Forgetting Romer

Susannah W. Pollvogt *

What are the implications of the Court’s decision to grant certiorari in Hollingsworth v. Perry? 1

No. 12-144, 2012 WL 3134429 (U.S. granted Dec. 7, 2012), granting cert. to Perry v. Brown, 671 F.3d 1052, 1063 (9th Cir. 2012) (Reinhardt, J.) (striking California’s Proposition 8, Cal. Const. art. I, § 7.5, as unconstitutional under the federal Equal Protection Clause). Proposition 8, enacted by popular referendum, prohibited the recognition of same-sex marriages, thereby reversing the effect of an earlier California Supreme Court decision, In re Marriage Cases, 43 Cal. 4th 757 (2008), which granted marriage rights to same-sex couples. See Perry, 671 F.3d at 1063, 1067.

On the same day that the Supreme Court granted certiorari in Hollingsworth, it also granted certiorari in United States v. Windsor, No. 12-307, 2012 WL 4009654 (U.S. granted Dec. 7, 2012), granting cert. to 699 F.3d 169 (2d Cir. 2012), which presents the constitutionality of the federal Defense of Marriage Act (DOMA), 1 U.S.C. § 7 (2011). See Petition for a Writ of Certiorari Before Judgment at I, Windsor, No. 12-307. For reasons beyond the scope of this Essay, I predict that the Court will strike down DOMA on narrow grounds that will not apply to the Court’s review of Proposition 8 or other state-law prohibitions against same-sex marriage. The Court’s reasoning regarding Proposition 8, by contrast, could govern the outcome of other marriage equality litigation.

Advocates of marriage equality may worry that the Court granted certiorari to overturn the decision. But they should also worry that the Court accepted certiorari to affirm the decision on the same narrow legal and factual grounds relied upon by the Ninth Circuit. Because, while the Ninth Circuit’s reasoning was good for marriage equality in California, it could be devastating to marriage equality efforts in other jurisdictions.

Why? The Ninth Circuit interpreted a key doctrine—unconstitutional animus 2 Animus is an elusive yet powerful concept in the Court’s equal protection jurisprudence. The essence of the doctrine is that public laws may not be used to enforce legal distinctions between social groups. See Susannah W. Pollvogt, Unconstitutional Animus, 81 Fordham L. Rev. 887, 926 (2012), available at Thus, laws that perform this function will be struck down under any level of equal protection scrutiny, including deferential rational basis review. —in a way that strips the concept of much of its justice-forcing power. 3 Cf. Nan D. Hunter, Animus Thick and Thin: The Broader Impact of the Ninth Circuit Decision in Perry v. Brown, 64 Stan. L. Rev. Online 111, 112 (2012), (asserting that the Ninth Circuit’s decision’s “elaboration of the role of animus in judicial review is an important contribution to equal protection doctrine”). It did so by (1) attaching the concept of unconstitutional animus to a narrow and unique set of facts, and (2) relying excessively on the Supreme Court’s decision in Romer v. Evans. 4 517 U.S. 620 (1996). The law at issue in Romer was Colorado’s Amendment 2. Colo. Const. art. II, § 30b (eliminating all antidiscrimination protection for lesbians, gays, and bisexuals), invalidated by Romer, 517 U.S. at 634-36 (concluding that Amendment 2’s sole purpose was to harm a politically unpopular group) (citing U.S. Dep’t of Agric. v. Moreno, 413 U.S. 528, 534 (1973)). This reliance is problematic because, although Romer is the Court’s most recent pronouncement on the doctrine of unconstitutional animus, its reasoning is irretrievably compromised. Romer was on the right side of history in terms of holding that sexual minorities were entitled to equal treatment under the law, but the Court reached this outcome in direct contradiction of the applicable precedent of the time. 5 Specifically, Bowers v. Hardwick, 478 U.S. 186 (1986), overruled by Lawrence v. Texas, 539 U.S. 558, 578 (2003), held that states could selectively criminalize acts of sodomy engaged in by homosexuals. See Bowers, 478 U.S. at 190-96. For this reason, the reasoning in Romer is tortured, opaque, and incomplete. Yet because the facts of Romer arguably parallel the facts the Ninth Circuit confronted in Perry, the Ninth Circuit made Romer the centerpiece of its analysis.

To get marriage equality right, the Supreme Court will have to both look past the unusual factual circumstances of Proposition 8 and put Romer in its proper place with respect to the Court’s broader animus jurisprudence. Ultimately, because Romer is irretrievably compromised by the historical moment at which it was decided, the Court must forget Romer.

Judicial Restraint in Perry

Judge Reinhardt’s opinion in Perry was rightly hailed as a major victory for marriage equality. Not only did he reach the “right” outcome from the perspective of marriage equality advocates, he did so in a thorough and carefully reasoned opinion that—unsurprisingly—manifested a sophisticated awareness of the dynamics of judicial review. In particular, Judge Reinhardt relied almost exclusively on uncontroversial, “so-called ‘adjudicative facts’” found by the federal district court, not on any more than one of the more controversial findings of “legislative facts.” Further, he emphasized that a substantial portion of these facts were binding interpretations of the California Constitution by the State’s highest court and thus beyond the scrutiny of the federal courts. 6 See Perry v. Brown, 671 F.3d 1052, 1075, 1079 (9th Cir. 2012) (declining to classify an important, but undisputed, fact as either adjudicative or legislative) (internal quotation marks omitted), cert. granted sub nom. Hollingsworth v. Perry, 2012 WL 3134429 (U.S. Dec. 7, 2012) (No. 12-144). In addition, he confined the court’s holding to the specific mechanism of California’s Proposition 8, not opining on the broader right to marriage. All of these maneuvers were presumably designed to make the Perry decision thoroughly defensible when subjected to Supreme Court review, if not cert-proof altogether.

But Judge Reinhardt’s cautious approach is problematic for marriage equality in other jurisdictions for two related reasons. First, by maintaining a narrow factual focus, the opinion can be interpreted as holding that unconstitutional animus is present only where a state grants rights to a group and then later takes those rights away. 7 Cf. Colo. Const. art. II, § 30b (withdrawing homosexuals’ previously viable rights), invalidated by Romer, 517 U.S. at 634-36. Thus, other courts could ignore the doctrine of unconstitutional animus altogether if the law at issue did not take away previously granted marriage rights. Judge Reinhardt likely took this tack for a number of reasons. As William Eskridge has pointed out, deciding cases on the narrowest grounds possible is an appropriate exercise of judicial restraint. 8 See William N. Eskridge Jr., The Ninth Circuit’s Perry Decision and the Constitutional Politics of Marriage Equality, 64 Stan. L. Rev. Online 93, 97 (2012), In addition, it is the apex of legal reasoning to draw a tight factual analogy to existing precedent.

But the downside of this approach is that Judge Reinhardt practically invited other courts to distinguish Perry on its facts, thereby severely limiting the precedential impact of the decision. And indeed, this is precisely what happened. Just a few months after Perry was handed down, the federal district court in Hawai‘i concluded that Perry was inapposite to the marriage equality challenge in that State because Hawai‘i had never previously granted marriage rights to same-sex couples, and thus that the Hawai‘ian legal regime was distinguishable from California’s Proposition 8. 9 See Jackson v. Abercrombie, Civ. No. 11-00734 ACK KSC, 2012 WL 3255201, at *18-21 (D. Haw. Aug. 8, 2012), appeal docketed, Nos. 12-16995, 12-16998 (9th Cir. Sept. 10, 2012). Indeed, having noted this factual distinction, the Hawai‘i district court did not even go on to examine whether a broader understanding of animus might be relevant to the challenge before it. 10 See id. at *18-22. The court further found that Romer was inapposite, reading Romer as standing for the proposition that animus is inferred only from unprecedented or “unusual” laws, and observing that bans on same-sex marriage were quite common, not unusual. Id. at *22.

In addition to suggesting this overly narrow factual prerequisite for a finding of animus, Judge Reinhardt presented a cabined view of animus because he relied excessively on the legal framework set forth in Romer. By lashing Perry to the mast of Romer, Judge Reinhardt minimized the grounds for reversing his decision and appealed to the sensibilities of Justice Kennedy, who authored Romer and still sits on the Court. 11 See, e.g., Rick Hasen, Initial Thoughts on 9th Circuit Prop. 8 Decision, Election Law Blog (Feb. 7, 2012, 11:23 AM), (“Judge Reinhardt wrote his opinion for an audience of one: Justice Kennedy.”). But in the process, Judge Reinhardt also reiterated Romer’s incomplete and ultimately incoherent understanding of the doctrine of unconstitutional animus.

Romer and Doctrinal Appeasement

In a sense, Romer’s incoherence is not its fault. The decision was stuck between a rock and a hard place, the rock being the Court’s contemptible 1986 ruling in Bowers v. Hardwick and the hard place being the seven years that would have to pass post-Romer before the Court overturned Bowers in Lawrence v. Texas. 12 See supra note 6. In short, because Bowers essentially held that it was permissible for states to criminalize not only homosexual conduct, but also homosexual identity, 13 See Lawrence v. Texas, 539 U.S. 558, 568-71 (2003) (distinguishing between traditional sodomy laws which criminalized all acts of non-procreative sex and contemporary sodomy laws which criminalized sodomy only when engaged in by homosexuals). the Romer Court had to perform analytical gymnastics to reach a pro-gay-rights outcome.

At issue in Romer was Colorado’s Amendment 2, which had been enacted by popular referendum and functioned to (1) repeal at every level within the state all existing antidiscrimination protections based on sexual orientation and (2) prohibit any such protections from being enacted in the future. 14 Romer v. Evans, 517 U.S. 620, 624 (1996) (citing Colo. Const. art. II, § 30b). In examining the constitutionality of the law, the Court did not address whether sexual orientation was a suspect classification or whether Amendment 2 implicated a fundamental right (a positive answer to either of these questions would have required the Court to apply some form of heightened scrutiny). Instead, the Court proceeded directly to rational basis review and concluded that the law could not survive under even this deferential standard. 15 Id. at 631-32.

Because Bowers stood for the proposition that naked antigay bias (when aligned with conventional morality) was a permissible basis for a law, the Romer Court could not point to the strongest evidence of unconstitutional animus available in that case—the ample direct evidence of antigay bias in Amendment 2 campaign literature. Instead, the Court invoked a more generic concept of animus and created a “novel structural analysis” for identifying the presence of animus that amounted to little more than an untethered assessment of the overall fairness of the law. 16 See Pollvogt, supra note 2, at 911-12. This analysis had three steps. First, the Court took quite some time to describe the impact of Amendment 2 and characterized that impact as vast. Second, it weighed the impact against the purported goals of the law—conserving governmental resources and protecting Coloradans’ freedom of association—and characterized both as relatively trivial. Finally, from the lack of fit between the law’s impact and its goals, the Court inferred the presence of unconstitutional animus, which it described as “the bare . . . desire to harm a politically unpopular group.” 17 Romer, 517 U.S. at 634-35 (quoting U.S. Dep’t of Agric. v. Moreno, 413 U.S. 528, 534 (1973)) (alteration in original).

Romer’s understanding of unconstitutional animus is both incomplete and misleading. First, although “the bare desire to harm a politically unpopular group” is one form of animus, it is not the only form. Indeed, nowhere else does the Supreme Court’s animus jurisprudence require a plaintiff to prove either that the challenged law was motivated by a “desire to harm” or that the plaintiff is a member of a “politically unpopular” group. Rather, the animus inquiry ultimately does not focus on the subjective intent motivating a law but on whether the law functions to enforce private bias. 18 This focus on the broader concept of private bias was stated most forcefully in the Court’s decision in Palmore v. Sidoti: “Private biases may be outside the reach of the law, but the law cannot, directly or indirectly, give them effect.” 466 U.S. 429, 433 (1984). The Court has recognized such diverse mindsets as fear of, negative attitudes toward, and stereotypes of a particular social group as being within the category of “private bias.” 19 See Pollvogt, supra note 2, at 924-26 (citing case support for the proposition that animus is present not only where there is a “desire to harm” a particular social group but any time a law functions to create and enforce distinctions between social groups). Where the evidence indicates that a law was enacted to express such sentiments or beliefs, the law may be struck down on that basis.

Second, the tortured reasoning in Romer strongly implies that direct evidence of social-group bias is insufficient or perhaps even irrelevant to proving the presence of unconstitutional animus. But again, when one examines the Court’s broader animus jurisprudence, this is clearly not the case. Rather, in cases where there is direct evidence of private bias motivating the enactment of a law, the Court has easily found the presence of animus and struck down the law. 20 See id. (discussing types of evidence the Supreme Court has traditionally accepted as proving the presence of animus).

Third, the Romer analysis leaves unanswered the critical question of the relationship between animus and rational basis review. The cases are clear that animus can never constitute a legitimate state interest sufficient to survive rational basis review. But what if there are other rational bases for a law independent of impermissible animus? The Romer decision suggests that the absence of a rational basis is itself evidence of animus—a framework that renders a finding of animus doctrinally gratuitous (that is, if a law lacks a rational basis, it fails rational basis review and need not be struck down on the separate ground of animus). Some commentators, and at least one Supreme Court Justice, have concluded that evidence of animus triggers “heightened rational basis review,” such that laws that normally would survive rational basis review may be invalidated when animus is afoot. In other words, the presence of animus requires the state to provide a more persuasive justification for the law. But this poses the question of whether there is any justification that would ever be sufficient to save a law when direct evidence of animus is present. As a historical matter, the Court has never found the presence of animus and then gone on to uphold the challenged law, suggesting that animus is a doctrinal silver bullet.

Thus, while Romer offers some guidance on the doctrine of unconstitutional animus, it is far from a comprehensive account. And in fixating on the factual parallels between Colorado’s Amendment 2 and California’s Proposition 8, Judge Reinhardt overlooked additional doctrinal resources presented by the full scope of the Court’s animus jurisprudence.

Getting Animus Right

Excessive reliance on Romer clouds the true nature and doctrinal potential of unconstitutional animus. Further, a correct understanding of animus matters for future marriage equality litigation because, in the majority of jurisdictions, there will be ample direct evidence that private bias against gays and lesbians was the primary motivating force behind same-sex marriage bans. This evidence, in turn, strongly suggests that the function of those laws is merely to enforce that private bias—something the public laws may not do.

By contrast, if animus is understood narrowly as existing only where a law withdraws previously granted rights, the majority of same-sex marriage bans will be upheld, despite the presence of blatant antigay propaganda surrounding enactment of these measures.

It is important to recognize that the Ninth Circuit’s decision in Perry did not purport to provide a comprehensive account of the doctrine of unconstitutional animus, and it should not be interpreted as doing so. Judicial restraint of the type exercised by Judge Reinhardt in Perry is indeed generally a virtue, but not in circumstances where it perpetuates doctrinal confusion. The marriage equality cases, including Perry, provide the Court with an opportunity to rationalize the doctrine of unconstitutional animus and articulate a clear, consistent, and principled standard for courts to apply going forward.