Breaking News

Commentary on marriage grants: Different ways of splitting the difference – the menu of options in Hollingsworth v. Perry

The blog is pleased to have commentary and analysis of Friday’s grants in the marriage cases from supporters of both sides.  This post has reactions from Kenji Yoshino.

On Friday, the Supreme Court granted review in both United States v. Windsor and Hollingsworth v. Perry. The first case is a federal challenge to an Act of Congress — the 1996 Defense of Marriage Act — which defines marriage for all federal purposes as a union of one man and one woman. The second is a federal challenge to a state ballot initiative — the 2008 amendment to the California Constitution known as Prop. 8 — which eliminated the pre-existing, if short-lived, right of same-sex couples to marry in California. Everyone assumed the Court would grant one of the eight petitions to review the constitutionality of DOMA, simply because the Supreme Court almost invariably reviews lower court decisions striking down Congressional statutes. The betting was much closer on whether the Court would grant review in Perry. Now that the Court has taken both, some have speculated that the Court will split the difference — striking down DOMA but upholding Prop. 8. I think this forecast is incorrect.

The logic undergirding this prediction is sufficiently intuitive that I should clarify where my disagreement lies. The first premise is that the Court will strike down DOMA. I agree. DOMA represents an intrusion of federal law into the traditional state domain of family law.  As the lower courts have pointed out through various formulations, invalidating DOMA would represent a triumph for state sovereignty as well as for gay rights. Justices on the right tend to favor state power (relative to federal power); Justices on the left tend to favor gay rights. The Justice in the middle — Justice Kennedy — has historically favored both.

The second premise is that the Court will wish to proceed incrementally — that it will not, in one Term, strike down DOMA and flip the forty-one states that do not currently recognize same-sex marriage. Here, too, I agree. In 1967, when the Court decided Loving v. Virginia, it only had to invalidate the laws of sixteen states. In general, the Court does not like to get too far in front of national consensus.

It might seem to follow from these two premises that the Court will split the baby between the cases. But the error lies in thinking that the Perry Court must require marriage in all fifty states or none. In fact, the Court can more narrowly invalidate Prop. 8 in at least three ways.

The procedural one-state solution

Under the procedural one-state solution, the Court would find that the proponents of Prop. 8 lack Article III standing to bring suit. The Court showed its interest in that argument by requiring the parties to brief this issue. It is, of course, axiomatic that a state may appeal adverse court decisions against it. But in Perry, the California state officials named in the suit declined to defend Prop. 8. In the 1987 case of Karcher v. May, the Court permitted state legislators to defend a law when the governmental officials named in the suit refused to do so. One open question in Perry was whether California law similarly authorized the proponents to bring suit. For this reason, the Ninth Circuit panel certified the issue to the California Supreme Court, which determined that proponents possessed such authority under state law. The case then returned to the Ninth Circuit, which determined that the proponents had established Article III standing as a matter of federal law.

To reach that result, however, the Ninth Circuit had to square its analysis with an earlier statement by the U.S. Supreme Court. In Arizonans for Official English v. Arizona, sponsors of a ballot initiative had also launched an appeal after state officials declined to do so. The Court expressed “grave doubts” about whether sponsors of a ballot initiative in Arizona had standing to bring suit. It observed that the sponsors of the initiative “are not elected representatives, and we are aware of no Arizona law appointing initiative sponsors as agents of the people of Arizona to defend, in lieu of public officials, the constitutionality of initiatives made law of the state.” However, given that it found the case moot, the Court did not need to resolve the standing question. The exact basis for its doubts was left unclear.

The Ninth Circuit panel in Perry found that this language in Arizonans was distinguishable. It noted that Arizona had not delegated authority to the sponsors of the initiative, while California had done so.  If the Court hews closely to the language quoted above, it will probably accept this reasoning. However, the Court’s exploratory dictum in Arizonans might express a broader set of concerns about the operation of unchecked direct democracy. If that is so, the proponents cannot assume that the Court’s “grave doubts” will be resolved in their favor.

A ruling that the proponents lacked standing would reinstate the district court’s opinion, given that it would mean that an improper party had appealed that decision. As the Ninth Circuit indicated during oral argument, the impact of a ruling based on standing would be limited to requiring the clerks of Alameda County and Los Angeles County — the only county clerks named in the complaint — to issue marriage licenses to same-sex couples. Plaintiffs’ attorney David Boies predicted that the Governor would then require other counties in California to operate in a manner that would ensure statewide consistency. He further averred that if the Governor refused to do so, the plaintiffs would sue to secure such uniformity. The upshot of this solution would be that only California would be affected.

The substantive one-state solution

The Court could also adopt a substantive one-state solution, which was the rule adopted by the Ninth Circuit panel. The Ninth Circuit held that under the Equal Protection Clause, a state could not grant an entitlement and then take it away without a legitimate reason. The panel stressed that prior to Prop. 8, the California Supreme Court had held that the state constitution required the state to allow same-sex couples to marry. According to the Ninth Circuit, Prop. 8’s withdrawal of that previously granted entitlement had no rational justification.

In reaching that conclusion, the Ninth Circuit panel relied heavily on the 1996 case of Romer v. Evans. The Romer Court invalidated a Colorado constitutional amendment enacted through a plebiscite that stated that there could be “no protected status” on the basis of gay, lesbian, or bisexual identity. The constitutional amendment in question (Amendment 2) superseded anti-discrimination laws protecting gays that had been passed by progressive cities in Colorado. Those cities were blue dots in a red state. Amendment 2 invalidated those municipal ordinances and, going forward, prohibited the state or any of its subdivisions from enacting a similar kind of anti-discrimination ordinance.

The Supreme Court struck down the Colorado amendment under the Equal Protection Clause without specifying a level of scrutiny. It observed that Amendment 2 was “at once too narrow and too broad” — it singled out a group of people on the basis of a single trait and then denied them protections across the board. The sweeping nature of the disability on a small group of people effectively made gay individuals “strangers” to the law. The Court stated that the imposition of such harm could only be explained by animus.

In the Perry case, the Ninth Circuit panel detected the same animus. One problem with the panel’s analysis was that Prop. 8 eliminated only one right — the right to marry. The panel acknowledged this distinction, saying that animus could be present even (or perhaps especially) when a state enacted legislation with “surgical precision.” Yet this interpretation extends Romer, which repeatedly emphasized the breadth of the harm imposed by Amendment 2 as part of what was constitutionally objectionable about it.

This objection, however, simply  means that Romer, standing alone, may not dispose of Perry. Because they are both gay-rights cases involving ballot measures, the understandable tendency has been to link Perry with Romer. Yet Romer is but one member of a family of cases in which the Court has invalidated laws that lacked a rational basis. Many of these cases invalidated laws much narrower than Prop. 8. In City of Cleburne v. Cleburne Living Center, the Court struck down a city zoning ordinance that discriminated against individuals with mental disabilities, applying equal protection rational basis review. Similarly, in United States Department of Agriculture v. Moreno, the Court struck down a portion of a federal food stamps program that discriminated against “hippies.” The best reading of these precedents is that the problem with legislation based solely on animus is the animus, not the breadth of the legislation. Such breadth is but one of the many possible indicia pointing to the existence of animus.

If the Court adopted this substantive rule, the decision would only affect the capacity of same-sex couples to marry in California, at least for the time being. No other state has permitted same-sex couples to marry before taking away the right. It would, however, preclude the nine states that currently permit same-sex marriage from withdrawing that right, unless they could show a rational basis for doing so.

The substantive eight-state solution

Yet another breakpoint on the spectrum would focus on the lack of justification for giving same-sex couples all the rights and responsibilities of marriage but withholding the word “marriage” from them. This resolution differs from the Ninth Circuit panel’s ruling because it removes any issue of retrogression from the analysis. What is important is not that California went all the way to same-sex marriage and then retreated, but rather that California went all the way to “everything but marriage.” Once it did so, it reached the point of no return. Currently, seven states besides California would be affected by such a ruling: Delaware, Hawaii, Illinois, Nevada, New Jersey, Oregon, and Rhode Island.

Such a ruling could have perverse effects, given that later legislatures that might otherwise have been willing to compromise at “everything but marriage” unions might switch to endorsing only weaker recognition or no recognition at all. Increasingly, however, the popularity of even “everything but marriage” arrangements appears to be waning. Indeed, reports from New Jersey and Rhode Island suggest that same-sex couples are unwilling to avail themselves of civil unions because they are holding out for full equality.


The purpose of this short intervention is not to endorse any of these three options — much more would need to be said about each. I seek to make a much narrower point, which is that Perry presents the Court with a menu of at least five options, not a binary choice between zero and fifty. For this reason, the impulse to proceed incrementally does not require that the Court rule against the Perry plaintiffs if it rules for the Windsor plaintiffs. There are plenty of options within Perry itself that would allow the Court to move at what one might call a stately pace.

Kenji Yoshino is the Chief Justice Earl Warren Professor of Constitutional Law at NYU School of Law. This essay reproduces or draws heavily on portions of his 2011 Leary Lecture at the University of Utah, which has just been published here.


Recommended Citation: Kenji Yoshino, Commentary on marriage grants: Different ways of splitting the difference – the menu of options in Hollingsworth v. Perry, SCOTUSblog (Dec. 8, 2012, 9:48 AM),