Is Textualism at War with Statutory Precedent?

Article - Volume 102 - Issue 4 - Volume 102


Textualism seems to have a tense relationship with statutory precedent. Jurists and scholars, including prominent textualists, have repeatedly insisted that stare decisis is largely incompatible with textualism.[1] As Justice Scalia and Bryan Garner put it in their treatise on statutory interpretation, “[s]tare decisis . . . is not a part of textualism. It is an exception to textualism . . . born not of logic but of necessity.”[2] Relatedly, some jurists and scholars suggest that textualists are—or should be—more willing than other interpreters to overrule statutory precedents.[3]

Accordingly, when purportedly “textualist” opinions use precedent, that practice can invite charges that the Justices are not engaging in true textualism. This critique has, for example, been leveled against Bostock v. Clayton County (2020),[4] which involved whether discrimination against a gay, lesbian, or transgender individual qualifies as “discriminat[ion] . . . because of such individual’s . . . sex” under Title VII of the Civil Rights Act of 1964.[5] Writing for the Court, self-proclaimed textualist Justice Gorsuch[6] turned to precedent to determine that “[i]n the language of law, . . . Title VII’s ‘because of’ test incorporates the ‘“simple”’ and ‘traditional’ standard of but-for causation.”[7] Justice Gorsuch later emphasized that the Court’s reading of the statute—to bar the disparate treatment of gay, lesbian, or transgender employees—was more consistent with prior Title VII sex discrimination cases than that offered by the dissenting opinions.[8] Some commentators have argued that this use of precedent demonstrates that the Bostock majority opinion relied on “non-textualist tools” or was, at best, a bad or “halfway” version of textualism.[9]

This Article aims to reexamine the relationship between textualism and statutory precedent. The Article asserts, contrary to prevailing assumptions, that textualism is quite compatible with important uses of statutory precedent. Moreover, the Article contends, a better understanding of the relationship between textualism and precedent sheds important light on descriptive and normative debates over the interpretive enterprise.

To make sense of all this, it is important to break down the analysis in two respects. First, the Article offers a typology of statutory precedent, demonstrating that case law can be and is used in markedly different respects in statutory analysis. Notably, this typology should be useful to interpretive theorists, whether or not they accept textualism. Second, the Article argues that, for textualists, these different types of precedent should matter at distinct stages of the statutory analysis.

So let’s begin with the typology. I will start first with a category that is crucially important but has thus far received only limited attention in the literature: Supreme Court precedent can help an interpreter determine the meaning of statutory terms and phrases.[10] I have already offered one illustration: Bostock’s reliance on precedent to determine that, “[i]n the language of law,” “because of” signals but-for causation.[11]

Second, statutory precedent can embody a specific holding—that a given statute applies (or does not apply) in a particular context. For example, in United Steelworkers of America v. Weber (1979),[12] the Supreme Court famously held that Title VII permits voluntary employer affirmative action programs.[13] This second category also encompasses the Court’s statutory implementation tests, such as the burden-shifting framework that the Court has established for some discrimination claims.[14] The Supreme Court has stated that a super-strong version of stare decisis applies to such statutory precedents.[15] This second type of precedent—and particularly the super-strong stare decisis approach—has been the focus of most commentary on statutory precedent.[16]

Third, a judge may seek to ensure—or at least to doublecheck—that there is consistency (“fit”) between the Court’s holding in the present case and the larger array of prior decisions in the area.[17] Justice Gorsuch also turned to this third type of statutory precedent in Bostock. The Court suggested that its holding—that Title VII prohibits the disparate treatment of gay, lesbian, and transgender employees—was more persuasive because it fit nicely into the broader array of the Court’s Title VII cases, such as those involving sexual harassment.[18]

Of these three types of statutory precedent, which one(s) could a textualist use? I argue that it depends on the stage of the statutory inquiry. So now it is important to break down the analysis in another respect. For textualists, statutory analysis consists of at least two distinct stages. First, interpreters aim to determine whether a statute has a plain meaning.[19] If not, then, second, they turn to additional sources to try to discern the best understanding of the statute, in the context of the case.[20]

Much of the literature on textualism focuses on the first stage—determining whether a statute has a plain meaning. But interestingly, textualists often define the first stage by identifying what is not relevant to the inquiry (for example: purpose, most substantive canons, and legislative history).[21] It is also important to consider what is part of the plain meaning analysis. This Article shows that, in addition to surrounding text and statutory structure, textualists often turn to the first category of statutory precedent—case law defining the meaning of terms and phrases—to determine the plain meaning of a law. That is, Bostock’s “because of” analysis is no outlier. Prominent textualists on the Supreme Court—including Justices Scalia, Thomas, Alito, Kavanaugh, and Kennedy[22]—have all relied on precedent not only to discern the meaning of recognized legal terms of art but also to inform the understanding of seemingly ordinary terms or phrases; to resolve a potential conflict among competing definitions; and to conclusively determine the meaning of terms or phrases in a single provision.

Moreover, the Article argues, this use of statutory precedent can be defended on textualist principles. Many textualists treat the search for plain meaning as involving a distinctively legal inquiry—as reflected by their focus on the perspective of a hypothetical “reasonable reader.” Accordingly, these textualists properly turn to legal sources, such as judicial precedent, at the first stage of the statutory analysis.

The second and third categories of precedent, by contrast, become significant in the second stage of the statutory analysis. If a textualist interpreter concludes that the operative provision does not have a plain meaning, other factors may come into play. Although textualists still tend to avoid resorting to legislative history, they may turn to statutory purpose or substantive canons.[23] At this stage, textualists can also be guided by case law holding that a statute applies (or does not apply) in a particular context and by statutory implementation tests. Textualists may further consider whether a particular holding is more consistent (has a closer “fit”) with the broader body of case law in a given area. That is, at this second stage, textualists can seek to “make sense rather than nonsense out of the corpus juris.”[24]

This breakdown gives us some insight into Justice Gorsuch’s discussion of past cases in Bostock. The Court used precedent at the first stage to determine that “because of” signals but-for causation. The Court then built on that definition to conclude that “taken together,” the phrase “discriminat[ion] . . . because of such individual’s . . . sex” prevents an employer from “intentionally treat[ing] a person worse because of sex—such as by firing the person for actions or attributes it would tolerate in an individual of another sex.”[25] The Court found that this principle prevents an employer from terminating a male employee who is romantically attracted to men, or from dismissing a female employee after she announces her transition from male to female.[26] That was enough to resolve the case on textualist principles—and to do so at the first stage, based on the plain meaning of the law. That the Court’s decision was also consistent with prior holdings was, to be sure, a nice touch. But it was unnecessary to the analysis—akin to “extra icing on a cake already frosted.”[27]

This Article helps inform our understanding of various issues of interpretive theory. First, the typology offered by this Article should be useful in any examination of statutory precedent. Indeed, one of the Article’s central contributions is to explore the first category: the use of precedent in determining the meaning of statutory terms and phrases. Second, the Article shows that the tension between textualism and precedent has been exaggerated (or at least insufficiently understood); textualists often rely heavily on Supreme Court precedent in identifying a law’s plain meaning. Third, this Article forms part of my broader effort to develop an affirmative picture of textualism. In addition to emphasizing what does not belong in the textualist inquiry, textualists should offer an account of what does belong.

This Article also sheds light on debates over the interpretive inquiry. Scholars disagree over whether the “ordinary meaning” sought by textualists refers to lawyerly meaning or lay meaning.[28] This Article demonstrates that to determine the “ordinary meaning” of a federal statute, textualists (as well as non-textualists) often look to judicial precedent—relying on case law to make sense of statutory terms and phrases.[29] This use of precedent strongly suggests that many textualists view the statutory analysis as involving a distinctively legal inquiry.

Finally, and relatedly, the analysis here brings to the surface an important division within textualist theory. Some scholarship suggests that the search for plain meaning is and should be largely a linguistic inquiry.[30] But as this Article demonstrates, judicial precedents are part and parcel of discerning plain meaning—for many textualists as well as other interpreters. For such textualists, plain meaning is not equivalent to linguistic meaning. These textualists seek to determine the meaning of statutory terms and phrases “[i]n the language of law.”[31]

At the outset, I want to clarify a few points. First, this Article focuses on the Supreme Court. As many scholars have observed, precedent of any kind exerts less force on a court of last resort, which in contrast to courts lower down in a judicial hierarchy, has the power to depart from its own precedents.[32] Nevertheless, as scholars have also acknowledged, precedent does still matter, even in such a court of last resort.[33] This Article seeks to show how different types of statutory precedent may impact the Court’s work—in an age where textualism appears to be the dominant approach.[34]

Second, in keeping with other literature, this Article uses the term “precedent” to encompass the reasoning of a past judicial opinion, including the opinion’s definition of words or phrases.[35] Some readers may argue that the term “precedent” should apply only to the specific holding of a past case, and to nothing else. But many judicial opinions and scholarly articles treat “precedent” more capaciously as encompassing (at least) the reasoning of a past opinion.[36] Moreover, and importantly, much of the commentary to which this Article responds has a similarly broad conception of precedent—as illustrated by the criticisms of Bostock for using “precedent” to define the phrase “because of.”[37]

Finally, this Article continues my effort to identify and explore divisions within interpretive theory, including specifically within textualism. In past work, I have described how textualism is split between a stricter, more formal version and a more relaxed, flexible version.[38] This Article’s analysis of statutory precedent highlights a separate divide: a dispute between those who treat the determination of meaning as a principally linguistic exercise and those who view it as involving legal and normative judgments. This latter division likely cuts across different interpretive methods—and may prove to be one of the most significant debates going forward.

The Article proceeds as follows. Part I explores the use of judicial precedent in identifying the plain meaning of a law. That Part further examines how this practice can be justified on textualist assumptions. Part II considers textualists’ approach to the second and third categories of statutory precedent and explores how the precedential force of a decision may depend on the category. Part III examines some implications of this argument, including for assumptions about ordinary meaning and the nature of the interpretive enterprise. The Article argues that the use of judicial precedent underscores how the interpretive inquiry can be seen as a largely legal and normative endeavor.

Chart: Typology of Precedent and Stages of Statutory Analysis
Category of Precedent Which Stage?
Category One:

Source of Meaning

First Stage: Determining Plain Meaning

Supreme Court precedent helps to define the meaning of statutory terms and phrases (and thus serves as one legal source for determining plain meaning).

Category Two:

and Implementation Tests

Second Stage: No Plain Meaning

When there is no plain meaning, a judge may turn to Supreme Court holdings (whether a statute applies or does not apply to a given factual scenario) and to statutory implementation tests.

Category Three:


Second Stage: No Plain Meaning

When there is no plain meaning, a judge may seek to ensure consistency (fit) between the present case and the broader array of precedents in the relevant area of law.

I. Precedent as a Means of Determining the Plain Meaning of a Law

For textualists, statutory analysis consists of at least two distinct stages. The first is determining whether a statute has a plain meaning.[39] Although textualists often focus on what is not relevant to the plain meaning inquiry,[40] it is important to consider what is part of the analysis. Modern textualists insist that they are not “literalists.”[41] They do not simply look at the “four corners” of a written document to divine statutory meaning.[42] So what can they look at? There seems to be broad agreement that textualists may legitimately look to the text and structure surrounding the operative provision at issue as well as dictionary definitions.[43] This Article highlights another common source of statutory meaning: Supreme Court precedent.

To be sure, the fact that many textualists use precedent in determining the meaning of statutory terms and phrases does not guarantee that such use is consistent with “textualism.” Indeed, as discussed below and in Part III, some theorists are likely to dispute the propriety of such use of precedent. But the fact that many prominent textualists use a particular method—and, indeed, use that method frequently and almost reflexively—gives us reason to at least explore whether the method can be seen as consistent with textualism. After surveying textualists’ use of precedent to define the meaning of terms and phrases, this Article argues that the use of this first category of statutory precedent can indeed be defended on textualist assumptions.

A. Statutory Precedent as a Source of Plain Meaning

Prominent textualists on the Supreme Court have repeatedly turned to precedent in determining the meaning of statutory provisions. Textualists use precedent not only to guide the analysis of recognized legal terms of art but also to inform the meaning of seemingly ordinary terms or phrases; to settle a potential conflict among competing definitions; and to resolve the meaning of terms or phrases in a single provision. In short, statutory precedent is a crucial part of determining the plain meaning of a law.

1. Not Just (Apparent) Legal Terms of Art

 In one respect, the use of precedent to determine statutory meaning may come as no surprise. Textualists, like other interpreters, acknowledge that judges should turn to common law cases to figure out the meaning of legal terms of art, such as “actual fraud.”[44] But the use of precedent does not stop with (what many of us would call) legal terms of art. Nor does it stop with the common law.

In fact, interpreters, including textualists, turn to the Supreme Court’s own precedents to make sense of seemingly ordinary terms and phrases, such as “or,”[45] “any,”[46] “now,”[47] “because of,”[48] and “in connection with.”[49] Thus, Justice Thomas’s opinion for the Court in Carcieri v. Salazar (2009)[50] turned to precedent to determine that “now” in the Indian Reorganization Act of 1934 referred to the time of statutory enactment, rather than to the time at which the Department of Interior sought to take land into trust.[51] “This definition,” Justice Thomas observed, “is consistent with interpretations given to the word ‘now’ by this Court, both before and after passage of the IRA . . . .”[52] Indeed, precedent may show that a seemingly ordinary term is, in the statutory context, a term of art. In Rimini Street, Inc. v. Oracle USA, Inc. (2019),[53] Justice Kavanaugh’s opinion for a unanimous Court declared that, under the Court’s precedents, the term “costs” is “a term of art that generally does not include expert fees.”[54]

The cases involving “because of” offer a vivid example of the Court’s use of precedent to make sense of seemingly ordinary terms and phrases. Certainly, the phrase “because of” can be used in ordinary conversation. My seven-year-old daughter often claims that she has misbehaved “because of Brother.” But much like “now” or “costs,” when the phrase “because of” is placed in a federal statute, it may take on a more distinctively legal meaning. The concept of causation runs throughout our legal system—encompassing ideas such as proximate cause, but-for cause, or sole cause.[55]

In several cases, the Supreme Court found that “because” or “because of” in certain employment discrimination statutes signaled but-for causation. Justice Kennedy’s opinion in University of Texas Southwestern Medical Center v. Nassar (2013)[56] involved a retaliation claim under Title VII of the Civil Rights Act of 1964, which prohibits an employer from “discriminat[ing] against any of his employees . . . because he has opposed any practice made an unlawful employment practice” by the statute.[57] In Nassar, the plaintiff claimed that his employer retaliated after he complained of discrimination based on race and religion.[58] To determine the meaning of “because,” Justice Kennedy relied in large part on precedent, particularly a 2009 case involving the Age Discrimination in Employment Act of 1967 (ADEA), which prohibits an employer from “discriminat[ing] against any individual . . . because of such individual’s age.”[59] In that earlier case, Gross v. FBL Financial Services, Inc.,[60] Justice Thomas turned to dictionaries from 1933 and 1966 to conclude that “because of” meant “by reason of: on account of.”[61] He then found that, under the Court’s precedents, “by reason of”—and, thus, “because of”—required a showing of but-for causation.[62]

In Nassar, Justice Kennedy reasoned that, given the textual similarities between the ADEA and Title VII, “the proper conclusion here, as in Gross, is that Title VII retaliation claims require proof that the desire to retaliate was the but-for cause of the challenged employment action.”[63] Justice Kennedy stated that this but-for test was consistent with tort law principles that pre-dated the enactment of the 1964 Civil Rights Act, as well as earlier Title VII precedents.[64] The Nassar Court pointed, for example, to City of Los Angeles Department of Water & Power v. Manhart (1978),[65] where female employees brought a Title VII sex discrimination suit, challenging a requirement that they pay more into a pension fund than male employees (on the stated ground that women tend to live longer than men).[66] In Manhart, the Court declared that “[s]uch a practice does not pass the simple test of whether the evidence shows ‘treatment of a person in a manner which but for that person’s sex would be different.’”[67]

The Court’s opinion in Bostock built on these cases. Relying on Nassar and Gross, Justice Gorsuch declared: “[A]s this Court has previously explained, ‘the ordinary meaning of “because of” is “by reason of” or “on account of.”’”[68] “In the language of law,” Justice Gorsuch continued, “this means that Title VII’s ‘because of’ test incorporates the ‘simple’ and ‘traditional’ standard of but-for causation.”[69]

2. Precedent as a Way to Choose Among Contending Meanings

 Precedent may be used to select among different meanings of a term, such as when dictionaries offer competing definitions. Pierce v. Underwood (1988),[70] for example, involved the Equal Access to Justice Act, which directs a court to award “fees and other expenses” to a party that prevails in a case against the United States “unless the court finds that the position of the United States was substantially justified or that special circumstances make an award unjust.”[71] The Court had to determine the meaning of the phrase “substantially justified.”

In an opinion by Justice Scalia, the Court noted that “the word ‘substantial’ can have two quite different—indeed, almost contrary—connotations.”[72] Dictionary definitions suggested that “substantial” could mean either “[c]onsiderable in amount, value, or the like; large,” or “‘in substance or in the main,’—as, for example, in the statement, ‘What he said was substantially true.’”[73] Justice Scalia declared: “Depending upon which connotation one selects, ‘substantially justified’ is susceptible of interpretations ranging from” the plaintiff’s call for a “high standard”[74] to the government’s view that its legal arguments must simply have “some substance and a fair possibility of success.”[75]

The Court found that case law could help clarify this uncertainty.[76] Under the Court’s precedents addressing whether agency action was supported by “substantial evidence,” the standard did “not mean a large or considerable amount of evidence, but rather ‘such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.’”[77] Building on these and other precedents,[78] the Court concluded that “as between the two commonly used connotations of the word ‘substantially,’ the one most naturally conveyed” is that the government’s position is “substantially justified” under the Equal Access to Justice Act when it has a “reasonable basis both in law and fact.”[79]

In Southwest Airlines Co. v. Saxon (2022),[80] the Court again relied on its own precedents to resolve a tension among potential statutory meanings. Latrice Saxon, an airline ramp supervisor who often engaged in loading and unloading cargo, brought a class action against Southwest Airlines, arguing that the airline had neglected to properly pay overtime.[81] Southwest contended that the case had to go to arbitration, but Saxon insisted that her case qualified for an exemption from the Federal Arbitration Act.[82] The Court had to determine whether an airline employee who loads cargo onto airplanes “belongs to a ‘class of workers engaged in foreign or interstate commerce’” that is exempt from arbitration under the Act.[83]

Southwest argued that the exemption did not apply: cargo loading “lack[ed] a necessary nexus to interstate commerce,”[84] given that “ramp-agent supervisors and ramp agents work only at the airport where they are based. Saxon worked solely at Chicago Midway International Airport.”[85] Saxon responded that her work was inherently bound up with transportation.[86] She supported her argument with several precedents that pre-dated the 1925 enactment of the Federal Arbitration Act.[87] “For decades,” Saxon insisted, “this Court had held that loading and unloading is commerce, because goods can’t cross state lines if they’re never loaded in the first place.”[88]

In an opinion by Justice Thomas, the Supreme Court unanimously agreed with the plaintiff Saxon.[89] The Court found that the pre-1925 precedents largely settled the matter: “[O]ur case law makes clear that airplane cargo loaders plainly do perform ‘activities within the flow of interstate commerce.’”[90] Relying on precedents interpreting the Hepburn Act of 1906 and the Federal Employers Liability Act of 1908, which held that loading and unloading goods for an interstate train shipment were “plain[ly] . . . so closely related to interstate transportation as to be practically a part of it,” the Saxon Court found it “equally plain” that airline cargo loaders “form ‘a class of workers engaged in foreign or interstate commerce.’”[91] This interpretation was supported by other precedents as well as the structure of the Act.[92]

Importantly, the Saxon Court made clear that this text, structure, and precedent all served to establish the plain meaning of the operative provision. For that reason, the Court rejected Southwest’s invitation to look at the purpose of the Federal Arbitration Act: to ensure the enforcement of arbitration agreements.[93] Justice Thomas declared: “[Section 1]’s plain text suffices to show that airplane cargo loaders are exempt from the FAA’s scope . . . [.] [W]e have no warrant to elevate vague invocations of statutory purpose over the words Congress chose.”[94]

3. Establishing the Meaning of a Single Provision

 As the previous sections demonstrate, the Justices, including the Court’s textualists, give significant weight to precedents that have defined the meaning of similar terms and phrases in other statutes and statutory provisions. One might expect the Court to give the most weight to past decisions that determined the meaning of a term or phrase in the very same provision. Indeed, that is the practice. The Supreme Court has repeatedly announced that it will read the same statutory provision in a uniform way across contexts (a rule that this Article will call the “one-meaning rule”).[95] Thus, for example, if a statute establishes both civil and criminal penalties for prohibited conduct, the Court will give the prohibition “the same construction” in a criminal prosecution and a civil enforcement action.[96]

Justice Scalia’s opinion for the Court in Clark v. Martinez (2005)[97] illustrates the principle. Clark was an immigration case involving the federal government’s authority to detain undocumented persons.[98] In general, under the Immigration and Nationality Act, once the federal government determines that an individual is subject to removal, “the Attorney General shall remove the alien from the United States within a period of 90 days.”[99] But the statute also provides:

An alien ordered removed who is inadmissible under section 1182 of this title, removable under section 1227(a)(1)(C), 1227(a)(2), or 1227(a)(4) of this title or who has been determined by the Attorney General to be a risk to the community or unlikely to comply with the order of removal, may be detained beyond the removal period and, if released, shall be subject to [the Attorney General’s supervision].[100]

Under the statute, then, three different groups of undocumented persons may be detained beyond the 90-day removal period: (1) those who were from the outset deemed inadmissible to the United States; (2) those who were lawfully admitted to the country but were later subject to removal; and (3) those who present a risk of future dangerousness.[101] The question before the Court in Clark was whether the phrase—“may be detained beyond the removal period”—gave the federal government broad discretion or only limited authority to detain the first group of individuals: those deemed inadmissible at the outset.[102]

Notably, the Court in Clark was not writing on a clean slate. The Court had previously interpreted this same language as applied to the second group: those who were lawfully admitted but later became subject to removal.[103] In Zadvydas v. Davis (2001),[104] Justice Breyer’s opinion for the Court found that the statutory phrase—“may be detained beyond the removal period”—did not permit unlimited detention.[105] That was in part because, Justice Breyer asserted, it would raise considerable constitutional difficulties to detain individuals—at least those who had initially been lawfully admitted—for an indefinite period.[106]

In Clark, the Court held that the provision must be interpreted in a uniform manner, thus making the Zadvydas Court’s view authoritative as to all three classes of undocumented persons.[107] That is, the federal government had only limited authority to detain any of these individuals. In reaching this conclusion, Justice Scalia insisted that the “[t]he operative language . . . , ‘may be detained beyond the removal period,’ applies without differentiation to all three categories of aliens that are its subject.”[108] So it did not matter that the constitutional concerns raised in Zadvydas may be less pressing in the context of individuals who had never been admitted to the country.[109] As illustrated by the cases reading criminal and civil prohibitions the same way, “[i]t is not at all unusual to give a statute’s ambiguous language a limiting construction called for by one of the statute’s applications . . . . The lowest common denominator, as it were, must govern.”[110] Otherwise, Justice Scalia admonished, the Court might “establish . . . the dangerous principle that judges can give the same statutory text different meanings in different cases.”[111]

There are several notable things about this decision. First, Justice Scalia treated as binding a precedent from which he himself had dissented. In Zadvydas, the dissenters insisted that the statutory language—“may be detained beyond the removal period”—was not ambiguous but clearly gave the Attorney General broad discretion to detain all three classes of undocumented persons.[112] Second, it is worth considering what the majority in Clark treated as precedential. The Clark majority viewed as binding both Zadvydas’s conclusion that the statutory language was ambiguous and its resolution of the ambiguity. Thus, once the Zadvydas Court had authoritatively interpreted that language, it was no longer ambiguous but binding in Clark.[113] Third, the Court’s textualists were not united in Clark; Justice Thomas believed the Court should either depart from or overrule Zadvydas.[114] But Justice Scalia’s opinion was supported by the longstanding legal rule that the very same statutory provision should mean the same thing across cases. Finally, and importantly, Justice Scalia’s decision in Clark vividly illustrates how a textualist jurist may be influenced or even bound by an arguably nontextual opinion—as long as that earlier opinion aimed to define the meaning of a statutory term or phrase.

B. Making Sense of the Use of Precedent to Define Plain Meaning

Textualists are not the only interpreters to use judicial precedent to define statutory meaning. That is in fact a longstanding practice by textualists and non-textualists alike.[115] But textualists’ use of this first category of precedent may surprise many readers. Indeed, as discussed, Justice Gorsuch’s reliance on precedent to define “because of” in Bostock led to accusations that the Court’s opinion was not true textualism.[116]

Although commentators have not clearly articulated how this use of precedent is “atextual,” there are likely a few concerns.[117] First, some scholars seem to rely on the widespread assumption that textualism is generally at odds with statutory precedent; on this view, any reference to precedent must be problematic. One of this Article’s primary goals is to distinguish among types of statutory precedent. The typology offered here should enable us to see more clearly that some uses of precedent may be more compatible with textualism than others. Second, some scholars may assume that the search for plain meaning is simply a linguistic exercise, so there is no reason to look at legal sources such as precedent. (I discuss this concern below and in Part III.) Finally, some criticism may rest on the presumption that textualists can never look at anything beyond the words on a page. Such a vision of textualism is impoverished; textualists have long acknowledged that other sources are relevant, such as the text and structure surrounding the operative provision as well as dictionary definitions.[118] Yet this critique does further underscore a point I emphasize in this Article: Textualists have for too long defined their method by emphasizing what cannot be considered. It is incumbent upon textualists going forward to articulate what evidentiary sources can be considered, particularly at the plain meaning stage.

This Article is part of my effort to do precisely that. As we have seen, many textualists have turned to the first category of precedent: relying on past cases in determining the meaning of statutory terms and phrases. I argue that this practice can be defended on textualist principles. First, statutory texts can be seen as embodying a special legal language, the content of which should be informed by legal sources such as precedent. That is how many textualists view statutes—as reflected in their reliance on a hypothetical “reasonable reader” to make sense of statutory language. Second, the use of the first category of statutory precedent, as well as legal rules such as that in Clark, comport with rule of law values and, relatedly, the interest of many textualists in cabining judicial discretion.

1. Reasonable Reader of a Federal Statute

 Let’s begin with the (relatively) uncontroversial case: terms of art.[119] Although there appears to be no agreed-upon definition of “terms of art,” the basic idea is that it can be challenging to comprehend certain technical terms without training in the relevant area—science, commercial trade, or other specialty.[120] For example, in McCaughn v. Hershey Chocolate Co. (1931),[121] the Hershey Company argued that chocolate was “food,” not “candy,” asserting that “candy” had a specialized industry definition limited to “confectionery, made principally of sugar or molasses, with or without the addition of coloring or flavoring matter.”[122] Thus, the company contended, chocolate should not be taxed at the higher “luxury” rate for “candy” under the Revenue Acts of 1918 and 1921.[123] Absent specialized knowledge of the candy industry, one would have difficulty comprehending the argument that chocolate is not “candy.”[124]

Legal terms of art function in a similar fashion. The importance of legal training is most evident when a term means entirely different things in legal parlance as opposed to ordinary speech. For example, early in my career, I told my (nonlawyer) mother that I was writing an article on “standing.” She replied: “Okay . . . . And your next paper will be on ‘sitting’ . . . and then ‘walking’?” As my mom’s response suggests, in ordinary parlance, “standing” refers to the upright position. But lawyers know that the term can also refer to one requirement for filing suit in court.[125] With legal training, particularly because these definitions of standing are so different, we can easily tell from context which type of “standing” a statute employs.[126]

Other legal terms of art, however, are less detached from their ordinary meaning. Consider the term “fraud.” In ordinary parlance, “fraud” can mean a trick or an imposter.[127] For example, if someone boasts—perhaps in an online dating profile—that he is a “25-year-old hunk,” and it turns out that he is in fact a few decades older than that (and not quite as handsome as the profile suggests), the boaster might be called a “fraud,” a “total fraud,” or perhaps an “actual fraud.” That lay concept is not entirely divorced from the legal concept of “fraud,” which also encompasses a misrepresentation of fact. But the legal concept contains additional elements—requiring not only such a misrepresentation but also (for example) that the misrepresentation induce reliance.[128]

In Field v. Mans (1995),[129] the Court understood the term “actual fraud” in the Bankruptcy Code to refer to the legal term of art, rather than to the ordinary meaning.[130] The Court did not explain that assumption,[131] but the Justices were unanimous on that point.[132] Nor did the litigants question that common law precedent should inform the meaning of the statutory term (although they disagreed on precisely which aspects of the common law the Court should adopt).[133] In short, no one argued that the ordinary meaning of “fraud”—as simply a misrepresentation of fact—should apply. And I suspect that most of us in the legal community—textualists and non-textualists alike—believe the Court was correct to turn to the legal concept and, accordingly, to look to common law cases. But why?

Let’s consider some possibilities. One potential explanation is that members of Congress, in enacting the Bankruptcy Code, subjectively opted for the legal concept. But few interpretive theories today focus on Congress’s subjective intentions; textualism in particular rejects them.[134] After all, it is doubtful that many members of Congress gave particular thought to this provision, much less to any differences between the ordinary and legal meaning of fraud. A second approach is to assume that we don’t actually know whether Congress has relied on a legal or an ordinary term, so we should turn to legislative history to figure it out. Although textualists are in general opposed to legislative history, some have expressed a willingness to look at such history to see if Congress has used a term of art.[135] But what if there is no legislative history on point? If (as seems likely) no one in Congress specifically considered whether to use the ordinary or legal meaning, the legislative history might not provide an answer. Yet a court still has to decide the case.

So let’s recall again that no one involved in the Field v. Mans litigation questioned that Congress was using a legal term. That is, no one argued that to prove fraud, the creditor simply needed to show some misrepresentation of fact. What accounts for this universal acceptance? I think that the answer is quite simple. As scholars of interpretive theory (including textualists) have repeatedly said, we must understand language in context.[136] The most basic context of a federal statute—indeed, so basic that it is too often overlooked—is that it is a legal document.

Several important features follow from the legal nature of a federal statute—again, seemingly so basic that they are often overlooked. First, in contrast to ordinary conversation, we do not expect federal statutes to use metaphor, irony, or innuendo.[137] Second, and relatedly, we do not expect federal statutes to contain puns or other humor.[138] Third, and perhaps most basic of all, we expect federal statutes to use legal concepts.[139] That is why it likely strikes many readers as odd even to ask the question whether the term “actual fraud”—in the Bankruptcy Code or elsewhere in the Statutes at Large—refers to the ordinary concept (any misrepresentation of fact) as opposed to the legal concept of fraud.

Once we focus on the legal nature of federal legislation, we also begin to see that in a federal statute, the line between ordinary terms and legal terms of art becomes fuzzier. Indeed, even terms and phrases that would strike many (lawyers and laypersons alike) as perfectly “ordinary” may take on a distinctively legal connotation as used in a federal statute. In federal legislation, “now” may no longer be the present time but the moment of the enactment of the statute.[140] The term “costs” may not mean any expenses, nor even any expenses related to litigation, but may exclude expert witness fees.[141] And the phrase “because of” may not be the ordinary concept invoked by my daughter to cast blame on her older brother. When “because of” is placed in a federal statute—for example, to prohibit an employer from discriminating against an employee “because of such individual’s race, color, religion, sex, or national origin”[142]—the phrase may take on a distinctively legal connotation.

That was a major aspect of the analyses in Nassar, Gross, and Bostock. Relying in part on dictionary definitions, the Court found that in the 1960s, when Title VII and the ADEA were enacted, the “ordinary meaning” of “because of” was “‘“[b]y reason of”’ or ‘“on account of.”’”[143] The Court then turned to precedent to specify the legal meaning of the phrase. As Justice Gorsuch put it in Bostock, “Title VII’s ‘because of’ test incorporates the . . . standard of but-for causation . . . [i]n the language of law.”[144] Ordinary terms and phrases may take on a distinctively legal hue when used in federal legislation.

Indeed, some terms and phrases may be more likely to strike us as legal concepts only if we have legal training. Consider the statutory language at issue in Saxon: “class of workers engaged in foreign or interstate commerce.”[145] A lay person could understand this phrase, without taking a course in law school. And such a person might quite reasonably conclude that a cargo loader, who never leaves Chicago’s Midway Airport, is not engaged in foreign or interstate commerce. But those of us with legal training know that there is a vast array of jurisprudence on foreign and interstate commerce.[146] With that background, we can better understand the Supreme Court’s conclusion in Saxon that the phrase had a distinctively legal meaning as used in the Federal Arbitration Act, such that pre-1925 precedent could inform the plain meaning of the law.[147]

Relatedly, precedent can help fill gaps left by dictionary definitions. As scholars (including textualists) have repeatedly observed, the dictionary is not a “fortress.”[148] Instead, dictionaries offer a range of possibilities as to how people might use a term or phrase.[149] To figure out which definition makes sense in the context of a federal statute, judges can look to legal sources, including judicial precedents. That is what the Court did in Nassar and Gross after examining dictionary definitions of “because of.” Likewise, Justice Scalia turned to precedent in Pierce v. Underwood to navigate between dictionary definitions of “substantial” in order to discern the legal meaning of “substantially justified” in the Equal Access to Justice Act.

Once we recognize statutory language as legal language, it makes a good deal of sense to look at legal sources to determine the meaning of terms and phrases. Such legal sources may, of course, include the surrounding statutory text and structure and related provisions.[150] Such sources also include judicial precedents.

Moreover, contrary to the concerns of some critics,[151] such reliance on precedent can be quite consistent with a search for the meaning of a statute at the time of enactment.[152] In Pierce and Saxon, the Court turned to Supreme Court precedent that predated the Equal Access to Justice Act and the Federal Arbitration Act. In Bostock, the Court relied on decisions (Nassar and Gross) that had themselves aimed to identify how “because” and “because of” were understood in the 1960s. (Notably, although scholars have written extensively about the Court’s causation analysis in Bostock,[153] neither dissenting opinion in that case called into question the Court’s adoption of a but-for test.) Supreme Court precedent was useful in those cases—not to invent a post-enactment meaning of a term or phrase, but to help specify the legal meaning of a term or phrase at the time of enactment. Case law guides judges by showing them how certain terms and phrases have been understood in “the language of law.”[154]

This notion of federal legislation as full of legal concepts accords with longstanding textualist principles. Many prominent textualists have argued that legal texts must be interpreted from the perspective of a hypothetical “reasonable reader.”[155] Justice Gorsuch, for example, has written that “the task in any case is to interpret and apply the law” from the standpoint of “a reasonable and reasonably well-informed citizen.”[156] In their treatise, Justice Scalia and Bryan Garner envision a highly sophisticated “reasonable reader”:

The interpretive approach we endorse is that of the “fair reading”: determining the application of a governing text to given facts on the basis of how a reasonable reader, fully competent in the language, would have understood the text at the time it was issued. The endeavor requires aptitude in language, sound judgment, the suppression of personal preferences regarding the outcome, and, with older texts, historical linguistic research.[157]

Likewise, John Manning asserts that “textualists interpret statutory language by asking how ‘a skilled, objectively reasonable user of words’ would have understood the statutory text, as applied to the problem before the judge.”[158] As both Manning and Justice Barrett have commented, “the statutory meaning derived by textualists” is thus “a construct.”[159]

This approach makes a good deal of sense if one understands that federal legislation includes legal concepts—and, indeed, that even seemingly ordinary terms and phrases, such as “costs” and “because of” (or “any” and “now”) may take on a distinctively legal meaning in a federal statute. A reasonable reader can be expected to look not only at dictionary definitions but also at the text and structure surrounding the operative provision at issue as well as judicial precedents construing similar terms. Indeed, this construct accords with the way that textualist Justices often conduct statutory analysis: only a hypothetical reasonable reader who is familiar with law can determine that “costs” is a term of art, that a cargo loader is engaged in interstate commerce, and that “because of” signals but-for causation.

To be sure, some scholars, including some self-described textualists, may argue that the search for “ordinary meaning” should be an empirical and linguistic exercise and thus resist this legalistic vision of legislation. I examine such concerns in Part III. My goal in this section is to show that reliance on statutory precedent in determining the meaning of terms and phrases can be defended on textualist principles. Textualists’ reasonable reader can be understood as recognizing that the “ordinary meaning” of statutory terms and phrases involves a good deal of law.

2. Legal Rules to Guide the Interpretation of Legal Documents

Once we view federal statutes as legal documents, we can also understand textualists’ (and other interpreters’) embrace of legal rules to govern interpretation. Such legal rules and presumptions are pervasive.[160] One example is the presumption that the same term means the same thing across a statute.[161] The one-meaning rule in Clark is an extension of that: If “[a] term appearing in several places in a statutory text is generally read the same way each time it appears,” there is “even stronger cause to construe a single formulation . . . the same way.”[162] Moreover, once we consider federal statutes as distinctively legal documents, it is not at all surprising that such legal rules and presumptions may not translate to ordinary conversation.

A few scholars have taken aim at the one-meaning rule, arguing that it makes little sense as a matter of linguistics. Jonathan Siegel contends that “as a purely linguistic matter, it is possible, though admittedly uncommon, for a single term in a single sentence to have multiple meanings.”[163] Siegel points to a syllepsis, such as in the sentence, “I ran ten miles on Monday and the Marathon Oil Company on Thursday.”[164] But as Siegel himself acknowledges, just as federal statutes tend not to use humor, they are also unlikely to employ a syllepsis.[165]

Ryan Doerfler goes further, insisting that “as a purely linguistic matter, multiple statutory meanings are not only possible but likely.”[166] Doerfler asserts, for example, that “the famous Uncle Sam poster that says ‘I Want YOU for U.S. Army,’” means “that Uncle Sam wants A for U.S. Army” if A reads the poster, and “that Uncle Sam wants B” for U.S. Army if B reads it.[167]

Assuming that “YOU” really does mean different things as a linguistic matter in this example,[168] there are good reasons to reject the view that the same federal statutory term should mean different things in different cases. When a statute such as Title VII provides that “an employer” may not discriminate because of certain traits,[169] it seems reasonable to apply that prohibition equally to all employers—rather than envision that different employers might read the prohibition in different ways. A uniform reading likewise seems appropriate when Congress declares in a single provision that undocumented individuals “may be detained beyond the removal period.”[170]

That is particularly true, given that Congress could have structured each statute differently. Congress could have created different classes of employers in Title VII.[171] Or Congress could have written separate detention rules for each class of undocumented persons—distinguishing those who were initially lawfully admitted from those who were not so admitted or who present a risk of future dangerousness. Interpreters respect Congress’s decision to craft a legal requirement in a single provision, as opposed to separate sections, by interpreting the same statutory language in the same way. Such respect for Congress’s structural choices accords with longstanding textualist assumptions.[172]

3. Guidelines and Rule of Law Values

Adherence to the first category of precedent seems likely to appeal to textualists who care about rule of law values, such as stability and predictability.[173] Relatedly, this first category should appeal to textualists who emphasize the importance of guiding and constraining judicial discretion (as many do).[174] Indeed, one central rule of law value is constraining those in positions of authority—including judges—by ensuring that such officials make decisions based on external factors, rather than their own personal preferences or ideology.[175]

Admittedly, precedent is never completely binding on a court of last resort. Such a court always has the option to depart from its prior decisions.[176] But as we have seen, statutory precedent on the meaning of terms and phrases can and does inform the analysis in later cases.

Of course, in order for the first category of precedent to serve these rule of law values, there must be guidelines for its use. Because jurists and scholars have not focused on this type of statutory precedent, they have not clearly articulated such guardrails. But one can see patterns in the existing case law. Drawing on this case law as well as textualist theory, I offer here some guiding principles.

Most importantly, there should be legal rules to guide interpreters on issues such as which statutory precedents should inform the meaning of terms and phrases. Given textualism’s general emphasis on the time of statutory enactment,[177] textualists should focus on two sets of judicial precedent. First, textualists should draw on cases that were issued before or around the time of the enactment of the relevant statutory provision. Such contemporaneous precedent can help inform the legal meaning of a term or phrase in the relevant era. Thus, Pierce and Saxon looked to precedents that predated the enactment of the Equal Access to Justice Act and the Federal Arbitration Act to capture the legal meaning of “substantially justified” and “class of workers engaged in foreign or interstate commerce.” Second, textualists also properly rely on precedents that themselves aimed to determine the meaning of terms and phrases at the time of enactment. Thus, the Bostock Court looked to Nassar and Gross, which sought to identify the meaning of “because” and “because of” in the 1960s, when Title VII and the ADEA became law.[178] Moreover, textualists should be drawn to legal rules, such as the one-meaning rule in Clark, that provide consistency and predictability in the interpretation of particular statutory provisions.

With such guardrails, the first category of statutory precedent can provide considerable guidance in future cases. Because of statutory precedent, judges have a better understanding of, for example, the legal meaning of “costs,” “substantially justified,” and “discrimination because of sex.” And the judicial precedents identifying the meaning of terms and phrases then influence subsequent interpretations.[179] Notably, Justice Alito’s dissenting opinion in Bostock argued that the decision would affect the interpretation of “[o]ver 100 other federal statutes” that also “prohibit discrimination because of sex.”[180] Justice Alito did not explain why Bostock was likely to have such an impact, but this Article’s discussion of the first type of statutory precedent offers an answer: Both textualist and non-textualist Justices are likely to be influenced by the meaning attributed to similar statutory language in prior decisions.

Indeed, textualists’ reliance on precedent to define statutory terms and phrases suggests how textualists may be influenced by past precedents that did not apply a strictly textualist approach. In Clark, for example, Justice Scalia announced as binding a decision from which he had dissented on textualist grounds.[181] As discussed further below, statutory precedents may thus exert a substantially greater influence, to the extent that those precedents engage with the statutory language.[182]

Finally, one can see how the use of the first category of precedent may cabin the (presumed) ideological leanings of judges—and thereby lead to unexpected results. For example, while commentators have criticized the recent tendency of a conservative Supreme Court majority to favor arbitration,[183] statutory precedent led a unanimous Court in Saxon—in an opinion authored by Justice Thomas—to reject an employer’s demand for arbitration. In Clark, Justice Scalia’s reliance on Zadvydas to issue a decision that protected a class of undocumented immigrants from unlimited detention was likely a surprise to many observers. Consider also in this regard Nassar and Gross, which found “because” and “because of” to signal but-for causation. In each case, the employee lost—in large part due to the but-for causation requirement.[184] It does not appear that many observers anticipated how that definition of “because of” would later be used in Bostock to lead to a monumental plaintiff employee victory.[185]

II. Textualists’ Use of Other Types of Statutory Precedent

Statutory precedent, it turns out, is an important component of the textualist inquiry. At the first stage of the analysis, textualists often rely on Supreme Court precedent in determining the plain meaning of a statutory provision. But that version of statutory precedent—case law to define the meaning of terms and phrases—is only one type. This Article also identifies two other categories. The second category includes the holdings of specific past cases (whether a statute applies or does not apply to a given factual scenario) as well as statutory implementation tests. The third category encompasses a judge’s effort to ensure consistency (“fit”) between the holding in the present case and the larger body of precedents in a given area.

This Article argues that textualists may properly rely on these latter two types of statutory precedent at the second stage of the interpretive inquiry—when the interpreter has determined that a federal statute lacks a plain meaning.[186] Just as a textualist may look to statutory purpose and substantive canons, textualists may be influenced by the Court’s holdings in prior cases. And a textualist may seek to ensure that a new holding fits more neatly within the broader legal framework.[187]

There are important implications of breaking down the analysis in this way. First, textualists properly rely on the second and third types of precedent only if a statute lacks a plain meaning. Conversely, if a textualist concludes at the first stage that a statute has a plain meaning, a textualist will naturally be skeptical of prior holdings and implementation tests that seem at odds with that meaning. Notably, as discussed below, that does not mean that a textualist should call for a reversal of such precedents, just that they warrant more scrutiny. Second, and crucially, this Article’s typology of precedent helps illuminate that the influence of a past precedent will depend in large part on the way in which the prior opinion was written. If a past precedent has defined a statutory term or phrase—thus creating the first category of statutory precedent—it will exert considerable influence on future textualist opinions. But if the statutory precedent has not so engaged with the text, the precedent may exert far less influence. This analysis has important lessons for understanding the precedential strength of one of the most-discussed cases in the interpretive literature: United Steelworkers of America v. Weber (1979).[188]

A. The Use of Precedent When There Is No Plain Meaning

When a statute lacks a plain meaning, textualist jurists have proven quite open to considering a range of materials. For example, in Robinson v. Shell Oil Co. (1997),[189] the Court examined whether the term “employee” in Title VII could encompass former employees, so that they could bring retaliation claims.[190] Writing for the Court, Justice Thomas concluded first that the statute contained no “plain and unambiguous meaning.”[191] Accordingly, the Court turned to purpose, concluding that the term “employees” included former employees, in large part because that was more consistent with the “primary purpose of antiretaliation provisions: Maintaining unfettered access to statutory remedial mechanisms.”[192] In United States v. Santos (2008),[193] Justice Scalia’s opinion first found that the federal money-laundering statute was ambiguous as to whether the term “proceeds” referred to “receipts” or “profits.”[194] Given the lack of a plain meaning, the Court applied a substantive canon—the rule of lenity—in favor of the defendant.[195]

Likewise, when a statute lacks a plain meaning, a textualist jurist may look to past judicial holdings and statutory implementation tests. Salman v. United States (2016)[196] offers an illustration. Maher Kara was an investment banker who passed along confidential information to his older brother, who then shared the information with Bassam Salman.[197] The issue in Salman was whether a tippee (Salman) could be held liable for insider trading, when the original tipper (Maher Kara) did not get any monetary benefit from sharing the information.[198]

Interestingly, in challenging his conviction, Salman argued that the “plain language” of the Securities and Exchange Act did not prohibit insider trading at all, much less impose liability on tippees[199]—an argument that, as the government pointed out, would “upend insider-trading law.”[200] But the Court did not accept Salman’s textual attack on insider trading. Instead, Justice Alito’s opinion for a unanimous Court led off by acknowledging that the Securities and Exchange Act, together with Rule 10b-5, “prohibit undisclosed trading on inside corporate information by individuals who are under a duty” not to “secretly us[e] such information for their personal advantage.”[201]

But neither the statutory text nor the regulation resolved the specific issue of tippee liability. For that, the Court turned to precedent.[202] A 1983 decision, Dirks v. SEC,[203] offered a test for tippee liability: “whether the insider personally will benefit, directly or indirectly, from his disclosure.”[204] One example, the Dirks Court advised, would be “when an insider makes a gift of confidential information to a trading relative or friend,” given that such a tip was similar to “trading by the insider himself followed by a gift of the profits to the recipient.”[205] In Salman, Justice Alito asserted that the test in Dirks, and that precedent’s discussion of gift giving, “resolve[d] [the] case” before the Court.[206] Because the tipper Maher presumptively benefitted from the “gift” of information to his brother, the tippee Salman could be held liable for insider trading.[207]

Just as a textualist may take account of past holdings and implementation tests, a textualist may seek to ensure consistency (“fit”) with the surrounding statutory precedents, when dealing with a provision that lacks a plain meaning. Notably, some scholars have argued that there was no plain meaning to guide the Supreme Court in Bostock.[208] On that view, the statutory prohibition on “discrimination . . . because of such individual’s . . . sex” did not clearly instruct the Court on how to view the disparate treatment of a gay, lesbian, or transgender employee.

For one who takes that position, it would have been entirely appropriate for an interpreter, including a textualist, to rely on the array of past cases involving Title VII. In Bostock, the Court pointed to several Title VII precedents.[209] For example, in Phillips v. Martin Marietta Corp. (1971),[210] the Court held that an employer engaged in sex discrimination when it was willing to hire men, but not women, with young children.[211] In City of Los Angeles Department of Water & Power v. Manhart (1978),[212] the Court likewise found sex discrimination when an employer required female employees to pay more into a pension fund than male employees.[213] As the Bostock Court observed, in those prior cases, the employers had tried to label the distinction at issue as something other than sex discrimination. In Phillips, the employer argued that it was making a distinction based on “motherhood,” not sex.[214] In Manhart, the employer asserted that its pension policy was motivated not by animosity toward women, just the simple fact that women tend to live longer than men.[215] Likewise, in Bostock, the employers argued that they were making distinctions on the basis of sexual orientation or gender identity, not sex. The Bostock Court reasoned: “[J]ust as labels and additional intentions or motivations didn’t make a difference in Manhart or Phillips, they cannot make a difference here.”[216]

The Bostock Court could have further relied on Price Waterhouse v. Hopkins (1989),[217] which established that Title VII bars sex stereotyping; in that case, an accounting firm denied a promotion to a woman who was deemed overly “aggressive.”[218] Along the same lines, scholars have argued, when an employer terminates (or otherwise penalizes) a gay, lesbian, or transgender employee, the employer is likely doing so on the ground that the employee is not conforming to traditional gender roles.[219]

My own view, as I have argued in past work, is that Bostock correctly held that there was a plain meaning in that case—that terminating a male employee who is romantically attracted to men, or dismissing a female employee after she announces her transition from male to female is “discriminat[ion] . . . because of such individual’s . . . sex.”[220] On that view, the Court’s discussion of cases such as Phillips and Manhart was “extra icing on a cake already frosted.”[221] But if the Bostock Court had found no plain meaning, it would have been entirely appropriate for the Court not simply to mention but to rely on this precedential landscape. Just as a textualist may consider statutory purpose and substantive canons to make sense of an unclear provision, a textualist may look to precedent—and seek to ensure that the decision in the present case is more consistent with the overall body of law in a particular area. Such reliance does not, as some have suggested, make an opinion “atextual.”[222] Instead, the textual interpreter turns to the second and third categories of precedent when the provision lacks a plain meaning.

B. What If There Is a Plain Meaning?

But what if a textualist decides that there is a plain meaning? More specifically, what if a textualist determines after the first stage of the analysis that a statute has a plain meaning that is difficult to reconcile with the second and third categories of precedent (past holdings and implementation tests, or the broader array of prior decisions)? I believe it would be challenging on textualist grounds to give super-strong stare decisis effect to such statutory precedents.

Consider the primary argument that the Supreme Court has offered in favor of super-strong stare decisis: Congress can override a judicial decision interpreting a federal statute. So, the argument goes, when Congress fails to overturn a statutory precedent, it has acquiesced in the Supreme Court’s earlier decision.[223]

Jurists and scholars of all interpretive stripes have raised important objections to this acquiescence argument—and more generally, to granting statutory precedents super-strong stare decisis effect.[224] For example, William Eskridge has forcefully argued that the acquiescence argument presumes that congressional “inaction” signals “approval”; yet we have no idea why a subsequent Congress did or did not enact a law.[225] Anita Krishnakumar has thoughtfully asserted that Congress may not be motivated to revise some statutory implementation tests, so the Supreme Court may need to do so.[226]

But for most textualists, the central problem with the acquiescence theory is that it relies on assumptions about the actions (or inactions) by the present-day Congress. The actions (or inactions) of subsequent legislatures tell us very little about the meaning of a statute enacted by an earlier Congress.[227]

That does not mean a textualist must ignore past holdings. As Caleb Nelson has recounted in impressive detail, historically, when a text-focused interpreter found fault with a prior statutory decision, the interpreter would seek to overturn the decision, unless there were reasons to preserve it, such as reliance interests.[228] Any statutory decision may create important reliance interests—among interested parties, government officials, or the general public, and that may be enough for a textualist jurist to conclude that the precedent should be retained.[229] Nevertheless, when a textualist determines that past holdings or implementation tests conflict with the plain meaning of the law, the textualist should at least look skeptically at these categories of statutory precedent.

Is this cause for concern? That normative question is difficult to answer in the abstract, given that for most observers, the value of stare decisis likely depends on the specific statutory holding(s) at issue. But I do want to address one common concern: Because textualism has long been associated with the conservative legal movement,[230] many observers assume that textualism could lead to a conservative revolution.[231] I argue, however, that textualist skepticism toward past decisions does not have a clear ideological valence. Indeed, textualism may offer opportunities to question statutory holdings and implementation tests that have long been criticized by progressives.

A few recent examples illustrate this point. In the wake of Bostock, Katie Eyer has argued that the but-for causation test offers a way to upend a good deal of Title VII case law that tends to undermine employment discrimination claims.[232] Deborah Widiss builds on Bostock’s textual analysis to contend that the longstanding McDonnell Douglas[233] framework for employment discrimination claims should be dismantled.[234] Along similar lines, scholars have argued that the limitations on civil rights claims under 42 U.S.C. § 1983, particularly the official immunity doctrines that shield government officials from personal liability, are difficult to reconcile with the text of that statute.[235]

My goal here is not to weigh in on whether these scholars are correct about the plain meaning of Title VII or Section 1983. Instead, I seek to point out that any attack on this jurisprudence confronts a heavy burden of stare decisis. The implementation test in McDonnell Douglas is fifty years old and has given birth to a web of subsidiary precedent.[236] Likewise, the Supreme Court’s qualified immunity jurisprudence dates back several decades.[237] Revisiting these precedents would seem to be a tall order, particularly to the extent that the Supreme Court accords super-strong stare decisis effect to its statutory holdings. But such reconsideration seems more plausible on textualist assumptions—if, as these scholars argue, these holdings and implementation tests are contrary to the plain meaning of the law.[238]

C. The Importance of Opinion Writing

One of the main goals of this Article is to distinguish among types of statutory precedent. It turns out that, in seeking the plain meaning of a provision, textualists often rely on precedents that define terms or phrases (a practice that, I argue, is defensible on textualist assumptions). Textualists properly turn to other precedents, such as past holdings and implementation tests, when a statute appears to have no plain meaning. This Article’s typology not only gives us a new way of understanding statutory precedent generally but also underscores the importance of opinion writing. The Justices can establish very different—and differently influential—precedents, depending on the way that they craft the opinion. United Steelworkers of America v. Weber (1979) provides a powerful example.

1. Weber and Johnson (as Written)

Weber involved a challenge to a voluntary affirmative action plan, which provided that half of those selected for a training program should be persons of color.[239] After failing to qualify for the program, Brian Weber, a white worker, brought suit under Title VII, alleging that the affirmative action plan constituted “discriminat[ion] . . . because of . . . race.”[240]

Writing for the Court, Justice Brennan found that the voluntary affirmative action plan did not violate Title VII.[241] Justice Brennan’s opinion made little effort to parse the terms of the statute. Instead, as Philip Frickey would later observe, the Court “essentially conceded that Title VII’s plain language supported Weber.”[242] But Justice Brennan insisted that Weber’s “reliance upon a literal construction of” Title VII was “misplaced.”[243] Turning to the Court’s decision in Church of the Holy Trinity v. United States (1892),[244] Justice Brennan announced that “[i]t is a ‘familiar rule, that a thing may be within the letter of the statute and yet not within the statute, because not within its spirit, nor within the intention of its makers.’”[245] The Court then looked to the legislative history and historical context of Title VII, which made “clear that an interpretation . . . that forbade all race-conscious affirmative action would ‘bring about an end completely at variance with the purpose of the statute’ and must be rejected.”[246]

There was an extremely critical reaction to Weber, particularly among conservative commentators.[247] But even more progressive observers found the decision “difficult to justify,” given that Justice Brennan had suggested that “the ‘spirit of the statute’ may trump seemingly plain statutory text and legislative intent.”[248] As Frickey put it, “In my legislation course, I tell my students that Holy Trinity Church is the case you always cite when the statutory text is hopelessly against you,” akin to “the ‘hail Mary’ pass in football.”[249]

Eight years after Weber, Justice Scalia called for the overruling of the case, largely on textualist grounds.[250] Johnson v. Transportation Agency (1987)[251] involved a challenge to a gender-based affirmative action program.[252] Paul Johnson alleged that he was passed over for a promotion in favor of a female employee, in violation of Title VII’s prohibition on “discriminat[ion] . . . because of . . . sex.”[253]

Writing for the Court, Justice Brennan upheld the validity of the affirmative action plan and—most relevant for present purposes—reaffirmed Weber.[254] The Court emphasized that “Congress has not amended the statute to reject our construction [in Weber], nor have any such amendments even been proposed, and we therefore may assume that our interpretation was correct.”[255] Although Justice Brennan acknowledged that congressional silence is not “acquiescence under all circumstances,” he stressed that such an assumption made sense in this context: “Weber . . . was a widely publicized decision that addressed a prominent issue of public debate. Legislative inattention thus is not a plausible explanation for congressional inaction.”[256]

Dissenting, Justice Scalia insisted that Weber was wrong and “should be overruled.”[257] “The language of [Title VII] . . . is unambiguous: it is an unlawful employment practice ‘ . . . to discriminate against any individual . . . because of such individual’s race . . . .’”[258] Yet “Weber disregarded the text of the statute, invoking instead its ‘spirit.’”[259] Justice Scalia further argued that any assumption that Congress had ratified Weber or any other decision by failing to overturn it “should be put to rest.”[260] That position was based on what Justice Scalia described as “the patently false premise that the correctness of statutory construction is to be measured by what the current Congress desires, rather than by what the law as enacted meant.”[261] Moreover, given the constitutional and other roadblocks to legislation, it was “impossible to assert with any degree of assurance” why Congress failed to enact legislation addressing Weber.[262] “[V]indication by congressional inaction,” Justice Scalia insisted, “is a canard.”[263]

2. Changing the Category of Statutory Precedent

 Justice Scalia was not the only one unhappy with the Court’s analysis in Weber. Commentators much more sympathetic to affirmative action have described the opinion as “a failure.”[264] Accordingly, scholars have offered ways to rewrite it. This Article’s typology underscores that such a different approach would not only make for better craftsmanship but should also affect the precedential impact of the decision.

Notably, as written, Weber established the second category of statutory precedent, holding that Title VII did not bar voluntary affirmative action programs. The Court in Weber did not attempt to define the meaning of any statutory term or phrase. Instead, Justice Brennan’s opinion “essentially conceded that Title VII’s plain language supported Weber.”[265]

Yet scholars have contended that the Weber Court did not need to punt quite so quickly. Instead, the Court might have interpreted the term “discriminate” in a way that did not clearly encompass voluntary affirmative action programs. As Eskridge, Frickey, and Ronald Dworkin have (separately) argued, the term “discriminate” contains—and contained in the 1960s—different dictionary definitions.[266] Although “discriminate” may refer to any differential treatment (as Brian Weber argued), the term may more narrowly refer to “invidious” distinctions—classifications that “reflect a desire to put one race at a disadvantage against another” or that are “arbitrary, because they serve no legitimate purpose.”[267] Under the latter view of “discrimination,” these scholars argue, efforts to expand opportunities to historically disadvantaged groups would not be “discrimination.”[268]

In this alternative formulation, Weber would become a very different statutory precedent. Weber would define the term “discriminate.” In the typology of this Article, Weber would fall into the first category of statutory precedent. As discussed in Part I, like other interpreters, textualists give significant weight to precedents that define the meaning of statutory terms and phrases. That is true, even if the prior precedent defining the relevant term or phrase was not itself a purely textualist opinion (as illustrated by Clark and Zadvydas). Indeed, under this alternative formulation, Weber would look a good deal like Zadvydas. The Court would have found that the relevant statutory term (discrimination) was ambiguous, and then resolved the ambiguity. Accordingly, Weber would be on much firmer ground as a precedent among textualists and non-textualists alike if the Court had engaged with the text and endeavored to define the relevant statutory terms.

I do not endeavor to say whether that would have been better as a normative matter. That likely depends both on one’s view of Weber itself and on the potential implications of a decision narrowing the definition of “discrimination.” As Eyer and others have thoughtfully observed, the broad definition of discrimination in Bostock as a difference in treatment, and the but-for causation test, seem to be in serious tension with the holding in Weber.[269] Yet the Bostock framework may bode well for employment discrimination plaintiffs going forward.[270] My goal here is to show that, once we grasp that there are different categories of statutory precedent, we can better understand the long-term impact of any given case.

III. Implications for Interpretive Debates

This Article aims to show that we can better understand statutory precedent, and its relationship to textualism, once we break down both the categories of precedent and the stages of analysis. Textualists often turn to one type of precedent—case law defining the meaning of terms and phrases—in determining the plain meaning of a law (a practice that, I argue, can be defended on textualist assumptions). Textualists may properly be guided by other types of precedent, such as past holdings and implementation tests, when they determine that a statute lacks a plain meaning. This Article not only clarifies the relationship between textualism and statutory precedent but also has implications for broader debates about “ordinary meaning” and the nature of the interpretive enterprise.

A. The “Ordinary Meaning” of a Federal Statute

There is a growing debate over whether the “ordinary meaning” sought by textualists refers to lawyerly meaning or lay meaning.[271] Many scholars seem to assume that textualists generally look for lay meaning—that is, how a person without any legal training would understand statutory terms and phrases.[272] This assumption has driven a good deal of recent theoretical and empirical literature on interpretive theory. William Eskridge, Victoria Nourse, and Anita Krishnakumar have, for example, criticized textualists’ use of legal terms of art, canons, and the common law to make sense of federal statutes, on the ground that such materials are largely inaccessible to the general public.[273] Meanwhile, many empirical scholars likewise assume that textualists aim to discern lay meaning; accordingly, these scholars suggest that corpus linguistics methods[274] or surveys of the general public can help determine the “ordinary meaning” of terms and phrases in federal statutes—or at least test whether textualists have properly determined that meaning.[275]

This Article’s exploration of statutory precedent sheds important light on this debate over “ordinary meaning.” Many textualists—like other interpreters—treat federal statutes as legal documents whose terms and phrases contain a legal meaning.[276] Accordingly, textualists often turn to legal sources, including the Supreme Court’s own precedents, to determine the plain meaning of even seemingly non-technical language in a statute. Thus, the ordinary meaning of “now” in federal legislation may be the time of statutory enactment, rather than the present-day; “costs” may exclude expert witness fees; and “because of” may trigger distinctively legal notions of causation. Likewise, textualists rely on legal rules such as the one-meaning rule in Clark that do not have a clear conversational analogue.

As discussed, such reliance on statutory precedent in determining the meaning of terms and phrases is consistent with longstanding textualist theory. Many textualists assert that statutory analysis must be done from the perspective of a reasonable and reasonably well-informed reader, not any actual reader—either in Congress or the general public.[277] Textualists thus often (albeit often implicitly) acknowledge what can be seen as the most basic context of a federal statute: that it is a legal document full of legal concepts. On this view, the ordinary meaning of a federal statute is not equivalent to a conversation on the street or an advertisement on television, radio, or social media. The ordinary meaning of a federal statute is full of law.

The analysis here connects to other scholarship—from a variety of methodological perspectives—that has pushed back on the assumption that legal texts can be equated with ordinary conversation.[278] As these commentators have argued, “[t]he assumption that legal interpretation should be modeled on the interpretation of ordinary conversation is problematic. Lawmaking has very different goals, presuppositions, and circumstances from ordinary conversation.”[279] The lawmaking context is “impersonal” and “less cooperative” than ordinary communication.[280] These differences help explain why legislation does not contain—and we do not expect it to contain—irony or humor. In the impersonal and uncooperative context of legislation, the relevant audience would be less likely to “get the joke.” Interpreting federal legislation is importantly different from understanding ordinary conversation—because of both the different setting and legislation’s reliance on law.

To be sure, the legalistic vision of federal statutes offered here may be contested—a point I explore further in the next section. But, interestingly, this more legalistic vision may better accord with the understanding of the general public. A recent empirical study by Kevin Tobia, Brian Slocum, and Victoria Nourse suggests that members of the public understand that the law is a special language.[281] According to the authors, even when a statute uses seemingly ordinary terms and phrases, such as “intent” or “because of,” lay people assume that the terms may take on a distinctively legal meaning.[282] For that reason, the authors assert, members of the public are inclined to defer to legal experts on statutory interpretive questions.[283]

B. Legal Analysis at the Plain Meaning Stage

As this Article has described, for textualists, statutory analysis involves two distinct stages. First, interpreters endeavor to determine whether a statute has a plain meaning. Second, if not, they turn to additional tools to make sense of the statute in the context of the case. This Article demonstrates that textualists often turn to legal sources, including surrounding text, structure, and judicial precedent, at the first stage: in determining the plain meaning of a statutory provision.

This legalistic vision of the plain meaning analysis is likely to be contested by some scholars.[284] Some prominent scholarship splits the statutory inquiry into two stages called “interpretation” and “construction.”[285] Although this terminology has been influential primarily in constitutional debates, scholars have also begun to use the terms to describe statutory analysis.[286] Some scholars may view what I have called the plain meaning inquiry as analogous to the “interpretation” stage (as often defined in this academic work) and may contest a legalistic vision of “interpretation.”

Scholars who discuss the interpretation–construction distinction often define “interpretation” as largely a search for linguistic meaning, while “construction” is the process of giving legal effect to that meaning.[287] Many commentators assume that “interpretation” looks at linguistic facts—facts about patterns of usage (such as syntax and grammar)—and thus is largely “value neutral” or “thinly normative.”[288] Most important for present purposes, “interpretation” seems to focus on determining a fact of the matter about communication and not to depend much, if at all, on legal or normative judgments.[289]

This Article calls into question this image of the first stage of the statutory inquiry as a primarily linguistic and empirical exercise. As an initial matter, the image does not seem descriptively accurate. As this Article has shown, to make sense of federal legislation, judges—including many textualist judges—rely on legal sources, such as the text and structure surrounding the operative provision at issue as well as judicial precedents defining the meaning of terms and phrases. Moreover, such statutory precedent may show that seemingly ordinary terms and phrases, such as “costs,” “now,” or “because of,” take on different meanings in legislation than what we might expect in ordinary conversation. Thus, “now” may not mean the present moment but the time of statutory enactment; “costs” may exclude expert fees; and “because of” may signal a specifically legal notion of but-for causation. Judges also apply legal rules—such as the one-meaning rule at issue in Clark—to make sense of statutory language.

In using these legal sources and legal rules to determine the plain meaning of federal statutes, judges do not simply examine linguistic practices and rules of grammar but also make legal and normative judgments. For example, in using the first category of statutory precedent to help discern the meaning of terms and phrases, judges must make determinations about which precedents count.

Is this practice normatively attractive? Some might worry about allowing judges to make these legal and normative judgments. Indeed, this practice may seem particularly problematic for textualism, a methodology that (according to many of its proponents) is designed to constrain judicial discretion.[290] But I believe that reliance on legal tools, including judicial precedent, is defensible—and, indeed, normatively preferable. The critique overlooks both the ability of interpreters to craft legal rules and guidelines to constrain their interpretive discretion in specific cases, and (relatedly) how the use of statutory precedent to define the meaning of terms and phrases can serve to constrain, rather than to expand, judicial discretion.

Let us take a step back and recall that constraint in specific cases seems to be of the utmost concern. The use of precedent is said to serve rule of law values, including stability and predictability, and ensuring that officials make decisions based on external factors, rather than their own personal preferences or ideology. Those concerns are greatest in specific cases. One might worry, for example, in an employment discrimination dispute, that a judge might be tempted to decide the case based on her background sympathy for (or against) corporate employers. Or one might be concerned, in a case involving the detention of undocumented immigrants, that a judge might be influenced by her background views on immigration. To further rule of law values, an interpretative approach—whether it be textualism or an alternative—should not depend on whether a judge is ruling on an employment dispute, an immigration question, or a securities fraud case. An interpretive approach should apply across the board.

Although no interpretive method can fully cabin judicial discretion, I believe that interpreters can craft legal rules to better constrain their discretion in specific cases. This basic insight applies to the first category of statutory precedent: the use of case law to define the meaning of terms and phrases at the first (plain meaning) stage of the statutory inquiry.

First, as I have suggested, interpreters, including textualists, can craft legal rules to govern issues such as which statutory precedents are relevant at the plain meaning stage.[291] Given textualism’s general emphasis on the time of statutory enactment, textualists should draw on cases that were issued prior to the enactment of the relevant statutory provision (as Pierce and Saxon did to make sense of phrases in the Equal Access to Justice Act and the Federal Arbitration Act), or on precedents that themselves aimed to determine the meaning of terms or phrases at the time of enactment (as illustrated by Bostock’s reliance on Nassar and Gross). The Justices can also adopt legal rules, such as the one-meaning rule in Clark, to guide their analysis of specific statutory provisions.

Second, and importantly, statutory precedent can then constrain judicial discretion in specific cases. As discussed, the first category of statutory precedent can provide considerable guidance in future cases.[292] Because of statutory precedent, judges have a better understanding of, for example, the legal meaning of “costs” and “discrimination because of sex.” Moreover, statutory precedent may lead judges to issue decisions that are contrary to (what we might view as) their ideological priors. Thus, Justice Thomas led a unanimous Court in Saxon to reject an employer’s demand for arbitration. And the one-meaning rule in Clark may be the most constraining. Justice Scalia relied on Zadvydas (a decision from which he himself had dissented) to issue a decision that protected a class of undocumented immigrants from unlimited detention.

In crafting legal rules and guidelines for applying the first category of statutory precedent, judges must make legal and normative judgments—and thereby exercise some discretion. It is, after all, a normative decision (albeit one shared by most textualists) that the relevant time period for interpreting a federal statute is the time of enactment.[293] Likewise, judges made a normative decision to adopt the one-meaning rule. But once interpreters adopt such legal rules and guidelines, they can provide considerable constraint in specific cases.

Moreover, it should be of some comfort to observers that such legal judgments are precisely the kinds of decisions that judges are most equipped to make. Scholars worry about the capacity of judges to engage in historical inquiry—with the concern that they may provide only “law office history.”[294] Likewise, scholars, including even some supporters of corpus linguistics, have expressed doubts about the capacity of judges to search electronic databases for past usage of terms or phrases.[295] But establishing and applying precedential rules is the kind of analysis for which law school and legal practice most clearly prepare judges. Indeed, it makes sense to allocate the statutory interpretive inquiry to judges, rather than, say, linguists, historians, or philosophers, to the extent that interpretation is, at bottom, about law.[296]


This Article complicates common assumptions about the relationship between textualism and statutory precedent. Once we recognize that there are different types of statutory precedent, we can see that textualists do make important use of such precedent. In identifying the plain meaning of a federal statute, textualists often rely on precedent that defines the meaning of terms and phrases—a practice that, this Article argues, can be defended on textualist principles. Textualists properly rely on past holdings and implementation tests and seek consistency with the general body of law in a given area when a statutory provision lacks a plain meaning. Conversely, textualists are properly more skeptical of such holdings and tests if they conflict with a statute’s plain meaning. This Article illuminates not only textualism’s relationship with statutory precedent but also the nature of the interpretive enterprise. Many textualists treat statutory provisions as legal documents that should be interpreted according to legal norms and conventions. For these textualists, like many other interpreters, the effort to identify the ordinary meaning of a federal statute is, at bottom, a legal and normative, not simply a linguistic, exercise.

  1. . See, e.g., Antonin Scalia & Bryan A. Garner, Reading Law 413–14 (2012) (“Stare decisis . . . is not a part of textualism. It is an exception to textualism . . . .”); Anita S. Krishnakumar, Textualism and Statutory Precedents, 104 Va. L. Rev. 157, 160–61, 164–65, 182 (2018) (asserting, based on an empirical study of the Roberts Court, that textualists are “far more willing” to overturn statutory precedents than “their purposivist counterparts”); Caleb Nelson, Stare Decisis and Demonstrably Erroneous Precedents, 87 Va. L. Rev. 1, 50–52 (2001) (urging that those who “believe[] in the determinacy of the underlying legal texts” may be less inclined to adhere to precedent); Lawrence B. Solum, Disaggregating Chevron, 82 Ohio St. L.J. 249, 295 (2021) (arguing that holdings “not based on textualist reasoning . . . are not entitled to stare decisis effect”); Josh Blackman, Justice Gorsuch’s Legal Philosophy Has a Precedent Problem, Atlantic (July 24, 2020), [] (contending that Justice Gorsuch’s approach of “profess[ing] to follow the original meaning of a text while in fact following precedents that ignored that meaning” undermines textualism). Then-Professor Amy Coney Barrett offered seemingly different views. Compare Amy Coney Barrett, Statutory Stare Decisis in the Courts of Appeals, 73 Geo. Wash. L. Rev. 317, 326 (2005) (suggesting a textualist commitment to statutory stare decisis), with Amy Coney Barrett, Precedent and Jurisprudential Disagreement, 91 Texas L. Rev. 1711, 1724–25 (2013) (suggesting that those who adopt text-based theories may be less committed to precedent).
  2. . Scalia & Garner, supra note 1, at 413–14; see also Richard H. Fallon Jr, The Meaning of Legal “Meaning” and Its Implications for Theories of Legal Interpretation, 82 U. Chi. L. Rev. 1235, 1284–85 (2015) (asserting that stare decisis presents a “dilemma” for textualists).
  3. . See, e.g., Gamble v. United States, 139 S. Ct. 1960, 1984–88 (2019) (Thomas, J., concurring) (“[F]ederal judges should . . . correct [a demonstrable] error.”); supra note 1; Margaret H. Lemos, The Politics of Statutory Interpretation, 89 Notre Dame L. Rev. 849, 898–900 (2013) (reviewing Scalia & Garner, supra note 1) (arguing that Justices Scalia and Thomas proved willing “to overturn or severely prune statutory precedents”). For a possible counterexample, in which Justice Kavanaugh suggested agreement with the Court’s “strict” adherence to statutory stare decisis, see Allen v. Milligan, 143 S. Ct. 1487, 1517 (2023) (Kavanaugh, J., concurring in part).
  4. . 140 S. Ct. 1731 (2020).
  5. . Id. at 1737–38, 1754 (quoting 42 U.S.C. § 2000e-2(a)(1)).
  6. . See Neil M. Gorsuch, A Republic, If You Can Keep It 131–32 (2019) (arguing for textualism).
  7. . Bostock, 140 S. Ct. at 1739 (quoting Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 346, 360 (2013)) (citing Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 176 (2009)).
  8. . See id. at 1743–44 (discussing Phillips v. Martin Marietta Corp., 400 U.S. 542 (1971) (per curiam)); City of Los Angeles Department of Water & Power v. Manhart, 435 U.S. 702 (1978); and Oncale v. Sundowner Offshore Services, Inc., 523 U.S. 75 (1998)).
  9. . See, e.g., Josh Blackman & Randy Barnett, Justice Gorsuch’s Halfway Textualism Surprises and Disappoints in the Title VII Cases, Nat’l Rev. (June 26, 2020, 6:30 AM), [] (criticizing Justice Gorsuch’s reliance on precedent to define “because of”); James Andrew Wynn, When Judges and Justices Throw Out Tools: Judicial Activism in Rucho v. Common Cause, 96 N.Y.U. L. Rev. 607, 640 (2021) (arguing that Justice Gorsuch used “non-textualist tools,” when he “relied, in part, on precedent to determine the plain meaning of Title VII”); William Baude, Conservatives, Don’t Give Up on Your Principles or the Supreme Court, N.Y. Times (July 9, 2020), [] (“What made Justice Gorsuch’s opinion most persuasive was not its textualist analysis but its use of precedents . . . .”); Erik Encarnacion, Text Is Not Law, 107 Iowa L. Rev. 2027, 2059–60, 2059 n.212 (2022) (noting that “textualists routinely rely on extratextual resources,” such as Bostock’s reliance on statutory precedent); William N. Eskridge, Jr. & Victoria F. Nourse, Textual Gerrymandering: The Eclipse of Republican Government in an Era of Statutory Populism, 96 N.Y.U. L. Rev. 1718, 1771–72 (2021) (arguing that “statutory precedents complemented and then overtook language analysis” in Bostock).
  10. . One thoughtful piece does note the practice. In an article exploring textualists’ disregard of statutory precedent, Anita Krishnakumar suggests that interpreters may rely on precedent to define the meaning of terms. But the piece does not focus on that practice and instead examines textualists’ apparent willingness to reject statutory implementation tests. See Krishnakumar, supra note 1, at 160–61, 164–65, 182 (urging that “the disregard for statutory stare decisis” is “a natural corollary to the textualist jurisprudential approach”).
  11. . Bostock, 140 S. Ct. at 1739.
  12. . 443 U.S. 193 (1979).
  13. . Id. at 197, 208–09.
  14. . See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802–03 (1973) (establishing a burden-shifting framework for discrimination claims that rely on circumstantial evidence).
  15. . See Patterson v. McLean Credit Union, 491 U.S. 164, 172–73 (1989) (Kennedy, J.) (“Considerations of stare decisis have special force in the area of statutory interpretation . . . .”), superseded by statute, Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071, as recognized in CBOCS West, Inc. v. Humphries, 553 U.S. 442 (2008); William N. Eskridge, Jr., Overruling Statutory Precedents, 76 Geo. L.J. 1361, 1409 (1988) (dubbing this practice a “super-strong presumption against overruling statutory precedents”).
  16. . Some jurists and scholars have advocated super-strong (or even absolute) statutory stare decisis. See Boys Mkts., Inc. v. Retail Clerks Union, Local 770, 398 U.S. 235, 257–58, 260 (1970) (Black, J., dissenting) (arguing that the Court should play a very minimal role in overturning statutory precedent); Frank E. Horack, Jr., Congressional Silence: A Tool of Judicial Supremacy, 25 Texas. L. Rev. 247, 251 (1947) (arguing that overruling statutory precedent would “chang[e] an established rule of law under which society has been operating”); Adrian Vermeule, Interpretive Choice, 75 N.Y.U. L. Rev. 74, 143–45 (2000) (advocating absolute statutory stare decisis as a means of reducing decision costs); Lawrence C. Marshall, “Let Congress Do It”: The Case for an Absolute Rule of Statutory Stare Decisis, 88 Mich. L. Rev. 177, 183, 215 (1989) (advocating absolute statutory stare decisis to make it more difficult for Congress to avoid overseeing statutory development). Others—textualists and non-textualists alike—doubt that statutory precedents deserve special protection. See Gamble v. United States, 139 S. Ct. 1960, 1987–88 (Thomas, J., concurring) (arguing that “the Court’s multifactor approach to stare decisis invites conflict with its constitutional duty”); Frank H. Easterbrook, Stability and Reliability in Judicial Decisions, 73 Cornell L. Rev. 422, 429–31, 433 (1988) (suggesting that it should not be easier to overturn constitutional decisions than statutory decisions); Eskridge, supra note 15, at 1409, 1426 (“In my view, statutory precedents are entitled only to normal stare decisis effect.”); see also Randy J. Kozel, Statutory Interpretation, Administrative Deference, and the Law of Stare Decisis, 97 Texas. L. Rev. 1125, 1130 (2019) (noting the debate over “elevated stare decisis”); Krishnakumar, supra note 1, at 219 (urging ordinary stare decisis for some implementation tests); Thomas R. Lee, Stare Decisis in Historical Perspective: From the Founding Era to the Rehnquist Court, 52 Vand. L. Rev. 647, 730–31 (1999) (stating that the heightened stare decisis test was a late-nineteenth-century development).
  17. . I borrow the term “fit” from Ronald Dworkin. E.g., Ronald Dworkin, Hard Cases, 88 Harv. L. Rev. 1057, 1094 (1975). One can see value in consistency among precedents and within the law, even if one does not accept a Dworkinian approach to interpretation.
  18. . See Bostock v. Clayton County, 140 S. Ct. 1731, 1743–44 (2020) (discussing three earlier cases).
  19. . To be sure, interpreters may disagree about when a statute has a “plain meaning.” For purposes of this Article’s typology, it is enough that textualists do sometimes find a statute to have and not to have a plain meaning. See infra note 186; infra subparts I(A), II(A).
  20. . Notably, some scholars might describe these stages as “interpretation” and “construction.” I discuss those labels below. See infra note 30; infra subpart III(B).
  21. . See Amy Coney Barrett, Substantive Canons and Faithful Agency, 90 B.U. L. Rev. 109, 164 (2010) (arguing that substantive canons that “permit a court to qualify clear text . . . are inconsistent with” textualism); John F. Manning, Foreword, The Means of Constitutional Power, 128 Harv. L. Rev. 1, 68 (2014) (noting that “the Court rejects legislative history when it conflicts with the statutory text”); John F. Manning, Competing Presumptions About Statutory Coherence, 74 Fordham L. Rev. 2009, 2010 (2006) (rejecting the use of purpose to “smooth over the details of an agreed-upon text”); see also Benjamin Eidelson & Matthew C. Stephenson, The Incompatibility of Substantive Canons and Textualism, 137 Harv. L. Rev. 515, 519–21 (2023) (urging textualists to abandon all substantive canons).
  22. . These Justices are viewed as textualist or textualist leaning. Anita S. Krishnakumar, Statutory History, 108 Va. L. Rev. 263, 284 n.59 (2022); John F. Manning, Separation of Powers as Ordinary Interpretation, 124 Harv. L. Rev. 1939, 1973 n.179 (2011).
  23. . See Barrett, supra note 21, at 109–10 (noting that textualists “continue to accept substantive canons”); John F. Manning, What Divides Textualists from Purposivists?, 106 Colum. L. Rev. 70, 75–76 (2006) (suggesting that modern textualists rely on statutory purpose, but not legislative history, when the text is ambiguous).
  24. . W. Va. Univ. Hosps., Inc. v. Casey, 499 U.S. 83, 100–01 (1991) (Scalia, J.), superseded by statute, Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071, as recognized in Landgraf v. USI Film Prods., 511 U.S. 244 (1994).
  25. . Bostock v. Clayton County, 140 S. Ct. 1731, 1740 (2020) (third alteration in original) (emphasis omitted) (quoting 42 U.S.C. § 2000e-2(a)(1)).
  26. . Id. at 1741–42.
  27. . Van Buren v. United States, 141 S. Ct. 1648, 1661 (2021) (Barrett, J.) (quoting Yates v. United States, 574 U.S. 528, 557 (2015) (Kagan, J., dissenting)). Indeed, that is how Justice Gorsuch characterized his discussion of the third type of statutory precedent. See Bostock, 140 S. Ct. at 1738–39, 1743–44 (pointing to precedent “[i]f more support . . . were required”).
  28. . Compare Anita S. Krishnakumar, The Common Law as a Statutory Backdrop, 136 Harv. L. Rev. 608, 660–61 (2022) (“[M]odern textualism increasingly has focused not just on the ordinary meaning of statutory language, but on the ordinary reader of statutory language.”), and James A. Macleod, Finding Original Public Meaning, 56 Ga. L. Rev. 1, 4–5 (2021) (arguing that textualists “ask how ordinary readers would have understood the relevant language”), and Brian G. Slocum, Introduction, in The Nature of Legal Interpretation 1, 7–8 (Brian G. Slocum ed., 2017) (observing that under ordinary meaning doctrine, “legal language is governed by the same general principles of language usage that apply outside of the law”), and Kevin P. Tobia, Testing Ordinary Meaning, 134 Harv. L. Rev. 726, 801 (2020) (asserting that textualist and originalist theories assume that ordinary meaning is “an empirical fact . . . grounded in what language communicates to ordinary people”), with Mitchell N. Berman & Guha Krishnamurthi, Bostock Was Bogus: Textualism, Pluralism, and Title VII, 97 Notre Dame L. Rev. 67, 97 (2021) (“[C]ommitted textualists have often insisted that ordinary meaning is . . . a partially normalized or idealized [inquiry].”), and Benjamin Eidelson, Dimensional Disparate Treatment, 95 S. Cal. L. Rev. 785, 845–46 (2022) (“[T]extualism gives legal weight to ‘ordinary meaning’ only in a specific sense: it asks what a reasonable member of the public would take Congress to have intended to communicate . . . .”), and Tara Leigh Grove, Foreword, Testing Textualism’s “Ordinary Meaning, 90 Geo. Wash. L. Rev. 1053, 1057, 1082–84 (2022) (arguing that “ordinary meaning” can be understood as a legal concept), and Felipe Jiménez, Textualism Without Trimmings 51–52 (Feb. 19, 2024) (unpublished manuscript) (on file with author) (agreeing that “ordinary meaning is a technical legal construct”), and Evan C. Zoldan, The Conversation Canon, 110 Ky. L.J. 1, 4, 6 (2022) (suggesting that textualists may disagree over whether it is proper to rely on conversational language to interpret statutes).
  29. . Scholars who view the textualist enterprise as a search for lay meaning do recognize an exception for legal terms of art. But the assumption appears to be that such terms of art are rare. See Thomas R. Lee & Stephen C. Mouritsen, Judging Ordinary Meaning, 127 Yale L.J. 788, 792 n.4 (2018) (treating legal terms of art as “extraordinary”); Macleod, supra note 28, at 56–57, 57 n.230 (indicating that a textualist’s preference to use the everyday meaning of words may not apply when “the terms at issue are terms of art”). This Article aims in part to show that there often is no sharp line between ordinary terms and legal terms of art in federal statutes. See infra Part I.
  30. . See infra subpart III(B). These scholars often split the statutory inquiry into two stages called “interpretation” and “construction.” This terminology has been most influential in constitutional theory. See Keith E. Whittington, Constitutional Construction 6 (1999) (“Allowing for both interpretation and construction expands the field of constitutional elaboration without shrinking the range of interpretation.”); Randy E. Barnett & Evan D. Bernick, The Letter and the Spirit: A Unified Theory of Originalism, 107 Geo. L.J. 1, 10–13 (2018) (“The interpretation–construction distinction became the second component of a defensible theory of originalism.”). But the concepts have begun to make headway into statutory debates as well. See Lawrence B. Solum & Cass R. Sunstein, Chevron as Construction, 105 Cornell L. Rev. 1465, 1468–70 (2020) (“Chevron involved a question of statutory construction.”); Solum, supra note 1, at 264–70 (describing four approaches to statutory interpretation and construction). Scholars who describe the inquiry in this fashion have suggested that “interpretation” is largely a search for linguistic meaning, while “construction” is the process of giving legal effect to that meaning. E.g., Lawrence B. Solum, The Interpretation-Construction Distinction, 27 Const. Comment. 95, 96 (2010); see also Slocum, supra note 28, at 4–8 (discussing the distinction between “interpretation,” which deals with “linguistic understanding,” and “‘construction,’ which represents instances where judges choose meanings that transcend interpretations”); Randy E. Barnett, Interpretation and Construction, 34 Harv. J.L. & Pub. Pol’y 65, 66 (2011) (“Interpretation is the activity of identifying the semantic meaning of a particular use of language in context. Construction is the activity of applying that meaning to particular factual circumstances.”); Jamal Greene, A Nonoriginalism for Originalists, 96 B.U. L. Rev. 1443, 1450 (2016) (noting the distinction between interpretation and construction). To the extent one views the “plain meaning” stage described in this Article as analogous to the “interpretation” stage, this Article has implications for our understanding of “interpretation.” See infra subpart III(B).
  31. . Bostock, 140 S. Ct. at 1739.
  32. . See, e.g., Randy J. Kozel, Settled Versus Right: Constitutional Method and the Path of Precedent, 91 Texas. L. Rev. 1843, 1849–51 (2013) (distinguishing precedent’s “hierarchical control” function from its “persuasive” function); Frederick Schauer, Has Precedent Ever Really Mattered in the Supreme Court?, 24 Ga. St. Univ. L. Rev. 381, 385, 393, 399 (2007) (arguing that precedent rarely matters in the Supreme Court). The Supreme Court can also narrow its own precedents, without overruling them. Cf. Richard M. Re, Precedent as Permission, 99 Texas. L. Rev. 907, 908–12 (2021) (questioning common assumptions about precedential constraint). Compare Barry Friedman, The Wages of Stealth Overruling (with Particular Attention to Miranda v. Arizona), 99 Geo. L.J. 1, 4–5 (2010) (criticizing “stealth overruling”), with Richard M. Re, Narrowing Precedent in the Supreme Court, 114 Colum. L. Rev. 1861, 1865–66 (2014) (defending “narrowing”), with Bill Watson, Obstructing Precedent, Nw. U. L. Rev. (forthcoming), (offering a limited defense of “obstructing” precedent).
  33. . That is true, even with respect to constitutional precedents—an area where the Supreme Court itself has asserted greater authority to correct mistakes. See Richard H. Fallon Jr., Constitutional Precedent Viewed Through the Lens of Hartian Positivist Jurisprudence, 86 N.C. L. Rev. 1107, 1147 (2008) (“[A]n ultimate rule of recognition authorizes the Justices to treat otherwise erroneous precedents either as binding or not on the basis of case-by-case considerations . . . .”); Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88 Colum. L. Rev. 723, 723–24 (1988) (suggesting that various features of the constitutional order “cannot be judicially overruled”).
  34. . See, e.g., Kevin Tobia, Brian G. Slocum & Victoria Nourse, Progressive Textualism, 110 Geo. L.J. 1437, 1439 (2022) (noting textualism’s influence at the Supreme Court). This Article sets aside debates over methodological precedent—whether judges should be bound by prior statements about how the statutory analysis should proceed. Many scholars assert that judges of all interpretive stripes tend to reject giving stare decisis effect to methodology. Sydney Foster, Should Courts Give Stare Decisis Effect to Statutory Interpretation Methodology?, 96 Geo. L.J. 1863, 1875 (2008); Nicholas Quinn Rosenkranz, Federal Rules of Statutory Interpretation, 115 Harv. L. Rev. 2085, 2144–45 (2002); see also Abbe R. Gluck, What 30 Years of Chevron Teach Us About the Rest of Statutory Interpretation, 83 Fordham L. Rev. 607, 613 (2014) (explaining that methodological precedent “is not viewed as ‘law’ for the next case”). But see Aaron-Andrew P. Bruhl, Eager to Follow: Methodological Precedent in Statutory Interpretation, 99 N.C. L. Rev. 101, 105, 132, 165 (2020) (arguing, based on a detailed survey, that there is “much more” methodological precedent than most recognize, particularly in the lower federal courts). I plan to explore methodological precedent in future work.
  35. . See infra note 36. As the examples in subpart I(A) make clear, the definition given to certain terms and phrases in past opinions is often necessary to the decision. With a different definition, the case would come out the other way. Cf. Seminole Tribe v. Florida, 517 U.S. 44, 67 (1996) (“When an opinion issues for the Court, it is not only the result but also those portions of the opinion necessary to that result by which we are bound.”).
  36. . See Randy J. Kozel, The Scope of Precedent, 113 Mich. L. Rev. 179, 191–97 (2014) (stating that “[s]ome conceptions of precedent are broad enough to treat supportive reasoning as carrying binding force” and that the Supreme Court often, albeit not uniformly, adopts that more capacious view); Krishnakumar, supra note 1, at 164, 185 (noting “precedents that closely parse the meaning of statutory text”). Indeed, some scholars have gone further than this Article and used the term “precedent” to refer to decisions that are not binding at all—for example, in describing how lower court “precedent” may influence the Supreme Court. Aaron-Andrew P. Bruhl, Following Lower-Court Precedent, 81 U. Chi. L. Rev. 851, 852–53 (2014). Scholars have further noted that the traditional distinction between “holding” and “dicta” is blurry in practice. See Kozel, supra, at 190–97 (arguing that precedent has a broad scope at the Supreme Court and giving examples); Allison Orr Larsen, Factual Precedents, 162 U. Pa. L. Rev. 59, 62 (2013) (noting that lower courts tend “to over-rely on Supreme Court opinions and to apply generalized statements of fact from old cases to new ones”); Michael C. Dorf, Dicta and Article III, 142 U. Pa. L. Rev. 1997, 2003–05 (1994) (discussing the difficulty of distinguishing holding from dicta); David Klein & Neal Devins, Dicta, Schmicta: Theory Versus Practice in Lower Court Decision Making, 54 Wm. & Mary L. Rev. 2021, 2024–27 (2013) (discussing the difficulty lower courts face in making the distinction).
  37. . See supra note 9.
  38. . See generally Tara Leigh Grove, Comment, Which Textualism?, 134 Harv. L. Rev. 265 (2020) (discussing competing strands of textualism and noting that “[s]cholarship on statutory interpretation has largely overlooked the divisions within textualism”); infra note 40.
  39. . This Article describes the first stage as a search for “plain meaning,” rather than as an effort to determine whether a provision is “ambiguous,” because textualists have at times discussed a “prima facie ambiguity” that can be resolved at the first stage of analysis. See, e.g., Manning, supra note 23, at 95–96 (arguing that semantic, rather than policy, context should inform the resolution of a “prima facie ambiguity”).
  40. . Statutory purpose, most substantive canons, and legislative history are excluded. See supra note 21 and accompanying text. There is a division among textualists as to whether to exclude at this first stage other evidence, such as the social context surrounding a statute’s enactment or the assumed practical consequences of a decision. See Grove, supra note 38, at 265–71, 279–90 (discussing the differences between a more “formalistic textualism” and a more “flexible textualism”).
  41. . Antonin Scalia, Common-Law Courts in a Civil-Law System: The Role of United States Federal Courts in Interpreting the Constitution and Laws, in A Matter of Interpretation 3, 24 (1997); Caleb Nelson, What Is Textualism?, 91 Va. L. Rev. 347, 376 & n.87 (2005).
  42. . John F. Manning, Textualism as a Nondelegation Doctrine, 97 Colum. L. Rev. 673, 696 (1997).
  43. . See, e.g., Frank H. Easterbrook, Text, History, and Structure in Statutory Interpretation, 17 Harv. J.L. & Pub. Pol’y 61, 64 (1994) (describing textualism as “preferring the language and structure of the law”); John F. Manning, Textualism and the Equity of the Statute, 101 Colum. L. Rev. 1, 109 (2001) (noting that textualists “often consult dictionaries”); Tara Leigh Grove, The Misunderstood History of Textualism, 117 Nw. U. L. Rev. 1033, 1097 (2023) (describing textualism as emphasizing the text and structure surrounding an operative provision). Some scholars suggest that judges should also use corpus linguistics methods. See Lee & Mouritsen, supra note 29, at 792, 828–30 (arguing that corpus linguistics can help to resolve linguistic uncertainty in the interpretation of legal texts). As suggested in Part III, this Article’s exploration of precedent may have some implications for the debate over corpus linguistics. For now, however, I simply note that corpus linguistics is not yet an accepted tool for identifying plain meaning on the Supreme Court.
  44. . E.g., Field v. Mans, 516 U.S. 59, 61, 69 (1995) (Souter, J.); see also Moskal v. United States, 498 U.S. 103, 121 (1990) (Scalia, J., dissenting) (“[W]hen a statute employs a term with a specialized legal meaning relevant to the matter at hand, that meaning governs.”); Manning, supra note 42, at 695 (“[I]n consulting a statute’s legal context, textualists often rely on extrinsic sources, such as judicial decisions and legal treatises, to determine the specific meaning of codified terms of art.”). Non-textualists also look to case law to define legal terms of art. See, e.g., Merck & Co. v. Reynolds, 559 U.S. 633, 644–48 (2010) (Breyer, J.) (relying on case law to determine the meaning of “discovery” “in the statute of limitations context”). This rule is longstanding. See G.A. Endlich, A Commentary on the Interpretation of Statutes 5 (Jersey City, Frederick D. Linn & Co. 1888) (stating the rule); Krishnakumar, supra note 28, at 668–69 (exploring the use of common law precedents).
  45. . See, e.g., Encino Motorcars, LLC v. Navarro, 138 S. Ct. 1134, 1141–42 (2018) (Thomas, J.) (observing, in interpreting the Fair Labor Standards Act’s exemption from overtime pay for “any salesman, partsman, or mechanic primarily engaged in selling or servicing automobiles,” 29 U.S.C. § 213(b)(10)(A), that “‘or’ is ‘almost always disjunctive’” (quoting United States v. Woods, 571 U.S. 31, 45 (2013))).
  46. . See, e.g., Ali v. Fed. Bureau of Prisons, 552 U.S. 214, 215–16, 218–20, 227–28 (2008) (Thomas, J.) (stating, in interpreting a Federal Tort Claims Act provision that declines to waive sovereign immunity for claims arising from the detention of property by “any officer of customs or excise or any other law enforcement officer,” 28 U.S.C. § 2680(c), that “[w]e have previously noted that ‘[r]ead naturally, the word “any” has an expansive meaning’” (second alteration in original) (quoting United States v. Gonzales, 520 U.S. 1, 5 (1997)) (citing Harrison v. PPG Indus., Inc., 446 U.S. 578 (1980))); see also Encino Motorcars, 138 S. Ct. at 1141 (relying on Ali’s expansive interpretation of “any”). The dissenting opinion in Ali relied on different precedents in arguing that the term “any” may have a narrower meaning in some contexts. See Ali, 552 U.S. at 234–35 (Kennedy, J., dissenting) (reviewing precedents in which the word “any” was read narrowly and in context). For a survey of statutes and cases using the term “any,” see generally James J. Brudney & Ethan J. Leib, “Any, 49 BYU L. Rev. 465 (2023).
  47. . See, e.g., Carcieri v. Salazar, 555 U.S. 379, 381–82, 388–90, 395 (2009) (Thomas, J.) (relying on 1933 and 1934 dictionaries and statutory structure, as well as Supreme Court precedent, to determine that the “plain meaning” of the term “now” in the Indian Reorganization Act of 1934 refers to “the time of the statute’s enactment”).
  48. . See, e.g., Bostock v. Clayton County, 140 S. Ct. 1731, 1739 (2020) (Gorsuch, J.) (“In the language of law . . . Title VII’s ‘because of’ test incorporates the ‘simple’ and ‘traditional’ standard of but-for causation.” (quoting Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 346, 360 (2013)) (citing Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 176 (2009))).
  49. . See, e.g., Mont v. United States, 139 S. Ct. 1826, 1832 (2019) (Thomas, J.) (noting that “[t]he Court has often recognized that ‘in connection with’ can bear a ‘broad interpretation,’” (quoting Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 547 U.S. 71, 85 (2006)) (citing United States v. Am. Union Transp., Inc., 327 U.S. 437, 443 (1946)), although concluding that the Court did not need to “consider the outer bounds” of the phrase, given that a “pretrial detention” was clearly “imprison[ment] in connection with a conviction” (alteration in original) (quoting 18 U.S.C. § 3624(e)), and thus tolled the supervised-release term). The dissenting opinion in Mont, written by Justice Sotomayor and joined by Justices Breyer, Kagan, and Gorsuch, relied in part on precedent to argue that the majority had misconstrued “imprisonment.” See id. at 1838–39 (Sotomayor, J., dissenting) (first citing Tapia v. United States, 564 U.S. 319, 327 (2011); and then citing Barber v. Thomas, 560 U.S. 474, 484 (2010)) (urging that other federal statutes and the Court’s precedents treated “imprisonment” as “post-trial detention”).
  50. . 555 U.S. 379 (2009).
  51. . See id. at 381–82, 388–90, 395–96 (citing to 25 U.S.C. § 479 (2006) (current version at 25 U.S.C. § 5129), which defines “Indian” to “include all persons of Indian descent who are members of any recognized Indian tribe now under Federal jurisdiction” (emphasis added)).
  52. . Id. at 388.
  53. . 139 S. Ct. 873 (2019) (Kavanaugh, J.).
  54. . Id. at 877–78 (quoting Arlington Cent. Sch. Dist. Bd. of Educ. v. Murphy, 548 U.S. 291, 297 (2006)) (relying on Crawford Fitting Co. v. J.T. Gibbons, Inc., 482 U.S. 437, 439 (1987), and West Virginia University Hospitals, Inc. v. Casey, 499 U.S. 83, 86 (1991), superseded by statute, Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071, as recognized in Landgraf v. USI Film Prods., 511 U.S. 244 (1994)). In Rimini, the Court went on to conclude, based on the text, structure, and history of the Copyright Act, that the term “full costs” did not change the calculus; the plaintiffs could not recover expert fees. Id. at 875–78. Some readers might argue that litigation “costs” is a legal term of art. As discussed below, the line between legal terms of art and ordinary terms is a good deal fuzzier in the statutory context than many commentators have appreciated. In this case, a person could understand the term “costs”—even costs related to litigation—without legal training. Such a lay person might be quite surprised to learn that “costs” does not include fees used to hire experts.
  55. . Cf. Husted v. A. Philip Randolph Inst., 138 S. Ct. 1833, 1842 (2018) (Alito, J.) (noting that a statute with “an undefined causation requirement” refers to one of these concepts).
  56. . 570 U.S. 338 (2013) (Kennedy, J.).
  57. . 42 U.S.C. § 2000e-3(a); Nassar, 570 U.S. at 343.
  58. . Nassar, 570 U.S. at 344–45.
  59. . 29 U.S.C. § 623(a)(1); see Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 169–70 (2009) (Thomas, J.) (presenting the question “whether a plaintiff must present direct evidence of age discrimination in order to obtain a mixed-motives jury instruction in a suit brought under the Age Discrimination in Employment Act of 1967”); Nassar, 570 U.S. at 349–52 (noting that Gross offered textual insights “concern[ing] the proper interpretation of the term ‘because’”).
  60. . 557 U.S. 167 (2009).
  61. . Id. at 176 (quoting Because of, Webster’s Third New International Dictionary (1966)) (citing Because, Oxford English Dictionary (1st ed. 1933)).
  62. . See id. at 176 (relying on Bridge v. Phoenix Bond & Indemnity Co., 553 U.S. 639, 653–54 (2008), and Safeco Insurance Co. of America v. Burr, 551 U.S. 47, 63–64, 64 n.14 (2007), in arguing that “the phrase, ‘by reason of,’ requires at least a showing of ‘but for’ causation”). Justice Thomas also pointed to past ADEA cases, which seemed to require plaintiffs to prove but-for causation. Id. at 177 (first citing Ky. Ret. Sys. v. EEOC, 554 U.S. 135, 139–43, 148–50 (2008); and then citing Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 141, 143 (2000)).
  63. . Nassar, 570 U.S. at 352.
  64. . See id. at 346–47 (concluding that Congress “is presumed to have incorporated [this legal context], absent an indication to the contrary”).
  65. . 435 U.S. 702 (1978).
  66. . Id. at 704; Nassar, 570 U.S. at 346.
  67. . Manhart, 435 U.S. at 711 (quoting Developments in the Law, Employment Discrimination and Title VII of the Civil Rights Act of 1964, 84 Harv. L. Rev. 1109, 1170 (1971)); see Nassar, 570 U.S. at 346 (quoting id.).
  68. . Bostock v. Clayton County, 140 S. Ct. 1731, 1739 (2020) (quoting Nassar, 570 U.S. at 350 (citing Gross v. FBL Fin. Servs., 557 U.S. 167, 176 (2009))).
  69. . Id. (quoting Nassar, 570 U.S. at 346, 360). Some scholars have wondered whether Bostock properly relied on Nassar and Gross. Ben Eidelson, for example, suggests that those earlier cases emphasized the “ordinary meaning” of Title VII’s “because of” language (and did so in mixed-motives scenarios), rather than a more “technical” legal meaning. See Eidelson, supra note 28, at 853 & n.292 (suggesting that the earlier cases relied on the meaning that would be applied “in ordinary speech”). I read those cases differently. Both cases, and particularly Nassar, turned not only to dictionary definitions but also to prior Supreme Court precedent and the common law to determine the meaning of “because” and “because of.” Nassar, 570 U.S. at 346–51; Gross, 557 U.S. at 176–77. Perhaps one might still ask whether those cases are distinguishable, on the ground that they were mixed-motive cases. But other scholars have agreed that Nassar and Gross properly informed the interpretation of “because of” in Bostock. See, e.g., Katie Eyer, The But-For Theory of Anti-Discrimination Law, 107 Va. L. Rev. 1621, 1646–47 (2021) (“[B]ecause of the advent of cases like Gross and Nassar, plaintiffs and their amici were able to argue that the but-for principle was textually mandated, and must control.”). And as noted below, none of the Justices (not even the dissenting Justices) doubted that a but-for test should apply. See infra note 153 and accompanying text.
  70. . 487 U.S. 552 (1988) (Scalia, J.).
  71. . Id. at 556 (quoting 28 U.S.C. § 2412(d)(1)(A)).
  72. . Id. at 564.
  73. . Id. (alteration in original) (citation omitted) (quoting Substantial, Webster’s New International Dictionary (2d ed. 1945)).
  74. . Brief for the Respondents at 28, Pierce, 487 U.S. 552 (No. 86-1512).
  75. . Pierce, 487 U.S. at 564 (quoting Brief for the Petitioner at 16, Pierce, 487 U.S. 552 (No. 86-1512)).
  76. . See id. (“We are not, however, dealing with a field of law that provides no guidance in this matter.”).
  77. . Id. at 564–65 (quoting Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)).
  78. . See id. at 565 (noting that the lower standard was also consistent with the rule applied in another “related” area: sanctions for resisting discovery under Federal Rule of Civil Procedure 37).
  79. . Id. at 565. Justice Scalia’s opinion went on to affirm the district court’s decision that the government’s position in the underlying litigation was not “substantially justified.” See id. at 570–71 (affirming the Ninth Circuit’s holding that the district court did not abuse its discretion); see also id. at 571–74 (discussing and remanding as to the amount of fees).
  80. . 142 S. Ct. 1783 (2022) (Thomas, J.).
  81. . Id. at 1787.
  82. . Id.
  83. . Id. (quoting 9 U.S.C. § 1).
  84. . Id. at 1792.
  85. . Brief for Petitioner at 8, 33, Saxon, 142 S. Ct. 1783 (No. 21-309) (citation omitted).
  86. . See Brief of Respondent at 1–2, 24, Saxon, 142 S. Ct. 1783 (No. 21-309) (arguing that “loading and unloading is, by definition, an activity ‘within the flow of interstate commerce’”).
  87. . Id. at 20–21.
  88. . Id. at 1–2; see also id. at 20–24 (citing precedent).
  89. . See Saxon, 142 S. Ct. at 1793 (holding that the exemption applied to Saxon).
  90. . Id. at 1792.
  91. . Id. at 1789 (first quoting Balt. & Ohio Sw. R.R. Co. v. Burtch, 263 U.S. 540, 544 (1924); and then quoting 9 U.S.C. § 1). Under the Federal Employers Liability Act (FELA), railroads were required to compensate employees who were injured while the railroad was “engaging in commerce” and the employee was “employed by such carrier in such commerce.” Act of Apr. 22, 1908, ch. 149, § 1, 35 Stat. 65, 65; see Burtch, 263 U.S. at 542–44 (finding train loaders engaged in commerce such that the case was governed by FELA rather than by state law).
  92. . See Saxon, 142 S. Ct. at 1789–90 (relying on statutory context; Circuit City Stores, Inc. v. Adams, 532 U.S. 105, 119 (2001), which held the FAA exemption applied only to transportation workers; and Erie Railroad Co. v. Shuart, 250 U.S. 465, 468 (1919), which further demonstrated that “[c]argo loaders” are “intimately involved with the commerce (e.g., transportation) of that cargo”); see also Erie Railroad, 250 U.S. at 467–68 (relying on a definition of “transportation” in the Hepburn Act in determining the contractual liability of a railroad).
  93. . See Saxon, 142 S. Ct. at 1792–93 (“Southwest falls back on statutory purpose.”).
  94. . Id. (emphasis added).
  95. . See, e.g., FCC v. Am. Broad. Co., 347 U.S. 284, 296 (1954) (“If we should give [the provision] the broad construction urged by the [FCC in this civil case], the same construction would likewise apply in criminal cases.”); United States ex rel. Marcus v. Hess, 317 U.S. 537, 541–42 (1943) (suggesting that “the same substantive language” will be treated the same way in criminal prosecutions and civil actions under the False Claims Act), superseded by statute, Act of Dec. 23, 1943, ch. 377, 57 Stat. 608, as recognized in Graham Cnty. Soil & Water Conservation Dist. v. United States ex rel. Wilson, 559 U.S. 280 (2010); see also Leocal v. Ashcroft, 543 U.S. 1, 11 n.8 (2004) (Rehnquist, C.J.) (“[W]e must interpret the statute consistently, . . . in a criminal or noncriminal context . . . .”). For scholarly studies, see Margaret V. Sachs, Harmonizing Civil and Criminal Enforcement of Federal Regulatory Statutes: The Case of the Securities Exchange Act of 1934, 2001 U. Ill. L. Rev. 1025, 1026–27, 1029–30, which argues that courts should read “prohibitions in hybrid statutes” uniformly. See also Lawrence M. Solan, Statutory Inflation and Institutional Choice, 44 Wm. & Mary L. Rev. 2209, 2218, 2272 (2003) (noting the practice of uniform interpretation).
  96. . FCC, 347 U.S. at 296.
  97. . 543 U.S. 371 (2005).
  98. . Id. at 373.
  99. . 8 U.S.C. § 1231(a)(1)(A). As the Court noted in Clark, the Attorney General’s authority to detain under this provision had been transferred to the Secretary of the Department of Homeland Security. See Clark, 543 U.S. at 377.
  100. . 8 U.S.C. § 1231(a)(6) (emphasis added); see also id. § 1231(a)(3) (establishing guidelines for supervised release).
  101. . See Clark, 543 U.S. at 377–78 (“By its terms, this provision applies to three categories . . . .”).
  102. . See id. at 373 (“These cases concern the Secretary [of Homeland Security]’s authority to continue to detain an inadmissible alien subject to a removal order after the 90-day removal period has elapsed.”).
  103. . Zadvydas v. Davis, 533 U.S. 678, 682 (2001).
  104. . 533 U.S. 678 (2001).
  105. . Id. at 682, 697.
  106. . See id. at 682, 688–93, 697, 699–700 (spelling out these concerns).
  107. . Clark, 543 U.S. at 377–78.
  108. . Id. at 378.
  109. . See id. at 380 (noting that even if “the constitutional concerns . . . are not present” in this case, “it cannot justify giving the same detention provision a different meaning”).
  110. . Id. at 380.
  111. . Id. at 386. The Court went on to hold that the government should have the same presumptive six-month period in which to remove an undocumented immigrant as was recognized in Zadvydas. Id. at 386–87.
  112. . Zadvydas v. Davis, 533 U.S. 678, 707–08, 710, 716 (2001) (Kennedy, J., dissenting, joined by Chief Justice Rehnquist and Justices Scalia and Thomas); see also id. at 710–11 (urging that it would not be “a plausible construction” to treat the three classes differently).
  113. . Indeed, that is also the pattern in the civil–criminal context. If the Court finds a statutory provision ambiguous in a civil enforcement action, the Court applies the rule of lenity—on the assumption that its interpretation will bind future criminal cases. See United States v. Thompson/Ctr. Arms Co., 504 U.S. 505, 518 n.10 (1992) (Souter, J.) (plurality opinion) (applying the rule of lenity because the “tax statute has criminal applications”).
  114. . Clark, 543 U.S. at 388 (Thomas, J., dissenting). The basis of Justice Thomas’s dissent is not entirely clear. He seemed to disagree with the majority’s conclusion that it was bound by a past interpretation of the same statutory language. But his dissent focused less on that principle and more on his view that the majority was either misreading—or should just overrule—Zadvydas. See id. at 388, 391–92, 401–04 (asserting that “the Court’s analysis cannot be squared with Zadvydas” and “even if it could be so squared, Zadvydas was wrongly decided and should be overruled”); see also id. at 393 (“If the majority is correct that the ‘lowest common denominator’ governs, then the careful distinction Zadvydas drew between admitted aliens and nonadmitted aliens was irrelevant at best and misleading at worst.”). As this Article shows, Justice Thomas has often relied on precedent in determining the meaning of statutory terms and phrases. So he clearly accepts the practice as a general matter.
  115. . See supra subpart I(A); see also, e.g., Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 547 U.S. 71, 85–86 (2006) (Stevens, J.) (“[W]hen ‘judicial interpretations have settled the meaning of an existing statutory provision, repetition of the same language in a new statute indicates, as a general matter, the intent to incorporate its . . . judicial interpretations as well.’” (second alteration in original) (quoting Bragdon v. Abbott, 524 U.S. 624, 645 (1998))); Felix Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. 527, 537 (1947) (“[I]f a word is obviously transplanted from another legal source, whether the common law or other legislation, it brings the old soil with it.”).
  116. . See supra subpart I(A).
  117. . This paragraph draws on the sources collected supra notes 1–3, 9, 28, 30.
  118. . See supra notes 21, 41–43 and accompanying text.
  119. . See supra note 44 (collecting sources). Anita Krishnakumar has criticized the Supreme Court’s use of the common law, but she acknowledges that such case law may help to understand legal terms of art. See Krishnakumar, supra note 28, at 668–69 (arguing for the use of common law meaning when the statute contains a legal term of art, on the assumption that such cases would be fairly small in number).
  120. . See Manning, supra note 43, at 112 (describing “terms of art” as “phrases that acquire specialized meaning through use over time as the shared language of specialized communities (legal, commercial, scientific, etc.)”); Lawrence B. Solum, Originalism and Constitutional Construction, 82 Fordham L. Rev. 453, 503–04 (2013) (exploring the meaning of “terms of art”); David Mellinkoff, The Language of the Law 16–17, 391 (1963) (discussing the use of “terms of art” and stating that a legal “term of art” is one that through recurrent use “conveys a workably clear notion to the lawyer”).
  121. . 283 U.S. 488 (1931).
  122. . Id. at 490–91.
  123. . The two revenue provisions were the same, except that candy was subject to a 5% luxury tax under the 1918 Act and a 3% tax under the 1921 Act. Revenue Act of 1918, Pub. L. No. 65-254, § 900(9), 40 Stat. 1057, 1122 (1919); Revenue Act of 1921, Pub. L. No. 67-98, § 900(6), 42 Stat. 227, 292; McCaughn, 283 U.S. at 489–90.
  124. . Ultimately, although the Supreme Court acknowledged that “the word ‘candy,’ . . . may be used in [the] narrower and more restricted sense” urged by the Hershey Company, the Court found that, in the context of the Revenue Acts, it was used “in a popular and more general sense” and embraced Hershey chocolate. McCaughn, 283 U.S. at 491.
  125. . Under current Article III standing doctrine, a private party must demonstrate a concrete injury that was caused by the defendant and that can be redressed by the requested relief. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560–61 (1992).
  126. . See, e.g., 2 U.S.C. § 288e(a) (“The [Senate] Counsel shall be authorized to intervene only if standing to intervene exists under . . . article III . . . .”).
  127. . See, e.g., Fraud, Oxford English Dictionary (2023) (defining “fraud” in part as “[o]ne who is not what he appears to be; an impostor, a humbug”); Fraud, Merriam-Webster Dictionary (2023) (defining “fraud” as “one that is not what it seems or is represented to be”).
  128. . See Restatement (Second) of Torts §§ 525–31 (Am. L. Inst. 1977) (describing the contours of the legal concept of “fraud”).
  129. . 516 U.S. 59 (1995).
  130. . Id. at 69 (Souter, J.).
  131. . Instead, the Court simply asserted that the statutory terms were “common-law terms” that “carr[ied] the acquired meaning of terms of art.” Id. at 69; see also id. at 70 (“[T]here is no reason to doubt Congress’s intent to adopt a common-law understanding of the terms . . . .”). The Court thus presupposed that the legal meaning, rather than the ordinary meaning, applied.
  132. . See id. at 79–80 (Breyer, J., joined by Scalia, J., dissenting) (agreeing that the common law meaning applied, and dissenting on the ground that the bankruptcy court actually applied the correct standard, even if it used the wrong words).
  133. . The dispute in the case was not over whether the common law concept applied but over one element of that common law concept: the type of reliance necessary. See Petitioners’ Brief at 8, 19–20, Field, 516 U.S. 59 (No. 94-967) (arguing for justifiable reliance); Respondents’ Brief on the Merits at 9, Field, 516 U.S. 59 (No. 94-967) (advocating reasonable reliance); see also Field, 516 U.S. at 74–75 (holding that 11 U.S.C. § 523(a)(2)(A) “requires justifiable, but not reasonable, reliance”).
  134. . See John F. Manning, Without the Pretense of Legislative Intent, 130 Harv. L. Rev. 2397, 2400 (2017) (noting various approaches, including purposivism and textualism, decline to focus on “what Congress intended about the problem at hand”); Philip P. Frickey, From the Big Sleep to the Big Heat: The Revival of Theory in Statutory Interpretation, 77 Minn. L. Rev. 241, 248 (1992) (noting the influence of the legal realist position that “legislative intent is a myth . . . . How can 535 legislators have an intent about anything?”).
  135. . See, e.g., John F. Manning, Why Does Congress Vote on Some Texts But Not Others?, 51 Tulsa L. Rev. 559, 570–71 (2016) (arguing that “legislative history may be particularly useful in helping a judge figure out the meaning of ‘a specialized term or phrase’”).
  136. . See, e.g., Scalia & Garner, supra note 1, at 15–16 (advocating a focus on “full context”); Manning, supra note 23, at 76 (“Textualists give precedence to semantic context . . . .”).
  137. . See Brian G. Slocum, Replacing the Flawed Chevron Standard, 60 Wm. & Mary L. Rev. 195, 252–53 (2018) (“Natural ‘[l]anguage is full of nonliteral meanings, such as metaphors, idioms, slang, and polite talk.’ . . . [S]uch usages of language are far less common in legal texts.” (first alteration in original) (footnote omitted) (quoting Brian G. Slocum, Ordinary Meaning 26 (2015))); Brian Flanagan, Revisiting the Contribution of Literal Meaning to Legal Meaning, 30 Oxford J. Legal Stud. 255, 260–61 (2010) (suggesting that there may be “a convention whereby” legislative utterances “lack ironic, counter-attributive or metaphorical meanings”); Saikrishna Prakash, Our Three Commerce Clauses and the Presumption of Intrasentence Uniformity, 55 Ark. L. Rev. 1149, 1158 (2003) (“We certainly do not expect double entendres (bawdy or otherwise) in the Constitution.”).
  138. . Theorists have understood this point for a long time. See Francis Lieber, Legal and Political Hermeneutics 96 n.1 (Boston, Charles C. Little & James Brown 1839) (noting that, in contrast to theatrical performances, “the object of law and politics is neither to amuse or touch”); Ryan D. Doerfler, Can a Statute Have More than One Meaning?, 94 N.Y.U. L. Rev. 213, 220 (2019) (noting that “legislative language typically lacks” “humorous effect”); Jonathan R. Siegel, The Polymorphic Principle and the Judicial Role in Statutory Interpretation, 84 Texas. L. Rev. 339, 366–67 (2005) (“Humor is not a quality one typically associates with statutes . . . .”).
  139. . Indeed, according to a recent empirical study, lay people expect even seemingly ordinary terms and phrases, such as “intent” and “because of,” to take on a distinctively legal meaning when used in a federal statute. See Kevin Tobia, Brian G. Slocum & Victoria Nourse, Ordinary Meaning and Ordinary People, 171 U. Pa. L. Rev. 365, 373, 430–31 (2023) (concluding that “to interpret ‘because of’ in line with its ordinary meaning . . . would be a disservice to the people who expect it to take its announced legal meaning”); infra subpart III(A).
  140. . See Carcieri v. Salazar, 555 U.S. 379, 381–82, 388–90, 395–96 (2009) (Thomas, J.) (interpreting the term “now” as used in the Indian Reorganization Act).
  141. . See Rimini St., Inc. v. Oracle USA, Inc., 139 S. Ct. 873, 877–78 (2019) (Kavanaugh, J.) (interpreting the term “costs” as used in the Copyright Act).
  142. . 42 U.S.C. § 2000e-2(a).
  143. . Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 350 (2013) (alteration in original) (quoting Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 176 (2009)).
  144. . Bostock v. Clayton County, 140 S. Ct. 1731, 1739 (2020).
  145. . 9 U.S.C. § 1.
  146. . Law school casebooks spend a good deal of space on the subject. See, e.g., Noah R. Feldman & Kathleen M. Sullivan, Constitutional Law 115–92 (21st ed. 2022).
  147. . See Sw. Airlines Co. v. Saxon, 142 S. Ct. 1783, 1787, 1792 (2022) (Thomas, J.) (resolving the question whether an employee was engaged in foreign or interstate commerce in part by relying on Baltimore & Ohio Southwestern Railroad Co. v. Burtch, 263 U.S. 540 (1924)).
  148. . Manning, supra note 43, at 111 (quoting Cabell v. Markham, 148 F.2d 737, 739 (2d Cir. 1945) (Hand, J.)); see also Cabell, 148 F.2d at 739 (stating that “a mature and developed jurisprudence” does not “make a fortress out of the dictionary”). Nevertheless, textualists’ use of dictionaries has drawn a good deal of criticism. See Ellen P. Aprill, The Law of the Word: Dictionary Shopping in the Supreme Court, 30 Ariz. St. L.J. 275, 280–81 (1998) (arguing that textualists’ use of dictionaries is “selective and inconsistent”); James J. Brudney & Lawrence Baum, Oasis or Mirage: The Supreme Court’s Thirst for Dictionaries in the Rehnquist and Roberts Eras, 55 Wm. & Mary L. Rev. 483, 489, 493–94 (2013) (“[D]ictionaries add at most modest value to the interpretive enterprise, and . . . they are being overused and often abused by the Court.”); cf. Abbe R. Gluck & Lisa Schultz Bressman, Statutory Interpretation from the Inside—An Empirical Study of Congressional Drafting, Delegation, and the Canons: Part I, 65 Stan. L. Rev. 901, 907 (2013) (finding that legislative staffers “do not consult dictionaries when drafting”).
  149. . See Easterbrook, supra note 43, at 67 (arguing that a dictionary “is a museum of words, an historical catalog rather than a means to decode the work of legislatures”); John F. Manning, The Absurdity Doctrine, 116 Harv. L. Rev. 2387, 2456–57, 2457 n.254 (2003) (“Although textualists (like other interpreters) frequently consult dictionaries as historical records of social meanings that speakers have attached to words, they do not (and could not) stop there.”). Indeed, in this respect, textualists do not differ from their purposivist counterparts. See Henry M. Hart, Jr. & Albert M. Sacks, The Legal Process 1375–76 (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994) (arguing that “[u]nabridged dictionaries are historical records,” which help show whether “a particular meaning is linguistically permissible”).
  150. . See supra note 43 and accompanying text. Some scholars have criticized the Court’s reliance on the “whole code” to make sense of statutory provisions. But these scholars still endorse reliance on related statutes (those in pari materia). E.g., William W. Buzbee, The One-Congress Fiction in Statutory Interpretation, 149 U. Pa. L. Rev. 171, 173–76, 246 (2000); Anita S. Krishnakumar, Cracking the Whole Code Rule, 96 N.Y.U. L. Rev. 76, 82–84, 153–55 (2021).
  151. . See, e.g., Blackman & Barnett, supra note 9 (criticizing Bostock’s reliance on cases from 2009 and 2013—Gross and Nassar—in examining the meaning of the 1964 Civil Rights Act).
  152. . To be sure, precedent-as-meaning (like any interpretive technique) can be misused. For that reason, I believe there should be guidelines for this practice. See infra section I(B)(3); infra subpart III(B).
  153. . E.g., Berman & Krishnamurthi, supra note 28, at 94; Eyer, supra note 69, at 1684; Eidelson, supra note 28, at 794–96; D’Andra Millsap Shu, The Coming Causation Revolution in Employment Discrimination Litigation, 43 Cardozo L. Rev. 1807, 1814 (2022).
  154. . Bostock v. Clayton County, 140 S. Ct. 1731, 1739 (2020).
  155. . Frank H. Easterbrook, The Role of Original Intent in Statutory Construction, 11 Harv. J.L. & Pub. Pol’y 59, 65 (1988) (emphasizing “the understanding of the objectively reasonable person”); see Scalia, supra note 41, at 17 (emphasizing “the intent that a reasonable person would gather from the text of the law”); Scalia & Garner, supra note 1, at 15–16 (“In their full context, words mean what they conveyed to reasonable people at the time they were written . . . .”); infra notes 156–59159 and accompanying text.
  156. . Gorsuch, supra note 6, at 51; see id. at 55–56 (arguing that the judge should ask, “What might a reasonable person have thought the law was at the time?”); Neil M. Gorsuch, Lecture, Of Lions and Bears, Judges and Legislators, and the Legacy of Justice Scalia, 66 Case W. Rsrv. L. Rev. 905, 910 (2016) (“As the founders understood it, the task of the judge is to interpret and apply the law as a reasonable and reasonably well-informed citizen might have done when engaged in the activity underlying the case or controversy . . . .”).
  157. . Scalia & Garner, supra note 1, at 33. The writers also state that such a reasonable reader may consider purpose but only as derived from the text itself. Id.
  158. . Manning, supra note 149, at 2458 (quoting Easterbrook, supra note 155, at 65).
  159. . Manning, supra note 23, at 83 (“[T]he statutory meaning derived by textualists is a construct. Textualists do not . . . claim that a constitutionally sufficient majority of legislators actually subscribed to the meaning that a textualist judge would ascribe to a hypothetical reasonable legislator conversant with the applicable social and linguistic conventions.”); see Amy Coney Barrett, Congressional Insiders and Outsiders, 84 U. Chi. L. Rev. 2193, 2196, 2200–04, 2211 (2017) (arguing that surveys of congressional staffers do not properly test textualism, because “textualists use the construct of a hypothetical reader,” not “the construct of a hypothetical writer of a statute,” and noting that textualists have “identified their construct as a skilled user of language, typically familiar with legal conventions”). Justice Barrett recognizes that textualists have not clearly explained whether they refer to “the perspective of the ‘ordinary lawyer’” or “the ordinary English speaker.” Id. at 2209; see also infra subpart III(A).
  160. . Consider the range of what are known as linguistic canons. See Scalia & Garner, supra note 1, at xii–xiv (listing a few dozen canons as “semantic,” “syntactic,” or “contextual”). Although some scholars assume that such canons are valid only to the extent that they mirror usage in ordinary conversation, my own view is that such canons may be useful in interpreting legal instruments, whether or not they accurately depict ordinary conversation. A full argument regarding the canons is beyond the scope of this Article, but I plan to address the issue in future work.
  161. . See, e.g., Milner v. Dep’t of the Navy, 562 U.S. 562, 569–70 (2011) (Kagan, J.) (finding that “personnel” means the same thing across provisions of the Freedom of Information Act); Brown v. Gardner, 513 U.S. 115, 118 (1994) (Souter, J.) (noting that “there is a presumption that a given term is used to mean the same thing throughout a statute, a presumption surely at its most vigorous when a term is repeated within a given sentence” (citation omitted)).
  162. . Ratzlaf v. United States, 510 U.S. 135, 143 (1994) (Ginsburg, J.).
  163. . Siegel, supra note 138, at 366.
  164. . Id.
  165. . Id. at 366–67.
  166. . Doerfler, supra note 138, at 222; see also id. at 215 n.5 (“This Article . . . assumes that what a statute ‘means’ in a linguistic sense corresponds, at least presumptively, to what it ‘means’ in a legal sense—that is, to its legal effect.”).
  167. . Id. at 218.
  168. . This example strikes me as questionable even as a linguistic matter. It seems that both A and B could understand that the advertisement is directed at any listener. Presumably, Uncle Sam would like as many people as possible to join the army. But that linguistic argument is not central here. For an article that agrees with Doerfler’s linguistic understanding and defends the singular reading, while recognizing that others may read “you” as addressed to a group, see Andy Egan, Billboards, Bombs and Shotgun Weddings, 166 Synthese 251, 261–65 (2009).
  169. . 42 U.S.C. § 2000e-2(a) (“It shall be an unlawful employment practice for an employer— . . . to discriminate against any individual . . . because of such individual’s race, color, religion, sex, or national origin . . . .”).
  170. . 8 U.S.C. § 1231(a)(6).
  171. . Congress could, for example, have created different requirements for small and large businesses. Instead, Title VII excludes any “employer” with fewer than fifteen employees. 42 U.S.C. § 2000e(b).
  172. . See Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 353 (2013) (Kennedy, J.) (“Just as Congress’ choice of words is presumed to be deliberate, so too are its structural choices.”); SAS Inst. Inc. v. Iancu, 138 S. Ct. 1348, 1355 (2018) (Gorsuch, J.) (stating that Congress’s structural choices are “deserving of judicial respect” (citing Nassar, 570 U.S. at 353)); John F. Manning, Federalism and the Generality Problem in Constitutional Interpretation, 122 Harv. L. Rev. 2003, 2007 (2009) (asserting that interpreters should respect Congress’s “specific choices . . . about how [statutory] goals are to be achieved”).
  173. . See Richard H. Fallon, Jr., “The Rule of Law” as a Concept in Constitutional Discourse, 97 Colum. L. Rev. 1, 8–9 (1997) (offering five elements of the rule of law, including predictability, efficacy, stability, “supremacy of legal authority,” and the need for “instrumentalities of impartial justice”); Jeremy Waldron, Stare Decisis and the Rule of Law: A Layered Approach, 111 Mich. L. Rev. 1, 3 (2012) (listing justifications for stare decisis, including stability and fairness); see also Lon L. Fuller, The Morality of Law 38–39, 42 (rev. ed. 1969) (describing eight requirements for a system of legal rules, including generality, publicity, prospectivity, clarity, coherence, stability, and “congruence between the rules as announced and their actual administration”). Some textualists argue that the interpretive approach serves rule of law values. See, e.g., Scalia & Garner, supra note 1, at xxix (“Textualism will not relieve judges of all doubts and misgivings about their interpretations. . . . But textualism will provide greater certainty in the law, and hence greater predictability and greater respect for the rule of law.”); Scalia, supra note 41, at 25 (arguing that textualism and the rule of law are formalistic); Manning, supra note 149, at 2391 (suggesting that a strict form of textualism that rejects the absurdity doctrine serves “a rule-of-law tradition”). That important question is beyond the scope of this Article, but my preliminary suspicion is that any rule of law argument for textualism very much depends on which version of textualism a judge adopts. See Grove, supra note 38, at 266–71, 279–90 (describing different versions of textualism).
  174. . See, e.g., Easterbrook, supra note 43, at 67, 69 (arguing that textualism serves as a basis for the proper relationship between the judiciary and legislature); Grove, supra note 38, at 290–307 (“Many (albeit not all) textualists put a premium on cabining judicial discretion.”); John F. Manning, Justice Scalia and the Idea of Judicial Restraint, 115 Mich. L. Rev. 747, 749–50, 770, 781 (2017) (reviewing Scalia, supra note 41) (asserting that Justice Scalia’s preference for textualism was driven by an “anti-discretion principle”); Nelson, supra note 41, at 403 (noting many textualists’ skepticism about judicial discretion); see also Adrian Vermeule, Judging Under Uncertainty 4–5 (2006) (advocating textualism based on concerns about the judiciary’s limited institutional capacities).
  175. . See Fuller, supra note 173, at 42 (noting that “[t]he demands of the inner morality of law” include the requirement that “your decisions as an official are guided by it”); Fallon, supra note 173, at 8 (identifying, as one crucial element of the rule of law, “the supremacy of legal authority,” for “[t]he law should rule officials, including judges, as well as ordinary citizens”); Waldron, supra note 173, at 3 (“The rule of law requires people in positions of authority to exercise their power under the authority, and within a constraining framework, of public norms (laws) rather than on the basis of their own preferences or ideology . . . .”); see also Kozel, supra note 32, at 1858 (“[T]he infusion of precedent with durability that outlasts the tenure of the issuing judges facilitates both the reality and appearance of decisionmaking that is driven by considerations beyond individual personalities.”).
  176. . See supra notes 32–33 and accompanying text. For example, the one-meaning rule did not prevent the Clark Court from reconsidering Zadvydas as to all three classes of undocumented immigrants. But with the exception of Justice Thomas, see supra note 114 and accompanying text, no member of the Court entertained that option, perhaps because the government did not request such an overruling. See Brief for the Petitioners at 13–14, 28, 36, Clark v. Martinez, 543 U.S. 371 (2005) (No. 03-878) (arguing, among other things, that the Court should not extend Zadvydas).
  177. . See Nelson, supra note 41, at 367–68 (“[T]he typical textualist judge seeks to unearth the statutes’ original meanings . . . .”). Notably, building on prior work, I refer to original meaning, not to how the public at the time of enactment might have expected a statute to apply. See Grove, supra note 38, at 303–04 (advocating formalistic textualism, which would instruct judges not to consider certain evidence, such as social context: past public understandings or expectations about how a statute would apply).
  178. . Clark’s reliance on Zadvydas is, in my view, consistent with this principle. Zadvydas considered whether “may be detained beyond the removal period” was ambiguous (such that the avoidance canon would apply), not whether the Court should “update” the statute.
  179. . That could lead a judge to be more thoughtful about the precedent(s) that she establishes. But whether or not precedent has that constraining impact, the new precedent will shape the future course of the law. See Frederick Schauer, Precedent, 39 Stan. L. Rev. 571, 572–73, 589 (1987) (asserting that “the conscientious decisionmaker” must consider how “future conscientious decisionmakers will treat” a precedent); Evan H. Caminker, Precedent and Prediction: The Forward-Looking Aspects of Inferior Court Decisionmaking, 73 Texas L. Rev. 1, 11–12 (1994) (“[T]he Court must judge not only what is best for today, but also how the current decision will affect the decision of other cases in the future.”); Waldron, supra note 173, at 2 (focusing on the “forward-looking perspective [of precedent] to explicate the relation between stare decisis and the rule of law”).
  180. . Bostock v. Clayton County, 140 S. Ct. 1731, 1778 (2020) (Alito, J., dissenting); see also id. (“The Court’s brusque refusal to consider the consequences of its reasoning is irresponsible.”).
  181. . Stare decisis is said to have its most important bite when it constrains judges who view the prior decision as incorrect. See Larry Alexander, Constrained by Precedent, 63 S. Cal. L. Rev. 1, 4 (1989) (stating that precedent requires a court to decide a case in front of it based on a prior decision although it believes the prior decision was incorrect); Schauer, supra note 179, at 575 (“[I]f we are truly arguing from precedent, then the fact that something was decided before gives it present value despite our current belief that the previous decision was erroneous.”).
  182. . See infra subpart II(C).
  183. . See Erwin Chemerinsky, Abandoning the Courts, Trial, July 2011, at 50, 52 (arguing that the Court’s trend of favoring arbitration “will have a devastating effect on many class actions”); Hila Keren, Divided and Conquered: The Neoliberal Roots and Emotional Consequences of the Arbitration Revolution, 72 Fla. L. Rev. 575, 601–08 (2020) (discussing the neoliberal nature of the so-called “arbitration revolution”).
  184. . See Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 362–63 (2013) (Kennedy, J.) (ruling against the plaintiff because the “text, structure, and history of Title VII” require a but-for showing); Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 180 (2009) (Thomas, J.) (holding that the lower courts should have required the plaintiff to prove but-for causation). For just one example of the criticisms of Gross, see, for example, Deborah A. Widiss, Undermining Congressional Overrides: The Hydra Problem in Statutory Interpretation, 90 Texas L. Rev. 859, 908–26 (2012).
  185. . For an article that did anticipate this application of Gross and Nassar, see Katie R. Eyer, Statutory Originalism and LGBT Rights, 54 Wake Forest L. Rev. 63, 78–79, 78 n.65 (2019). Cf. Katie Eyer, Disentangling Textualism and Originalism, 13 ConLawNow, no. 1, 2021–2022, at 115, 130 (arguing that, absent consistency across cases such as Gross, Nassar, and Bostock, textualism cannot maintain a reputation for “neutrality and . . . restrain[t]”).
  186. . To be sure, there will be disagreements about when a statute has a “plain meaning”—in part because there are debates among textualists as to what evidence goes into the first stage of the statutory analysis. See supra note 40 and accompanying text. For purposes of this Article’s typology, it is enough that textualists do sometimes find a statute to have and not to have a plain meaning, as illustrated by the cases discussed in subparts I(A) and II(A).
  187. . I do not claim, as a normative matter, that a textualist must always rely on precedent at the second stage. At times, there may be no case on point. I argue that reliance on the second and third categories is permissible, when a textualist concludes that there is no plain meaning.
  188. . 443 U.S. 193 (1979).
  189. . 519 U.S. 337 (1997).
  190. . Id. at 339.
  191. . Id. at 340–41.
  192. . Id. at 345–46.
  193. . 553 U.S. 507 (2008).
  194. . See id. at 509, 511–14 (plurality opinion) (construing in the defendant’s favor under the rule of lenity because, “[f]rom the face of the statute, there is no more reason to think that ‘proceeds’ means ‘receipts’ than there is to think that ‘proceeds’ means ‘profits’”).
  195. . Id. at 514 (“Under a long line of our decisions, the tie must go to the defendant.”). Ben Eidelson and Matthew Stephenson have recently argued that textualists should abandon substantive canons. See Eidelson & Stephenson, supra note 21, at 519–21 (“[S]ubstantive canons are generally just as incompatible with textualists’ jurisprudential commitments as they first appear.”). Their argument seems to focus on what I call the first stage of the statutory analysis—the plain meaning stage. On that understanding, their argument is not inconsistent with the discussion here.
  196. . 137 S. Ct. 420 (2016) (Alito, J.).
  197. . See id. at 424 (noting that Maher sought to “appease” his brother, “who pestered him incessantly” for the information).
  198. . See id. at 425 (explaining the Court’s intent to resolve a circuit split over whether a tipper must have received a personal benefit or additional monetary gain in order for a tippee to be held liable).
  199. . Brief for Petitioner at 20–22, Salman, 137 S. Ct. 420 (No. 15-628) (“If the Court were inclined to reconsider its prior cases, it could readily hold, based on the plain language of the statute, that § 10(b) does not prohibit insider trading at all . . . .”).
  200. . Brief for the United States at 12, Salman, 137 S. Ct. 420 (No. 15-628); see also id. (urging that the prohibition “does not reflect judge-made law unmoored from the text”).
  201. . Salman, 137 S. Ct. at 423; see also 15 U.S.C. § 78j(b) (prohibiting “any manipulative or deceptive device” “in connection with the purchase or sale of any security” in violation of SEC rules); 17 C.F.R. § 240.10b-5(a), (b) (2022) (making it unlawful “[t]o employ any device, scheme, or artifice to defraud,” and “[t]o make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in the light of the circumstances under which they were made, not misleading”).
  202. . See Salman, 137 S. Ct. at 423, 427 (relying on Dirks v. SEC, 463 U.S. 646 (1983)).
  203. . 463 U.S. 646 (1983).
  204. . Id. at 662.
  205. . Id. at 664.
  206. . Salman, 137 S. Ct. at 427–28.
  207. . See id. at 429 (“[A]s Salman conceded below, this evidence is sufficient to sustain his conviction under our reading of Dirks.”).
  208. . See Anuj C. Desai, Text Is Not Enough, 93 U. Colo. L. Rev. 1, 2–3 (2022) (asserting that Bostock had “nothing to do with textualism” or “semantic meaning”); Cary Franklin, Living Textualism, 2020 Sup. Ct. Rev. 119, 139–40, 151 (arguing that there was no single “ordinary meaning” in Bostock).
  209. . Bostock v. Clayton County, 140 S. Ct. 1731, 1738–39, 1743–44 (2020).
  210. . 400 U.S. 542 (1971) (per curiam).
  211. . Id. at 544. The Court, however, permitted the employer to show that hiring only men with young children was a “bona fide occupational qualification.” Id. (quoting Civil Rights Act of 1964, Pub. L. No. 88-352, § 703(e), 78 Stat. 241, 256); see also Cary Franklin, Inventing the “Traditional Concept” of Sex Discrimination, 125 Harv. L. Rev. 1307, 1356 (2012) (“[W]hat the Court gave [in Phillips], it then took away . . . .”).
  212. . 435 U.S. 702 (1978).
  213. . Id. at 711.
  214. . Bostock, 140 S. Ct. at 1744. Notably, the lower court in Phillips accepted that argument. See Phillips v. Martin Marietta Corp., 411 F.2d 1, 4 (5th Cir. 1969) (rejecting the sex discrimination claim, given that “[t]he discrimination was based on a two-pronged qualification, i.e., a woman with pre-school age children”), vacated, 400 U.S. 542 (1971) (per curiam).
  215. . Bostock, 140 S. Ct. at 1744.
  216. . Id.
  217. . 490 U.S. 228 (1989) (Brennan, J.), superseded by statute, Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071, as recognized in Comcast Corp. v. Nat’l Ass’n of Afr. Am.-Owned Media, 140 S. Ct. 1009 (2020).
  218. . Id. at 231–32, 235, 250–51 (plurality opinion).
  219. . See, e.g., William N. Eskridge Jr., Title VII’s Statutory History and the Sex Discrimination Argument for LGBT Workplace Protections, 127 Yale L.J. 322, 362, 381 (2017) (arguing that Title VII is meant to protect LGBT “employees against employer insistence upon conforming to old-fashioned, rigid gender roles”); Andrew Koppelman, Bostock, LGBT Discrimination, and the Subtractive Moves, 105 Minn. L. Rev. Headnotes 1, 9–11 (2020) (arguing that discrimination against LGBT employees is based on their gender nonconformity).
  220. . See Grove, supra note 38, at 266 (alterations in original) (quoting 42 U.S.C. § 2000e-2(a)(1)) (arguing that the text of Title VII “appeared to strongly favor the plaintiffs”).
  221. . Van Buren v. United States, 141 S. Ct. 1648, 1661 (2021) (quoting Yates v. United States, 574 U.S. 528, 557 (2015) (Kagan, J., dissenting)).
  222. . See sources cited supra note 9.
  223. . See, e.g., Johnson v. Transp. Agency, 480 U.S. 616, 629 n.7 (1987) (Brennan, J.) (“Congress has not amended the statute to reject our construction, . . . and we therefore may assume that our interpretation was correct.”); cf. Neal v. United States, 516 U.S. 284, 295 (1996) (Kennedy, J.) (“One reason that we give great weight to [statutory] stare decisis . . . is that ‘Congress is free to change this Court’s interpretation of its legislation.’” (quoting Ill. Brick Co. v. Illinois, 431 U.S. 720, 736 (1977))).
  224. . See sources cited supra note 16.
  225. . See Eskridge, supra note 15, at 1364, 1409–11 (describing the Court’s reliance on acquiescence as “a fallacious argument”); see also William N. Eskridge, Jr., Interpreting Legislative Inaction, 87 Mich. L. Rev. 67, 94–95, 98–99 (1988) (asserting that “legislative inaction rarely tells us much about relevant legislative intent”).
  226. . Krishnakumar, supra note 1, at 219.
  227. . Cf. Nelson, supra note 41, at 367 (“[T]he typical textualist judge seeks to unearth the statutes’ original meanings . . . .”).
  228. . See Nelson, supra note 1, at 4–5, 8, 21, 45 (tracing the history of the relationship between perceptions of indeterminacy and stare decisis).
  229. . For valuable scholarship exploring which reliance interests should count for stare decisis purposes, see generally Randy J. Kozel, Precedent and Reliance, 62 Emory L.J. 1459 (2013), and Nina Varsava, Precedent, Reliance, and Dobbs, 136 Harv. L. Rev. 1845 (2023). A full exploration of reliance is beyond the scope of this Article.
  230. . See Neal Devins & Lawrence Baum, The Company They Keep 117 (2019) (“The Federalist Society is also a leading propagator of textualism and originalism, two linked theories of legal interpretation that have . . . strengthened the conservative legal movement.”); Steven M. Teles, The Rise of the Conservative Legal Movement 145 (2008) (discussing the role the Federalist Society played in the evolution of textualism and originalism).
  231. . See Neil H. Buchanan & Michael C. Dorf, A Tale of Two Formalisms: How Law and Economics Mirrors Originalism and Textualism, 106 Cornell L. Rev. 591, 640 (2021) (arguing that statutory textualism and constitutional originalism are “a rhetorical smokescreen for extremely conservative results”); William N. Eskridge, Jr. & Philip P. Frickey, Foreword, Law as Equilibrium, 108 Harv. L. Rev. 26, 77 (1994) (arguing that “the new, tougher version of textualism advocated by Justices Scalia and Thomas exacerbates the tension between democracy and the rule of law and ultimately serves as a cover for the injection of conservative values into statutes”); Lemos, supra note 3, at 851 (observing that “textualism is widely regarded as a politically conservative methodology”); Alexander Volokh, Choosing Interpretive Methods: A Positive Theory of Judges and Everyone Else, 83 N.Y.U. L. Rev. 769, 771–75 (2008) (recognizing, but questioning, the “conventional wisdom” that “[t]extualism is a ‘conservative’ method”).
  232. . See Eyer, supra note 69, at 1624–27, 1661 (advocating a textual approach to Title VII to challenge “pathologies that currently plague” anti-discrimination law).
  233. . McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973).
  234. . See id. at 802–03 (establishing a burden-shifting framework for discrimination claims); Deborah A. Widiss, Proving Discrimination by the Text, 106 Minn. L. Rev. 353, 355–58 (2021) (arguing that McDonnell Douglas is “deeply in tension with” the statutory language).
  235. . See Alexander A. Reinert, Qualified Immunity’s Flawed Foundation, 111 Calif. L. Rev. 201, 208, 216–17 (2023) (arguing that “no qualified immunity doctrine at all should apply in Section 1983 actions, if courts stay true to the text adopted by the enacting Congress”); William Baude, Is Qualified Immunity Unlawful?, 106 Calif. L. Rev. 45, 77–78 (2018) (contending that qualified immunity under § 1983 is “far removed from ordinary principles of legal interpretation”); Jack M. Beermann, A Critical Approach to Section 1983 with Special Attention to Sources of Law, 42 Stan. L. Rev. 51, 70 (1989) (“The text of the statute plays no role, since it does not mention immunities.”). For a small sample of the (many) other criticisms of qualified immunity, see, for example, Katherine Mims Crocker, Qualified Immunity and Constitutional Structure, 117 Mich. L. Rev. 1405 (2019); Alexander Reinert, Joanna C. Schwartz & James E. Pfander, New Federalism and Civil Rights Enforcement, 116 Nw. U. L. Rev. 737, 752–54 (2021); and Joanna C. Schwartz, The Case Against Qualified Immunity, 93 Notre Dame L. Rev. 1797 (2018).
  236. . Eyer and Widiss recognize that stare decisis presents an obstacle. See Eyer, supra note 69, at 1683–85, 1694–95 (acknowledging that precedent “can be used to argue that a simple focus on the but-for principle is inappropriate in anti-discrimination cases”); Widiss, supra note 234, at 410 (“Calling for the Supreme Court to consider abandoning . . . ‘the most important case’ in employment discrimination law—as well as a significant number of subsidiary precedents—is no small matter.” (footnote omitted) (quoting Sandra F. Sperino, McDonnell Douglas: The Most Important Case in Employment Discrimination Law (2020))).
  237. . See Aaron L. Nielson & Christopher J. Walker, A Qualified Defense of Qualified Immunity, 93 Notre Dame L. Rev. 1853, 1855 (2018) (arguing that the attacks on qualified immunity face a heavy burden of stare decisis).
  238. . Cf. Eyer, supra note 69, at 1685 (asserting that “the time may have come for progressives to cut loose of their long-standing opposition to textualism”); Widiss, supra note 234, at 358–59 (“A fair reading of a progressive statute will often . . . advance progressive objectives.”).
  239. . United Steelworkers of Am. v. Weber, 443 U.S. 193, 197 (1979).
  240. . Id. at 199, 201 (second and third alterations in original) (quoting Civil Rights Act of 1964, Pub. L. No. 88-352, § 703(a)(1), (d), 78 Stat. 241, 255–56).
  241. . Id. at 197, 208–09.
  242. . Frickey, supra note 134, at 246.
  243. . Weber, 443 U.S. at 201.
  244. . 143 U.S. 457 (1892).
  245. . Weber, 443 U.S. at 201 (quoting Holy Trinity, 143 U.S. at 459). Holy Trinity involved an 1885 statute that prohibited “any person” from entering a “contract or agreement” to bring “any foreigner . . . into the United States . . . to perform labor or service of any kind.” Act of Feb. 26, 1885, ch. 164, § 1, 23 Stat. 332, 332 (repealed 1952). The question in the case was whether the law prohibited the Holy Trinity Church from contracting with a pastor from England. Holy Trinity, 143 U.S. at 457–58. Although the Court acknowledged that the statutory language was “broad enough to reach” the pastor, it found that such an interpretation would violate the law’s “spirit.” Id. at 459, 463, 472.
  246. . Weber, 443 U.S. at 201–02 (quoting United States v. Pub. Utils. Comm’n, 345 U.S. 295, 315 (1953)); see id. at 202–04 (examining Title VII’s legislative history). Justice Brennan did later point to one statutory provision—and its legislative history—as support for the view that Title VII did not prohibit voluntary affirmative action programs. See id. at 204–07 (discussing Civil Rights Act of 1964, Pub. L. 88-352, § 703(j), 78 Stat. 241, 257, which provides that “[n]othing contained in this title shall be interpreted to require any employer . . . to grant preferential treatment,” and does not say Title VII would not “permit racially preferential integration efforts”). As scholars have observed, that provision seems directed at courts and government agencies—prohibiting them from requiring affirmative action programs—and does not appear to comment on voluntary programs. See Bernard D. Meltzer, The Weber Case: The Judicial Abrogation of the Antidiscrimination Standard in Employment, 47 U. Chi. L. Rev. 423, 445–46 (1980) (asserting that this “section was added to prevent courts and agencies from imposing quotas”). In any event, the Court in Weber clearly did not view this text as central to the analysis.
  247. . See, e.g., Terry Eastland & William J. Bennett, Counting by Race 197–210 (1979) (discussing and criticizing Weber); see also Nancy MacLean, Freedom Is Not Enough 251, 255, 304 (2006) (discussing conservative opposition, and noting that the Reagan Justice Department “pledged to find a case . . . to overthrow” Weber).
  248. . Frickey, supra note 134, at 247; see infra section II(C)(2).
  249. . Frickey, supra note 134, at 247.
  250. . See Johnson v. Transp. Agency, 480 U.S. 616, 657–58, 670–71 (1987) (Scalia, J., dissenting) (attacking Weber); Frickey, supra note 134, at 255 (noting Scalia’s textual attack on Weber).
  251. . 480 U.S. 616 (1987).
  252. . See id. at 619–22 (noting that the challenged program “authorized the consideration of . . . sex as a factor when evaluating qualified candidates for jobs”).
  253. . 42 U.S.C. § 2000e-2(a)(1); see Johnson, 480 U.S. at 623–25 (detailing the case’s factual background).
  254. . Johnson, 480 U.S. at 619–20, 627, 629 n.7, 641–42. Notably, as Justice O’Connor observed in her concurrence, no party asked the Court to overrule Weber. Id. at 648 (O’Connor, J., concurring in the judgment).
  255. . Id. at 629 n.7 (majority opinion).
  256. . Id. William Eskridge has argued, by contrast, that there was insufficient political will in Congress to overrule Weber. See Eskridge, supra note 15, at 1410–11 (asserting that powerful interests supported the decision, while there was only diffuse opposition).
  257. . Johnson, 480 U.S. at 673 (Scalia, J., dissenting).
  258. . Id. at 670 (quoting 42 U.S.C. § 2000e-2(a)(1)).
  259. . Id. (quoting United Steelworkers of Am. v. Weber, 443 U.S. 193, 201 (1979)); see also id. (noting that Weber relied on and quoted Holy Trinity).
  260. . Id. at 671.
  261. . Id.
  262. . Id. at 671–72 (arguing that a “congressional failure to act” could mean “(1) approval of the status quo, . . . (2) inability to agree upon how to alter the status quo, (3) unawareness of the status quo, (4) indifference to the status quo, or even (5) political cowardice”).
  263. . Id. at 672.
  264. . E.g., Philip P. Frickey, John Minor Wisdom Lecture, Wisdom on Weber, 74 Tul. L. Rev. 1169, 1177 (2000).
  265. . Frickey, supra note 134, at 246.
  266. . See Frickey, supra note 264, at 1180 (relying on a 1968 dictionary); Ronald Dworkin, A Matter of Principle 318 (1985) (asserting that “‘discriminate against someone because of race’ . . . may be used . . . so that any racial classification whatsoever is included” or “in an evaluative way, to mark off racial classifications that are invidious”); William N. Eskridge, Jr., Dynamic Statutory Interpretation, 135 U. Pa. L. Rev. 1479, 1489 (1987) (urging that “discriminate” could apply to “any and every differential treatment of employees on racial grounds,” but could also “penalize only discrimination which is invidious”).
  267. . Dworkin, supra note 266, at 318; see also Eskridge, supra note 266, at 1489 (noting that “discriminate” could refer to “discrimination which is invidious”).
  268. . See supra note 266.
  269. . See Eyer, supra note 69, at 1685 (acknowledging the “obvious tension” between affirmative action and the but-for principle); see also, e.g., Jeannie Suk Gersen, Could the Supreme Court’s Landmark L.G.B.T-Rights Decision Help Lead to the Dismantling of Affirmative Action?, New Yorker (June 27, 2020), [] (“Bostock surely stands to affect other categories that Title VII addresses, most importantly race.”).
  270. . See supra subpart II(B).
  271. . See sources cited supra note 28. Although textualist opinions regularly reference “ordinary meaning,” there is little textualist scholarship exploring the concept. See, e.g., BP P.L.C. v. Mayor & City Council, 141 S. Ct. 1532, 1537 (2021) (Gorsuch, J.) (stating that the Court looks for “the ordinary meaning of [statutory] terms at the time of their adoption” (citing Niz-Chavez v. Garland, 141 S. Ct. 1474, 1480–81 (2021)). That is why, in a recent essay, I examine the concept of ordinary meaning. Grove, supra note 28, at 1082–84. This Article’s exploration of textualists’ use of statutory precedent offers considerable support for my instincts in that essay—that ordinary meaning can be seen as largely a legal concept.
  272. . Scholars who suggest that ordinary meaning is generally equivalent to lay meaning do recognize an exception for terms of art. But they seem to treat that category as limited. See sources cited supra note 29.
  273. . See Eskridge & Nourse, supra note 9, at 1727–28 (questioning the claim by textualists that “they were faithful to democratic values”); Krishnakumar, supra note 28, at 663 (arguing that “[t]he Court’s emphasis . . . on how ‘most people’ or one’s ‘friends’ talk in everyday contexts is in notable tension with its reliance . . . on” the common law).
  274. . See, e.g., Lee & Mouritsen, supra note 29, at 813–14, 818 (characterizing interpretation as “the law’s search for ordinary communicative content” and claiming that “we should seek to assess the public’s understanding of the law at the time it was passed”). Corpus linguistics involves the use of datasets to study linguistic phenomena, including searching databases to determine the frequency with which a word appears alongside other words in a given time period. See id. at 828–30 (describing and advocating the method).
  275. . See, e.g., Macleod, supra note 28, at 4–6, 8–10 (using surveys to test the results in Bostock); Kevin Tobia & John Mikhail, Two Types of Empirical Textualism, 86 Brook. L. Rev. 461, 483–85 (2021) (also using surveys to test the results in Bostock but cautioning against “any uncritical reliance” on that approach). To be sure, not all recent survey work has the goal of testing the results of specific decisions. For a general survey of empirical work on not only statutory interpretation but also common law concepts, see generally Kevin Tobia, Experimental Jurisprudence, 89 U. Chi. L. Rev. 735 (2022).
  276. . That is true, even though textualists (like other interpreters) do sometimes use examples from ordinary conversation (what Eskridge and Nourse have dubbed “homey examples”). See Eskridge & Nourse, supra note 9, at 1728, 1762, 1777, 1780–82 (using the phrase repeatedly). I discuss the use of homey examples in separate work. Grove, supra note 28, at 1082–84.
  277. . See supra subpart II(B).
  278. . See, e.g., Richard H. Fallon, Jr., The Statutory Interpretation Muddle, 114 Nw. U. L. Rev. 269, 327 (2019) (rejecting the assumption that statutes contain “ordinary utterances eligible for interpretation through the model of conversational interpretation”); Mark Greenberg, Legislation as Communication? Legal Interpretation and the Study of Linguistic Communication, in Philosophical Foundations of Language in the Law 217, 252 (Andrei Marmor & Scott Soames eds., 2011) [hereinafter Greenberg, Legislation as Communication?] (emphasizing “differences between the prototypical personal communication context and the legislative context”); Zoldan, supra note 28, at 7 (“suggest[ing] the deficiencies of a rule of statutory interpretation that emphasizes the nonlegal meaning of language”); see also William Baude & Stephen E. Sachs, The Law of Interpretation, 130 Harv. L. Rev. 1079, 1088–92 (2017) (“[O]ur legal system does seem to misread texts quite deliberately—to apply rules and conventions on a regular basis that don’t track any coherent theory of linguistic meaning. . . . [C]onventional legal rules seem to depart from the actual intentions and understandings of actual people.”); cf. Cass R. Sunstein, There Is Nothing that Interpretation Just Is, 30 Const. Comment. 193, 212 (2015) (“The meaning of the Constitution must be made rather than found . . . .”). Greenberg suggests that the assumption that language in legal texts is equivalent to that in ordinary conversation is part of what he calls the Standard Picture. Mark Greenberg, The Standard Picture and Its Discontents, in 1 Oxford Studies in Philosophy of Law 39, 48, 54 (Leslie Green & Brian Leiter eds., 2011). Greenberg further argues that textualists have endorsed the Standard Picture. See Mark Greenberg, Legal Interpretation and Natural Law, 89 Fordham L. Rev. 109, 111–24 (2020) [hereinafter Greenberg, Legal Interpretation] (criticizing textualism on this basis). I consider myself a textualist who does not endorse at least this aspect of the Standard Picture—that is, the conflation of legal and ordinary language.To be sure, Greenberg’s important work encompasses more than concerns about equating statutory meaning with ordinary conversation. The Article does not aim to comment on (or endorse) other aspects of Greenberg’s theory. For one recent comment on the broader aspects of Greenberg’s theory, see Bill Watson, In Defense of the Standard Picture: What the Standard Picture Explains that the Moral Impact Theory Cannot, 28 Legal Theory 59 (2022).
  279. . Greenberg, Legal Interpretation, supra note 278, at 122–23.
  280. . Greenberg, Legislation as Communication?, supra note 278, at 252.
  281. . See Tobia et al., supra note 139, at 372–73, 431 (reporting that people understand law to use “technical” language).
  282. . Id. at 373.
  283. . Id. at 397–98.
  284. . I focus here on arguments about whether the first stage of the statutory analysis should involve not simply linguistic meaning but also legal and normative concerns. Notably, the legalistic vision of statutory analysis advocated in this Article may also raise fair notice concerns. See Grove, supra note 28, at 1084–86 (briefly discussing this issue). Critics of textualism have at times raised questions about whether the method provides “fair notice.” See, e.g., Benjamin Minhao Chen, Textualism as Fair Notice?, 97 Wash. L. Rev. 339, 342 (2022) (“Fair notice does not ineluctably require textualism.”); Jesse M. Cross, The Fair Notice Fiction, 75 Ala. L. Rev. 487, 522 (2023) (doubting that “fair notice principles” require “textualist methodologies”). Yet, interestingly, textualists themselves have rarely focused on “fair notice” as a justification for their method (although one thoughtful student note endeavored to do so: Note, Textualism as Fair Notice, 123 Harv. L. Rev. 542, 542–43, 557–58 (2009)). That may be because, as Caleb Nelson has pointed out, all interpretive methods aim to provide “fair notice.” See Nelson, supra note 41, at 352–53 (asserting that “[t]extualists and intentionalists alike give every indication of caring . . . about the need for readers to have fair notice”). A full exploration of fair notice is beyond the scope of this Article, but I plan to examine the issue in future work.
  285. . See supra note 30.
  286. . Because this Article focuses on statutory analysis, I make no claim about how the arguments here might carry over to the constitutional context. For scholarship questioning the interpretation–construction distinction in that realm, see, for example, Curtis A. Bradley & Neil S. Siegel, Constructed Constraint and the Constitutional Text, 64 Duke L.J. 1213, 1217 (2015); John O. McGinnis & Michael B. Rappaport, Original Methods Originalism: A New Theory of Interpretation and the Case Against Construction, 103 Nw. U. L. Rev. 751, 751–53 (2009); and John O. McGinnis & Michael B. Rappaport, The Power of Interpretation: Minimizing the Construction Zone, 96 Notre Dame L. Rev. 919, 921–22 (2021). See also J. Joel Alicea, Liberalism and Disagreement in American Constitutional Theory, 107 Va. L. Rev. 1711, 1724 (2021) (noting that “[t]he interpretation/construction distinction is controversial among originalists”).
  287. . See sources cited supra note 30.
  288. . See, e.g., Anya Bernstein, Saying What the Law Is, 48 Law & Soc. Inquiry 14, 26 (2023) (describing “interpretation,” as understood by those who endorse the interpretation-construction distinction, as “a kind of scientific investigation that reveals the true meaning of the text. . . . Interpretation is thus ‘“value neutral,” or only “thinly normative”’” (quoting Solum, supra note 30, at 104) and concluding that “[i]nterpretation, on this view, can be grounded in semantics—the part of word meaning that travels from context to context unchanged—while construction requires recourse to pragmatics—the meaning-making context”); Sunstein, supra note 278, at 204–05 (noting that “some people have insisted on the importance of making a distinction between ‘interpretation’ and ‘construction,’” and suggested that interpretation is “value-free or only ‘thinly normative,’ in the sense that our normative views about what the law should be do not determine whether an interpretation is correct” (footnote omitted) (quoting Solum, supra note 30, at 104)).
  289. . Lawrence Solum states, in discussing “the interpretation-construction distinction,” that:Interpretation is an empirical inquiry. The communicative content of a text is determined by linguistic facts (facts about conventional semantic meanings and syntax) and by facts about the context in which the text was written. Interpretations are either true or false—although in some cases we may not have sufficient evidence to show that a particular interpretation is true or false. Constructions are justified by normative considerations. . . . [that serve] to give legal effect to the undeterminate text.

    Lawrence B. Solum, The Fixation Thesis: The Role of Historical Fact in Original Meaning, 91 Notre Dame L. Rev. 1, 12 (2015); see also Tun-Jen Chiang & Lawrence B. Solum, The Interpretation-Construction Distinction in Patent Law, 123 Yale L.J. 530, 548 (2013) (“The essence of the interpretation-construction distinction is to place under the rubric of interpretation the issue of discerning the linguistic meaning, and then to address the remaining issues . . . .”).

  290. . See supra note 174 and accompanying text; see also William N. Eskridge, Jr., Brian G. Slocum & Kevin Tobia, Textualism’s Defining Moment, 123 Colum. L. Rev. 1611, 1611–12, 1625, 1692 (2023) (arguing, to the extent textualists make choices about, for example, the relevance of certain text, context, history, or precedent, textualism is less constraining than advertised).
  291. . See supra section I(B)(3).
  292. . See supra section I(B)(3).
  293. . An interpreter who rejected an emphasis on original meaning would presumably adopt different guidelines. Cf. Frederick Schauer, Unoriginal Textualism, 90 Geo. Wash. L. Rev. 825, 828–29 (2022) (suggesting an “unoriginal” textualist approach to statutory interpretation).
  294. . See, e.g., Jack N. Rakove, Original Meanings 11 (1997) (expressing skepticism that “this method of interpretation can provide the faithful and accurate application of the original constitutional understandings its advocates promise”); Saul Cornell, Heller, New Originalism, and Law Office History: “Meet the New Boss, Same as the Old Boss, 56 UCLA L. Rev. 1095, 1098 (2009) (criticizing “law office history” as “results oriented”); Larry D. Kramer, When Lawyers Do History, 72 Geo. Wash. L. Rev. 387, 389 (2003) (discussing how “different scholars looking at basically the same materials can reach such disparate conclusions”). For more charitable perspectives, see, for example, William Baude & Stephen E. Sachs, Originalism and the Law of the Past, 37 Law & Hist. Rev. 809, 810–11 (2019); Mark Tushnet, Interdisciplinary Legal Scholarship: The Case of History-in-Law, 71 Chi.-Kent L. Rev. 909, 934–35 (1996).
  295. . See, e.g., Stefan Th. Gries & Brian G. Slocum, Ordinary Meaning and Corpus Linguistics, 2017 BYU L. Rev. 1417, 1423–24, 1440–42, 1470–71 (favoring corpora but suggesting that judges may need to rely on experts); Aaron-Andrew P. Bruhl, Interpreting State Statutes in Federal Court, 98 Notre Dame L. Rev. 61, 113–14 (2022) (doubting federal judges’ capacity to use corpus linguistics). For other criticisms of corpus linguistics, see, for example, Carissa Byrne Hessick, Corpus Linguistics and the Criminal Law, 2017 BYU L. Rev. 1503, 1505, 1514–15; Matthew Jennejohn, Samuel Nelson & D. Carolina Núñez, Hidden Bias in Empirical Textualism, 109 Geo. L.J. 767, 770–71 (2021); Tobia, supra note 28, at 734.
  296. . My goal here is not to suggest that there is no distinction between “interpretation” and “construction.” That is, I do not adopt the position of Fred Schauer, who has argued that, to the extent law is a “technical language,” there really is no difference. See Frederick Schauer, Is Law a Technical Language?, 52 San Diego L. Rev. 501, 503, 513 (2015) (“[I]dentifying even the possibility that legal language may be a pervasively technical language has the potential for seriously undercutting the ubiquitous interpretation–construction distinction.”); Frederick Schauer, Constructing Interpretation, 101 B.U. L. Rev. 103, 108–09, 123–27 (2021) (“The existence of technical legal language creates a problem for the distinction between interpretation and construction.”). I believe there are separate stages of the statutory inquiry: (1) discerning whether the provision at issue has a plain meaning, and then (2) figuring out how to resolve the case in the absence of a plain meaning. In my view, both stages involve law, but they involve resort to different legal sources, including different categories of precedent. Thus, some legal sources are relevant at the first stage of determining the plain meaning of the law: the text and structure surrounding the operative provision at issue, related statutes, and the first category of statutory precedent (cases identifying the meaning of terms and phrases). Other legal sources are relevant if the interpreter finds no plain meaning: purpose, substantive canons, and the second and third categories of precedent (past holdings and implementation tests, and an effort for consistency across an area of law). So I believe there are distinct stages of the statutory inquiry. But both stages, in my view, involve legal analysis.