In the days since the U.S. Supreme Court handed down its decision in Bostock v. Clayton County, many have heralded the fact that all three of the opinions issued in the case—the majority plus two dissents—took a textualist approach to interpreting Title VII. As a statutory interpretation teacher and scholar, I find this celebration curious. In my view, the fact that the majority and dissenting opinions took a textualist approach but reached different outcomes undermines one of the key justifications textualists offer for their interpretive approach—that it minimizes judicial discretion and ideological decisionmaking by pointing clearly to the correct reading of a statute. Indeed, the Bostock opinions highlight something my research has long shown to be true: that textualism is no more constraining of judicial discretion than are other interpretive approaches. If we look closely, the Bostock opinions provide other, perhaps surprising, lessons about textualism as well. Below is a short list:

1. Textualism does not necessarily lead to a single correct reading of a statute.
2. Textualists care about practical consequences and other extratextual considerations.
3. Textualism often involves some speculation about legislative intent or purpose.

Let us consider each lesson in turn:

1. Textualism does not necessarily lead to a single correct interpretation of a statute.

My own research has shown that the members of the Roberts Court regularly disagree about the plain or “ordinary” meaning of a statute’s text—and that majority and dissenting opinions in the same case often both claim that their conflicting constructions are clearly dictated by the statute’s text. The Bostock opinions illustrate how this can happen. At issue was a provision in Title VII that prohibits discrimination in hiring, firing, or other terms of employment “because of … sex.” Both the majority and dissenting opinions accepted the employers’ dictionary definition of “sex” to mean “status as either male or female [as] determined by reproductive biology.” But Justice Gorsuch’s majority opinion construed the ordinary meaning of the phrase “because of … sex” to clearly cover discrimination on the basis of sexual orientation or gender identity, whereas the dissenting opinions authored by Justices Alito and Kavanagh did not.

How is this possible? Well, Justice Gorsuch insisted that “it is impossible to discriminate against a person for being homosexual or transgender without discriminating against that individual based on sex” and offered numerous hypothetical examples of employer action to illustrate his view. (Full disclosure: This is a reading I agree with, and I joined an amicus brief in the case that said so.) Justice Alito’s dissenting opinion just as vociferously insisted that “the concept of discrimination because of ‘sex’ is different from discrimination because of ‘sexual orientation’ or ‘gender identity’” and criticized Justice Gorsuch’s hypotheticals as well as offered a few of its own. Finally, Justice Kavanagh’s dissent argued that Justice Gorsuch’s opinion impermissibly focused on the “literal” rather than the “ordinary” meaning of the phrase “discriminate because of sex” and insisted that the “ordinary meaning” of “discriminate because of sex” does not encompass discrimination because of sexual orientation.

Notably, Justice Gorsuch backed up his ordinary meaning argument with a long line of precedents applying Title VII. Justice Alito backed up his reading of the statute by pointing out that when Title VII was enacted in 1964 no “living American” thought that discrimination because of sex meant discrimination because of sexual orientation let alone gender identity, that the EEOC and lower courts for decades interpreted Title VII not to cover such discrimination, and that the United States military for years refused to hire homosexuals. And Justice Kavanagh backed up his ordinary meaning argument with references to common parlance, historical differences between the womens’ rights and gay rights movements, past Executive Branch practice, and other federal and state statutes that separately prohibit sexual orientation discrimination.

There are multiple subtle disagreements about the proper approach to textualism and the scope of ordinary meaning analysis at work in these conflicting textualist opinions. One is a temporal disagreement over “which ordinary meaning” courts should apply when interpreting statutes—the ordinary meaning of the statute as of the present day, or the ordinary meaning of the statute as it would be interpreted by a reasonable reader at the time of its enactment. Justice Alito insists that the latter is the proper frame of reference, but this is a question that remains uresolved in the Court’s statutory interpretation jurisprudence. Another is a disagreement about whether Title VII should be interpreted on its own terms, as Justice Gorsuch’s opinion does, or in light of the surrounding legal landscape, including executive orders and other federal and state statutes enacted by different legislators at different points in time, as Justice Kavanagh’s opinion advocates. (I have written about this latter, “whole code” approach in some detail.)

In the end, the majority opinion and two dissenting opinions all insisted that they were applying the statute’s “unmistakable” or “perfectly clear” ordinary meaning—while reaching opposing results. It is, on the one hand, tempting to view conservative Justice Gorsuch’s and Justice Roberts’ votes in favor of the homosexual and transgender plaintiffs in Bostock as evidence that textualism constrains judges and promotes non-ideological decisionmaking. But viewed from another angle, the fact that the majority and dissenting opinions reached such different outcomes—while relying on the same dictionary definitions and applying what each considered to be the statute’s “ordinary meaning”—highlights the fact that textualism leaves judges with significant discretion to determine just what a statute means.

2. Textualists care about practical consequences.

Textualism purports to focus on the statute’s text and to ignore extratextual considerations such as a statute’s purpose, congressional intent, or the outcome a particular interpretation will produce. And yet, my own and other scholars’ research has shown that the Court—including ardent textualists such as Justice Scalia—regularly pay attention to, and construe statutes based on, the practical consequences that will flow from a particular interpretation. The Bostock opinions highlight this tendency as well. Justice Alito’s dissenting opinion is explicit and transparent about the fact that the interpretation of Title VII adopted by the majority will have “far-reaching consequences” and it faults the majority’s failure to acknowledge these consequences as “irresponsible.” Indeed, Justice Alito is so troubled by the practical consequences of the majority’s ruling that his opinion spends eleven pages discussing them, and another sixteen pages listing other statutes whose meaning he fears will be altered by the Court’s ruling.

In addition, it is worth noting that all three of the Bostock opinions backed up their assertions about the statute’s ordinary meaning by citing extratextual interpretive aids, including prior judicial interpretations of Title VII, past practice by government agencies, employers, and the executive branch, the historical treatment of homosexuality as a mental disorder in the DSM, differences between the women’s rights and gay rights movements, and the terminology used by unrelated state and federal statutes. Thus, Justice Gorsuch’s claim that, “When the express terms of a statute give us one answer and extratextual considerations suggest another, it’s no contest” is a bit of a chimera: The reality is that all of the opinions in the case employed extratextual considerations alongside textual arguments, as is common on the Roberts Court.

3. Textualism often involves some speculation about legislative intent or purpose.

Textualism presents itself as a formalist interpretive method that focuses on the text of the statute in part because only the enacted text follows the bicameralism and presentment requirements prescribed in Article I, Section 7 of the Constitution. Indeed, textualists reject legislative history as an interpretive resource in part because it has not gone through this formal enactment process. Relatedly, textualists are skeptical of arguments based on legislative intent—arguing that it is the words Congress has enacted, rather than the intentions of its legislators, that govern statutory meaning. And they insist that, in any event, it is the enacted words of a statute, rather than comments made by a subset of the legislative body, that constitute the best evidence of Congress’s intent. It is noteworthy, then, that both of the textualist dissenting opinions in Bostock—Justice Alito’s and Justice Kavanagh’s—invoked a form of legislative history that neither survived bicameralism and presentment nor provides any clear indication of Congress’s intent: rejected legislative proposals. That is, both dissenting opinions pointed to the fact that Congress, specifically the House of Representatives, has over the years considered legislative proposals to add “sexual orientation” to Title VII’s list of prohibited bases for discrimination—but that none of those proposals, to date, has become law.

The implication is that Congress must not think Title VII’s current language covers “sexual orientation” discrimination, or it would not seek to add such a prohibition to the statute. Failed legislative proposals obviously have no formal claim to authority, having survived neither bicameralism nor presentment. Moreover, inferring meaning from such rejected proposals is a highly speculative endeavor—one that requires judicial conjecture and presumptions about Congress’s reasons for abandoning a proposal and ignores the fact that different legislators may have had different reasons for declining to vote for the proposal. Even if Justices Alito and Kavanagh were citing the rejected legislative proposals only as evidence of English language usage in the United States (as Justice Kavanagh claims he was), that argument rests on an assumption that Congress sought to add sexual orientation discrimination to Title VII because its members thought the ordinary meaning of the term “sex” did not cover such discrimination.

Further, Justice Alito’s dissenting opinion, in refuting arguments raised by the parties and by lower courts, openly discusses legislative intent and purpose. His opinion notes, for example, the legislative history of how “sex” came to be added to Title VII (as a killer amendment by civil rights opponent Howard Smith). It also distinguishes cases holding that Title VII prohibits discrimination on the basis of interracial marriage on the grounds that bans on interracial marriage were integrally connected to discrimination on the basis of race and a means by which hierarchical distinctions between the black and white races were maintained. Both of these arguments use history and background circumstances to speculate about Congress’s intended scope and purpose in enacting the prohibition against sex discrimination—in ways that go beyond straightforward textual analysis.

In the end, the Court’s trio of textualist opinions in Bostock have much to teach us about what textualism means on the modern Court. And while textualists can certainly celebrate that all three opinions “flew a textualist flag” as Justice Alito put it, they also should take note that textual analysis did not produce a single clear-cut reading of the statute—and that all three opinions took into account extratextual considerations as well.

Anita S. Krishnakumar is the Mary C. Daly Professor of Law at St. John’s University School of Law.