So I’ve finally gotten around to reading the gay marriage decision, and, as (to my reading) it depends more on the Declaration than on the Constitution, July 4th weekend feels like an appropriate time to air my thoughts thereon.

I will admit, when I first heard about the decision I was torn between being very happy about the outcome and not being thrilled about what I understood to be the reasoning. I am not a big fan of natural rights reasoning, not a big fan of substantive due process, not a big fan of sweeping principle-declaring decisions, and not a big fan of Justice Kennedy. I had figured the Court would come to a narrower conclusion that effectively nationalized same-sex marriage based on the Full Faith and Credit clause while still allowing states formally to define marriage as they individually chose. And, I admit, I hoped such a conclusion could garner a six-Justice majority rather than a majority of five, and thus produce something like a social consensus.

But after reading the decision, and re-reading Loving v. Virginia, the key precedent case, I have come to a more meditative conclusion. I understand the logic of Kennedy’s opinion, and see how it flows from the body of precedent – and how, while other decisions would surely have been more conservative, they would still have been innovative. But mostly I’ve tried to be a realist about what those other, alternative decisions would actually mean in practice, and how they would differ from what is likely to flow from Obergefell.

My (partial) defense of Kennedy’s opinion begins with the following thought experiment. Imagine that Loving had been decided the opposite way, upholding miscegenation statutes, and that, in response, an amendment to the Constitution had been passed with the following wording:

The family being the fundamental basis of society, the right to matrimony shall not be infringed.

The passage of this amendment would surely have overturned miscegenation statutes nationally – as it would have been intended to do. It would also have made it clear that prisoners, the mentally handicapped, the carriers of genetic diseases – that none of these can be denied access to matrimony. How, though, would it be applied today in the context of same-sex marriage? How should it be applied?

The answer hinges on the question of what marriage is. At the time of the passage of the amendment, it’s true, only a few would have argued that it encompassed same-sex unions. But in 2015 a great many people thought it did, and many states had come to express that view in their laws (whether prompted by the state-level judiciary or not). Once such a view is current, it becomes necessary for the Court to decide whether or not it is correct – because it is necessary to determine whether the definition of marriage restricting it to unions between men and women is, in fact, an infringement on a fundamental right. This is particularly the case when states have undertaken explicitly to define marriage as exclusively a male-female bond, and not merely done so implicitly.

That’s basically the situation the Court found itself in if it took the Loving precedent seriously. Loving clearly established the right to marry as fundamental, pre-political, and central to the Declaration of Independence’s concept of the “pursuit of happiness.” Note that there is nothing traditional about this idea. Traditionally, marriage was a matter better arranged by your parents than by you, and love was something you hoped would grow within and sustain happiness in marriage as opposed to marriage’s origin. Traditionally and cross-culturally, regulation or prohibition of exogamy has been more the rule than the exceptionLoving certainly didn’t invent the idea of the love match, but it did raise it to the level of Constitutional principle.

Assuming the Court did not want retrospectively to limit the scope of that earlier decision, Loving provides quite firm ground to stand on for rejecting most of the arguments against same-sex marriage, as well as the argument that marriage is traditionally a state-level matter raising no Federal issues. The Court did not have the luxury of dealing with abstractions. It had to deal with individual gay families demanding recognition, and individual states denying them that recognition, and claiming that denial is not an infringement on their rights because the people seeking to marry are themselves confused about what marriage essentially is.

My point, basically, is that the Court, assuming it did not want to limit Loving, was faced with a new question in 2015 presented by new facts. In 1967, nobody disputed that a marriage between a black man and a white woman was a marriage; the dispute was over whether such a marriage could be prohibited for reasons of the purported social good. The Court determined it could not, both because marriage was a fundamental right and because that purported social good (preserving the white race) was not a legitimate state end. In 2015, the Court had to opine on what marriage is in order to resolve whether two men or two women being denied a license to marry were being denied something they were due. It would certainly have been more conservative of the Court to say: we don’t profess to know what marriage is; the states seem to disagree about what marriage is; the debate about the meaning of marriage is relatively novel; therefore we decline to register an opinion other than to demand that various states respect each other’s decisions on the matter (the Full Faith and Credit approach). But that’s not the same as saying it is illegitimate for the Court to decide that it needs to have an answer to the question of what marriage is because there is a fundamental right at issue. Which, per Loving, there is.

And, stripping away the high-flown rhetoric, both about freedom and about the glories of marriage, that’s what Kennedy’s opinion for the Court holds.

What would have been different had the Court held differently? In practice, I suspect not much. Consider first three other possible routes to a similar substantive result. As noted above, the Court could have declined to say anything about marriage, but to require the states to respect each other’s decisions, as they do with differing laws on age of consent, degrees of consanguinity, and divorce. The result would be effective nationalization of same-sex marriage, the only difference being the requirement for some Americans to travel. Undoubtedly at some point in the future this would be deemed an unfair burden on those without the means to travel, and same-sex marriage would be formally nationalized.

Another alternative would have been to declare that sexual orientation is a “suspect classification” requiring more heightened scrutiny for exclusion. This would have been a somewhat awkward way into opening up marriage specifically, since nothing in earlier marriage law actually refers to sexual orientation, but it is at least plain that the intent of the various laws and state constitutional amendments defining marriage as a male-female bond that the intent is to exclude same-sex couples from marriage. The Court has, in the past, declined to define a sexual orientation as analogous to race in this way, and doing so could have far-reaching implications – but many of those implications are being reached anyway by a jurisprudence that declares discrimination against gay people to be “irrational” on its face.

Yet another alternative would have been to strike down traditional-marriage-preservation statutes on the basis of gender discrimination – which was the oldest argument in favor of same-sex marriage, and, not incidentally, the most telling, since the teleological arguments for the necessity of complementarity in marriage all derive from a conception of gender that values essential differences between male and female. Such a finding might also have far-reaching implications, but again, probably not very different from those we face now.

And what about an alternative world in which a five-Justice majority ruled that marriage was the province of the several states, and that radical innovations (like same-sex marriage), did not require recognition under the Full Faith and Credit clause? How long would such a decision last in the face of changing views across the nation? Not long at all, I should think.

Of course, it might be all to the good for the Court to have said: the law doesn’t say this – and to watch the people change it so that it does. I remain very proud of the New York legislature for doing its proper job and changing the law to say what they thought it ought to say.

But to say that the Court stole the people’s limelight is not the same as to say it became a tyrant. We have, for better or worse, gotten accustomed since Carolene Products to a Court that sees itself as the ultimate vindicator of individual rights, and we differ with each other mainly in terms of which rights we want to see vindicated. We the people could change that any time if we wanted to. We just don’t really want to.