Photo by Bill Mason on Unsplash

Bostock and Originalism

Mark Tushnet

On June 15, 2020, the Supreme Court decided Bostock v. Clayton County. Dividing 5-4, the Court held that the ban on employment discrimination “because of sex” in Title VII of the 1964 Civil Rights Act extended to discrimination against gays, lesbians, and transgender people. Remarkably, both Justice Neil Gorsuch for the majority and Justice Alito in dissent said that they were interpreting the statutory term in light of its meaning as of 1964. In doing so they brought to the fore serious questions about some prominent arguments for textualism in statutory interpretation and, by implication, originalism in constitutional interpretation.

An important argument for textualism and originalism is that those methods prevent judges from “enacting” their personal policy preferences in the guise of interpreting the words of statutes and the Constitution—or, to avoid the imputation of nefarious personal motivations, the methods are said to constrain judges quite substantially so that they do no more or less than what Congress or the Constitution’s drafters actually did.

Close observers of discussions of these methods have known for a long time that they come in different flavors, so to speak. Several figure prominently in what judges do in practice. That fact alone suggests that the methods cannot be all that constraining—because by choosing which flavor of textualism and originalism to use, a judge can effectively choose an outcome consistent with the judge’s policy preferences.

The most prominent academic version, endorsed in form by Justices Gorsuch and Alito, has been labeled “original public meaning” originalism. Here a judge asks what a reasonably well-informed member of the public would have understood the statutory or constitutional terms to mean at the time they were written down: What would such a person have understood the words “right to keep and bear arms” to mean in 1791? What would such a person have understood “because of sex” to mean in 1964?

One difficulty with original public meaning originalism can be called “quantitative.” When you try to figure out what a reasonable person would have understood the words to mean, you rapidly find out, almost always, that some reasonable people understood the words to mean one thing while other reasonable people understood them to mean something different. You can’t do a survey to find out which side had more adherents, because it’s too late.

Even if you could, you’d have to decide whether you should adopt the meaning understood by a 60% majority or that understood by a substantial minority. This appears to be true about whether the Second Amendment confers a right on individuals to bear arms. It might well be that a majority in 1791 understood “keep and bear arms” to refer to such a right, but it’s clear that a substantial group of people understood it to refer to using weapons in a state-organized militia.

This poses a real difficulty in constitutional interpretation: Suppose you could find out what the proportions were, and it turned out that today’s Congress had done something consistent with the meaning understood by a substantial minority—say, enacted substantial restrictions on gun ownership. Then add to the mix that, as originally understood, the Constitution was meant generally to endorse decisions made by democratically elected majorities. Does that principle add enough to allow the originalist judge to interpret the constitutional provision to allow the gun regulation—even though we’re assuming that in 1791 more people understood the term in a way that would make the regulation unconstitutional?

Some scholars avoid the quantitative problem by noting that it typically arises only when the statutory or constitutional terms are somewhat abstract—“freedom of speech,” “equal protection of the laws,” or “because of sex.” Rather than trying to count noses, proponents of this flavor of originalism argue that the abstract terms enact general principles whose content is to be worked out over time. Sometimes they say that the meaning is “constructed” bit by bit as legislators and judges work with the general principle—by modifying the statute or interpreting the constitutional provision case by case. Sometimes they say that the judges’ task in interpreting terms that enact general principles is to decide what the best contemporary understanding of those principles is.

In Bostock Justice Gorsuch says that “because of sex” is a general term. He argues that, as of 1964, people would have understood the term to mean that a decision that would be made one way if the employee were male and another were the employee female would have been made “because of sex.” Justice Alito would have compared the employer’s action when an employee was gay (defined as a category including only men) to its action when the employee was a lesbian (only women): The employer would fire both, and so, for Justice Alito, the action was not “because of sex.”

How do you explain Justice Gorsuch’s choice of what to compare? He says that Justice Alito’s way of asking the question sneaks a gender distinction—that between male gays and female lesbians—into the basis for the comparison, and that makes the employer act “because of sex.” But that’s only because there’s an underlying principle of gender equality that we today understand to make the gendered basis of the distinction between gays and lesbians irrelevant.

So far I’ve described a couple of flavors of originalism defended by academics and judges. There’s another one, though, which features quite prominently in public discussions of originalism—on blogs and in op-eds, for example. It is called “original expected applications” originalism. Here you ask, “What would people in 1791 or 1964 have done if they were presented—at those times—with the problem now at hand?” Justice Alito’s dissent offers a pretty straight-forward version of this: No one in 1964—pre-gay rights movement—would have said that discriminating against gays was discrimination because of sex. And it’s a common trope in other contexts: “How could the Eighth Amendment’s ban on cruel and unusual punishment be interpreted to make the death penalty unconstitutional when everyone in 1791 obviously thought that capital punishment was constitutionally permissible?”

What’s striking is that “original expected applications” originalism has almost no defenders in the legal academy—or, put another way, academic originalists have abandoned it almost completely. The reasons are complex but can be captured in the observation that “original expected applications” originalism has enormous difficulty in explaining why the Supreme Court’s decision in Brown v. Board of Education, finding racial segregation in public schools unconstitutional, was correct; other flavors of originalism can readily accommodate that outcome.

Bostock is a good case about originalism, then, because it shows how the existence of different flavors of originalism undermines the argument that textualism and originalism strongly constrain judges.


Mark Tushnet is William Nelson Cromwell Professor of Law at Harvard Law School. His previous books include Why the Constitution Matters and In the Balance: Law and Politics on the Roberts Court.


Further Reading:

Recent Posts

All Blogs

Categories