was an “overbroad generalization[ ]” about males and females and was based on the “obsolescing view” about unwed fathers,²⁹ the Court concluded that the citizenship provision’s “discrete duration-of-residency requirements for unwed mothers and fathers who have accepted parental responsibility [was] stunningly anachronistic.”³⁰

In response to what the lower court had described as the “most vexing problem” in the case,³¹ the Morales-Santana Court, in crafting a remedy for the equal protection violation, deviated from the presumption that “extension, rather than nullification” of the denied benefit is generally the “proper course.”³² The Court observed that Congress had established derivative citizenship rules that varied depending upon whether one or both parents were U.S. citizens and whether the child was born in or outside marriage.³³ Justice Ruth Bader Ginsburg writing for the majority concluded that extending the much-shorter physical presence requirement applicable to unwed U.S. citizen mothers to unwed U.S. citizen fathers would run significantly counter to Congress’s intentions when it established this statutory scheme, because such a remedy would result in a longer physical presence requirement for a married U.S. citizen who had a child abroad than for a similarly situated unmarried U.S. citizen.³⁴ As a result, the Court held that the longer physical presence requirement for unwed U.S. citizen fathers governed, as that is the remedy that “Congress likely would have chosen had it been apprised of the constitutional infirmity.”³⁵

Another area presenting some difficulty is that of the relationship of pregnancy classifications to gender discrimination. In Cleveland Board of Education v. LaFleur,³⁶ which was decided upon due process grounds, two school systems requiring pregnant school teachers to leave work four and five months respectively, before the expected childbirths were found to have acted arbitrarily and irrationally in establishing rules not supported by anything more weighty than administrative convenience buttressed with some possible embarrassment of the school boards in the face of pregnancy. On the other hand, the exclusion of pregnancy from a state financed program of payments to persons disabled from employment was upheld against equal protection attack as supportable by legitimate state interests in the maintenance of a self-sustaining program with rates low enough to permit the participation of low-income workers at affordable levels.³⁷ The absence of supportable reasons in one case and their presence in the other may well have made the significant difference.

Amdt14.S1.8.8.4 Facially Non-Neutral Laws Benefiting Women

Cases of “benign” discrimination, that is, statutory classifications that benefit women and disadvantage men in order to overcome the effects of past societal discrimination against women, have presented the Court with some difficulty. Although the first two cases were reviewed under apparently traditional rational basis scrutiny, the more recent cases appear to subject these classifications to the same intermediate standard as any other sex classification. Kahn v. Shevin¹ upheld a state property tax exemption allowing widows but not widowers a $500 exemption. In justification, the state had presented extensive statistical data showing the substantial economic and employment disabilities of women in relation to men. The provision, the Court found, was “reasonably designed to further the state policy of cushioning the financial impact of spousal loss upon the sex for whom that loss imposes a disproportionately heavy burden.”² And, in Schlesinger v. Ballard,³ the Court sustained a provision requiring the mandatory discharge from the Navy of a male officer who has twice failed of promotion to certain levels, which in Ballard’s case meant discharge after nine years of service, whereas women officers were entitled to thirteen years of service before mandatory discharge for want of promotion. The difference was held to be a rational recognition of the fact that male and female officers were dissimilarly situated and that women had far fewer promotional opportunities than men had.

Although in each of these cases the Court accepted the proffered justification of remedial purpose without searching inquiry, later cases caution that “the mere recitation of a benign, compensatory purpose is not an automatic shield which protects against any inquiry into the actual purposes underlying a statutory scheme.”⁴ Rather, after specifically citing the heightened scrutiny that all sex classifications are subjected to, the Court looks to the statute and to its legislative history to ascertain that the scheme does not actually penalize women, that it was actually enacted to compensate for past discrimination, and that it does not reflect merely “archaic and overbroad generalizations” about women in its moving force. But where a statute is “deliberately enacted to compensate for particular economic disabilities suffered by women,” it serves an important governmental objective and will be sustained if it is substantially related to achievement of that objective.⁵

Many of these lines of cases converged in Mississippi University for Women v. Hogan,⁶ in which the Court stiffened and applied its standards for evaluating claimed benign distinctions benefitting women and additionally appeared to apply the intermediate standard itself more strictly. The case involved a male nurse who wished to attend a female-only nursing school located in the city in which he lived and worked; if he could not attend this particular school he would have had to commute 147 miles to another nursing school that did accept men, and he would have had difficulty doing so and retaining his job. The state defended on the basis that the female-only policy was justified as providing “educational affirmative action for females.” Recitation of a benign purpose, the Court said, was not alone sufficient. “[A] State can evoke a compensatory purpose to justify an otherwise discriminatory classification only if members of the gender benefited by the classification actually suffer a disadvantage related to the classification.”⁷ But women did not lack opportunities to obtain training in nursing; instead they dominated the field. In the Court’s view, the state policy did not compensate for discriminatory barriers facing women, but it perpetuated the stereotype of nursing as a woman’s job. “[A]lthough the State recited a ‘benign, compensatory purpose,’ it failed to establish that the alleged objective is the actual purpose underlying the discriminatory classification.”⁸ Even if the classification was premised on the proffered basis, the Court concluded, it did not substantially and directly relate to the objective, because the school permitted men to audit the nursing classes and women could still be adversely affected by the presence of men.⁹