UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT
February 7, 2012
KRISTIN M. PERRY; SANDRA B. STIER; PAUL T. KATAMI; JEFFREY J. ZARRILLO, PLAINTIFFS-APPELLEES, CITY AND COUNTY OF SAN FRANCISCO, INTERVENOR-PLAINTIFF-APPELLEE,
EDMUND G. BROWN, JR., IN HIS OFFICIAL CAPACITY AS GOVERNOR OF CALIFORNIA; KAMALA D. HARRIS, IN HER OFFICIAL CAPACITY AS ATTORNEY GENERAL OF CALIFORNIA; MARK B. HORTON, IN HIS OFFICIAL CAPACITY AS DIRECTOR OF THE CALIFORNIA DEPARTMENT OF PUBLIC HEALTH & STATE REGISTRAR OF VITAL STATISTICS; LINETTE SCOTT, IN HER OFFICIAL CAPACITY AS DEPUTY DIRECTOR OF HEALTH INFORMATION & STRATEGIC PLANNING FOR THE CALIFORNIA DEPARTMENT OF PUBLIC HEALTH; PATRICK O'CONNELL, IN HIS OFFICIAL CAPACITY AS CLERK-RECORDER FOR THE COUNTY OF ALAMEDA; DEAN C. LOGAN, IN HIS OFFICIAL CAPACITY AS REGISTRAR-RECORDER/COUNTY CLERK FOR THE COUNTY OF LOS ANGELES, DEFENDANTS, HAK-SHING WILLIAM TAM, INTERVENOR-DEFENDANT, AND DENNIS HOLLINGSWORTH; GAIL J. KNIGHT; MARTIN F. GUTIERREZ; MARK A. JANSSON; PROTECTMARRIAGE.COM-YES ON 8, A PROJECT OF CALIFORNIA RENEWAL, AS OFFICIAL PROPONENTS OF PROPOSITION 8, INTERVENOR-DEFENDANTS-APPELLANTS. KRISTIN M. PERRY; SANDRA B. STIER; PAUL T. KATAMI; JEFFREY J. ZARRILLO, PLAINTIFFS-APPELLEES, CITY AND COUNTY OF SAN FRANCISCO, INTERVENOR-PLAINTIFF-APPELLEE,
EDMUND G. BROWN, JR., IN HIS OFFICIAL CAPACITY AS GOVERNOR OF CALIFORNIA; KAMALA D. HARRIS, IN HER OFFICIAL CAPACITY AS ATTORNEY GENERAL OF CALIFORNIA; MARK B. HORTON, IN HIS OFFICIAL CAPACITY AS DIRECTOR OF THE CALIFORNIA DEPARTMENT OF PUBLIC HEALTH & STATE REGISTRAR OF VITAL STATISTICS; LINETTE SCOTT, IN HER OFFICIAL CAPACITY AS DEPUTY DIRECTOR OF HEALTH INFORMATION & STRATEGIC PLANNING FOR THE CALIFORNIA DEPARTMENT OF PUBLIC HEALTH; PATRICK O'CONNELL, IN HIS OFFICIAL CAPACITY AS CLERK-RECORDER FOR THE COUNTY OF ALAMEDA; DEAN C. LOGAN, IN HIS OFFICIAL CAPACITY AS REGISTRAR-RECORDER/COUNTY CLERK FOR THE COUNTY OF LOS ANGELES, DEFENDANTS, HAK-SHING WILLIAM TAM, INTERVENOR-DEFENDANT, AND DENNIS HOLLINGSWORTH; GAIL J. KNIGHT; MARTIN F. GUTIERREZ; MARK A. JANSSON; PROTECTMARRIAGE.COM-YES ON 8, A PROJECT OF CALIFORNIA RENEWAL, AS OFFICIAL PROPONENTS OF PROPOSITION 8, INTERVENOR-DEFENDANTS-APPELLANTS.
Appeal from the United States District Court for the Northern District of California Vaughn R. Walker, Chief District Judge, Presiding (No. 10-16696) James Ware, Chief District Judge, Presiding (No. 11-16577) D.C. No. 3:09-cv-02292-VRW D.C. No. 3:09-cv-02292-JW
The opinion of the court was delivered by: Reinhardt, Circuit Judge
Argued and Submitted December 6, 2010 San Francisco, California
Submission Withdrawn January 4, 2011
Resubmitted February 7, 2012 No. 11-16577:
Argued and Submitted December 8, 2011 San Francisco, California
Before: Stephen Reinhardt, Michael Daly Hawkins, and N. Randy Smith, Circuit Judges.
Opinion by Judge Reinhardt; Partial Concurrence and Partial Dissent by Judge N.R. Smith
Prior to November 4, 2008, the California Constitution guaranteed the right to marry to opposite-sex couples and same-sex couples alike. On that day, the People of California adopted Proposition 8, which amended the state constitution to eliminate the right of same-sex couples to marry. We consider whether that amendment violates the Fourteenth Amendment to the United States Constitution. We conclude that it does.
Although the Constitution permits communities to enact most laws they believe to be desirable, it requires that there be at least a legitimate reason for the passage of a law that treats different classes of people differently. There was no such reason that Proposition 8 could have been enacted. Because under California statutory law, same-sex couples had all the rights of opposite-sex couples, regardless of their marital status, all parties agree that Proposition 8 had one effect only. It stripped same-sex couples of the ability they previously possessed to obtain from the State, or any other authorized party, an important right-the right to obtain and use the designation of 'marriage' to describe their relationships. Nothing more, nothing less. Proposition 8 therefore could not have been enacted to advance California's interests in child-rearing or responsible procreation, for it had no effect on the rights of same-sex couples to raise children or on the procreative practices of other couples. Nor did Proposition 8 have any effect on religious freedom or on parents' rights to control their children's education; it could not have been enacted to safeguard these liberties.
All that Proposition 8 accomplished was to take away from same-sex couples the right to be granted marriage licenses and thus legally to use the designation of 'marriage,' which symbolizes state legitimization and societal recognition of their committed relationships. Proposition 8 serves no purpose, and has no effect, other than to lessen the status and human dignity of gays and lesbians in California, and to officially reclassify their relationships and families as inferior to those of opposite-sex couples. The Constitution simply does not allow for "laws of this sort." Romer v. Evans, 517 U.S. 620, 633 (1996).
"Broader issues have been urged for our consideration, but we adhere to the principle of deciding constitutional questions only in the context of the particular case before the Court." Sweatt v. Painter, 339 U.S. 629, 631 (1950). Whether under the Constitution same-sex couples may ever be denied the right to marry, a right that has long been enjoyed by opposite-sex couples, is an important and highly controversial question. It is currently a matter of great debate in our nation, and an issue over which people of good will may disagree, sometimes strongly. Of course, when questions of constitutional law are necessary to the resolution of a case, courts may not and should not abstain from deciding them simply because they are controversial. We need not and do not answer the broader question in this case, however, because California had already extended to committed same-sex couples both the incidents of marriage and the official designation of 'marriage,' and Proposition 8's only effect was to take away that important and legally significant designation, while leaving in place all of its incidents. This unique and strictly limited effect of Proposition 8 allows us to address the amendment's constitutionality on narrow grounds.
Thus, as a result of our "traditional reluctance to extend constitutional interpretations to situations or facts which are not before the Court, much of the excellent research and detailed argument presented in th[is] case[ ] is unnecessary to [its] disposition." Id. Were we unable, however, to resolve the matter on the basis we do, we would not hesitate to proceed to the broader question-the constitutionality of denying same-sex couples the right to marry.
Before considering the constitutional question of the validity of Proposition 8's elimination of the rights of same-sex couples to marry, we first decide that the official sponsors of Proposition 8 are entitled to appeal the decision below, which declared the measure unconstitutional and enjoined its enforcement. The California Constitution and Elections Code endow the official sponsors of an initiative measure with the authority to represent the State's interest in establishing the validity of a measure enacted by the voters, when the State's elected leaders refuse to do so. See Perry v. Brown, 134 Cal. Rptr. 3d 499 (2011). It is for the State of California to decide who may assert its interests in litigation, and we respect its decision by holding that Proposition 8's proponents have standing to bring this appeal on behalf of the State. We therefore conclude that, through the proponents of ballot measures, the People of California must be allowed to defend in federal courts, including on appeal, the validity of their use of the initiative power. Here, however, their defense fails on the merits. The People may not employ the initiative power to single out a disfavored group for unequal treatment and strip them, without a legitimate justification, of a right as important as the right to marry. Accordingly, we affirm the judgment of the district court.
We also affirm-for substantially the reasons set forth in the district court's opinion-the denial of the motion by the official sponsors of Proposition 8 to vacate the judgment entered by former Chief Judge Walker, on the basis of his purported interest in being allowed to marry his same-sex partner.
Upon its founding, the State of California recognized the legal institution of civil marriage for its residents. See, e.g., Cal. Const. of 1849, art. XI, §§ 12, 14 (discussing marriage contracts and marital property); Cal. Stats. 1850, ch. 140 ("An Act regulating Marriages"). Marriage in California was understood, at the time and well into the twentieth century, to be limited to relationships between a man and a woman. See In re Marriage Cases, 183 P.3d 384, 407-09 (Cal. 2008). In 1977, that much was made explicit by the California Legislature, which amended the marriage statute to read, "Marriage is a personal relation arising out of a civil contract between a man and a woman, to which the consent of the parties capable of making that contract is necessary." Cal. Stats. 1977, ch. 339, § 1. The 1977 provision remains codified in California statute. See Cal. Fam. Code § 300(a).
Following the enactment of the Defense of Marriage Act of 1996, Pub. L. 104-199, 110 Stat. 2419 (codified in relevant part at 1 U.S.C. § 7), which expressly limited the federal definition of marriage to relationships between one man and one woman, dozens of states enacted similar provisions into state law. See Andrew Koppelman, The Difference the Mini-DOMAs Make, 38 Loy. U. Chi. L.J. 265, 265-66 (2007). California did so in 2000 by adopting Proposition 22, an initiative statute, which provided, "Only marriage between a man and a woman is valid or recognized in California." Cal. Fam. Code § 308.5. The proposition ensured that same-sex marriages performed in any state that might permit such marriages in the future would not be recognized in California, and it guaranteed that any legislative repeal of the 1977 statute would not allow same-sex couples to marry within the State, because the Legislature may not amend or repeal an initiative statute enacted by the People. See Marriage Cases, 183 P.3d at 409-10.
Meanwhile, however, California had created the designation "domestic partnership" for "two adults who have chosen to share one another's lives in an intimate and committed relationship of mutual caring." Cal. Stats. 1999, ch. 588, § 2 (codified at Cal. Fam. Code § 297(a)). At first, California gave registered domestic partners only limited rights, such as hospital visitation privileges, id. § 4, and health benefits for the domestic partners of certain state employees, id. § 3. Over the next several years, however, the State substantially expanded the rights of domestic partners. By 2008, "California statutory provisions generally afford[ed] same-sex couples the opportunity to . . . obtain virtually all of the benefits and responsibilities afforded by California law to married opposite-sex couples." Marriage Cases, 183 P.3d at 417-18. The 2003 Domestic Partner Act provided broadly: "Registered domestic partners shall have the same rights, protections, and benefits, and shall be subject to the same responsibilities, obligations, and duties under law, whether they derive from statutes, administrative regulations, court rules, government policies, common law, or any other provisions or sources of law, as are granted to and imposed upon spouses." Cal. Stats. 2003, ch. 421, § 4 (codified at Cal. Fam. Code § 297.5(a)). It withheld only the official designation of marriage and thus the officially conferred and societally recognized status that accompanies that designation.
In 2004, same-sex couples and the City and County of San Francisco filed actions in California state courts alleging that the State's marriage statutes violated the California Constitution. Proposition 22 was among the statutes challenged, because as an initiative statutory enactment, it was equal in dignity to an enactment by the Legislature and thus subject to the restrictions of the state constitution.*fn1 The consolidated cases were eventually decided by the California Supreme Court, which held the statutes to be unconstitutional, for two independent reasons.
First, the court held that the fundamental right to marry provided by the California Constitution could not be denied to same-sex couples, who are guaranteed "the same substantive constitutional rights as opposite-sex couples to choose one's life partner and enter with that person into a committed, officially recognized, and protected family relationship that enjoys all of the constitutionally based incidents of marriage." Marriage Cases, 183 P.3d at 433-34. The court began by reaf-firming that "the right to marry is an integral component of an individual's interest in personal autonomy protected by the privacy provision of article I, section 1 [of the California Constitution], and of the liberty interest protected by the due process clause of article I, section 7." Id. at 426 (emphasis omitted). It then held "that an individual's homosexual orientation is not a constitutionally legitimate basis for withholding or restricting the individual's legal rights." Id. at 429. The court acknowledged that although such an inclusive understanding of the right to marry was one that had developed only "in recent decades," as the State extended greater recognition to same-sex couples and households, it was "apparent that history alone does not provide a justification for interpret-ing the constitutional right to marry as protecting only one's ability to enter into an officially recognized family relationship with a person of the opposite sex," because " '[f]undamental rights, once recognized, cannot be denied to particular groups on the ground that these groups have historically been denied those rights.' " Id. at 428-30 (quoting Hernandez v. Robles, 7 N.Y.3d 338, 381 (2006) (Kaye, C.J., dissenting)).
The court concluded its due process analysis by rejecting the argument that the availability of domestic partnerships satisfied "all of the personal and dignity interests that have traditionally informed the right to marry," because "[t]he current statutes-by drawing a distinction between the name assigned to the family relationship available to opposite-sex couples and the name assigned to the family relationship available to same-sex couples, and by reserving the historic and highly respected designation of 'marriage' exclusively to opposite-sex couples while offering same-sex couples only the new and unfamiliar designation of domestic partnership-pose a serious risk of denying the official family relationship of same-sex couples the equal dignity and respect that is a core element of the constitutional right to marry." Id. at 434-35.
Second, the court held that "[t]he current statutory assignment of different names for the official family relationships of opposite-sex couples on the one hand, and of same-sex couples on the other" violated the equal protection clause in article I, section 7 of the California Constitution. Id. at 435, 452-53. The court determined that the State had no interest in reserving the name 'marriage' for opposite-sex couples; "the historic and well-established nature of this limitation" could not itself justify the differential treatment, and the court found no reason that restricting the designation of 'marriage' to opposite-sex couples was necessary to preserve the benefits of marriage enjoyed by opposite-sex couples or their children. Id. at 450-52. The court noted specifically that "the distinction in nomenclature between marriage and domestic partnership cannot be defended on the basis of an asserted difference in the effect on children of being raised by an opposite-sex couple instead of by a same-sex couple," because "the governing California statutes permit same-sex couples to adopt and raise children and additionally draw no distinction between married couples and domestic partners with regard to the legal rights and responsibilities relating to children raised within each of these family relationships." Id. at 452 n.72. Restricting access to the designation of 'marriage' did, however, "work[ ] a real and appreciable harm upon same-sex couples and their children," because "providing only a novel, alternative institution for same-sex couples" constituted "an official statement that the family relationship of same-sex couples is not of comparable stature or equal dignity to the family relationship of opposite-sex couples." Id. at 452. Consequently, the court determined that withholding only the name 'marriage' from same-sex couples violated the California Constitution's guarantee of equal protection.
The court remedied these constitutional violations by striking the language from the marriage statutes "limiting the designation of marriage to a union 'between a man and a woman,' " invalidating Proposition 22, and ordering that the designation of 'marriage' be made available to both opposite-sex and same-sex couples. Id. at 453. Following the court's decision, California counties issued more than 18,000 marriage licenses to same-sex couples.
Five California residents-defendants-intervenors-appellants Dennis Hollingsworth, Gail J. Knight, Martin F. Gutierrez, Hak-Shing William Tam, and Mark A. Jansson (collectively, "Proponents")-collected voter signatures and filed petitions with the state government to place an initiative on the November 4, 2008, ballot. Unlike Proposition 22, this was an initiative constitutional amendment, which would be equal in effect to any other provision of the California Consti- tution, rather than subordinate to it. The Proponents' measure, designated Proposition 8, proposed to add a new provision to the California Constitution's Declaration of Rights, immediately following the Constitution's due process and equal protection clauses. The provision states, "Only marriage between a man and a woman is valid or recognized in California." According to the official voter information guide, Proposition 8 "[c]hanges the California Constitution to eliminate the right of same-sex couples to marry in California." Official Voter Information Guide, California General Election (Nov. 4, 2008), at 54. Following a contentious campaign, a slim majority of California voters (52.3 percent) approved Proposition 8. Pursuant to the state constitution, Proposition 8 took effect the next day, as article I, section 7.5 of the California Constitution.
Opponents of Proposition 8 then brought an original action for a writ of mandate in the California Supreme Court. They contended that Proposition 8 exceeded the scope of the People's initiative power because it revised, rather than amended, the California Constitution. The opponents did not raise any federal constitutional challenge to Proposition 8 in the state court. The state officials named as respondents refused to defend the measure's validity, but Proponents were permitted to intervene and do so. Following argument, the court upheld Proposition 8 as a valid initiative but construed the measure as not nullifying the 18,000-plus marriages of same-sex couples that had already been performed in the State. Strauss v. Horton, 207 P.3d 48, 98-110, 119-22 (Cal. 2009).
The court also explained Proposition 8's precise effect on California law: "[T]he measure carves out a narrow and limited exception to the[ ] state constitutional rights [articulated in the Marriage Cases], reserving the official designation of the term 'marriage' for the union of opposite-sex couples as a matter of state constitutional law, but leaving undisturbed all of the other extremely significant substantive aspects of a same-sex couple's state constitutional right to establish an officially recognized and protected family relationship and the guarantee of equal protection of the laws." Id. at 61; see also id. at 75. In other words, after Proposition 8, "[s]ame-sex couples retain all of the fundamental substantive components encompassed within the constitutional rights of privacy and due process, with the sole (albeit significant) exception of the right to equal access to the designation 'marriage.' " Id. at 116. Proposition 8 accomplished this result not by "declar[ing] the state of the law as it existed when the Marriage Cases decision was rendered, but instead [by] establish[ing] a new substantive state constitutional rule that became effective once Proposition 8 was approved by the voters." Id. at 115; see also id. at 63.
Two same-sex couples-plaintiffs Kristin Perry and Sandra Stier, and Paul Katami and Jeffrey Zarrillo-filed this action under 42 U.S.C. § 1983 in May 2009, after being denied marriage licenses by the County Clerks of Alameda County and Los Angeles County, respectively. Alleging that Proposition 8 violates the Fourteenth Amendment to the United States Constitution, they sought a declaration of its unconstitutionality and an injunction barring its enforcement. The City and County of San Francisco ("San Francisco") was later permitted to intervene as a plaintiff to present evidence of the amendment's effects on its governmental interests. The defendants-the two county clerks and four state officers, including the Governor and Attorney General-filed answers to the complaint but once again refused to argue in favor of Proposition 8's constitutionality. As a result, the district court granted Proponents' motion to intervene as of right under Federal Rule of Civil Procedure 24(a) to defend the validity of the proposition they had sponsored.*fn2
The district court held a twelve-day bench trial, during which it heard testimony from nineteen witnesses and, after giving the parties a full and fair opportunity to present evidence and argument, built an extensive evidentiary record.*fn3 In a thorough opinion in August 2010, the court made eighty findings of fact and adopted the relevant conclusions of law. Perry v. Schwarzenegger (Perry IV), 704 F. Supp. 2d 921 (N.D. Cal. 2010).*fn4 The court held Proposition 8 unconstitu-tional under the Due Process Clause because no compelling state interest justifies denying same-sex couples the fundamental right to marry. Id. at 991-95. The court also determined that Proposition 8 violated the Equal Protection Clause, because there is no rational basis for limiting the designation of 'marriage' to opposite-sex couples. Id. at 997-1003. The court therefore entered the following injunction: "Defendants in their official capacities, and all persons under the control or supervision of defendants, are permanently enjoined from applying or enforcing Article I, § 7.5 of the California Constitution."*fn5 Doc. 728 (Permanent Injunction), Perry v. Schwarzenegger, No. 09-cv-02292 (N.D. Cal. Aug. 12, 2010).*fn6
Proponents appealed immediately, and a motions panel of this court stayed the district court's injunction pending appeal.
The motions panel asked the parties to discuss in their briefs, as a preliminary matter, whether the Proponents had standing to seek review of the district court order. After considering the parties' arguments, we concluded that Proponents' standing to appeal depended on the precise rights and interests given to official sponsors of an initiative under California law, which had never been clearly defined by the State's highest court. We therefore certified the following question to the California Supreme Court:
Whether under Article II, Section 8 of the California Constitution, or otherwise under California law, the official proponents of an initiative measure possess either a particularized interest in the initiative's validity or the authority to assert the State's interest in the initiative's validity, which would enable them to defend the constitutionality of the initiative upon its adoption or appeal a judgment invalidating the initiative, when the public officials charged with that duty refuse to do so.
Perry v. Schwarzenegger (Perry V), 628 F.3d 1191, 1193 (9th Cir. 2011). The state court granted our request for certification in February 2011, and in November 2011 rendered its decision. See Perry v. Brown (Perry VII), 134 Cal. Rptr. 3d 499( 2011). We now resume consideration of this appeal.*fn7
We begin, as we must, with the issue that has prolonged our consideration of this case: whether we have jurisdiction over an appeal brought by the defendant-intervenor Proponents, rather than the defendant state and local officers who were directly enjoined by the district court order.*fn8 In view of Proponents' authority under California law, we conclude that they do have standing to appeal.
For purposes of Article III standing, we start with the premise that "a State has standing to defend the constitutionality of its [laws]." Diamond v. Charles, 476 U.S. 54, 62 (1986). When a state law is ruled unconstitutional, either the state or a state officer charged with the law's enforcement may appeal that determination. Typically, the named defendant in an action challenging the constitutionality of a state law is a state officer, because sovereign immunity protects the state from being sued directly. See Ex parte Young, 209 U.S. 123, 157-58 (1908); L.A. County Bar Ass'n v. Eu, 979 F.2d 697, 704 (9th Cir. 1992). In such cases, if a court invalidates the state law and enjoins its enforcement, there is no question that the state officer is entitled to appeal that determination. See, e.g., Ysursa v. Pocatello Educ. Ass'n, 555 U.S. 353 (2009) (Idaho Secretary of State and Attorney General appealed decision striking down an Idaho law on First Amendment grounds); Stenberg v. Carhart, 530 U.S. 914 (2000) (Nebraska Attorney General appealed decision holding unconstitutional a Nebraska abortion law). Moreover, there is no reason that a state itself may not also choose to intervene as a defendant, and indeed a state must be permitted to intervene if a state officer is not already party to an action in which the constitutionality of a state law is challenged. See 28 U.S.C. § 2403(b); Fed. R. Civ. P. 5.1; cf. Fed. R. App. P. 44(b). When a state does elect to become a defendant itself, the state may appeal an adverse decision about the constitutionality of one of its laws, just as a state officer may. See, e.g., Caruso v. Yamhill County ex rel. County Comm'r, 422 F.3d 848, 852-53 & n.2 (9th Cir. 2005) (sole appellant was the State of Oregon, which had intervened as a defendant in the district court). In other words, in a suit for an injunction against enforcement of an allegedly unconstitutional state law, it makes no practical difference whether the formal party before the court is the state itself or a state officer in his official capacity. Cf. Pennhurst State Sch. & Hosp. v. Halderman, 465 U.S. 89, 114 n.25 (1984) (discussing the "fiction" of Ex parte Young); see also Idaho v. Coeur d'Alene Tribe of Idaho, 521 U.S. 261, 269-70 (1997) (same).
Whether the defendant is the state or a state officer, the decision to assert the state's own interest in the constitutionality of its laws is most commonly made by the state's executive branch-the part of state government that is usually charged with enforcing and defending state law. See, e.g., Ysursa, 555 U.S. at 354 (Idaho state officers represented by state Attorney General); Caruso, 422 F.3d at 851 (State of Oregon represented by Oregon Department of Justice). Some sovereigns vest the authority to assert their interest in litigation exclusively in certain executive officers. See, e.g., 28 U.S.C. §§ 516-19; 28 C.F.R. § 0.20.
The states need not follow that approach, however. It is their prerogative, as independent sovereigns, to decide for themselves who may assert their interests and under what circumstances, and to bestow that authority accordingly. In Karcher v. May, 484 U.S. 72 (1987), for example, the Supreme Court held that the State of New Jersey was properly represented in litigation by the Speaker of the General Assembly and the President of the Senate, appearing on behalf of the Legislature, because "the New Jersey Legislature had authority under state law to represent the State's interests." Id. at 82 (citing In re Forsythe, 450 A.2d 499, 500 (N.J. 1982)).*fn9 Prin- ciples of federalism require that federal courts respect such decisions by the states as to who may speak for them: "there are limits on the Federal Government's power to affect the internal operations of a State." Va. Office for Protection & Advocacy v. Stewart, 131 S. Ct. 1632, 1641 (2011). It is not for a federal court to tell a state who may appear on its behalf any more than it is for Congress to direct state law-enforcement officers to administer a federal regulatory scheme, see Printz v. United States, 521 U.S. 898 (1997), to command a state to take ownership of waste generated within its borders, see New York v. United States, 505 U.S. 144 (1992), or to dictate where a state shall locate its capital, see Coyle v. Smith, 221 U.S. 559 (1911). Who may speak for the state is, necessarily, a question of state law. All a federal court need determine is that the state has suffered a harm sufficient to confer standing and that the party seeking to invoke the jurisdiction of the court is authorized by the state to represent its interest in remedying that harm.
Proponents claim to assert the interest of the People of California in the constitutionality of Proposition 8, which the People themselves enacted. When faced with a case arising in a similar posture, in which an Arizona initiative constitutional amendment was defended only by its sponsors, the Supreme Court expressed "grave doubts" about the sponsors' standing given that the Court was "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." Arizonans for Official English v. Arizona (Arizonans), 520 U.S. 43, 65-66 (1997). Absent some conferral of authority by state law, akin to the authority that the New Jersey legislators in Karcher had as "elected representatives," the Court suggested that proponents of a ballot measure would not be able to appeal a decision striking down the initiative they sponsored. Id. at 65.
 Here, unlike in Arizonans, we do know that California law confers on "initiative sponsors" the authority "to defend, in lieu of public officials, the constitutionality of initiatives made law of the State." The California Supreme Court has told us, in a published opinion containing an exhaustive review of the California Constitution and statutes, that it does. In answering our certified question, the court held that when the public officials who ordinarily defend a challenged state law or appeal a judgment invalidating the law decline to do so, under article II, section 8 of the California Constitution and the relevant provisions of the Elections Code, the official proponents of a voter-approved initiative measure are authorized to assert the state's interest in the initiative's validity, enabling the proponents to defend the constitutionality of the initiative and to appeal a judgment invalidating the initiative.
Perry VII, 134 Cal. Rptr. 3d at 536-37. "[T]he role played by the proponents in such litigation," the court explained, "is comparable to the role ordinarily played by the Attorney General or other public officials in vigorously defending a duly enacted state law and raising all arguable legal theories upon which a challenged provision may be sustained." Id. at 525. The State's highest court thus held that California law provides precisely what the Arizonans Court found lacking in Arizona law: it confers on the official proponents of an initiative the authority to assert the State's interests in defending the constitutionality of that initiative, where the state officials who would ordinarily assume that responsibility choose not to do so.
We are bound to accept the California court's determination. Although other states may act differently, California's conferral upon proponents of the authority to represent the People's interest in the initiative measure they sponsored is consistent with that state's unparalleled commitment to the authority of the electorate: "No other state in the nation carries the concept of initiatives as 'written in stone' to such lengths as" does California. People v. Kelly, 222 P.3d 186, 200 (Cal. 2010) (internal quotation marks omitted). Indeed, California defines the initiative power as "one of the most precious rights of our democratic process," and considers "the sovereign people's initiative power" to be a "fundamental right" under the state constitution. Assoc. Home Builders v. City of Livermore, 557 P.2d 473, 477 (Cal. 1976); Brosnahan v. Brown, 651 P.2d 274, 277 (Cal. 1982); Costa v. Super. Ct., 128 P.3d 675, 686 (Cal. 2006). As the California Supreme Court explained in answering our certified question, "[t]he initiative power would be significantly impaired if there were no one to assert the state's interest in the validity of the measure when elected officials decline to defend it in court or to appeal a judgment invalidating the measure." Perry VII, 134 Cal. Rptr. 3d at 523. The authority of official proponents to "assert[ ] the state's interest in the validity of an initiative measure" thus "serves to safeguard the unique elements and integrity of the initiative process." Id. at 533.
It matters not whether federal courts think it wise or desirable for California to afford proponents this authority to speak for the State, just as it makes no difference whether federal courts think it a good idea that California allows its constitution to be amended by a majority vote through a ballot measure in the first place. Cf. Pac. States Tel. & Tel. Co. v. Oregon, 223 U.S. 118 (1912) (holding non-justiciable a Guaranty Clause challenge to Oregon's initiative system). The People of California are largely free to structure their system of governance as they choose, and we respect their choice. All that matters, for federal standing purposes, is that the People have an interest in the validity of Proposition 8 and that, under California law, Proponents are authorized to represent the People's interest. That is the case here.
In their supplemental brief on the issue of standing, Plaintiffs argue for the first time that Proponents must satisfy the requirements of third-party standing in order to assert the interests of the State of California in this litigation. Litigants who wish "to bring actions on behalf of third parties" must satisfy three requirements. Powers v. Ohio, 499 U.S. 400, 410-11 (1991). First, they "must have suffered an 'injury in fact,' thus giving [them] a 'sufficiently concrete interest' in the outcome of the issue in dispute." Id. at 411. Second, they "must have a close relation to the third party." Id. Third, "there must exist some hindrance to the third party's ability to protect his or her own interests." Id. Plaintiffs contend that Proponents cannot satisfy these requirements with respect to the State of California as a third party.
The requirements of third-party standing, however, are beside the point: the State of California is no more a "third party" relative to Proponents than it is to the executive officers of the State who ordinarily assert the State's interest in litigation. As the California Supreme Court has explained, "the role played by the proponents" in litigation "regarding the validity or proper interpretation of a voter-approved initiative measure . . . is comparable to the role ordinarily played by the Attorney General or other public officials in vigorously defending a duly enacted state law." Perry VII, 134 Cal. Rptr. 3d at 525. When the Attorney General of California appears in federal court to defend the validity of a state statute, she obviously need not satisfy the requirements of third-party standing; she stands in the shoes of the State to assert its interests in litigation. For the purposes of the litigation, she speaks to the court as the State, not as a third party. The same is true of Proponents here, just as it was true of the presiding legislative officers in Karcher, 484 U.S. at 82. The requirements of third-party standing are therefore not relevant.
 Nor is it relevant whether Proponents have suffered a personal injury, in their capacities as private individuals. Although we asked the California Supreme Court whether "the official proponents of an initiative measure possess either a particularized interest in the initiative's validity or the authority to assert the State's interest in the initiative's validity," Perry V, 628 F.3d at 1193 (emphasis added), the Court chose to address only the latter type of interest. Perry VII, 134 Cal. Rptr. 3d at 515 ("Because [our] conclusion [that proponents are authorized to assert the State's interest] is sufficient to support an affirmative response to the question posed by the Ninth Circuit, we need not decide whether, under California law, the official proponents also possess a particularized interest in a voter-approved initiative's validity."). The exclusive basis of our holding that Proponents possess Article III standing is their authority to assert the interests of the State of California, rather than any authority that they might have to assert particularized interests of their own. Just as the Attorney General of California need not satisfy the requirements of third-party standing when she appears in federal court to defend the validity of a state statute, she obviously need not show that she would suffer any personal injury as a result of the statute's invalidity. The injury of which she complains is the State's, not her own. The same is true here. Because "a State has standing to defend the constitutionality of its [laws]," Diamond, 476 U.S. at 62, Proponents need not show that they would suffer any personal injury from the invalidation of Proposition 8. That the State would suffer an injury, id., is enough for Proponents to have Article III standing when state law authorizes them to assert the State's interests.
To be clear, we do not suggest that state law has any "power directly to enlarge or contract federal jurisdiction." Duchek v. Jacobi, 646 F.2d 415, 419 (9th Cir. 1981). "Standing to sue in any Article III court is, of course, a federal question which does not depend on the party's . . . standing in state court." Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 804 (1985). State courts may afford litigants standing to appear where federal courts would not,*fn10 but whether they do so has no bearing on the parties' Article III standing in federal court.
 State law does have the power, however, to answer questions antecedent to determining federal standing, such as the one here: who is authorized to assert the People's interest in the constitutionality of an initiative measure? Because the State of California has Article III standing to defend the constitutionality of Proposition 8, and because both the California Constitution and California law authorize "the official proponents of [an] initiative . . . to appear and assert the state's interest in the initiative's validity and to appeal a judgment invalidating the measure when the public officials who ordinarily defend the measure or appeal such a judgment decline to do so," Perry VII, 134 Cal. Rptr. 3d at 505, we conclude that Proponents are proper appellants here. They possess Article III standing to prosecute this appeal from the district court's judgment invalidating Proposition 8.
We review the district court's decision to grant a permanent injunction for abuse of discretion, but we review the determinations underlying that decision by the standard that applies to each determination. Accordingly, we review the court's conclusions of law de novo and its findings of fact for clear error. See Ting v. AT&T, 319 F.3d 1126, 1134-35 (9th Cir. 2003); Fed. R. Civ. P. 52(a).
Plaintiffs and Proponents dispute whether the district court's findings of fact concern the types of "facts"-so-called "adjudicative facts"-that are capable of being "found" by a court through the clash of proofs presented in adjudication, as opposed to "legislative facts," which are generally not capable of being found in that fashion. "Adjudicative facts are facts about the parties and their activities . . . , usually answering the questions of who did what, where, when, how, why, with what motive or intent"-the types of "facts that go to a jury in a jury case," or to the factfinder in a bench trial. Marshall v. Sawyer, 365 F.2d 105, 111 (9th Cir. 1966) (quoting Kenneth Culp Davis, The Requirement of a Trial-Type Hear- ing, 70 Harv. L. Rev. 193, 199 (1956)) (internal quotation marks omitted). "Legislative facts," by contrast, "do not usually concern [only] the immediate parties but are general facts which help the tribunal decide questions of law, policy, and discretion." Id.
It is debatable whether some of the district court's findings of fact concerning matters of history or social science are more appropriately characterized as "legislative facts" or as "adjudicative facts." We need not resolve what standard of review should apply to any such findings, however, because the only findings to which we give any deferential weight- those concerning the messages in support of Proposition 8 that Proponents communicated to the voters to encourage their approval of the measure, Perry IV, 704 F. Supp. 2d at 990-91-are clearly "adjudicative facts" concerning the parties and " 'who did what, where, when, how, why, with what motive or intent.' " Marshall, 365 F.2d at 111. Aside from these findings, the only fact found by the district court that matters to our analysis is that "[d]omestic partnerships lack the social meaning associated with marriage"-that the difference between the designation of 'marriage' and the designation of 'domestic partnership' is meaningful. Perry IV, 704 F. Supp. 2d at 970. This fact was conceded by Proponents during discovery. Defendant-Intervenors' Response to Plaintiffs' First Set of Requests for Admission, Exhibit No. PX 0707, at 2 ("Proponents admit that the word 'marriage' has a unique meaning."); id. at 11 (Proponents "[a]dmit that there is a significant symbolic disparity between domestic partnership and marriage"). Our analysis therefore does not hinge on what standard we use to review the district court's findings of fact. Cf. Lockhart v. McCree, 476 U.S. 162, 168 n.3 (1986) ("Because we do not ultimately base our decision today on the [validity or] invalidity of the lower courts' 'factual' findings, we need not decide the 'standard of review' issue"-whether "the 'clearly erroneous' standard of Rule 52(a) applies to the kind of 'legislative' facts at issue here.").
We now turn to the merits of Proposition 8's constitutionality.
The district court held Proposition 8 unconstitutional for two reasons: first, it deprives same-sex couples of the fundamental right to marry, which is guaranteed by the Due Process Clause, see Perry IV, 704 F. Supp. 2d at 991-95; and second, it excludes same-sex couples from state-sponsored marriage while allowing opposite-sex couples access to that honored status, in violation of the Equal Protection Clause, see id. at 997-1003. Plaintiffs elaborate upon those arguments on appeal.
Plaintiffs and Plaintiff-Intervenor San Francisco also offer a third argument: Proposition 8 singles out same-sex couples for unequal treatment by taking away from them alone the right to marry, and this action amounts to a distinct constitutional violation because the Equal Protection Clause protects minority groups from being targeted for the deprivation of an existing right without a legitimate reason. Romer, 517 U.S. at 634-35. Because this third argument applies to the specific history of same-sex marriage in California, it is the narrowest ground for adjudicating the constitutional questions before us, while the first two theories, if correct, would apply on a broader basis. Because courts generally decide constitutional questions on the narrowest ground available, we consider the third argument first. See Plaut v. Spendthrift Farm, Inc., 514 U.S. 211, 217 (1995) (citing Ashwander v. Tenn. Valley Auth., 297 U.S. 288, 347 (1936) (Brandeis, J., concurring)).
 Proposition 8 worked a singular and limited change to the California Constitution: it stripped same-sex couples of the right to have their committed relationships recognized by the State with the designation of 'marriage,' which the state constitution had previously guaranteed them, while leaving in place all of their other rights and responsibilities as partners- rights and responsibilities that are identical to those of married spouses and form an integral part of the marriage relationship. In determining that the law had this effect, "[w]e rely not upon our own interpretation of the amendment but upon the authoritative construction of [California's] Supreme Court." Romer, 517 U.S. at 626. The state high court held in Strauss that "Proposition 8 reasonably must be interpreted in a limited fashion as eliminating only the right of same-sex couples to equal access to the designation of marriage, and as not otherwise affecting the constitutional right of those couples to establish an officially recognized family relationship," which California calls a 'domestic partnership.' 207 P.3d at 76. Proposition 8 "leaves intact all of the other very significant constitutional protections afforded same-sex couples," including "the constitutional right to enter into an officially recognized and protected family relationship with the person of one's choice and to raise children in that family if the couple so chooses." Id. at 102. Thus, the extent of the amend-ment's effect was to "establish[ ] a new substantive state constitutional rule," id. at 63, which "carves out a narrow and limited exception to these state constitutional rights," by "re-serving the official designation of the term 'marriage' for the union of opposite-sex couples as a matter of state constitutional law," id. at 61.*fn11
Both before and after Proposition 8, same-sex partners could enter into an official, state-recognized relationship that affords them "the same rights, protections, and benefits" as an opposite-sex union and subjects them "to the same responsibilities, obligations, and duties under law, whether they derive from statutes, administrative regulations, court rules, government policies, common law, or any other provisions or sources of law, as are granted to and imposed upon spouses." Cal. Fam. Code § 297.5(a). Now as before, same-sex partners may:
* Raise children together, and have the same rights and obligations as to their children as spouses have, see Cal. Fam. Code § 297.5(d);
* Enjoy the presumption of parentage as to a child born to either partner, see Elisa B. v. Super. Ct., 117 P.3d 660, 670 (Cal. 2005); Kristine M. v. David P., 135 Cal. App. 4th 783 (2006); or adopted by one partner and raised jointly by both, S.Y. v. S.B., 201 Cal. App. 4th 1023 (2011);
* Adopt each other's children, see Cal. Fam. Code § 9000(g);
* Become foster parents, see Cal. Welf. & Inst. Code § 16013(a);
* Share community property, see Cal. Fam. Code § 297.5(k);
* File state taxes jointly, see Cal. Rev. & Tax. Code § 18521(d);
* Participate in a partner's group health insurance policy on the same terms as a spouse, see Cal. Ins. Code § 10121.7;
* Enjoy hospital visitation privileges, see Cal. Health & Safety Code § 1261;
* Make medical decisions on behalf of an incapacitated partner, see Cal. Prob. Code § 4716;
* Be treated in a manner equal to that of a widow or widower with respect to a deceased partner, see Cal. Fam. Code § 297.5(c);
* Serve as the conservator of a partner's estate, see Cal. Prob. Code §§ 1811-1813.1; and
* Sue for the wrongful death of a partner, see Cal. Civ. Proc. Code § 377.60-among many other things.
Proposition 8 did not affect these rights or any of the other " 'constitutionally based incidents of marriage' " guaranteed to same-sex couples and their families. Strauss, 207 P.3d at 61 (quoting Marriage Cases, 183 P.3d at 434). In adopting the amendment, the People simply took the designation of 'marriage' away from lifelong same-sex partnerships, and with it the State's authorization of that official status and the societal approval that comes with it.
By emphasizing Proposition 8's limited effect, we do not mean to minimize the harm that this change in the law caused to same-sex couples and their families. To the contrary, we emphasize the extraordinary significance of the official designation of 'marriage.' That designation is important because 'marriage' is the name that society gives to the relationship that matters most between two adults. A rose by any other name may smell as sweet, but to the couple desiring to enter into a committed lifelong relationship, a marriage by the name of 'registered domestic partnership' does not. The word 'marriage' is singular in connoting "a harmony in living," "a bilateral loyalty," and "a coming together for better or for worse, hopefully enduring, and intimate to the degree of being sacred." Griswold v. Connecticut, 381 U.S. 479, 486 (1965). As Proponents have admitted, "the word 'marriage' has a unique meaning," and "there is a significant symbolic disparity between domestic partnership and marriage." It is the designation of 'marriage' itself that expresses validation, by the state and the community, and that serves as a symbol, like a wedding ceremony or a wedding ring, of something profoundly important. See id. at 971.
We need consider only the many ways in which we encounter the word 'marriage' in our daily lives and understand it, consciously or not, to convey a sense of significance. We are regularly given forms to complete that ask us whether we are "single" or "married." Newspapers run announcements of births, deaths, and marriages. We are excited to see someone ask, "Will you marry me?", whether on bended knee in a restaurant or in text splashed across a stadium Jumbotron. Certainly it would not have the same effect to see "Will you enter into a registered domestic partnership with me?". Groucho Marx's one-liner, "Marriage is a wonderful institution . . . but who wants to live in an institution?" would lack its punch if the word 'marriage' were replaced with the alternative phrase. So too with Shakespeare's "A young man married is a man that's marr'd," Lincoln's "Marriage is neither heaven nor hell, it is simply purgatory," and Sinatra's "A man doesn't know what happiness is until he's married. By then it's too late." We see tropes like "marrying for love" versus "marrying for money" played out again and again in our films and literature because of the recognized importance and permanence of the marriage relationship. Had Marilyn Monroe's film been called How to Register a Domestic Partnership with a Millionaire, it would not have conveyed the same meaning as did her famous movie, even though the underlying drama for same-sex couples is no different. The name 'marriage' signifies the unique recognition that society gives to harmonious, loyal, enduring, and intimate relationships. See Knight v. Super. Ct., 128 Cal. App. 4th 14, 31 (2005) ("[M]arriage is considered a more substantial relationship and is accorded a greater stature than a domestic partnership."); cf. Griswold, 381 U.S. at 486.
The official, cherished status of 'marriage' is distinct from the incidents of marriage, such as those listed in the California Family Code. The incidents are both elements of the institution and manifestations of the recognition that the State affords to those who are in stable and committed lifelong relationships. We allow spouses but not siblings or roommates to file taxes jointly, for example, because we acknowledge the financial interdependence of those who have entered into an "enduring" relationship. The incidents of marriage, standing alone, do not, however, convey the same governmental and societal recognition as does the designation of 'marriage' itself. We do not celebrate when two people merge their bank accounts; we celebrate when a couple marries. The designation of 'marriage' is the status that we recognize. It is the principal manner in which the State attaches respect and dignity to the highest form of a committed relationship and to the individuals who have entered into it.*fn12
 We set this forth because we must evaluate Proposition 8's constitutionality in light of its actual and specific effects on committed same-sex couples desiring to enter into an officially recognized lifelong relationship. Before Proposition 8, California guaranteed gays and lesbians both the incidents and the status and dignity of marriage. Proposition 8 left the incidents but took away the status and the dignity. It did so by superseding the Marriage Cases and thus endorsing the "official statement that the family relationship of same-sex couples is not of comparable stature or equal dignity to the family relationship of opposite-sex couples." Marriage Cases, 183 P.3d at 452. The question we therefore consider is this: did the People of California have legitimate reasons for enacting a constitutional amendment that serves only to take away from same-sex couples the right to have their lifelong relationships dignified by the official status of 'marriage,' and to compel the State and its officials and all others authorized to perform marriage ceremonies to substitute the label of 'domestic partnership' for their relationships?
Proponents resist this framing of the question. They deem it irrelevant to our inquiry that the California Constitution, as interpreted by the Marriage Cases, had previously guaranteed same-sex couples the right to use the designation of 'marriage,' because In re Marriage Cases was a "short-lived decision," and same-sex couples were allowed to marry only during a "143-day hiatus" between the effective date of the Marriage Cases decision and the enactment of Proposition 8. Proponents' Reply Br. 75, 79-80. According to Proponents, a decision to "restore" the "traditional definition of marriage" is indistinguishable from a decision to "adhere" to that definition in the first place. Id. at 79-80. We are bound, however, by the California Supreme Court's authoritative interpretation of Proposition 8's effect on California law, see Romer, 517 U.S. at 626: Proposition 8 "eliminat[ed] . . . the right of same-sex couples to equal access to the designation of marriage" by "carv[ing] out a narrow and limited exception to these state constitutional rights" that had previously guaranteed the designation of 'marriage' to all couples, opposite-sex and same-sex alike. Strauss, 207 P.3d at 61, 76.
Even were we not bound by the state court's explanation, we would be obligated to consider Proposition 8 in light of its actual effect, which was, as the voters were told, to "eliminate the right of same-sex couples to marry in California." Voter Information Guide at 54. The context matters. Withdrawing from a disfavored group the right to obtain a designation with significant societal consequences is different from declining to extend that designation in the first place, regardless of whether the right was withdrawn after a week, a year, or a decade. The action of changing something suggests a more deliberate purpose than does the inaction of leaving it as it is. As the California Supreme Court held, "Proposition 8 [did] not 'readjudicate' the issue that was litigated and resolved in the Marriage Cases." Strauss, 207 P.3d at 63. Rather than "declar[ing] the state of the law as it existed under the California Constitution at the time of the Marriage Cases," Proposition 8 "establishe[d] a new substantive state constitutional rule that took effect upon" its adoption by the electorate. Id. (emphasis added). Whether or not it is a historical accident, as Proponents argue, that Proposition 8 postdated the Marriage Cases rather than predating and thus preempting that decision, the relative timing of the two events is a fact, and we must decide this case on its facts.
This is not the first time the voters of a state have enacted an initiative constitutional amendment that reduces the rights of gays and lesbians under state law. In 1992, Colorado adopted Amendment 2 to its state constitution, which prohibited the state and its political subdivisions from providing any protection against discrimination on the basis of sexual orientation. See Colo. Const. art. II, § 30b. Amendment 2 was proposed in response to a number of local ordinances that had banned sexual-orientation discrimination in such areas as housing, employment, education, public accommodations, and health and welfare services. The effect of Amendment 2 was "to repeal" those local laws and "to prohibit any governmental entity from adopting similar, or more protective statutes, regulations, ordinances, or policies in the future." Evans v. Romer, 854 P.2d 1270, 1284-85 (Colo. 1993). The law thus "withdr[ew] from homosexuals, but no others, specific legal protection . . . , and it forb[ade] reinstatement of these laws and policies." Romer, 517 U.S. at 627.
 The Supreme Court held that Amendment 2 violated the Equal Protection Clause because "[i]t is not within our constitutional tradition to enact laws of this sort"-laws that "singl[e] out a certain class of citizens for disfavored legal status," which "raise the inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected." Id. at 633-34. The Court considered possible justifications for Amendment 2 that might have overcome the "inference" of animus, but it found them all lacking. It therefore concluded that the law "classifie[d] homosexuals not to further a proper legislative end but to make them unequal to everyone else." Id. at 635.*fn13
 Proposition 8 is remarkably similar to Amendment 2. Like Amendment 2, Proposition 8 "single[s] out a certain class of citizens for disfavored legal status . . . ." Id. at 633. Like Amendment 2, Proposition 8 has the "peculiar property," id. at 632, of "withdraw[ing] from homosexuals, but no others," an existing legal right-here, access to the official designation of 'marriage'-that had been broadly available, notwithstanding the fact that the Constitution did not compel the state to confer it in the first place. Id. at 627. Like Amendment 2, Proposition 8 denies "equal protection of the laws in the most literal sense," id. at 633, because it "carves out" an "exception" to California's equal protection clause, by removing equal access to marriage, which gays and lesbians had previously enjoyed, from the scope of that constitutional guarantee. Strauss, 207 P.3d at 61. Like Amendment 2, Proposition 8 "by state decree . . . put[s] [homosexuals] in a solitary class with respect to" an important aspect of human relations, and accordingly "imposes a special disability upon [homosexuals] alone." Romer, 517 U.S. at 627, 631. And like Amendment 2, Proposition 8 constitutionalizes that disability, meaning that gays and lesbians may overcome it "only by enlisting the citizenry of [the state] to amend the State Constitution" for a second time. Id. at 631. As we explain below, Romer compels that we affirm the judgment of the district court.
To be sure, there are some differences between Amendment 2 and Proposition 8. Amendment 2 "impos[ed] a broad and undifferentiated disability on a single named group" by "identif[ying] persons by a single trait and then den[ying] them protection across the board." Romer, 517 U.S. at 632-33. Proposition 8, by contrast, excises with surgical precision one specific right: the right to use the designation of 'marriage' to describe a couple's officially recognized relationship. Proponents argue that Proposition 8 thus merely "restor[es] the traditional definition of marriage while otherwise leaving undisturbed the manifold rights and protections California law provides gays and lesbians," making it unlike Amendment 2, which eliminated various substantive rights. Proponents' Reply Br. 77.
These differences, however, do not render Romer less applicable. It is no doubt true that the "special disability" that Proposition 8 "imposes upon" gays and lesbians has a less sweeping effect on their public and private transactions than did Amendment 2. Nevertheless, Proposition 8 works a meaningful harm to gays and lesbians, by denying to their committed lifelong relationships the societal status conveyed by the designation of 'marriage,' and this harm must be justified by some legitimate state interest. Romer, 517 U.S. at 631. Proposition 8 is no less problematic than Amendment 2 merely because its effect is narrower; to the contrary, the surgical precision with which it excises a right belonging to gay and lesbian couples makes it even more suspect. A law that has no practical effect except to strip one group of the right to use a state-authorized and socially meaningful designation is all the more "unprecedented" and "unusual" than a law that imposes broader changes, and raises an even stronger "inference that the disadvantage imposed is born of animosity toward the class of persons affected," id. at 633-34. In short, Romer governs our analysis notwithstanding the differences between Amendment 2 and Proposition 8.
 There is one further important similarity between this case and Romer. Neither case requires that the voters have stripped the state's gay and lesbian citizens of any federal constitutional right. In Romer, Amendment 2 deprived gays and lesbians of statutory protections against discrimination; here, Proposition 8 deprived same-sex partners of the right to use the designation of 'marriage.' There is no necessity in either case that the privilege, benefit, or protection at issue be a constitutional right. We therefore need not and do not consider whether same-sex couples have a fundamental right to marry, or whether states that fail to afford the right to marry to gays and lesbians must do so. Further, we express no view on those questions.*fn14
Ordinarily, "if a law neither burdens a fundamental right nor targets a suspect class, we will uphold the legislative classification so long as it bears a rational relation to some legitimate end." Romer, 517 U.S. at 631. Such was the case in Romer, and it is the case here as well. The end must be one that is legitimate for the government to pursue, not just one that would be legitimate for a private actor. See id. at 632, 635. The question here, then, is whether California had any more legitimate justification for withdrawing from gays and lesbians its constitutional protection with respect to the official designation of 'marriage' than Colorado did for withdrawing from that group all protection against discrimination generally.
Proposition 8, like Amendment 2, enacts a " '[d]iscrimination[ ] of an unusual character,' " which requires " 'careful consideration to determine whether [it] [is] obnoxious to the' " Constitution. Id. at 633 (quoting Louisville Gas & Elec. Co. v. Coleman, 277 U.S. 32, 37-38 (1928)). As in Romer, therefore, we must consider whether any legiti-mate state interest constitutes a rational basis for Proposition 8; otherwise, we must infer that it was enacted with only the constitutionally illegitimate basis of "animus toward the class it affects." Romer, 517 U.S. at 632.
Before doing so, we briefly consider one other objection that Proponents raise to this analysis: the argument that because the Constitution "is not simply a one-way ratchet that forever binds a State to laws and policies that go beyond what the Fourteenth Amendment would otherwise require," the State of California-" 'having gone beyond the requirements of the Federal Constitution' " in extending the right to marry to same-sex couples-" 'was free to return . . . to the standard prevailing generally throughout the United States.' " Proponents' Reply Br. 76 (quoting Crawford v. Bd. of Educ., 458 U.S. 527, 542 (1982)). Proponents appear to suggest that unless the Fourteenth Amendment actually requires that the designation of 'marriage' be given to same-sex couples in the first place, there can be no constitutional infirmity in taking the designation away from that group of citizens, whatever the People's reason for doing so.
 Romer forecloses this argument. The rights that were repealed by Amendment 2 included protections against discrimination on the basis of sexual orientation in the private sphere. Those protections, like any protections against private discrimination, were not compelled by the Fourteenth Amendment.*fn15 Rather, "[s]tates ha[d] chosen to counter discrimination by enacting detailed statutory schemes" prohibiting discrimination in employment and public accommodations, among other contexts, and certain Colorado jurisdictions had chosen to extend those protections to gays and lesbians. Romer, 517 U.S. at 628 (emphasis added). It was these elective protections that Amendment 2 withdrew and forbade.*fn16
The relevant inquiry in Romer was not whether the state of the law after Amendment 2 was constitutional; there was no doubt that the Fourteenth Amendment did not require antidiscrimination protections to be afforded to gays and lesbians. The question, instead, was whether the change in the law that Amendment 2 effected could be justified by some legitimate purpose.
The Supreme Court's answer was "no"-there was no legitimate reason to take away broad legal protections from gays and lesbians alone, and to inscribe that deprivation of equality into the state constitution, once those protections had already been provided. We therefore need not decide whether a state may decline to provide the right to marry to same-sex couples. To determine the validity of Proposition 8, we must consider only whether the change in the law that it effected- eliminating by constitutional amendment the right of same-sex couples to have the official designation and status of 'marriage' bestowed upon their relationships, while maintaining that right for opposite-sex couples-was justified by a legitimate reason.
 This does not mean that the Constitution is a "oneway ratchet," as Proponents suggest. It means only that the Equal Protection Clause requires the state to have a legitimate reason for withdrawing a right or benefit from one group but not others, whether or not it was required to confer that right or benefit in the first place. Thus, when Congress, having chosen to provide food stamps to the poor in the Food Stamp Act of 1964, amended the Act to exclude households of unrelated individuals, such as "hippies" living in "hippie communes," the Supreme Court held the amendment unconstitutional because "a bare congressional desire to harm a politically unpopular group cannot constitute a legitimate governmental interest." U.S. Dep't of Agric. v. Moreno, 413 U.S. 528, 534 (1973). In both Romer and Moreno, the constitutional violation that the Supreme Court identified was not the failure to confer a right or benefit in the first place; Congress was no more obligated to provide food stamps than Colorado was to enact antidiscrimination laws. Rather, what the Supreme Court forbade in each case was the targeted exclusion of a group of citizens from a right or benefit that they had enjoyed on equal terms with all other citizens. The constitutional injury that Romer and Moreno identified-and that serves as a basis of our decision to strike down Proposition 8-has little to do with the substance of the right or benefit from which a group is excluded, and much to do with the act of exclusion itself. Proponents' reliance on Crawford v. Board of Education, 458 U.S. 527 (1982), is therefore misplaced. In Crawford, the Court affirmed Proposition 1, a California initiative constitutional amendment that barred state courts from ordering school busing or pupil-assignment plans except when necessary to remedy a federal constitutional violation. Id. at 531-32. Like Proposition 8, Proposition 1 was adopted in response to a decision of the California Supreme Court under the state constitution, which had held that state schools were obligated to take "reasonably feasible steps," including busing and pupil-assignment plans, "to alleviate school segregation." Crawford v. Bd. of Educ., 551 P.2d 28, 45 (Cal. 1976). The Supreme Court "reject[ed] the contention that once a State chooses to do 'more' than the Fourteenth Amendment requires, it may never recede."*fn17 Crawford, 458 U.S. at 535.
That conclusion was consistent with the principle that states should be free "to experiment" with social policy, without fear of being locked in to "legislation that has proved unworkable or harmful when the State was under no obligation to adopt the legislation in the first place." Id. at 535, 539-40.
Critically, however, the Court noted that Proposition 1 did not itself draw any classification; "[i]t simply forb[ade] state courts" from ordering specific remedies under state law "in the absence of a Fourteenth Amendment violation," while maintaining the state constitution's more robust "right to desegregation than exists under the Federal Constitution." Id. at 537, 542 (emphasis added); see also id. at 544 (noting that other remedies remained available). Most important, the prop-osition's purported benefit, "neighborhood schooling," was "made available regardless of race." Id. There was no evidence that the "purpose of [the] repealing legislation [was] to disadvantage a racial minority," which would have made the proposition unconstitutional. Id. at 539 n.21, 543-45 (citing Reitman v. Mulkey, 387 U.S. 369 (1967)). Because Proposition 1 did not establish any classification, and because it was supported by permissible policy preferences against specific court remedies, the Supreme Court held that it was valid. On the same day, by contrast, the Court struck down a similar Washington initiative, because it had been "drawn for racial purposes" in a manner that "impose[d] substantial and unique burdens on racial minorities" and accordingly violated the Fourteenth Amendment. Washington v. Seattle Sch. Dist. No. 1, 458 U.S. 457, 470-71 (1982).
Romer, not Crawford, controls where a privilege or protection is withdrawn without a legitimate reason from a class of disfavored individuals, even if that right may not have been required by the Constitution in the first place. Although Colorado presented before the Supreme Court an argument regarding Crawford identical to the one that Proponents present here, that argument did not persuade the Court.*fn18 Neither Proposition 8 nor Amendment 2 was a law of general applicability that merely curtailed state courts' remedial powers, as opposed to a single group's rights. Rather, both Proposition 8 and Amendment 2 "carve[d] out" rights from gays and lesbians alone. Unlike the measure in Crawford, Proposition 8 is a "discrimination of an unusual character" that requires "careful consideration" of its purposes and effects, whether or not the Fourteenth Amendment required the right to be provided ab initio. Following Romer, we must therefore decide whether a legitimate interest exists that justifies the People of California's action in taking away from same-sex couples the right to use the official designation and enjoy the status of 'marriage'-a legitimate interest that suffices to overcome the "inevitable inference" of animus to which Proposition 8's discriminatory effects otherwise give rise.
We first consider four possible reasons offered by Proponents or amici to explain why Proposition 8 might have been enacted: (1) furthering California's interest in childrearing and responsible procreation, (2) proceeding with caution before making significant changes to marriage, (3) protecting religious freedom, and (4) preventing children from being taught about same-sex marriage in schools. To be credited, these rationales "must find some footing in the realities of the subject addressed by the legislation." Heller v. Doe, 509 U.S. 312, 321 (1993). They are, conversely, not to be credited if they "could not reasonably be conceived to be true by the governmental decisionmaker." Vance v. Bradley, 440 U.S. 93, 111 (1979).*fn19 Because Proposition 8 did not further any of these interests, we conclude that they cannot have been rational bases for this measure, whether or not they are legitimate state interests.
The primary rationale Proponents offer for Proposition 8 is that it advances California's interest in responsible procreation and childrearing. Proponents' Br. 77-93. This rationale appears to comprise two distinct elements. The first is that children are better off when raised by two biological parents and that society can increase the likelihood of that family structure by allowing only potential biological parents-one man and one woman-to marry. The second is that marriage reduces the threat of "irresponsible procreation"-that is, unintended pregnancies out of wedlock-by providing an incentive for couples engaged in potentially procreative sexual activity to form stable family units. Because same-sex couples are not at risk of "irresponsible procreation" as a matter of biology, Proponents argue, there is simply no need to offer such couples the same incentives. Proposition 8 is not rationally related, however, to either of these purported interests, whether or not the interests would be legitimate under other circumstances.
 We need not decide whether there is any merit to the sociological premise of Proponents' first argument-that families headed by two biological parents are the best environments in which to raise children-because even if Proponents are correct, Proposition 8 had absolutely no effect on the ability of same-sex couples to become parents or the manner in which children are raised in California. As we have explained, Proposition 8 in no way modified the state's laws governing parentage, which are distinct from its laws governing marriage. See Strauss, 207 P.3d at 61. Both before and after Proposition 8, committed opposite-sex couples ("spouses") and same-sex couples ("domestic partners") had identical rights with regard to forming families and raising children. See Cal. Fam. Code § 297.5(d) ("The rights and obligations of registered domestic partners with respect to a child of either of them shall be the same as those of spouses."). Similarly, Proposition 8 did not alter the California adoption or presumed-parentage laws, which continue to apply equally to same-sex couples. Cf. Elisa B., 117 P.3d at 667-71 (applying the presumed parentage statutes to a lesbian couple); Sharon S. v. Super. Ct., 73 P.3d 554, 570 (Cal. 2003) (applying the adoption laws to a lesbian couple). In order to be rationally related to the purpose of funneling more childrearing into families led by two biological parents, Proposition 8 would have had to modify these laws in some way. It did not do so.*fn20
Moreover, California's "current policies and conduct . . . recognize that gay individuals are fully capable of . . . responsibly caring for and raising children." Marriage Cases, 183 P.3d at 428. And California law actually prefers a non-biological parent who has a parental relationship with a child to a biological parent who does not; in California, the parentage statutes place a premium on the "social relationship," not the "biological relationship," between a parent and a child. See, e.g., Susan H. v. Jack S., 30 Cal. App. 4th 1435, 1442-43 (1994). California thus has demonstrated through its laws that Proponents' first rationale cannot "reasonably be conceived to be true by the governmental decisionmaker," Vance, 440 U.S. at 111. We will not credit a justification for Proposition 8 that is totally inconsistent with the measure's actual effect and with the operation of California's family laws both before and after its enactment.
Proponents' second argument is that there is no need to hold out the designation of 'marriage' as an encouragement for same-sex couples to engage in responsible procreation, because unlike opposite-sex couples, same-sex couples pose no risk of procreating accidentally. Proponents contend that California need not extend marriage to same-sex couples when the State's interest in responsible procreation would not be advanced by doing so, even if the interest would not be harmed, either. See Johnson v. Robison, 415 U.S. 361, 383 (1974) ("When . . . the inclusion of one group promotes a legitimate governmental purpose, and the addition of other groups would not, we cannot say that the statute's classification of beneficiaries and nonbeneficiaries is invidiously discriminatory."). But Plaintiffs do not ask that marriage be extended to anyone. As we have by now made clear, the question is whether there is a legitimate governmental interest in withdrawing access to marriage from same-sex couples. We therefore need not decide whether, under Johnson, California would be justified in not extending the designation of 'marriage' to same-sex couples; that is not what Proposition 8 did. Johnson concerns decisions not to add to a legislative scheme a group that is unnecessary to the purposes of that scheme, but Proposition 8 subtracted a disfavored group from a scheme of which it already was a part.*fn21
Under Romer, it is no justification for taking something away to say that there was no need to provide it in the first place; instead, there must be some legitimate reason for the act of taking it away, a reason that overcomes the "inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected." Romer, 517 U.S. at 634. In order to explain how rescinding access to the designation of 'marriage' is rationally related to the State's interest in responsible procreation, Proponents would have had to argue that opposite-sex couples were more likely to procreate accidentally or irresponsibly when same-sex couples were allowed access to the designation of 'marriage.' We are aware of no basis on which this argument would be even conceivably plausible. There is no rational reason to think that taking away the designation of 'marriage' from same-sex couples would advance the goal of encouraging California's opposite-sex couples to procreate more responsibly. The Johnson argument, to put it mildly, does not help Proponents' cause.
Given the realities of California law, and of human nature, both parts of Proponents' primary rationale simply "find [no] footing in the realities of the subject addressed by the legislation," and thus cannot be credited as rational. Heller, 509 U.S. at 321. Whatever sense there may be in preferring biological parents over other couples-and we need not decide whether there is any-California law clearly does not recognize such a preference, and Proposition 8 did nothing to change that circumstance. The same is true for Proponents' argument that it is unnecessary to extend the right to use the designation of 'marriage' to couples who cannot procreate, because the purpose of the designation is to reward couples who procreate responsibly or to encourage couples who wish to procreate to marry first. Whatever merit this argument may have-and again, we need not decide whether it has any-the argument is addressed to a failure to afford the use of the designation of 'marriage' to same-sex couples in the first place; it is irrelevant to a measure withdrawing from them, and only them, use of that designation.
The same analysis applies to the arguments of some amici curiae that Proposition 8 not only promotes responsible procreation and childrearing as a general matter but promotes the single best family structure for such activities. See, e.g., Br. Amicus Curiae of High Impact Leadership Coalition, et al. 14 ("Society has a compelling interest in preserving the institution that best advances the social interests in responsible procreation, and that connects procreation to responsible child-rearing."); Br. Amicus Curiae of Am. Coll. of Pediatricians 15 ("[T]he State has a legitimate interest in promoting the family structure that has proven most likely to foster an optimal environment for the rearing of children."). As discussed above, Proposition 8 in no way alters the state laws that govern child-rearing and procreation. It makes no change with respect to the laws regarding family structure. As before Proposition 8, those laws apply in the same way to same-sex couples in domestic partnerships and to married couples. Only the designation of 'marriage' is withdrawn and only from one group of individuals.
 We in no way mean to suggest that Proposition 8 would be constitutional if only it had gone further-for example, by also repealing same-sex couples' equal parental rights or their rights to share community property or enjoy hospital visitation privileges. Only if Proposition 8 had actually had any effect on childrearing or "responsible procreation" would it be necessary or appropriate for us to consider the legitimacy of Proponents' primary rationale for the measure.*fn22 Here, given all other pertinent aspects of California law, Proposition 8 simply could not have the effect on procreation or child-bearing that Proponents claim it might have been intended to have. Accordingly, an interest in responsible procreation and childbearing cannot provide a rational basis for the measure.
We add one final note. To the extent that it has been argued that withdrawing from same-sex couples access to the designation of 'marriage'-without in any way altering the substantive laws concerning their rights regarding childrearing or family formation-will encourage heterosexual couples to enter into matrimony, or will strengthen their matrimonial bonds, we believe that the People of California "could not reasonably" have "conceived" such an argument "to be true." Vance, 440 U.S. at 111. It is implausible to think that denying two men or two women the right to call themselves married could somehow bolster the stability of families headed by one man and one woman. While deferential, the rational-basis standard "is not a toothless one." Mathews v. Lucas, 427 U.S. 495, 510 (1976). "[E]ven the standard of rationality . . . must find some footing in the realities of the subject addressed by the legislation." Heller, 509 U.S. at 321. Here, the argument that withdrawing the designation of 'marriage' from same-sex couples could on its own promote the strength or stability of opposite-sex marital relationships lacks any such footing in reality.
 Proponents offer an alternative justification for Proposition 8: that it advances California's interest in "proceed[ing] with caution" when considering changes to the definition of marriage. Proponents' Br. 93. But this rationale, too, bears no connection to the reality of Proposition 8. The amendment was enacted after the State had provided same-sex couples the right to marry and after more than 18,000 couples had married (and remain married even after Proposition 8, Strauss, 207 P.3d at 122).*fn23
Perhaps what Proponents mean is that California had an interest in pausing at 18,000 married same-sex couples to evaluate whether same-sex couples should continue to be allowed to marry, or whether the same-sex marriages that had already occurred were having any adverse impact on society. Even if that were so, there could be no rational connection between the asserted purpose of "proceeding with caution" and the enactment of an absolute ban, unlimited in time, on same-sex marriage in the state constitution.*fn24 To enact a constitutional prohibition is to adopt a fundamental barrier: it means that the legislative process, by which incremental policymaking would normally proceed, is completely foreclosed. Cf. Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483, 489 (1955) (observing that legislatures may rationally reform policy "one step at a time"). Once Proposition 8 was enacted, any future steps forward, however cautious, would require "enlisting the citizenry of [California] to amend the State Constitution" once again. Romer, 517 U.S. at 631.
Had Proposition 8 imposed not a total ban but a time-specific moratorium on same-sex marriages, during which the Legislature would have been authorized to consider the question in detail or at the end of which the People would have had to vote again to renew the ban, the amendment might plausibly have been designed to "proceed with caution." In that case, we would have had to consider whether the objective of "proceed[ing] with caution" was a legitimate one. But that is not what Proposition 8 did. The amendment superseded the Marriage Cases and then went further, by prohibiting the Legislature or even the People (except by constitutional amendment) from choosing to make the designation of 'marriage' available to same-sex couples in the future. Such a permanent ban cannot be rationally related to an interest in proceeding with caution.
 In any event, in light of the express purpose of Proposition 8 and the campaign to enact it, it is not credible to suggest that "proceed[ing] with caution" was the reason the voters adopted the measure. The purpose and effect of Proposition 8 was "to eliminate the right of same-sex couples to marry in California"-not to "suspend" or "study" that right. Voter Information Guide at 54 (Proposition 8, Official Title and Summary) (emphasis added).*fn25 The voters were told that Proposition 8 would "overturn[ ]" the Marriage Cases "to RESTORE the meaning of marriage." Id. at 56 (Argument in Favor of Proposition 8). The avowed purpose of Proposition 8 was to return with haste to a time when same-sex couples were barred from using the official designation of 'marriage,' not to study the matter further before deciding whether to make the designation more equally available.
 We briefly consider two other potential rationales for Proposition 8, not raised by Proponents but offered by amici curiae. First is the argument that Proposition 8 advanced the State's interest in protecting religious liberty. See, e.g., Br. Amicus Curiae of the Becket Fund for Religious Liberty (Becket Br.) 2. There is no dispute that even before Proposition 8, "no religion [was] required to change its religious policies or practices with regard to same-sex couples, and no religious officiant [was] required to solemnize a marriage in contravention of his or her religious beliefs." Marriage Cases, 183 P.3d at 451-52; see Becket Br. 4-5 (acknowledging this point). Rather, the religious-liberty interest that Proposition 8 supposedly promoted was to decrease the likelihood that religious organizations would be penalized, under California's antidiscrimination laws and other government policies concerning sexual orientation, for refusing to provide services to families headed by same-sex spouses. But Proposition 8 did nothing to affect those laws. To the extent that California's antidiscrimination laws apply to various activities of religious organizations, their protections apply in the same way as before. Amicus's argument is thus more properly read as an appeal to the Legislature, seeking reform of the State's antidiscrimination laws to include greater accommodations for religious organizations. See, e.g., Becket Br. 8 n.6 ("Unlike many other states, California has no religious exemptions to its statutory bans on gender, marital status, and sexual orientation discrimination in public accommodations."). This argument is in no way addressed by Proposition 8 and could not have been the reason for Proposition 8.
 Second is the argument, prominent during the campaign to pass Proposition 8, that it would "protect[ ] our children from being taught in public schools that 'same-sex marriage' is the same as traditional marriage." Perry IV, 704 F. Supp. 2d at 930, 989-90 (quoting the Voter Information Guide at 56) (emphasis omitted); see Br. Amicus Curiae for the Hausvater Project 13-15. Yet again, California law belies the premise of this justification. Both before and after Proposition 8, schools have not been required to teach anything about same-sex marriage. They "may . . . elect[ ] to offer comprehensive sexual health education"; only then might they be required to "teach respect for marriage and committed relationships." Cal. Educ. Code § 51933(a)-(b), (b)(7). Both before and after Proposition 8, schools have retained control over the content of such lessons. And both before and after Proposition 8, schools and individual teachers have been prohibited from giving any instruction that discriminates on the basis of sexual orientation; now as before, students could not be taught the superiority or inferiority of either same- or opposite-sex marriage or other "committed relationships." Cal. Educ. Code §§ 51500, 51933(b)(4). The Marriage Cases therefore did not weaken, and Proposition 8 did not strengthen, the rights of schools to control their curricula and of parents to control their children's education.
There is a limited sense in which the extension of the designation 'marriage' to same-sex partnerships might alter the content of the lessons that schools choose to teach. Schools teach about the world as it is; when the world changes, lessons change. A shift in the State's marriage law may therefore affect the content of classroom instruction just as would the election of a new governor, the discovery of a new chemical element, or the adoption of a new law permitting no-fault divorce: students learn about these as empirical facts of the world around them. But to protest the teaching of these facts is little different from protesting their very existence; it is like opposing the election of a particular governor on the ground that students would learn about his holding office, or opposing the legitimation of no-fault divorce because a teacher might allude to that fact if a course in societal structure were taught to graduating seniors. The prospect of children learning about the laws of the State and society's assessment of the legal rights of its members does not provide an independent reason for stripping members of a disfavored group of those rights they presently enjoy.
Proposition 8's only effect, we have explained, was to withdraw from gays and lesbians the right to employ the designation of 'marriage' to describe their committed relationships and thus to deprive them of a societal status that affords dignity to those relationships. Proposition 8 could not have reasonably been enacted to promote childrearing by biological parents, to encourage responsible procreation, to proceed with caution in social change, to protect religious liberty, or to control the education of schoolchildren. Simply taking away the designation of 'marriage,' while leaving in place all the substantive rights and responsibilities of same-sex partners, did not do any of the things its Proponents now suggest were its purposes. Proposition 8 "is so far removed from these particular justifications that we find it impossible to credit them." Romer, 517 U.S. at 635. We therefore need not, and do not, decide whether any of these purported rationales for the law would be "legitimate," id. at 632, or would suffice to justify Proposition 8 if the amendment actually served to further them.
We are left to consider why else the People of California might have enacted a constitutional amendment that takes away from gays and lesbians the right to use the designation of 'marriage.' One explanation is the desire to revert to the way things were prior to the Marriage Cases, when 'marriage' was available only to opposite-sex couples, as had been the case since the founding of the State and in other jurisdictions long before that. This purpose is one that Proposition 8 actually did accomplish: it "restore[d] the traditional definition of marriage as referring to a union between a man and a woman." Strauss, 207 P.3d at 76. But tradition alone is not a justification for taking away a right that had already been granted, even though that grant was in derogation of tradition. In Romer, it did not matter that at common law, gays and lesbians were afforded no protection from discrimination in the private sphere; Amendment 2 could not be justified on the basis that it simply repealed positive law and restored the "traditional" state of affairs. 517 U.S. at 627-29. Precisely the same is true here.
Laws may be repealed and new rights taken away if they have had unintended consequences or if there is some con- ceivable affirmative good that revocation would produce, cf. Crawford, 458 U.S. at 539-40, but new rights may not be stripped away solely because they are new. Tradition is a legitimate consideration in policymaking, of course, but it cannot be an end unto itself. Cf. Williams v. Illinois, 399 U.S. 235, 239-40 (1970). "[T]he fact that the governing majority in a State has traditionally viewed a particular practice as immoral is not a sufficient reason for upholding a law prohibiting the practice; neither history nor tradition could save a law prohibiting miscegenation from constitutional attack." Lawrence v. Texas, 539 U.S. 558, 577-78 (2003); see Loving v. Virginia, 388 U.S. 1 (1967) (noting the historical pedigree of bans on interracial marriage but not even considering tradition as a possible justification for Virginia's law). If tradition alone is insufficient to justify maintaining a prohibition with a discriminatory effect, then it is necessarily insufficient to justify changing the law to revert to a previous state. A preference for the way things were before same-sex couples were allowed to marry, without any identifiable good that a return to the past would produce, amounts to an impermissible preference against same-sex couples themselves, as well as their families.
 Absent any legitimate purpose for Proposition 8, we are left with "the inevitable inference that the disadvantage imposed is born of animosity toward," or, as is more likely with respect to Californians who voted for the Proposition, mere disapproval of, "the class of persons affected." Romer, 517 U.S. at 634. We do not mean to suggest that Proposition 8 is the result of ill will on the part of the voters of California. "Prejudice, we are beginning to understand, rises not from malice or hostile animus alone." Bd. of Trustees of Univ. of Ala. v. Garrett, 531 U.S. 356, 374 (2001) (Kennedy, J., concurring). Disapproval may also be the product of longstanding, sincerely held private beliefs. Still, while "[p]rivate biases may be outside the reach of the law, . . . the law cannot, directly or indirectly, give them effect." Palmore v. Sidoti, 466 U.S. 429, 433 (1984). Ultimately, the "inevitable infer- ence" we must draw in this circumstance is not one of ill will, but rather one of disapproval of gays and lesbians as a class. "[L]aws singling out a certain class of citizens for disfavored legal status or general hardships are rare." Romer, 517 U.S. at 633. Under Romer, we must infer from Proposition 8's effect on California law that the People took away from gays and lesbians the right to use the official designation of 'marriage' -and the societal status that accompanies it-because they disapproved of these individuals as a class and did not wish them to receive the same official recognition and societal approval of their committed relationships that the State makes available to opposite-sex couples.
 It will not do to say that Proposition 8 was intended only to disapprove of same-sex marriage, rather than to pass judgment on same-sex couples as people. Just as the criminalization of "homosexual conduct . . . is an invitation to subject homosexual persons to discrimination both in the public and in the private spheres," Lawrence, 539 U.S. at 575, so too does the elimination of the right to use the official designation of 'marriage' for the relationships of committed same-sex couples send a message that gays and lesbians are of lesser worth as a class-that they enjoy a lesser societal status. Indeed, because laws affecting gays and lesbians' rights often regulate individual conduct-what sexual activity people may undertake in the privacy of their own homes, or who is permitted to marry whom-as much as they regulate status, the Supreme Court has "declined to distinguish between status and conduct in [the] context" of sexual orientation. Christian Legal Soc'y v. Martinez, 130 S. Ct. 2971, 2990 (2010). By withdrawing the availability of the recognized designation of 'marriage,' Proposition 8 enacts nothing more or less than a judgment about the worth and dignity of gays and lesbians as a class.
 Just as a "desire to harm . . . cannot constitute a legitimate governmental interest," Moreno, 413 U.S. at 534, neither can a more basic disapproval of a class of people. Romer, 517 U.S. at 633-35. "The issue is whether the majority may use the power of the State to enforce these views on the whole society" through a law that abridges minority individuals' rights. Lawrence, 539 U.S. at 571. It may not. Without more, "[m]oral disapproval of [a] group, like a bare desire to harm the group, is an interest that is insufficient to satisfy rational basis review under the Equal Protection Clause." Id. at 582 (O'Connor, J., concurring). Society does sometimes draw classifications that likely are rooted partially in disapproval, such as a law that grants educational benefits to veterans but denies them to conscientious objectors who engaged in alternative civilian service. See Johnson, 415 U.S. at 362-64. Those classifications will not be invalidated so long as they can be justified by reference to some independent purpose they serve; in Johnson, they could provide an incentive for military service and direct assistance to those who needed the most help in readjusting to post-war life, see id. at 376-83. Enacting a rule into law based solely on the disapproval of a group, however, "is a classification of persons undertaken for its own sake, something the Equal Protection Clause does not permit." Romer, 517 U.S. at 635. Like Amendment 2, Proposition 8 is a classification of gays and lesbians undertaken for its own sake.
The "inference" that Proposition 8 was born of disapproval of gays and lesbians is heightened by evidence of the context in which the measure was passed.*fn26 The district court found that "[t]he campaign to pass Proposition 8 relied on stereo-types to show that same-sex relationships are inferior to opposite-sex relationships." Perry IV, 704 F. Supp. 2d at 990. Television and print advertisements "focused on . . . the concern that people of faith and religious groups would somehow be harmed by the recognition of gay marriage" and "conveyed a message that gay people and relationships are inferior, that homosexuality is undesirable and that children need to be protected from exposure to gay people and their relationships." Id. These messages were not crafted accidentally. The strategists responsible for the campaign in favor of Proposition 8 later explained their approach: " '[T]here were limits to the degree of tolerance Californians would afford the gay community. They would entertain allowing gay marriage, but not if doing so had significant implications for the rest of society,' " such as what children would be taught in school. Id. at 988 (quoting Frank Schubert & Jeff Flint, Passing Prop 8, Politics, Feb. 2009, at 45-47). Nor were these messages new; for decades, ballot measures regarding homosexuality have been presented to voters in terms designed to appeal to stereotypes of gays and lesbians as predators, threats to children, and practitioners of a deviant "lifestyle." See Br. Amicus Curiae of Constitutional Law Professors at 2-8. The messages presented here mimic those presented to Colorado voters in support of Amendment 2, such as, "Homosexual indoctrination in the schools? IT'S HAPPENING IN COLORADO!" Colorado for Family Values, Equal Rights-Not Special Rights, at 2 (1992), reprinted in Robert Nagel, Playing Defense, 6 Wm. & Mary Bill Rts. J. 167, 193 (1997).
 When directly enacted legislation "singl[es] out a certain class of citizens for disfavored legal status," we must "insist on knowing the relation between the classification adopted and the object to be attained," so that we may ensure that the law exists "to further a proper legislative end" rather than "to make the[ ] [class] unequal to everyone else." Romer, 517 U.S. at 632-33, 635. Proposition 8 fails this test. Its sole purpose and effect is "to eliminate the right of same-sex couples to marry in California"-to dishonor a disfavored group by taking away the official designation of approval of their committed relationships and the accompanying societal status, and nothing more. Voter Information Guide at 54. "It is at once too narrow and too broad," for it changes the law far too little to have any of the effects it purportedly was intended to yield, yet it dramatically reduces the societal standing of gays and lesbians and diminishes their dignity. Romer, 517 U.S. at 633. Proposition 8 did not result from a legitimate "Kulturkampf" concerning the structure of families in California, because it had no effect on family structure, but in order to strike it down, we need not go so far as to find that it was enacted in "a fit of spite." Id. at 636 (Scalia, J., dissenting). It is enough to say that Proposition 8 operates with no apparent purpose but to impose on gays and lesbians, through the public law, a majority's private disapproval of them and their relationships, by taking away from them the official designation of 'marriage,' with its societally recognized status. Proposition 8 therefore violates the Equal Protection Clause.
Finally, we address Proponents' motion to vacate the district court's judgment. On April 6, 2011, after resigning from the bench, former Chief Judge Walker disclosed that he was gay and that he had for the past ten years been in a relationship with another man. Proponents moved shortly thereafter to vacate the judgment on the basis that 28 U.S.C. § 455(b)(4) obligated Chief Judge Walker to recuse himself, because he had an "interest that could be substantially affected by the outcome of the proceeding," and that 28 U.S.C. § 455(a) obligated him either to recuse himself or to disclose his potential conflict, because "his impartiality might reasonably be questioned." Chief Judge Ware, to whom this case was assigned after Chief Judge Walker's retirement, denied the motion after receiving briefs and hearing argument.
The district court properly held that it had jurisdiction to hear and deny the motion under Fed. R. Civ. P. 62.1(a), that the motion was timely, and that Chief Judge Walker had no obligation to recuse himself under either § 455(b)(4) or § 455(a) or to disclose any potential conflict. As Chief Judge Ware explained, the fact that a judge "could be affected by the outcome of a proceeding[,] in the same way that other members of the general public would be affected, is not a basis for either recusal or disqualification under Section 455(b)(4)." Perry v. Schwarzenegger, 790 F. Supp. 2d 1119, 1122 (N.D. Cal. 2011); see In re City of Houston, 745 F.2d 925, 929-30 (5th Cir. 1984) ("We recognize that 'an interest which a judge has in common with many others in a public matter is not sufficient to disqualify him.' "). Nor could it possibly be "reasonable to presume," for the purposes of § 455(a), "that a judge is incapable of making an impartial decision about the constitutionality of a law, solely because, as a citizen, the judge could be affected by the proceeding." 790 F. Supp. 2d at 1122; see United States v. Alabama, 828 F.2d 1532, 1541-42 (11th Cir. 1987). To hold otherwise would demonstrate a lack of respect for the integrity of our federal courts.
 The denial of the motion to vacate was premised on Chief Judge Ware's finding that Chief Judge Walker was not obligated to recuse himself. "We review the district court's denial of a motion to vacate the judgment for an abuse of discretion." Jeff D. v. Kempthorne, 365 F.3d 844, 850 (9th Cir. 2004). Our standard for abuse of discretion requires us to (1) "look to whether the trial court identified and applied the correct legal rule to the relief requested"; and, if the trial court applied the correct legal rule, to (2) "look to whether the trial court's resolution . . . resulted from a factual finding that was illogical, implausible, or without support in inferences that may be drawn from the facts in the record." United States v. Hinkson, 585 F.3d 1247, 1263 (9th Cir. 2009) (en banc). Here, Chief Judge Ware did not incorrectly apply the law. He identified and applied § 455(b)(4) and § 455(a), the correct legal rules, as well as the relevant precedents. His application of the law, determining whether Chief Judge Walker was obligated to recuse himself, was discretionary. See United States v. Johnson, 610 F.3d 1138, 1147-48 (9th Cir. 2010). His resolution of the issue on the basis of the facts was not illogical, implausible, or without support in inferences that may be drawn from the facts in the record. Thus, we affirm Chief Judge Ware's decision not to grant the motion to vacate.
By using their initiative power to target a minority group and withdraw a right that it possessed, without a legitimate reason for doing so, the People of California violated the Equal Protection Clause. We hold Proposition 8 to be unconstitutional on this ground. We do not doubt the importance of the more general questions presented to us concerning the rights of same-sex couples to marry, nor do we doubt that these questions will likely be resolved in other states, and for the nation as a whole, by other courts. For now, it suffices to conclude that the People of California may not, consistent with the Federal Constitution, add to their state constitution a provision that has no more practical effect than to strip gays and lesbians of their right to use the official designation that the State and society give to committed relationships, thereby adversely affecting the status and dignity of the members of a disfavored class. The judgment of the district court is
N.R. SMITH, Circuit Judge, concurring in part and dissenting in part:
I agree with the majority's analysis and decisions in parts III and VI of its opinion, determining that (1) the Proponents have standing to bring this appeal; and (2) the Motion to Vacate the Judgment should be denied. Because I do not agree with the majority's analysis of other topics regarding the constitutionality of Proposition 8, I have chosen to write separately. Ultimately, I am not convinced that Proposition 8 is not rationally related to a legitimate governmental interest. I must therefore respectfully dissent.
Before addressing the issues now presented before our panel, I want to emphasize a distinguishing point in my analysis from what may be anticipated by the reader. Similar to the California Supreme Court in its prior opinion concerning Proposition 8, our panel was not tasked with determining whether this constitutional amendment "is wise or sound as a matter of policy or whether we, as individuals, believe it should be a part of the California Constitution." Strauss v. Horton, 207 P.3d 48, 59 (Cal. 2009). Our personal views regarding the political and sociological debate on marriage equality are irrelevant to our task. Instead, we are only asked to consider the constitutional validity of Proposition 8 under the federal Constitution. The California Supreme Court has already interpreted and applied "the principles and rules embodied in the California Constitution" to Proposition 8 and found it valid. Strauss, 207 P.3d 48.
Proponents and their supporting amici (hereinafter Proponents) argue that the United States Supreme Court's summary dismissal in Baker v. Nelson, 409 U.S. 810 (1972) (mem.), "mandates reversal of the district court's ruling." According to Proponents, the claims raised here are the same as those rejected in Baker, and the claims are therefore foreclosed by that decision. The majority dispenses with Baker in a footnote. However, other federal courts have indicated that Baker, if it is not controlling, at least stands for exercising "restraint" when it comes to addressing due process and equal protection challenges against laws prohibiting marriage by same-sex couples. Citizens for Equal Protection v. Bruning, 455 F.3d 859, 870 (8th Cir. 2006); see also Wilson v. Ake, 354 F. Supp. 2d 1298, 1305 (M.D. Fla. 2005) ("Baker v. Nelson is binding precedent upon this Court . . . ."). But see In re Kandu, 315 B.R. 123, 138 (Bankr. W.D. Wash. 2004) (concluding that "Baker is not binding precedent on the issues presented" because the case centered on federal Defense of Marriage Act and because "doctrinal developments" indicated Baker was no longer binding). Because Baker is binding United States Supreme Court precedent and may foreclose Plaintiffs' claims, one must follow it or distinguish it.
In Baker v. Nelson, two men were denied a marriage license by a Minnesota county clerk. 191 N.W.2d 185, 185 (Minn. 1971). Because they were denied the license, the two men filed suit asking that the court force the clerk to grant the license. Id. In Minnesota Statutes c. 517, the Minnesota state legislature had codified that the state "d[id] not authorize marriage between persons of the same sex. . . ." Id. at 186. On appeal, the Minnesota Supreme Court addressed several issues, including whether the Minnesota statutes prohibiting marriage by same-sex couples denied the petitioners "the equal protection of the laws" as guaranteed by the Fourteenth Amendment. Id. The Minnesota Supreme Court held that "[t]he equal protection clause of the Fourteenth Amendment . . . is not offended by the state's classification of persons authorized to marry." Id. at 187. On appeal to the United States Supreme Court, the Court summarily dismissed the appeal "for want of a substantial federal question." Baker v. Nelson, 409 U.S. 810.
Though not stated in the summary dismissal in Baker, the Supreme Court decision has long standing precedent supporting it. Throughout our nation's history, the States have had "the absolute right to prescribe the conditions upon which the marriage relation between its own citizens shall be credited . . . ." Pennoyer v. Neff, 95 U.S. 714, 734-35 (1878), reaf-firmed in Sosna v. Iowa, 419 U.S. 393, 404 (1975).
Marriage, as creating the most important relation in life, as having more to do with the morals and civilization of a people than any other institution, has always been subject to the control of the legislature. That body prescribes the age at which parties may contract to marry, the procedure or form essential to constitute marriage, the duties and obligations it creates, its effects upon the property rights of both, present and prospective, and the acts which may constitute grounds for its dissolution.
Maynard v. Hill, 125 U.S. 190, 205 (1888).
As Justice Stewart opined in his concurrence in Zablocki v. Redhail, a State may in many circumstances absolutely prohibit [marriage]. Surely, for example, a State may legitimately say that no one can marry his or her sibling, that no one can marry who is not at least 14 years old, that no one can marry without first passing an examination for venereal disease, or that no one can marry who has a living husband or wife.
434 U.S. 374, 392 (1978) (Stewart, J., concurring).
The summary dismissal of an appeal for want of a substantial federal question is a decision on the merits. Hicks v. Miranda, 422 U.S. 332, 344 (1975). "[U]nless and until the Supreme Court should instruct otherwise, inferior federal courts had best adhere to the view that if the Court has branded a question as unsubstantial, it remains so except when doctrinal developments indicate otherwise . . . ." Id. (internal quotation marks omitted). "[L]ower courts are bound by summary decisions by [the Supreme] Court until such time as the Court informs (them) that (they) are not." Id. at 344-45 (internal quotation marks omitted). "Summary . . . dismissals for want of a substantial federal question . . . reject the specific challenges presented in the statement of jurisdiction and do leave undisturbed the judgment appealed from. They do prevent lower courts from coming to opposite conclusions on the precise issues presented and necessarily decided by those actions." Mandel v. Bradley, 432 U.S. 173, 176 (1977) (per curiam). Thus, "[a] summary disposition affirms only the judgment of the court below, and no more may be read into [the] action than was essential to sustain that judgment." Ill. State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 182-83 (1979) (citation omitted). "Questions which 'merely lurk in the record' are not resolved, and no resolution of them may be inferred." Id. at 183 (citation omitted).
The jurisdictional statements presented to the United States Supreme Court in Baker v. Nelson were as follows:
1. Whether appellee's refusal to sanctify appellants' marriage deprives appellants of their liberty to marry and of their property without due process of law under the Fourteenth Amendment.
2. Whether appellee's refusal, pursuant to Minnesota marriage statutes, to sanctify appellants' marriage because both are of the male sex violates their rights under the equal protection clause of the Fourteenth Amendment.
3. Whether appellee's refusal to sanctify appellants' marriage deprives appellants of their right to privacy under the Ninth and Fourteenth Amendments.
See In re Kandu, 315 B.R. at 137.
Here, we must address whether the question before us involves "the precise issues presented and necessarily decided by" Baker v. Nelson, such that the Supreme Court's summary dismissal would have precedential effect here. Alternatively, the question before us could be one that "merely lurk[ed] in the record" of Baker, and the present case would not be resolved by the Supreme Court's summary dismissal.
In this case, the following issues were presented for review:
1. Whether [Proponents] have standing to appeal the district court's judgment.
2. Whether Proposition 8 violates the Due Process Clause of the Fourteenth Amendment to the United States Constitution.
3. Whether Proposition 8 violates the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution.
Plaintiff-Intervenor City and County of San Francisco (hereinafter San Francisco) presented the following additional issue for review:
1. Whether Proposition 8, a constitutional amendment adopted after a plebiscite campaign that played on fears and prejudices about lesbians and gay men, violates the Equal Protection Clause of the federal Constitution where its effect is to remove the honored title "marriage" but not the incidents of marriage from same-sex couples, and its purpose is to remove the taint that its supporters believed the inclusion of lesbian and gay couples worked on the institution of marriage.
The equal protection question raised in this case seems to be distinguishable from the precise issues presented and necessarily decided in Baker, especially when the equal protection issue is framed as San Francisco advocates.*fn28 The equal protection issue decided in Baker rested on whether Minnesota's "refusal, pursuant to Minnesota marriage statutes, to sanctify appellants' marriage . . . violates their rights under the equal protection clause . . . ." In re Kandu, 315 B.R. at 137. Here, San Francisco presents the issue of whether Proposition 8's effect of "remov[ing] the honored title 'marriage' but not the incident of marriage from same-sex couples" violates equal protection. This Proposition 8 issue may have "merely lurk[ed] in the record" of Baker. Unlike Minnesota, California granted same-sex couples rights to both the designation and the incidents of marriage, before withdrawing the right of access to the designation through Proposition 8. Therefore, the constitutionality of withdrawing from same-sex couples the right of access to the designation of marriage does not seem to be among the "specific challenges" raised in Baker. If so, though the precedential effect of Baker v. Nelson is not challenged by this decision, such precedent is distinguishable from the decision of the district court here.
In deciding this case, one should be mindful that generally state governance over marriage is not challenged easily. However, while "marriage is a social relation subject to the State's police power," this does not mean that the State's "powers to regulate marriage are unlimited notwithstanding the commands of the Fourteenth Amendment." Loving v. Virginia, 388 U.S. 1, 7 (1967). A marriage regulation "containing racial classifications," such as the one at issue in Loving, is subject to "the very heavy burden of justification which the Fourteenth Amendment has traditionally required of state statutes drawn according to race." Id. at 9. However, not "every state regulation which relates in any way to the incidents of or prerequisites for marriage must be subjected to rigorous scrutiny." Zablocki, 434 U.S. at 386 (majority opinion). Proposition 8 does not involve such a suspect classification and therefore should not be analyzed under any heightened scrutiny, but we must still ask "whether there is any rational foundation for the discrimination[ ] . . . ." See Loving, 388 U.S. at 9.
The Plaintiffs, San Francisco, and their supporting amici (hereinafter Plaintiffs) challenge Proposition 8 under the Equal Protection Clause of the Fourteenth Amendment. However, because Proposition 8 is "a classification neither involving fundamental rights nor proceeding along suspect lines," Heller v. Doe, 509 U.S. 312, 319 (1993), I do not address the application of strict scrutiny review to Proposition 8. Under strict scrutiny review, the government would need to establish that the classification is necessary to achieve a compelling governmental interest, and there must not be a less onerous available alternative. The United States Supreme Court has not recognized that the fundamental right to marry includes a fundamental right to gay marriage. See Lawrence, 539 U.S. at 578. Gays and lesbians are not a suspect or quasi-suspect class. High Tech Gays v. Def. Indus. Sec. Clearance Office, 895 F.2d 563, 573 (9th Cir. 1990).
I also do not address intermediate scrutiny because Supreme Court precedent thus far has never held that sexual orientation is a "quasi-suspect classification." See City of Cle-burne v. Cleburne Living Ctr., 473 U.S. 432, 441-42 (1985). Under that standard of review, generally applied in illegitimacy and gender cases, the government would need to establish that the classification is substantially related to an important governmental interest. See id. at 441.
Thus, Proposition 8 is subject to rational basis review rather than to any heightened scrutiny. See id. at 440-42.
"The Fourteenth Amendment's promise that no person shall be denied the equal protection of the laws must coexist with the practical necessity that most legislation classifies for one purpose or another, with resulting disadvantage to various groups or persons." Romer v. Evans, 517 U.S. 620, 631 (1996). Thus, when assessing the constitutionality of most government measures, we use rational basis review in an attempt "to reconcile the principle with the reality." Id. Under rational basis review, "we will uphold the legislative classification so long as it bears a rational relation to some legitimate end." Id.
In equal protection analysis, rational basis review "is not a license for courts to judge the wisdom, fairness, or logic of legislative choices." Heller, 509 U.S. at 319 (internal quotation marks omitted). A classification "neither involving fundamental rights nor proceeding along suspect lines is accorded a strong presumption of validity." Id. "Such a classification cannot run afoul of the Equal Protection Clause if there is a rational relationship between the disparity of treatment and some legitimate governmental purpose." Id. at 320. The government is not required to "actually articulate at any time the purpose or rationale supporting its classification"; rather, a classification "must be upheld against equal protection challenge if there is any reasonably conceivable state of facts that could provide a rational basis for the classification." Id. (internal quotation marks omitted).
Additionally, the government "has no obligation to provide evidence to sustain the rationality of a statutory classification." Id. The measure at issue "is not subject to courtroom factfinding and may be based on rational speculation unsupported by evidence or empirical data." Id. (internal quotation marks omitted). "[T]he burden is on the one attacking the legislative arrangement to negative every conceivable basis which might support it . . . ." Id. (internal quotation marks omitted). Further, a legislature's generalizations may pass rational basis review "even when there is an imperfect fit between means and ends." Id. at 321. In sum, the measure need only have "arguable" assumptions underlying its "plausible rationales" to survive constitutional challenge. Id. at 333.
However, "even the standard of rationality . . . must find some footing in the realities of the subject addressed by the legislation." Id. at 321. Also, some interests are not legitimate governmental interests. E.g., Romer, 517 U.S. at 634 (stating that "animosity toward the class of persons affected" is not a legitimate governmental interest); Cleburne, 473 U.S. at 448 (stating that "mere negative attitudes, or fear" are not legitimate governmental interests); U.S. Dep't of Agric. v. Moreno, 413 U.S. 528, 534 (1973) (stating that a "bare . . . desire to harm a politically unpopular group" is not a legitimate governmental interest).
As a general rule, states may use their police power to regulate the "morals" of their population. See, e.g., Berman v. Parker, 348 U.S. 26, 32 (1954). In his dissent in Lawrence, 539 U.S. at 589-91 (Scalia, J., dissenting), Justice Scalia argued that "[c]ountless judicial decisions and legislative enactments have relied on the ancient proposition that a governing major-ity's belief that certain sexual behavior is 'immoral and unacceptable' constitutes a rational basis for regulation." Id. at 589. He then suggested that the Supreme Court has relied on morality as the basis for its decision making and states, "[s]tate laws against bigamy, same-sex marriage, adult incest, prostitution, masturbation, adultery, fornication, bestiality, and obscenity are likewise sustainable only in light of validation of laws based on moral choices." Id. at 590.
However, Justice O'Connor articulated a different perspective in determining whether moral disapproval may serve as a rational basis for equal protection. She outlined that moral disapproval is not a legitimate state interest to justify, by itself, a statute that bans homosexual conduct. She stated that "[m]oral disapproval of this group, like a bare desire to harm the group, is an interest that is insufficient to satisfy rational basis review under the Equal Protection Clause." Id. at 582 (O'Connor, J., concurring). She continued: "Indeed, we have never held that moral disapproval, without any other asserted state interest, is a sufficient rationale under the Equal Protection Clause to justify a law that discriminates among groups of persons." Id. The Lawrence majority opinion seems to have implicitly agreed with Justice O'Connor, when it stated that a court's "obligation is to define the liberty of all, not to mandate its own moral code." Id. at 559 (majority opinion) (internal quotation mark omitted).
Therefore, such interests (e.g., animus, negative attitudes, fear, a bare desire to harm, and moral disapproval) alone will not support the constitutionality of a measure, because the Equal Protection Clause does not permit a "status-based enactment divorced from any factual context from which [the courts] could discern a relationship to legitimate state interests," or a "classification of persons undertaken for its own sake . . . ." Romer, 517 U.S. at 635.
The majority concludes that "Romer governs our analysis notwithstanding the differences between Amendment 2 and Proposition 8," because of the similarities between the measures at issue in Romer and in the present case. However, the differences between Amendment 2 and Proposition 8 indicate that Romer does not directly control our analysis of the constitutionality of Proposition 8.
Before comparing Amendment 2 to Proposition 8, I want to attempt to clarify the extent of the Plaintiffs' interest asserted here. One must understand the unique manner in which California defines this interest. Because the California Supreme Court defined and clarified that interest in its Strauss v. Horton opinion, I quote liberally from it.
Proposition 8 "properly must be understood as having a considerably narrower scope and more limited effect" than what might be the case in other states. Strauss, 207 P.3d at 61. "Proposition 8 does not entirely repeal or abrogate the aspect of a same-sex couple's state constitutional right to . . . choose one's life partner and enter with that person into a committed, officially recognized, and protected family relationship that enjoys all of the constitutionally based incidents of marriage." Id. (internal quotation marks omitted).
Nor does Proposition 8 fundamentally alter the meaning and substance of state constitutional equal protection principles . . . . Instead, the measure carves out a narrow and limited exception to these state constitutional rights, reserving the official designation of the term "marriage" for the union of opposite-sex couples as a matter of state constitutional law, but leaving undisturbed all of the other extremely significant substantive aspects of a same-sex couple's state constitutional right to establish an officially recognized and protected family relationship and the guarantee of equal protection of the laws. Id.
Further, the California Supreme Court continued, "as a qualitative matter, the act of limiting access to the designation of marriage to opposite-sex couples [through Proposition 8] does not have a substantial or, indeed, even a minimal effect on the governmental plan or framework of California that existed prior to the amendment." Id. at 62.
However, the California Supreme Court was also quick to point out that this differentiation did not diminish or minimize "the significance of the official designation of 'marriage,' " which they characterized as "a vital factor" in their prior decision holding that failing to provide access to this designation to same-sex couples "impinged upon the privacy and due process rights of same-sex couples and violated those couples' right to the equal protection of the laws guaranteed by the California Constitution." Id. at 59, 61.
Therefore, "Proposition 8 reasonably must be interpreted in a limited fashion as eliminating only the right of same-sex couples to equal access to the designation of marriage, and as not otherwise affecting the constitutional right of those couples to establish an officially recognized family relationship." Id. at 76.
Accordingly, although Proposition 8 eliminates the ability of same-sex couples to enter into an official relationship designated "marriage," in all other respects those couples continue to possess, under the state constitutional privacy and due process clauses, "the core set of basic substantive legal rights and attributes traditionally associated with marriage," including, "most fundamentally, the opportunity of an individual to establish-with the person with whom the individual has chose to share his or her life-an officially recognized and protected family possessing mutual rights and responsibilities and entitled to the same respect and dignity accorded a union traditionally designated as marriage." Like opposite-sex couples, same-sex couples enjoy this protection not as a matter of legislative grace, but of constitutional right.
Id. at 77 (citation omitted).
In Romer, Colorado voters adopted Amendment 2 to the State Constitution, which "prohibits all legislative, executive, or judicial action at any level of state or local government designed to protect . . . gays and lesbians.' " 517 U.S. at 624. Amendment 2 was passed in response to municipal ordinances enacted in various Colorado cities that protected "per-sons discriminated against by reason of their sexual orientation." Id. The Supreme Court examined Amendment 2 under rational basis review, where "if a law neither burdens a fundamental right nor targets a suspect class, we will uphold the legislative classification so long as it bears a rational relation to some legitimate end." Id. at 631. The Supreme Court held that Amendment 2 failed rational basis review for two reasons. Id. at 632. "First, the amendment has the peculiar property of imposing a broad and undifferentiated disability on a single named group, an exceptional and . . . invalid form of legislation." Id. "Second, its sheer breadth is so discontinuous with the reasons offered for it that the amendment seems inexplicable by anything but animus toward the class if affects; it lacks a rational relationship to legitimate state interests." Id.
There are several ways to distinguish Romer from the present case. First, in Romer, the Supreme Court stated that "[t]he change Amendment 2 works in the legal status of gays and lesbians in the private sphere is far reaching, both on its own terms and when considered in light of the structure and operation of modern anti-discrimination laws." Id. at 627. Here, "Proposition 8 reasonably must be interpreted in a limited fashion as eliminating only the right of same-sex couples to equal access to the designation of marriage, and as not otherwise affecting the constitutional right of those couples to establish an officially recognized family relationship." Strauss, 207 P.3d at 76. Thus, Romer is inapposite, because Proposition 8 eliminates the right of access to the designation of marriage from same-sex couples, rather than working a far reaching change in their legal status.
Second, Amendment 2's "sheer breadth is so discontinuous with the reasons offered for it that the amendment seems inexplicable by anything but animus toward the class it affects." Romer, 517 U.S. at 632. Again, Proposition 8 "carves out a narrow and limited exception to [the] state constitutional rights" of privacy and due process. Strauss, 207 P.3d at 61. Proposition 8 therefore lacks the "sheer breadth" that prompted the Supreme Court to raise the inference of animus in Romer.
The effect of animus is also unclear. In Romer, the Supreme Court stated that "laws of the kind now before us raise the inevitable inference that the disadvantage imposed is born of animosity towards the class of persons affected." 517 U.S. at 634. The Supreme Court indicated that Amendment 2 was constitutionally invalid, because its only purpose was animus; Amendment 2 was not "directed to any identifiable legitimate purpose or discrete objective." Id. at 635. In short, Romer was a case where the only basis for the measure at issue was animus. However, in a case where the measure at issue was prompted both by animus and by some independent legitimate purpose, the measure may still be constitutionally valid. The Supreme Court has stated that while "negative attitudes," "fear" or other biases "may often accompany irrational (and therefore unconstitutional) discrimination, their presence alone does not a constitutional violation make." Bd. of Trustees of Univ. of Ala. v. Garrett, 531 U.S. 356, 367 (2001) (discussing Cleburne, 473 U.S. at 448). If "animus" is one such bias, its presence alone may not make Proposition 8 invalid if the measure also rationally relates to a legitimate governmental interest.
Finally, gays and lesbians were burdened by Amendment 2, because it "operate[d] to repeal and forbid all laws or policies providing specific protection for gays or lesbians from discrimination by every level of Colorado government." Romer, 517 U.S. at 629. In contrast, "although Proposition 8 eliminates the ability of same-sex couples to enter into an official relationship designated 'marriage,' in all other respects those couples continue to possess, under the state constitutional privacy and due process clauses, the core set of basic substantive legal rights and attributes traditionally associated with marriage . . . ." Strauss, 207 P.3d at 77 (internal quotation marks omitted). Put otherwise, Proposition 8 does not burden gays and lesbians to the same extent Amendment 2 burdened gays and lesbians in Colorado.
Proponents argue that the fact that Proposition 8 withdrew from same-sex couples the existing right of access to the designation of marriage should be significant in our constitutional analysis. However, Supreme Court equal protection cases involving challenges to measures withdrawing an existing right do not indicate that the withdrawal should affect our analysis. Instead, it seems that the court has upheld legislation that withdraws, rather than reserves, some legal right. E.g., U.S. R.R. Ret. Bd. v. Fritz, 449 U.S. 166, 176-77 (1980) (applying "traditional" principles of rational basis review to Congress's determination "that some of those who in the past received full windfall benefits would not continue to do so"); City of New Orleans v. Dukes, 427 U.S. 297, 303-05 (1976) (per curiam) (concluding that city's elimination of rights of some pushcart food vendors, but not others, was "not constitutionally impermissible"). In fact, in its decision in Romer, the Supreme Court does not base its decision on this contention. Rather, it mentioned withdrawing specific legal protections from gays and lesbians only in the context of referring to the irrational targeting of that group when compared to the sweeping change Amendment 2 created in the law.*fn29 Romer, 517 U.S. at 627.
The above differences between Amendment 2 and Proposition 8 indicate that Romer does not directly control here. In Romer, the Supreme Court found that animus alone was the purpose behind Amendment 2. Here, the majority backs into its inference of animus, first determining that all other bases for Proposition 8 are constitutionally invalid. Assuming animus or moral disapproval were one of the purposes of Proposition 8, the measure would still survive rational basis review if there were also a valid rational basis behind Proposition 8. Only if there were no other basis would Proposition 8 fail rational basis review. Thus, our task is to determine whether Proposition 8 rationally relates to any independent legitimate governmental interest.
In our case, Proponents argue that Proposition 8, defining marriage as the union of one man and one woman, is rationally related to a legitimate governmental interest for several reasons. Some of those reasons have already been discussed in the majority opinion and need no further discussion here. However, two of those reasons deserve more discussion, because they have been credited by other courts: (1) a responsible procreation theory, justifying the inducement of marital recognition only for opposite-sex couples, because it "steers procreation into marriage" because opposite-sex couples are the only couples who can procreate children accidentally or irresponsibly; and (2) an optimal parenting theory, justifying the inducement of marital recognition only for opposite-sex couples, because the family structure of two committed biological parents-one man and one woman-is the optimal partnership for raising children. See, e.g., Citizens for Equal Protection, 455 F.3d at 867-68.
Proponents argue that Proposition 8, defining marriage as the union of one man and one woman, preserves the fundamental and historical purposes of marriage. They argue that, if the definition of marriage between a man and a woman is changed, it would fundamentally redefine the term from its original and historical procreative purpose. This shift in purpose would weaken society's perception of the importance of entering into marriage to have children, which would increase the likelihood that couples would choose to cohabitate rather than to get married. They also argue that irresponsible procreation, by accident or willfully in a cohabitation relationship, will result in less stable circumstances for children and that same-sex couples do not present this threat of irresponsible procreation. They argue that, in the case of unintended pregnancies, the question is not whether the child will be raised by two opposite-sex parents, but rather whether it will be raised, on the one hand by two parents, or on the other hand by its mother alone (often with the assistance of the state). "Proposition 8 seeks to channel potentially procreative conduct into relationships where that conduct is likely to further, rather than harm, society's interest in responsible procreation and childrearing."
Proponents also argue the "optimal parenting" rationale serves as a rational basis for Proposition 8. The optimal parenting rationale posits that Proposition 8 promotes the optimal setting for the responsible raising and care of children-by their biological parents in a stable marriage relationship. Proponents offer many judicial decisions and secondary authorities supporting both rationales.
In sum, Proponents argue that Proposition 8 is rationally related to legitimate governmental interests.
The first requirement of rational basis review is that there must be some conceivable legitimate governmental interest for the measure at issue.*fn30
The California Supreme Court indicated that responsible procreation is a legitimate governmental interest:
Whether or not the state's interest in encouraging responsible procreation properly can be viewed as a reasonably conceivable justification for the statutory limitation of marriage to a man and a woman for purposes of the rational basis equal protection standard, this interest clearly does not provide an appropriate basis for defining or limiting the scope of the constitutional right to marry. . . . [A]lthough the state undeniably has a legitimate interest in promoting "responsible procreation," that interest cannot be viewed as a valid basis for defining or limiting the class of persons who may claim the protection of the fundamental constitutional right to marry.
In re Marriage Cases, 183 P.3d 384, 432 (Cal. 2008) (emphasis added), superseded by constitutional amendment as stated in Strauss, 207 P.3d 48.
With regard to the optimal parenting rationale, the California Supreme Court stated the following about "the state's interest in fostering a favorable environment for the procreation and raising of children":
[A]lthough promoting and facilitating a stable environment for the procreation and raising of children is unquestionably one of the vitally important purposes underlying the institution of marriage and the constitutional right to marry, past cases make clear that this right is not confined to, or restrictively defined by, that purpose alone. As noted above, our past cases have recognized that the right to marry is the right to enter into a relationship that is the center of the personal affections that ennoble and enrich human life-a relationship that is at once the most socially productive and individually fulfilling relationship that one can enjoy in the course of a lifetime. The personal enrichment afforded by the right to marry may be obtained by a couple whether or not they choose to have children, and the right to marry never has been limited to those who plan or desire to have children. . . . [T]he state constitutional right to marry . . . cannot properly be defined by or limited to the state's interest in fostering a favorable environment for the procreation and raising of children.
Marriage Cases, 183 P.3d at 432 (citations and internal quotation marks omitted). Thus, the California Supreme Court discussed "the state's interest in fostering a favorable environment for the protection and raising of children" without using the "legitimate interest" and "for the purposes of the rational basis equal protection standard" language used to discuss "responsible procreation." See id. a.
Plaintiffs argue that the optimal parenting rationale cannot be a legitimate governmental interest because same-sex couples in domestic partnerships have all the substantive parenting rights opposite-sex couples in marriages enjoy. Additionally, California family law does not give any official preferences to opposite-sex parents.*fn31 Proposition 8 did not change this factual situation, because it "leav[es] undisturbed . . . a same-sex couple's state constitutional right to establish an officially recognized and protected family relationship and the guarantee of equal protection of the laws." Strauss, 207 P.3d at 61. "This state's current policies and conduct regarding homosexuality . . . recognize that gay individuals are fully capable of entering into the kind of loving and enduring committed relationships that may serve as the foundation of a family and of responsibly caring for and raising children." Marriage Cases, 183 P.3d at 428.
The parties argue about whether this analysis subjects Proposition 8 to heightened scrutiny rather than rational basis review. In my view, while Plaintiffs may give a correct accounting of California law, it does not necessarily follow that the optimal parenting rationale is an illegitimate govern-mental interest, because it contradicts existing laws on parenting and the family. For example, a posited reason offered by one lawmaking body after being rejected by another lawmaking body can "provide[ ] a conceivable basis" for a measure. FCC v. Beach Comm'ns, Inc., 508 U.S. 307, 318 (1993). In Beach Communications, the Supreme Court accepted a posited reason for a federal agency regulation, even though Congress had previously rejected that purpose and the regulation presented a conflict in the statutory scheme.*fn32 Id. Thus, even if California's legislature previously rejected the optimal parenting rationale in its parenting laws (and Proposition 8 is inconsistent with its statutory scheme), that does not prevent the people of California from adopting Proposition 8 under that rationale.
In Heller, the Supreme Court stated that "legislative choice is not subject to courtroom factfinding and may be based on rational speculation unsupported by evidence or empirical data." 509 U.S. at 320 (citations omitted). However, the Supreme Court went on to state that "even the standard of rationality as we so often have defined it must find some footing in the realities of the subject addressed by the legislation." Id. at 321.
Under rational basis review, the challenger has the burden to "negative every conceivable basis which might support" the measure. Id. at 320. In light of this burden, Plaintiffs have offered many secondary authorities to support their argument that the optimal parenting rationale cannot be a legitimate governmental interest. "Against [a] background of more than 100 peer-reviewed studies, the State of California could not reasonably accept as a true-or even debatable-statement of fact Proponents' view that only opposite-sex couples can create an 'ideal' childrearing environment." Thus, "[i]t is not an end that the State rationally could adopt as its own and therefore cannot sustain Proposition 8."
Although Proponents were not required to put on any evidence under rational basis review, they also produced evidence. They argue that their evidence shows that married biological parents are the optimal parenting structure. Further, they argue "Plaintiffs fail to cite to a single study comparing outcomes for the children of married biological parents and those of same-sex parents. Thus, Plaintiffs have failed to undermine, let alone remove 'from debate,' the studies show- ing that married biological parents provide the best structure for raising children."
After review, both sides offer evidence in support of their views on whether the optimal parenting rationale is a legitimate governmental interest. Both sides also offer evidence to undermine the evidence presented by their opponents. However, the standard only requires that the optimal parenting rationale be based on "rational speculation" about married biological parents being the best for children. Heller, 509 U.S. at 320. Considering "the question is at least debatable," id. at 326 (internal quotation marks omitted), the optimal parenting rationale could conceivably be a legitimate governmental interest.*fn33
Having a conceivable legitimate governmental interest is, alone, not sufficient for rational basis review. To survive rational basis review, a measure must also have a rational relationship to the posited legitimate governmental interest. In determining whether there is a rational relationship, one should bear in mind "the nature of rational-basis scrutiny, which is the most relaxed and tolerant form of judicial scru-tiny under the Equal Protection Clause."*fn34 Dallas, 490 U.S. at 26.
The Eighth Circuit credited the responsible procreation and optimal parenting rationales in Citizens for Equal Protection, where Nebraska had enacted a constitutional amendment prohibiting recognition of marriages by same-sex couples and other official same-sex relationships:
The State argues that the many laws defining marriage as the union of one man and one woman and extending a variety of benefits to married couples are rationally related to the government interest in "steering procreation into marriage." By affording legal recognition and a basket of rights and benefits to married heterosexual couples, such laws "encourage procreation to take place within the socially recognized unit that is best situated for raising children." . . . The argument is based in part on the traditional notion that two committed heterosexuals are the optimal partnership for raising children, which modern-day homosexual parents understandably decry. But it is also based on a "responsible procreation" theory that justifies conferring the inducements of marital recognition and benefits on opposite-sex couples, who can otherwise produce children by accident, but not on same-sex couples, who cannot. Whatever our personal views regarding this political and sociological debate, we cannot conclude that the State's justification "lacks a rational relationship to legitimate state interests."
455 F.3d at 867-68 (citations omitted).
The factual context in California is distinguishable from the one the Eighth Circuit faced in Nebraska. Unlike the Nebraska constitutional amendment, which prohibited the recognition of both marriages by same-sex couples and other same-sex relationships, Proposition 8 left California's existing domestic partnership laws intact. In California, same-sex couples in domestic partnerships still enjoy the same substantive rights and benefits as opposite-sex couples in marriages. Thus, it cannot be said that Proposition 8 "confer[s] the inducements of marital . . . benefits on opposite-sex couples . . . , but not on same-sex couples . . . ." See id. at 867. However, this distinction may not be dispositive, because the Eighth Circuit was considering both the substantive legal benefits as well as the designation of marriage.
That leaves the question of whether withdrawing from same-sex couples the right to access the designation of marriage, alone, rationally relates to the responsible procreation and optimal parenting rationales.
Regarding the responsible procreation rationale, Plaintiffs argue that Proponents suggest no reason to believe prohibiting same-sex couples from entering relationships designated "marriage" will make it more likely that opposite-sex couples in California will marry. Put differently, Plaintiffs argue that, because Proposition 8 does not bestow an honor on opposite-sex couples but instead withdraws an honor from same-sex couples, the responsible procreation rationale could be credited only if it is rational to believe that opposite-sex couples will be less likely to raise children in a marital family if the stature of marriage is also available to same-sex couples. Further, Plaintiffs argue that Proponents' failure to describe how Proposition 8 rationally relates to the responsible procreation rationale indicates that the rationale lacks the required "footing in the realities of the subject addressed by the legislation." Heller, 509 U.S. at 321.
In response, Proponents argue that, "[b]ecause only sexual relationships between men and woman can produce children, such relationships have the potential to further-or harm- this interest in a way that other types of relationships do not." Thus, "it follows that the commonsense distinction that our law has always drawn between opposite-sex couples, on the one hand, and all other types of relationships-including same-sex couples-on the other hand, plainly bears a rational relationship to the government interest in steering procreation into marriage."
However, Proposition 8 is not a "distinction that [California] law has always drawn," because it "establishes a new substantive state constitutional rule that became effective once Proposition 8 was approved by the voters." Strauss, 207 P.3d at 115. Also,
[n]one of the past cases discussing the right to marry -and identifying this right as one of the fundamental elements of personal autonomy and liberty protected by our Constitution-contains any suggestion that the constitutional right to marry is possessed only by individuals who are at risk of producing children accidentally, or implies that this constitutional right is not equally important for and guaranteed to responsible individuals who can be counted upon to take appropriate precautions in planning for parenthood.
Marriage Cases, 183 P.3d at 432. In this particular context, the fact that Proposition 8 established a new rule, instead of continuing a "distinction that [California] law has always drawn," weakens Proponents' argument that Proposition 8 "plainly bears a rational relationship" to the responsible procreation rationale.
Regarding the optimal parenting rationale, Plaintiffs argue that, because Proposition 8 does not change California's substantive laws governing childraising, procreation, or the family structure, Proposition 8 cannot be rationally related to the optimal parenting rationale. To channel more childrearing into families led by married biological parents, they argue that Proposition 8 would have had to change those laws somehow. Rather, Proposition 8 only singles out gays and lesbians, as a group, as inferior.
Proponents contend that this argument subjects Proposition 8 to heightened scrutiny review, and that the standard for rational basis review does not require the classification be substantially related to an important governmental interest. Instead, for rational basis review, the classification must only
(1) serve some conceivable governmental interest; (2) have a plausible reason for the enactment; (3) remain debatable; and (4) not be totally arbitrary. Their argument continues that, in California's unique context, Proposition 8 only deals with the designation of the term "marriage" but leaves undisturbed all of the other significant substantive aspects of recognized and protected family relationships. Proponents' theory only increases the likelihood that children are born and raised in a family structure of biological parents by encouraging such parents to marry; the designation of marriage for only that union would make it more likely that opposite-sex couples will want to enter into marriage and then subsequently raise their own biological offspring, rather than implying that any other union could not be good parents. Proponents claim this interest does not depend on any judgment about the relative parenting capabilities of opposite-sex and same-sex couples; it only confirms the instinctive, commonsense belief that married biological parents provide the optimal environment for raising children. Lastly, they argue there can be no requirement of narrow tailoring where there would be a perfect fit with the governmental interest and the law. If the state denied same-sex couples significant benefits under the law, the law would be more likely to fail equal protection by denying important government rights, thus increasing the burden of the test.
"[C]courts are compelled under rational-basis review to accept a legislature's generalizations even when there is an imperfect fit between means and ends." Heller, 509 U.S. at 321. Here, the people of California might have believed that withdrawing from same-sex couples the right to access the designation of marriage would, arguably, further the interests in promoting responsible procreation and optimal parenting. "The assumptions underlying these rationales may be erroneous, but the very fact that they are 'arguable' is sufficient, on rational-basis review, to 'immuniz[e]' the congressional choice from constitutional challenge." Beach Commc'ns, 508 U.S. at 320 (alteration in original).
] Plaintiffs argue that Proposition 8 could only advance the offered rationales through encouraging opposite-sex couples to marry, who otherwise would not marry because they disapprove of same-sex couples having the right of access to the designation of marriage and the stature that comes with the designation. Therefore, Proposition 8 impermissibly gives effect to those "private biases." See Palmore v. Sidoti, 466 U.S. 429, 433 (1984). However, Supreme Court precedent does not suggest that a measure is invalid under rational basis review simply because the means by which its purpose is accomplished rest on such biases.*fn35 Rather, precedent indicates that such biases invalidate a measure if they are the only conceivable ends for the measure. See, e.g., Romer, 517 U.S. at 535. Again, in determining whether there is a rational relationship, one must bear in mind that rational basis review "is the most relaxed and tolerant form of judicial scrutiny under the Equal Protection Clause." Dallas, 490 U.S. at 26. Thus, I cannot conclude that Proposition 8 is "wholly irrelevant" to any legitimate governmental interests. Heller, 509 U.S. at 324 (internal quotation marks omitted).
Given the presumption of validity accorded Proposition 8 for rational basis review, I am not convinced that Proposition 8 lacks a rational relationship to legitimate state interests. Precedent evidences extreme judicial restraint in applying rational basis review to equal protection cases.
Only by faithful adherence to this guiding principle of judicial review of legislation is it possible to preserve to the legislative branch its rightful independence and its ability to function. . . . [R]estraints on judicial review have added force where the legislature must necessarily engage in a process of line-drawing. Defining the class of persons subject . . . inevitably requires that some persons who have an almost equally strong claim to favored treatment be placed on different sides of the line, and the fact that the line might have been drawn differently at some points is a matter for legislative, rather than judicial, consideration.
Beach Commc'ns, 508 U.S. at 315-16 (alteration, citations, and internal quotation marks omitted). Thus, the judiciary faces a conspicuous limit on our judicial role in applying equal protection to legislative enactments, because
[t]he Court has held that the Fourteenth Amendment permits States a wide scope of discretion in enacting laws which affect some groups of citizens differently than others. The constitutional safeguard is offended only if classification rests on grounds wholly irrelevant to the achievement of the State's objective. State legislatures are presumed to have acted within their constitutional power despite the fact that, in practice, their laws result in some inequality. A statutory discrimination will not be set aside if any state of facts reasonably may be conceived to justify it.
McGowan v. Maryland, 366 U.S. 420, 425-26 (1961). A law must be upheld unless the government's judgment "is 'clearly wrong, a display of arbitrary power, [or] not an exercise of judgment.' " Mathews v. DeCastro, 429 U.S. 181, 185 (1976).
Applying rational basis review in these circumstances also requires such restraint. As the Eighth Circuit said, in Citizens for Equal Protection, 455 F.3d at 870:
In the nearly one hundred and fifty years since the Fourteenth Amendment was adopted, to our knowledge no Justice of the Supreme Court has suggested that a state statute or constitutional provision codifying the traditional definition of marriage violates the Equal Protection Clause or any other provision of the United States Constitution. Indeed, in Baker v. Nelson, . . . when faced with a Fourteenth Amendment challenge to a decision by the Supreme Court of Minnesota denying a marriage license to a same-sex couple, the United States Supreme Court dismissed "for want of a substantial federal question." There is good reason for this restraint.