In today’s column, the first in a two-part series, we begin to analyze and assess an important decision handed down last week by the United States Court of Appeals for the Ninth Circuit concerning discrimination against would-be jurors who happen to be gay or lesbian. In SmithKline Beecham Corp. v. Abbott Laboratories, a unanimous three-judge panel ruled that it violates the Fourteenth Amendment’s Equal Protection Clause for a lawyer to “strike” (that is, remove) individuals from a jury panel on account of their sexual orientation. As one of us explained more fully in an earlier column previewing the Ninth Circuit oral argument last fall, the antitrust lawsuit involved HIV medications, and an attorney for one of the companies (Abbott) exercised a so-called “peremptory strike” (also known as a “peremptory challenge”)—effectively removing an individual from inclusion in the jury—because the would-be juror was “or appears to be, could be, homosexual.” Peremptory challenges allow each side of a case to strike an equal number of would-be jurors for no supportable reason, solely because of a lawyer’s hunches or intuitions about how a particular person might behave and decide as a juror.
In holding that judicial acceptance of Abbott’s peremptory challenge would violate the Constitution, the Ninth Circuit opinion, authored by Judge Reinhardt, made a number of analytic moves. First, the panel had to determine whether an earlier three-judge Ninth Circuit panel ruling from 2008, Witt v. Department of the Air Force—which held that governmental actions discriminating on the basis of sexual orientation need only satisfy the lowest, most deferential, “rational basis” standard in order to be upheld under the Equal Protection Clause—is still good law. If Witt’s teaching that sexual orientation discrimination is not, as a general matter, subject to any kind of beefed-up constitutional scrutiny is still good law, it would be binding on the SmithKline panel judges, and the panel would have been required to come out the other way in SmithKline, since the Supreme Court has stated that peremptories may be used to remove individuals who are members of a class that is protected only by rational-basis review.
The SmithKline panel instead decided that Witt’s rational-basis-review approach is inconsistent with—and thus no longer binding because of—the Supreme Court’s ruling in United States v. Windsor, the case decided last summer that struck down the Defense of Marriage Act (DOMA), the federal law denying recognition to same-sex marriages. The Ninth Circuit found Witt to be inconsistent with Windsor because the Court in Windsor, whether the Justices admitted it or not, effectively applied “heightened scrutiny” (rather than rational-basis review) in holding that the DOMA violated equal protection guarantees. The Ninth Circuit’s reading of Windsor was crucial to getting around Witt. In this vein, the Ninth Circuit had to (and did) decide not only that Windsor applied heightened scrutiny to DOMA, but also that Windsor is not limited by the facts or context of that case, and instead stands for the legal proposition that heightened scrutiny now applies to all government actions that discriminate on the basis of sexual orientation. Based on that conclusion, the Ninth Circuit went on to determine the result of heightened-scrutiny review of sexual orientation-based peremptory challenges under the Fourteenth Amendment.
In the space below, we focus on the beginning of Judge Reinhardt’s analysis and examine the Ninth Circuit’s initial moves—the notion that subsequent Supreme Court authority could, by implication, reverse earlier clear Ninth Circuit precedent, the determination that Windsor is best read as applying heightened scrutiny to DOMA, and the contention that Windsor effectively requires heightened scrutiny to be applied to all government discrimination on the basis of sexual orientation. In Part Two of this series, we examine some of the later moves that the Ninth Circuit made—including the application of heightened scrutiny in the peremptory challenge context—and also explore where the case may go from here.
Was the Ninth Circuit Right That Supreme Court Authority Can Implicitly Erase Earlier Circuit Precedent?
No one would deny that a Supreme Court ruling that directly contradicts a Circuit Court opinion, whether the Circuit opinion is recent or old, would effectively eliminate the precedential authority of the lower court’s ruling. But the issue in SmithKline is a little different. The Ninth Circuit conceded that Windsor (the recent Supreme Court case) doesn’t directly contradict Witt (the older Ninth Circuit ruling), but determined instead that the reasoning of Windsor is in tension with Witt—that the two cases seem to reflect different mindsets—so that if we had to predict how today’s Supreme Court would decide Witt, we might bet that the Court would reach a result that is different from that reached by the Ninth Circuit in Witt.
One possible problem with the Ninth Circuit’s getting around Witt by noting some analytic tension between it and Windsor’s reasoning is the Supreme Court’s own admonition that lower courts should not “underrule” older cases based on predictions about the direction in which the Supreme Court is headed. As the Court put the point in 1989 in Rodriguez de Quijas v. Shearson/American Express, Inc.: “If a precedent of this Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the [lower courts] should follow the case which directly controls, leaving to this Court the prerogative of overruling its own decisions.”
One might think that this principle prevents the Ninth Circuit from ignoring Witt in favor of the new direction that is reflected by Windsor. But, in fact, the Shearson/American Express principle may have no application in this setting, because Witt (the older case that is arguably on point) is not a Supreme Court case, but rather simply a prior three-judge panel case from the Ninth Circuit. It is not obvious why three-judge panels of a Circuit court should bind subsequent three-judge panels of the same court in any event. The best answer is probably the need for some stability so that persons within a Circuit can know “what the law is.” But these intra-court stability concerns are less weighty than the reasons that explain why Circuit courts have to obey older rulings from a superior court—the Supreme Court (or an en banc panel of the Circuit). “Vertical” hierarchy and obedience to precedents of higher courts implicate different concerns than “horizontal” stare decisis (whereby a court pays deference, even strict deference, to its own rulings). So the Ninth Circuit was correct to carefully examine Windsor to assess the level of tension between it and Witt. (If the Supreme Court had, in some prior case, clearly ruled that sexual orientation discrimination implicates only minimum-rationality review and not heightened review, then the Shearson/American Express principle might apply here. But the clear holding in Witt—that minimum rationality review applies—came from a three-judge Ninth Circuit panel, not the Supreme Court.)
Did the Court Properly Read Windsor as a Case Employing Heightened Scrutiny?
Judge Reinhardt concedes, as he must, that the Windsor opinion does not explicitly state the level of scrutiny that the Court is employing to strike down DOMA. Accordingly, he looks to three factors that Ninth Circuit precedent requires him to consider in order to determine whether the level of scrutiny utilized in Windsor was, in fact, more rigorous than the highly deferential, rational-basis standard of review. The three factors are: (1) whether the Windsor opinion considered post-hoc rationalizations for DOMA—hypothetical purposes that might conceivably justify the law—or instead focused only on the actual goals Congress relied upon in enacting the statute; (2) whether the Windsor opinion required that there be a “legitimate” state interest to “justify” the effect of the law; and (3) whether the cases that were cited in Windsor themselves applied rational-basis review or heightened scrutiny.
With regard to the first factor, Judge Reinhardt correctly concludes that Justice Kennedy’s opinion in Windsor examines the actual purpose of DOMA in considerable detail, and that an emphasis on the legislature’s actual purpose strongly suggests that some form of heightened scrutiny is being applied.
An analysis of the second factor—the requirement that there must be a legitimate state interest to justify the challenged law—is more complicated, however, because, in some circumstances, we think the insistence on “legitimate” purposes for upholding a law is consistent with an application of rational basis review. And in other contexts, doubts about the legitimacy of the state’s purpose may undermine the validity of the law, but they do so outside of the framework of rational basis review, intermediate level scrutiny, or strict scrutiny. For these reasons, the use of the word “legitimate” is not a signal of heightened scrutiny.
When courts focus on the state’s interest in equal protection cases, what differentiates rational basis review from intermediate-level scrutiny or strict scrutiny, as a formal doctrinal matter, is that the latter two standards require, respectively, an “important” or “compelling” state interest. Conversely, a modest or even marginal state interest can satisfy rational basis review. But under all three standards, the state’s interest must be “legitimate.” Suppose, for example, that Congress had adopted DOMA for the express purpose of complying with what Congress understood to be divine law condemning same-sex marriage. Obedience to religious requirements is not a legitimate state purpose. Even under rational basis review, a court cannot use this purpose as a post-hoc rationalization to sustain a law.
Another state interest that lacks legitimacy, and which is relied on in cases cited in Windsor, such as Texas v. Lawrence and Moreno v. Department of Agriculture, is the goal of degrading or demeaning a group because of bare animus toward its members. The nature and scope of this characterization of a purpose as “illegitimate” remain unclear. Certainly, the goal of demeaning and punishing drug dealers would not undermine the constitutionality of statutes criminalizing drug trafficking.
Moreover, and more directly relevant here, the conclusion that a law serves an illegitimate state interest does not trigger the application of heightened scrutiny so much as it does an automatic invalidation of the law without further review. Heightened scrutiny is often described as mechanism that enables courts to flesh out impermissible state purposes. If a court determines at the outset that a law serves illegitimate purposes through an independent analysis, there is no reason to “flesh out” what the court already knows.
Indeed, if we examine this section of Judge Reinhardt’s opinion more closely, it appears that his emphasis on an inquiry into the legitimacy of a law does not really pertain to questions about whether a law serves impermissible purposes. Instead, Judge Reinhardt seems primarily concerned that the Windsor opinion discussed the harm caused by DOMA and the government’s need to justify such consequences if the law was to withstand the equal protection challenge directed against it. Reinhardt argues that this kind of implicit balancing of interests suggests more rigorous scrutiny than the highly deferential rational basis standard of review.
Indeed, it does. But does this necessarily mean that the Windsor opinion was applying heightened scrutiny to DOMA? As Judge Reinhardt acknowledges in his discussion of the third factor—an examination of the cases cited in Windsor to determine whether or not they themselves employed heightened scrutiny—sometimes the Supreme Court has applied rational basis review to a law, but the scrutiny it employed was more rigorous than the conventional leniency associated with a rational basis analysis. Moreno, dealing with discrimination against “hippie” households, was one such case. Romer v. Evans, involving discrimination against gays and lesbians, also cited in Windsor, is another, as is Cleburne v. Cleburne Living Center, Inc., (a case that wasn’t cited in Windsor) involving discrimination against the mentally retarded. One can certainly argue with some justification that the analysis in these hybrid rational basis cases arguably parallels the analysis in Windsor.
But it is less clear to us than it is to Judge Reinhardt that these “rational basis with teeth” cases should be characterized as applying heightened scrutiny, such that other laws burdening the same groups at issue in those cases in other settings would be subjected to rigorous review. As a general matter, we see no evidence that Cleburne or Moreno has changed the level of review applied to laws discriminating against the mentally retarded or “hippie” groups. True, these not-so-deferential rational-basis cases cannot easily be incorporated into the multi-tier system of review that the Court utilizes in equal protection cases. But at least for now, until the Supreme Court tells us otherwise, they are rational-basis cases.
As Judge Reinhardt acknowledges (and as we noted earlier), the Supreme Court has stated that “‘(p)arties may . . . exercise their peremptory challenges to remove from the venire any group or class of individuals normally subject to ‘rational basis’ review.’” Accordingly, one may argue with considerable force that as long the Court considers cases like Romer, Moreno, and Cleburne to have been decided under rational basis review, lower courts adjudicating equal protection claims against the use of peremptory challenges to strike gays, hippies, and the disabled from juries have no discretion to invalidate these jury selection decisions.
Even if Windsor Applied Heightened Scrutiny, Did It Do So Under Equal Protection Alone?
Finally, Judge Reinhardt’s opinion is as notable for what it omits as for what it says. There is virtually no mention in SmithKline of the federalism argument that makes up so much of Justice Kennedy’s opinion in Windsor. Clearly, Justice Kennedy’s majority writing in Windsor emphasizes the fact that the institution of marriage has been, and should be, a subject of state, rather than federal regulation. What is less clear in Windsor, however, is how this federalism analysis fits together with Justice Kennedy’s equal protection arguments. The uncertainty about precisely how the structural (federalism) and rights-based (equal protection) arguments fit together to support the Court’s holding in Windsor may have contributed to Judge Reinhardt’s decision to ignore the federalism aspect of Windsor altogether, and discuss the equal protection analysis in isolation.
Yet we think there may be a meaningful way to integrate the federalism and rights-based arguments in Windsor. There is sometimes a structural dimension to equal protection doctrine. In equal protection cases involving discrimination against non-citizens, for example, the Court sharply distinguishes between the level of review applicable to state laws discriminating against non-citizens, and the standard applicable to federal laws involving similar discrimination. Because the power to regulate immigration and naturalization is vested in the national government, state laws discriminating against non-citizens are more problematic and suspicious than discriminatory federal legislation. Accordingly, state laws discriminating against non-citizens receive much more rigorous review. Even if the federal government can permissibly regulate where resident aliens may live in the United States, a state has no business burdening their mobility.
A similar but mirror image analysis arguably applies in Windsor. Because marriage is quintessentially a matter of state sovereignty and control, it is federal laws discriminating against couples a state deems to be married that seem suspicious and problematic and warrant at least rational basis with teeth review. Under this analysis, however, it is harder to read Windsor as holding that all laws discriminating against gays and lesbians should receive heightened scrutiny, where there is no structural basis for distinguishing between the exercise of federal or state sovereignty in the government’s actions.
In Part Two of this series, which will appear on this site on February 14, we will continue our analysis of Judge Reinhardt’s approach, and also discuss the likelihood that the Supreme Court would be interested in this case.