In the past two weeks, there has been a landslide of new same-sex marriage decisions. In particular, the Arkansas Supreme Court as well as federal district courts in Oregon and Pennsylvania all concluded that bans against same-sex marriage are unconstitutional.
At this point, the outcome in these cases is not surprising–by all appearances the tide has turned and the arguments attempting to justify marriage bans have been exhausted. (At least at the level of the lower courts–this is not a prediction about how the Supreme Court will ultimately decide on state-level marriage bans.)
What is surprising is the utter lack of consistency in terms of the reasoning employed by these lower courts. As I have written about previously, before United States v. Windsor, there were multiple paths available through an equal protection analysis of same-sex marriage. Possible avenues included:
- Applying strict scrutiny because such laws implicated the fundamental right to marriage.
- Applying intermediate scrutiny because such laws discriminate on the basis of sexual orientation, while deeming sexual orientation a new, quasi-suspect classification.
- Applying intermediate scrutiny because such laws discriminate on the basis of gender, an established quasi-suspect classification.
- Applying so-called “heightened rational basis review,” and finding that such laws do not serve the interests they purport to.
- Applying traditional rational basis review, but concluding such laws advanced no legitimate state interest.
- Concluding that such laws are based in unconstitutional animus such that they fail under any level of scrutiny.
Even this lengthy list overly simplifies the menu of doctrinal choices available to courts addressing the marriage equality issue.
And what is fascinating is that all of these options remain available–and are being used by lower courts–despite the fact that the highest court in the nation has ruled on marriage equality.
The Decisions in Arkansas, Oregon and Pennsylvania
This lack of consolidation is manifest in the most recent spate of marriage equality decisions.
MARRIAGE AS FUNDAMENTAL RIGHT
For example, on the issue of whether state-level marriage bans implicate a fundamental right, the Arkansas court concluded that this was self-evident, and did not engage in a tortured analysis of whether the plaintiffs were seeking to access the traditional, recognized right of marriage, or claiming that there was a fundamental right specifically to same-sex marriage. Because the challenged laws implicated a fundamental right, strict scrutiny would apply–but the court instead executed the by-now familiar move of not actually applying this level of scrutiny, instead concluding that the law would fail under deferential rational basis review in any case.
By contrast, the Pennsylvania court engaged in a lengthy analysis of whether the fundamental right to marriage encompassed a right to same-sex marriage, and concluded that it did. But rather than applying strict scrutiny, the court concluded that this made the marriage laws per se unconstitutional, which is wrong as a matter of either due process or equal protection analysis.
SEXUAL ORIENTATION AS SUSPECT CLASSIFICATION
On this point, the Arkansas court performed a traditional suspect-classification analysis (looking at the relevance of the trait to ability; immutability; and political powerlessness) and concluded that sexual orientation was a quasi-suspect classification, such that intermediate scrutiny would be required for laws discriminating on that basis.
The Pennsylvania court made an interesting move following the lead of the 9th Circuit in the recent SmithKline decision, by declining to engage in a substantive suspect classification analysis, but concluding that sexual orientation classifications were subject to heightened scrutiny simply because the Windsor decision applied something like heightened scrutiny in its analysis. (The Oregon court flirted with this analysis but ultimately determined that there was insufficient guidance to commit to it.)
It would take some familiarity with equal protection jurisprudence and a lot more unpacking of this analysis to demonstrate how odd–if not unprecedented–this move is. In short, the level of scrutiny applied in any particular case is supposed to reflect substantive concerns about the type of discrimination at issue, but here, a level of scrutiny is assigned which then dictates the level of judicial concern. This is a truly a matter of the level-of-scrutiny-tail wagging the merits-dog.
MARRIAGE BANS AS GENDER DISCRIMINATION
Another key point of discord in these cases was whether marriage bans constitute discrimination on the basis of gender. As I have argued previously, of course they do. But yet again, there is no consensus among the lower courts.
The Oregon court was particularly adamant that marriage bans could not be considered discrimination on the basis of gender. This line of reasoning flies in the face of Loving v. Virginia, which held that a law “classifies” on the basis of race when it makes rights dependent on race–one racial group need not be treated more favorably than the other. The district court distinguished Loving precisely because it dealt with race rather than gender, but this is a distinction without a difference. The difference between race and gender classifications is the level of scrutiny each receives. There is no reason to think that the test for whether a law relies on a classification would change based on the nature of the classification.
The Pennsylvania court reached the same conclusion, although it confined its discussion of the matter to a footnote. But there are seeds of controversy even in its brief discussion. The court determined that marriage bans do not constitute gender discrimination because “the intentional discrimination occurring in this case has nothing to do with gender-based prejudice or stereotypes.”
This is wrong as a matter of fact because one of the primary justifications for bans against same-sex marriage is the supposed need for male and female role models in the ideal parenting configuration. This could not be a more blatant reliance on gender stereotypes. (I would also argue that there is obviously gender prejudice wrapped up in prejudice against sexual minorities, but this is a more involved argument).
This is also wrong as a matter of law because it verges on suggesting that, to make a claim of discrimination, one must demonstrate not only the presence of a facial classification, but also discriminatory or invidious intent behind the use of that classification. This would truly turn equal protection jurisprudence on its head.
In short, the same-sex marriage decisions are providing an unlikely (and perhaps dangerously under-examined) vehicle for an evolving understanding of gender discrimination. This is a complicated and important topic that I will dissect in a separate post.
Between the cases already in the appeal process and these new decisions, it is uncertain how the issue of state-level marriage bans will best be teed up for the Supreme Court, on a doctrinal level. The obvious candidate seems to be the doctrine of animus, following in the footsteps of Windsor, but it remains unclear what exactly animus is and how the Court identifies its presence. Further, the evidence around the issue of animus may be very different in any given state-level challenge–different from the evidence available in Windsor, and different from the evidence available in other state-level challenges. (Raising the question of whether the determination of animus is properly considered a question of fact or a question of law.)
Lower courts have been playing with a variety of other options–primarily the idea that marriage bans interfere with a fundamental right and the idea that sexual orientation is a quasi-suspect classification. Both of these conclusions would mandate the application of heightened scrutiny. But the Supreme Court has shown no inclination to travel down these potentially controversial paths.