Download PDF

Perry v. Schwarzenegger: Is Traditional Marriage Unconstitutional?Note from the Editor:

This article and the article in this issue by Mark Strasser provide different perspectives on the main issue at stake in the Ninth Circuit case Perry v. Schwarzenegger, namely whether California Proposition 8, an amendment to the state’s constitution providing that a marriage is a union between a man and a woman, violates the U.S. Constitution. We hope that publishing these articles helps contribute to the debate over this and other challenges to state constitutional provisions. The Federalist Society takes no position on particular legal or public policy initiatives. We welcome your responses to these articles; to join the debate, you can e-mail us at [email protected].


Until 2000 the legal institution of civil marriage was understood to be available only to one man and one woman. In 2000 Californians passed an initiative statute (Proposition 22) reaffirming that understanding. The California legislature then enacted a law authorizing domestic partnerships for same-sex couples that offer the same legal treatment as marriage under a different name.1 In 2008 the California Supreme Court nullified Proposition 22 and construed the state constitution to mandate that marriage be redefined to be available to same-sex couples.2

At the next opportunity, just five months later, the people of California approved Proposition 8, which added to the state constitution: “Only marriage between a man and a woman is valid or recognized in California.” The initiative did not affect domestic partnerships.

Two same-sex couples who were denied marriage licenses after passage of Proposition 8 sued, challenging its constitutionality. The Governor, Attorney General, and other state officials refused to defend the law. Sponsors of Proposition 8 intervened to defend it. Judge Vaughn Walker of the U.S. District Court for the Northern District of California held that the intervenors had standing to defend the law and that Proposition 8 violates both the Due Process and Equal Protection Clauses of the U.S. Constitution.3

The defendant-intervenors appealed to the U.S. Court of Appeals for the Ninth Circuit. A three-judge panel of that court heard oral argument on the case in December 2010.4

I. Defendants’ Standing

An initial question is the standing of the defendant-intervenors. In Arizonans for Official English v. Arizona5 the majority opinion by Justice Ginsburg expressed in dictum “grave doubts” whether sponsors of a ballot initiative have standing to defend it if elected officials refuse to do so. However, the purpose of ballot initiatives is to enable voters to enact laws that government officials refuse to adopt. To deny sponsors of initiatives standing to defend them would in effect privilege officials to nullify this democratic process. It is unlikely that the court of appeals or Supreme Court will allow such nullification.

II. Findings of Fact

In reaching his decision Judge Walker made several crucial—essentially dispositive—determinations that he labeled findings of fact. Ordinary findings of fact are reversed only if found on appeal to be clearly erroneous.6 However, the legislative and executive branches of government must constantly make findings of fact in order to formulate and enforce laws and regulations, and these “legislative facts” cannot be ignored by a trial court and are not subject to the “clearly erroneous” standard but to de novo review.

“Legislative choice is not subject to courtroom factfinding and may be based on rational speculation unsupported by evidence or empirical data.”7 The burden is “on the one attacking the legislative arrangement to negative every conceivable basis which might support it, whether or not the basis has a foundation in the record.”8 The issue, then, is whether the law satisfies the relevant standard of review, and this is an issue to be decided de novo by the appellate court, with due deference to democratic processes.

III. Issues Specific to Proposition 8

All but a few states have laws limiting marriage to one man and one woman. This suit certainly could have bearing on all those laws. Hesitant to overreach, the plaintiffs have struggled to identify particulars to differentiate Proposition 8 from other state laws and thereby narrow the scope (and the threat) of a ruling in their favor. One such particular is that Proposition 8 was adopted after the California Supreme Court mandated recognition of same-sex marriages. Thus, it is claimed, Proposition 8 differs from other state marriage laws because it deprived same-sex couples of an existing right rather than simply withholding a right they never had. However, if a right is not constitutionally mandated, how can it be unconstitutional for a state that has granted the right to change its mind and withdraw it?

In a few cases, the Supreme Court has overturned laws that withdrew constitutionally-discretionary rights because the Court found that the laws were impelled by an impermissible motive. In Romer v. Evans9 for example, the Court struck down a Colorado constitutional amendment adopted by voter initiative that withdrew from the state legislature and local governments the power to enact laws against sexual-orientation discrimination. The majority said that the law was motivated by animus—a bare desire to harm—because it was not “directed to any identifiable legitimate purpose or discrete objective.”10 The majority further objected to the law because it amended the state constitution and thereby precluded those seeking laws against sexual-orientation discrimination from attaining them through ordinary legislation.

As the minority pointed out, the very purpose of constitutional provisions is to erect a barrier against ordinary legislation. Further, in many states constitutional provisions are the only laws that can be adopted by voter initiative. Therefore, to nullify such an initiative in effect deprives citizens of any power to act on a particular matter, even though the goal they seek is permissible under the Federal Constitution. Romer seems to make much of state constitutions unconstitutional.

The minority in Romer also noted that the Colorado initiative merely overrode local laws that were not constitutionally mandatory. However, the holding that the initiative had no “legitimate purpose” seems to mean that there would also be no legitimate purpose for not having laws against sexual-orientation discrimination to begin with, which would mean that such laws are constitutionally mandatory. Neither Romer nor any other Supreme Court decision, though, has so held.

Moreover, it is easy to find legitimate purposes for the initiative in Romer. In a free society, people are generally free to choose with whom to deal, even if others might consider one’s choices irrational or improper. Discrimination is forbidden only on a few select grounds. The people of Colorado might plausibly have believed that discrimination based on sexual orientation is tolerable or that any problems it creates were not serious enough to require the heavy burden of government intrusion through antidiscrimination laws.

One cannot claim that the whole structure of marriage, recognized by every civilization throughout history, was contrived solely to harm homosexuals. As discussed below, it is also easy to find a legitimate purpose for Proposition 8. However, Romer seems to be a constitutional wild card—a precedent with no firm meaning that can simply be played whenever five Justices feel like striking down a law they do not like but in which there is no constitutional flaw.

IV. Standard of Constitutional Scrutiny

The central issue of Perry is the constitutional validity of laws restricting legal marriage to a woman and a man. A key subsidiary issue is the standard of constitutional scrutiny by which such laws should be reviewed under the Equal Protection and Due Process Clauses. All laws create distinctions and thus treat people unequally, and all laws limit rights either by forbidding some kind of behavior or by granting benefits for some persons or conduct and not others. In general, however, a law satisfies the Equal Protection and Due Process Clauses if the distinctions it makes have a rational basis. As noted above, this means only that there is some conceivable rational basis, even if that basis is not found in the record.

However, a few areas—those involving fundamental rights or distinctions that create a “suspect class”—are subject to strict scrutiny, requiring the state to show that the law serves a compelling interest that cannot be achieved by less discriminatory means.11 The paradigm category of strict scrutiny is supposed to be race12 because the Equal Protection Clause was adopted in the wake of the Civil War to protect the rights of former slaves who were being reduced to virtual serfdom by racially-discriminatory laws adopted in the former Confederate states.

Strict scrutiny is not necessarily fatal to discriminatory state action. For example, the Court has upheld racial preferences in university admissions on the factually dubious ground that racial diversity improves the quality of education.13

A strong argument can be made that traditional marriage serves a compelling state interest. The family is society’s most basic institution, and traditional marriage has always been considered crucial to the successful functioning of the family. The suffering of children in our neighborhoods where marriage has lost its prestige and has ceased to be the norm certainly argues for a compelling need to retain and promote traditional marriage.

However, strict scrutiny should not be the applicable standard. The Supreme Court has never applied anything more stringent than the rational basis standard to sexual orientation. The Court’s only decisions overturning laws based on sexual orientation are Romer, where the Court found no legitimate purpose for the law, and Lawrence v. Texas, which held only that disapproval of homosexual acts could not be enforced “through operation of the criminal law.”14 In both Lawrence and Romer the Court applied the rational basis standard. The Court in Lawrence said that Texas’s criminal sodomy law “furthers no legitimate state interest which can justify its intrusion into the personal and private life of the individual.”15 Perry involves no “intrusion into the personal and private li[ves]” of homosexuals. The Court in Lawrence said expressly that the case did “not involve whether the government must give formal recognition to any relationship that homosexual persons seek to enter.”16

The Equal Protection Clause was not intended originally to protect homosexuality. Lawrence might be defended on the ground that there had evolved a national consensus to tolerate private homosexual acts. Most states had repealed their criminal sodomy laws, and the few remaining laws were rarely enforced. In effect, these laws had ceased to be expressions of public morality and had become tools of arbitrary police harassment. It is not surprising that the Court would find such laws irrational.

Public attitudes about homosexual marriage are very different. Even if the meaning of equal protection can change with public consensus, that has not occurred here. Referenda on this issue have been held in thirty-one states, and in every one traditional marriage has been affirmed—usually by a large margin. Interracial marriage offers an instructive contrast. Prior to the Supreme Court’s decision in Loving v. Virginia (discussed further below), some state courts had struck down antimiscegenation laws. In none of these states was there a serious effort to restore the law by ballot initiative.

The Court has sometimes suggested that strict scrutiny applies to groups with “an immutable characteristic determined solely by birth.”17 This could be a very broad concept. Intelligence, for example, is to some significant extent hereditary, yet many state actions (like college admissions and matriculation) discriminate on the basis of intelligence. Even if homosexual  orientation is “an immutable characteristic,” traditional marriage is intended to encourage responsible procreation, a purpose that is clearly irrelevant to homosexual conduct.

Further, the concept arguably does not apply to homosexual marriage. First, it is unclear that sexual orientation is “determined solely by birth.” The American Psychiatric Association says “there are no replicated scientific studies supporting any specific biological etiology for homosexuality.”18 In at least some cases sexual inclinations may be influenced by experience.

Second, the issue in Perry is not sexual orientation but sexual behavior, which is not immutable. Some cultures have condoned some forms of homosexual activity,19 and in these cultures such activity has been more common than in cultures where that activity is severely condemned. Californians could reasonably decide that they do not want to encourage homosexual conduct by honoring same-sex marriages.

Strict scrutiny may also apply to laws that discriminate against “discrete and insular minorities” that are subject to widespread discrimination.20 Homosexuals certainly have experienced discrimination, although there is some question how widespread this discrimination is now. Unlike African-Americans, for example, homosexuals do not have lower average incomes than other Americans. California’s authorization of domestic partnerships with the same legal rights as marriage and the ability of opponents of Prop. 8 to raise more money for their campaign than its proponents did show further that in California homosexuals have significant political influence.

The passage of Prop. 8 alone does not establish that homosexuals in California are a powerless, oppressed minority needing constant judicial insulation from democracy. Every substantive law (and the rejection of every proposal for a new law) creates winners and losers. Supporters of many causes lose repeatedly, but not every such group is entitled to the privilege of strict scrutiny.

Most American marriage laws, for example, exclude not only same-sex couples, but also marriage of close relatives (“endogamy”) and marriage of groups of more than two persons (“polygamy” or “polyamory”). Supporters of these forms of marriage have not succeeded in any state, nor have they attained approval of civil unions or domestic partnerships for their relationships. Most advocates of same-sex marriage (including the plaintiffs in Perry) have not argued that these groups are “discrete and insular minorities” whose exclusion from marriage demands strict scrutiny review, yet it is hard to see why they (whose practices find more support in other societies than does same-sex marriage) do not merit as much judicial solicitude as homosexuals.

This inconsistency points to a more fundamental problem with the equal protection claim here. All parties in Perry agree that marriage is a privileged status. Plaintiffs do not challenge that status; they simply want same-sex couples to be eligible for it. However, if traditional marriage is unconstitutionally discriminatory, can any privileged status for marriage be upheld? The defense of traditional marriage is that it promotes responsible bearing and raising of children. If that defense is constitutionally inadequate, what constitutional justification is there for privileging marriage at all? Proponents of same-sex marriage do not answer this question.

V. The Fundamental Right to Marry

Judge Walker held that Proposition 8 violates the Due Process Clause by denying homosexuals a fundamental right of people to marry as they please. The Supreme Court has sometimes recognized a constitutional right to marry.21 However, this right has always been limited, and the Court has never held or even hinted in dictum that the right extended to same-sex couples.

Same-sex marriage is very different from the cases where the Court has recognized a right to marry. In Loving v. Virginia,22  the Court overturned a law forbidding interracial marriage. However, California does not forbid homosexual marriage; it simply does not license it, but leaves it as a private matter. Further, California offers homosexuals all the legal benefits of marriage, withholding only the label. The Court has never suggested that there is a fundamental right to the label “marriage.”

As noted, the limitation of marriage to opposite-sex couples is only one of several traditional restrictions on marriage. The parties must be unmarried, i.e., no polygamy. The parties must not be too closely-related, i.e., no endogamy. And the parties must be adults, i.e., no child marriage. Unlike the same-sex requirement, all these practices have been condoned in many societies. If there is a fundamental right to same-sex marriage, a fortiori all the other practices must be permitted. It seems unlikely that the Supreme Court wants to take such a step.

The fundamental right to marry should mean the right to enter into a relationship that falls within the traditional definition of marriage, not to legal recognition of whatever arrangement some person or group of people wants to label marriage. The law struck down in Loving was a sharp departure from the traditional definition of marriage in Western civilization, which never forbade interracial marriages.23 If anything, then, Loving is a precedent for adhering to, rather than nullifying, the exclusion of same-sex marriage because, unlike interracial marriages, Western civilization has never recognized same-sex relationships as marriages.

VI. The Case for (Traditional) Marriage

Judge Walker held that the “purported rationales” for the non-recognition of same-sex marriage “are nothing more than post-hoc justifications” by Prop. 8’s proponents. As with his finding that Prop. 8 was motivated by a desire to harm homosexuals, this conclusion seems to rest on the premise that the institution of marriage was fabricated in every culture throughout history for the sole purpose of stigmatizing homosexuals.

Our society generally leaves adults free to arrange their own affairs. However, a woman and a man can create children who cannot protect their own interests. Marriage practices have varied among cultures in myriad ways. However, whatever else marriage is about (e.g., caste or property), it has always been centrally concerned with the bearing and raising of children. As Bertrand Russell said: “But for children, there would be no need for any institution concerned with sex. . . . [I]t is through children alone that sexual relations become of importance to society.”24

Marriage both memorializes and solemnizes the relationship of a man and woman and provides the basis for an enforceable legal commitment among them and their children. Marriage both reminds the parties and informs the world that they have entered into a relationship with responsibilities to each other and to the human lives they may create. In so doing, it encourages them to plan for responsible procreation, which includes not only conception but everything that might affect children. We know, for example, that married men work longer hours, commit fewer crimes, and abuse drugs and alcohol less than unmarried men. In other words, marriage works.

The district court held that “same-sex parents and opposite-sex parents are of equal quality.”25 It is not clear exactly what this means or what is the basis for the statement. However, there have been no studies comparing same-sex parents with married, biological parents. If the district court’s statement means that the two are the same, it has no basis in fact. If it means something else, it is not relevant to the constitutionality of marriage.

Moreover, there are good reasons to believe that the finding is inaccurate. Innumerable studies have found the traditional family to be better for children than families with a single parent or cohabiting couples. If the district court is right, then cohabiting same-sex couples are better parents than cohabiting opposite-sex couples. The district court pointed to no studies purporting to support such a finding.

Same-sex couples can be allowed to adopt, but adoption is a legal event, not a biological act as is reproduction. Adopted children can be better protected through adoption proceedings and custody regulation than by fitting the square peg of same-sex relationships into the round hole of marriage.

The district court declared that “[p]ermitting same-sex couples to marry will not affect the number of opposite-sex couples who marry, divorce, cohabit, have children outside of marriage or otherwise affect the stability of opposite-sex marriages.”26 The only empirical basis the court gave for that finding was that recognition of same-sex marriages in Massachusetts supposedly has not affected rates of marriage and divorce.

This is not a firm basis on which to brand traditional marriage as irrational bigotry. First, Judge Walker’s empirical finding may not be correct. Defendant-intervenors cited studies showing that divorce rates rose and marriage rates fell in Massachusetts from 2004 to 2007.27 Similarly, marriage rates have declined and illegitimacy rates have risen in the Netherlands since it recognized same-sex marriages.

Further, the Massachusetts law has been in effect only briefly. More time is needed to determine its long-term effects, such as whether it will influence the raising of children and the use of artificial reproduction. Moreover, Massachusetts is a small state in a large country, in nearly all of which traditional marriage still prevails. Massachusetts may be atypical. Even its own residents may consider its law an aberration, not a general change of the meaning of marriage.

There are also substantial reasons to think that recognizing homosexual marriage would impair the social prestige of marriage. At trial both plaintiffs and defendant-intervenors introduced expert statements that validating same-sex marriage would radically alter the institution.28 Some gay activists support same-sex marriage for the express purpose of destroying its social standing.29

Recognizing same-sex marriage would transform marriage from its immemorial function as an arrangement centrally concerned with children to one primarily for the gratification of adults. As discussed above, thirty-one states have held referenda on the issue, and in every one the voters favored traditional marriage, usually by large margins. It is hard to believe that a Supreme Court decision branding the majority of Americans (and, indeed, virtually all human beings who have ever lived) as irrational bigots because they believe there is something special about the ability of a woman and a man to create human life would not diminish public respect for marriage.

Much of the legal benefit of marriage is achieved through the expressive function of law—the effect of the law in promoting certain norms by the law’s symbolic support. Perhaps the esteem for marriage generated by the law’s symbolic support would be impaired by extending it to intrinsically sterile relationships, but  this esteem  may be less impaired if a different label is used. Whether one thinks that California domestic partnerships go too far or not far enough in recognizing same-sex relationships, that approach is not irrational.

VII. Future Proceedings

If, as expected, the Ninth Circuit panel affirms the district court’s decision, the defendants could seek an en banc rehearing, or head for the Supreme Court as quickly as possible. The latter would set the stage for a Supreme Court hearing and decision in its 2011-12 Term, thus making the case a potential issue in the 2012 presidential and other elections.


The Constitution confers no right to legal validation of same-sex marriage. As Judge Richard Posner has said, “If there is such a right, it will have to be manufactured by the justices out of whole cloth.”30 For the Supreme Court to do so would gravely damage its legitimacy and invite efforts to change the composition of the Court. However justified the public anger at the obliteration of traditional marriage, such moves would create a dangerous precedent. It is hoped that the Court will not provoke such action.


* Schott-van den Eynden Professor of Law, Case Western Reserve University School of Law. 



1  Cal. Fam. Code § 297.5.

In re Marriage Cases, 183 P.3d 384 (Cal. 2008).

3  Perry v. Schwarzenegger, 704 F. Supp. 2d 921 (N.D. Cal. 2010).

4  Defendant-intervenors argue that the case is governed by Baker v. Nelson, 409 U.S. 810 (1972), in which the Court summarily rejected an appeal from a decision upholding the Minnesota marriage law that excluded homosexual couples. The Supreme Court generally gives little precedential force to such summary denials of appeals. See John E. Nowak & Ronald D. Rotunda, Constitutional Law § 2.5(c) (8th ed. 2010).

5  520 U.S. 43 (1997).

6  Fed. R. Civ. Pro. 52(a)(6).

7  Heller v. Doe, 509 U.S. 312, 320 (1993).

Id. at 320-21.

9  517 U.S. 620 (1996).

10  Id. at 636.

11  United States v. Carolene Products Co., 304 U.S. 144, 152 n.4 (1938).

12  See Nowak & Rotunda, supra note 4, § 14.3(a)(iii).

13  Grutter v. Bollinger, 539 U.S. 306 (2003).

14  539 U.S. 558, 571 (2003).

15  Id. at 578.

16  Id.

17  Frontiero v. Richardson, 411 U.S. 677, 686 (1973) (plurality op.).

18  American Psychiatric Association, Gay/Lesbian/Bisexuals (2009).

19  Significantly, however, the homosexual conduct condoned has usually been the sodomizing of boys by adult men or of slaves or other social inferiors by their masters or social superiors. Almost never have same-sex relationships between equal adults been approved.

20  United States v. Carolene Products Co., 304 U.S. 144 (1938).

21  E.g., Zablocki v. Redhail, 434 U.S. 374 (1978) (overturning a law forbidding marriage to persons who are in violation of a court order of child support).

22  388 U.S. 1 (1967).

23  See Irving G. Tragen, Statutory Prohibitions Against Interracial Marriage, 32 Cal. L. Rev. 269, 269 & n.2 (1944) (“[A]t common law there was no ban on interracial marriage . . . .”).

24  Bertrand Russell, Marriage and Morals 77, 156 (1929).

25  Perry v. Schwarzenegger, 704 F. Supp. 2d 921, slip op. at 127 (N.D. Cal. 2010).

26  Id. at 83.

27  Ctr. for Disease Control, Divorce Rates by State: 1990, 1995, and 1999-2009, available at; Ctr. for Disease Control, Marriage Rates by State: 1990, 1995, and 199-2009, available at

28  See, e.g., Witherspoon Inst., Marriage and the Public Good: Ten Principles 18-19 (2008).

29  See, e.g., Michelangelo Signorile, Bridal Wave, Out Mag., Dec./Jan. 1994.

30  Richard Posner, Wedding Bell Blues, New Republic Online, Dec. 19, 2003, available at