Text Is Not Law


Bostock v. Clayton County purports to apply a textualist reading of Title VII of the Civil Rights Act to conclude that discrimination because of sexual orientation or gender identity counts as sex discrimination.1 It quickly became conventional wisdom that the case will have wide-ranging influence on law beyond Title VII employment protections.2 More immediately, the decision was seen as an unambiguous victory for LGBTQ+ employees.3

Textualists have also found reason to celebrate, finding further evidence for Justice Kagan’s familiar claim that “we’re all textualists now.”4 To be sure, some textualists disagree with Justice Gorsuch’s approach.5 But insofar as both Justice Gorsuch’s majority and Justice Alito’s dissent purport to engage in textualist reasoning, some commentators think the Supreme Court has moved beyond the question of whether textualism is the best approach to statutory interpretation to the question of how best to engage in textualist reasoning.6

This Article does not seek to rehearse familiar disputes about whether textualism: is theoretically coherent,7 is incompatible with the faithful agent model of judicial interpretation,8 presents a genuine alternative to other theories as opposed to merging with them,9 harms democratic decision-making,10 is successfully constrained by canons of construction,11 or whether textualism is otherwise theoretically “bankrupt.”12 Nor does this Article take sides in the dispute between the majority and dissenting approaches to textualism in Bostock itself.13 Textualism is not this Article’s target.14

Instead, this Article criticizes a common claim frequently made by prominent textualists in characterizing their own commitments, and which surfaced in Justice Gorsuch’s majority opinion in Bostock.15 Specifically, Justice Gorsuch writes, “[o]nly the written word is the law, and all persons are entitled to its benefit,”16 adding later that “only the words on the page constitute the law adopted by Congress and approved by the President.”17 These claims are not new. Similar sentiments appeared a year earlier in his book, A Republic, If You Can Keep It, which argued that “textualism honors only what’s survived bicameralism and presentment—and not what hasn’t. The text of the statute and only the text becomes law.”18

This Article argues that conflating text and law is a potentially harmful mistake that both textualists and nontextualists alike should reject. To set the stage, Part I shows that Justice Gorsuch is not alone. Many prominent textualists have conflated text and law in their writings about statutory interpretation.19 Part II, however, shows that this conflation rests on a conceptual error—a category mistake.20 As Section II.A shows, text and law are different kinds of things. Texts, as Mitchell Berman has observed, are organized structures of meaning-bearing symbols, while laws are norms.21 But, as Section II.B shows, textualists aren’t necessarily committed to this mistake, given that textualists presuppose a relation between a text’s linguistic meaning and the law’s content.22 By appealing to some such relation, textualists can avoid confusing statutory text with law.

The fact that prominent textualists routinely conflate text and law may seem much ado about nothing, like a careless but harmless oversight. Part III argues otherwise. Textualists themselves have reason to resist conflating the two. Section III.A argues that the conflation invites mischaracterizations and uncharitable understandings of textualism. Section III.B adds that judges who do embrace the notion that text is law will feel empowered to engage selectively in literalism, or even purport to embrace it, as an ad hoc response to statutory analyses—textualist ones included—that rely heavily on appeals to extratextual context. Indeed, Section III.B also argues that Justice Gorsuch adopted precisely this response to Justice Kavanaugh’s accusation that the majority opinion in Bostock was too literalist. Section III.C argues that conflating text and law perpetuates myths about appellate adjudication. Lastly, Part IV speculates about the popularity of textualists’ casual conflation of text with law, arguing that the identification provides rhetorical advantages that allow textualist judges to dodge thorny questions that any theory of interpretation must grapple with. These advantages will make the trope that “text is law” an appealing mainstay of judicial rhetoric despite being false and misleading.

Rhetorically appealing shortcuts have their place. Pithy slogans that summarize theoretical orientations can be illuminating. But given the misunderstandings that conflating text and law encourage, both to our understanding of textualism and to popular understanding of how judges make law, and given that the conflation also facilitates glib and ad hoc literalism, this shortcut should be abandoned. Text is not law.

Textualists Often Conflate Text and Law

Again, Justice Gorsuch is not alone. Many prominent textualists conflate statutory text and law. Justice Brett Kavanaugh has asserted, in both a book review and keynote address, “[t]he text of the law is the law.”23 Judge Frank Easterbrook made a similar assertion years earlier, arguing that because “[o]nly the text survived the complex process for proposing, amending, adopting, and obtaining the President’s signature,”24 it follows that “[t]he text of the statute, and not the private intent of the legislators, is the law.”25 And that was not the only time he made the claim.26

Few have done more to popularize the conflation of text and law than the late Justice Antonin Scalia.27 In A Matter of Interpretation, after making his now-familiar complaints about Holy Trinity Church v. United States,28 Scalia argues that “the decision was wrong because it failed to follow the text,” adding, “[t]he text is the law, and it is the text that must be observed.”29 Years later, in his co-authored volume, Reading Law, Justice Scalia and Bryan Garner doubled down on the conflation, asserting that it is “[t]he traditional view . . . that an enacted text is itself the law.”30

Textualist-friendly courts have latched onto these slogans. Several courts cite Judge Easterbrook’s dictum approvingly.31 Even more U.S. Courts of Appeals,32 federal district courts,33 and state court judges have explicitly quoted the above-referenced passages from Justice Scalia.34 Other courts invoke the claim that “the text is law” or its variants without explicitly citing the leading textualists.35 This suggests that the text-is-law assertion is taking root, perhaps even as an unremarkable platitude.

Although textualist judges appear especially prone to this conflation, legal scholars, not all of whom are textualists, also assert that text is law. Saikrishna Prakash appears to have taken it for granted, asserting that “what counts as law is the text.”36 In an exchange with Jonathan Siegel that this Article will revisit later on, Ilya Somin writes, “[t]extualists are indeed committed to the proposition that the text is the law.”37 A. Benjamin Spencer likewise approves of Scalia’s oft-cited dictum that text is law.38 And Cass Sunstein agrees that text is law, denying only that it tells us anything important about how to interpret text.39

Given both the familiar criticisms against competitor theories and the main argument proffered in favor of textualism, the tendency of textualists to conflate statutory text with law might not be surprising. Roughly, textualists have criticized intentionalism for inviting judges to perform the impossible task of seeking out a coherent collective intent that does not exist40 and purposivism for inducing judges to illegitimately pursue abstract policy objectives instead of respecting concrete legislative compromises encoded in statutory text.41 Conflating text and law may serve as a forceful reminder that the text, rather than speculative purposes or nonexistent intent, ought to guide judges as faithful agents of the legislature.42  

The conflation also traces to bicameralism-and-presentment arguments frequently proffered in favor of textualism.43 Textualists have argued that because only statutory texts survive the Constitutional process requiring bills to be approved by both houses of Congress and the president, judges should not treat extratextual sources like legislative history as though they are binding sources of law.44 As Justice Gorsuch remarks in Bostock, “only the words on the page constitute the law adopted by Congress and approved by the President.”45 Years earlier, Judge Easterbrook makes the same point in Continental Can Company v. Chicago Truck Drivers, arguing that because “[o]nly the text survived the complex process for proposing, amending, adopting, and obtaining the President’s signature,” it follows that “[t]he text of the statute, and not the private intent of the legislators, is the law.”46 Conflating text and law seems all but inevitable once one accepts that a statutory text is the only product of a lawmaking body that counts as binding authority.

It is somewhat surprising that Justice Gorsuch’s recent writings emphasize this bicameralism-and-presentment argument, because some textualists have since cabined its import in the face of serious criticisms.47 First, the argument proves too much. Courts and administrative agencies routinely promulgate new rules or specifications of congressional mandates that, while having the force of law, are not themselves approved by Congress and signed into law by the president.48 This observation calls into question the idea that only statutory text that survives bicameralism and presentment counts as a source of law.49 This argument proves too much in a second way. If the argument entails that courts should not be guided by extratextual sources, then textualists themselves cannot seek out extratextual guidance, including—but not limited to—dictionaries, canons of construction, and precedent.50 As Victoria Nourse observes, “the bicameralism argument reaches so far that it undermines everything but the text of the statute, including all canons and every other traditional method of statutory interpretation.”51

Forbidding judges from relying on any extratextual evidence of statutory meaning is unworkable. True, John Manning has restated the bicameralism-and-presentment argument to defend a narrower conclusion: that judges should avoid or heavily discount legislative history as a source of authority.52 But it remains puzzling why the discredited bicameralism-and-presentment argument continues to find a receptive audience for the positive thesis that only text counts as law—at least as evidenced by Justice Gorsuch’s popular writings.53 It is puzzling because the claim embodies a basic conceptual mistake, one that textualists are not in principle committed to making.

Why Text Is Not Law, Even for Textualists

So, the idea that statutory text counts as law is widespread. With Gorsuch’s landmark opinion in Bostock, the claim may prove more influential still. As Section A shows, however, the claim is false. But Section B explains what may already be obvious to those familiar with debates about textualism: that textualists needn’t endorse the falsehood.

Why Text Is Not Law

Sentences like “text is law” can be construed in several ways. They may express a strict identity relation (i.e., something along the lines of: statutory text t = legal norm l). Or we could construe them as conditional identity claims, something to the effect that if x counts as a duly enacted statutory text, then x also counts as a legal norm.54 Interpreted either way, the casual conflation of text and law is what philosophers call a “category mistake.”55 Although defining category mistakes is not easy, roughly speaking they involve ascribing to an object properties that it cannot have because they belong in different “ontological categories.”56

To illustrate, suppose that a campus tour guide points to classrooms, buildings where academic departments are housed, dormitories, and administrative offices. Now suppose that the prospective students on tour later ask the guide to see the university. The visitors would be making a category mistake because they incorrectly assumed that the university is an institution housed in an independent brick-and-mortar building like the others encountered on tour.57 Although universities do bear some relation to certain people and certain physical facilities, they are not necessarily the type of thing that is housed in a single such facility.58 They are constituted in part by a collection of people occupying certain roles, the buildings already encountered on tour where they work, budgets, and other facilities.

Conflating text and law involves a similar category mistake.59 Surprisingly few scholars have noticed this mistake, or have bothered to say anything about it. Mitchell Berman is a rare and illuminating exception.60 Writing about constitutional interpretation, Berman explains:

Obviously, . . . the constitutional text matters a great deal. Now, stating precisely how it matters is no mean feat. Scholars and theorists of varied jurisprudential and ideological stripes often proclaim that “the constitutional text is the law.” But, that’s a category mistake that, however harmless in most contexts, can breed confusion if taken literally. Roughly: text is an arrangement of signs and symbols, while law is the set of norms—rights, duties, powers, permissions—that a legal system delivers or comprises.61

When Berman says that conflating text and law involves a category mistake, this means, roughly, that when Justice Scalia says that “text is law”62 or when Justice Gorsuch claims that “[o]nly the written word is the law,”63 they attribute to texts properties that they cannot have—i.e., they attribute to texts or words the properties that law has.64 Text and law belong to different ontological categories—they are different kinds of things. Texts are arrangements of signs and symbols, whereas “a law is a normative entity­—a prohibition, permission, power, and so on.”65 To illustrate this conceptual point, Berman points out that “nonidentical texts can correspond to the same law and because a statutory text can be amended (say, to make things clearer) without changing the law, text is not law.”66 Indeed, at least one textualist, Judge Easterbrook, has similarly observed that, “[t]o propose a change in a law’s text is not necessarily to propose a change in the law’s effect.”67

Plenty of examples bolster Berman’s point, as independently recognized (in effect) by Judge Easterbrook.68 But scrivener’s errors provide the most salient illustration. These errors are drafting mistakes that inadvertently remain in statutes that legislatures approve.69 In Scurto v. Le Blanc, for example, the text of a Louisiana statute permitted a litigant to secure “the testimony given by his opponent on cross-examination, ‘in any unlawful way.’”70 Although that text itself amended a prior statute, the Louisiana Supreme Court took “cognizance of the fact—which is obvious—that this substitution of the word ‘unlawful’ for the word ‘lawful’ was an accident,” and thus ignored the word “unlawful” and decided instead to “continue to read the law as it was originally written.”71 After all, the legislature had not intended to permit litigants to break the law to elicit testimony, notwithstanding the text. So, the Court ignored the new text.72 In other words, new text was added without new law being made.

Although scrivener’s errors provide the clearest counterexample to the claim that text is law, many textualists themselves famously accept scrivener’s errors as an exception to textualism’s core commitments.73 Justice Scalia has embraced the scrivener’s error doctrine, because where “the objective import of such a statute is clear enough,” it is permissible “to give the totality of context precedence over a single word.”74 In their joint work, Bryan Garner and Justice Scalia continued to make an exception for these mistakes, asserting that this exception does not undermine core textualist principles.75 Other textualists concur.76 What they fail to acknowledge is that allowing this exception concedes that text is not law, and that this in turn undermines Scalia’s own assertion to the contrary.77 As Jonathan Siegel observes, “[i]f statutory text is the law—if the constitutional process of enactment imbues the statutory text with legal force regardless of what anyone intended—then the text cannot cease to be the law when it is absurd or erroneous.”78

The point is not merely an academic concern. In Cameron v. Auto Club Ins. Co.,79 the Supreme Court of Michigan addressed the question of whether, by relying on a tolling statute, someone could sue an insurance company for benefits falling outside a statutory deadline.80 The majority refused to apply the tolling statute, which extended by one year a deadline “to make an entry or bring an action” after their “disability” was “removed.”81 At issue was whether this statute tolled claims for lawsuits alone or also tolled reimbursement claims by insureds that fell outside that one-year window.82

The majority argued for the lawsuit-only interpretation.83 This conclusion in effect empowered insurance companies to stall their reimbursement procedures or delay correspondence with their insureds, thereby minimizing damages that would fall inside the one-year cut off.84 Although acknowledging that the result was “harsh,” the court waxed poetic about honoring legislative compromise—just as Justice Barrett and Dean Manning, both textualists, have emphasized elsewhere.85 And the court declined to recognize that the result, although harsh, rose to the level of absurdity.86

Setting aside the merits, notice that then-Judge Marilyn J. Kelly, writing in dissent, excoriated the majority for what she regarded as its crabbed reading of the tolling and deadline statutes, one that she traced to Justice Scalia’s influence.87 She then turned directly to Scalia’s writings, explicitly recognizing the outright contradiction between the text-is-law trope and Scalia’s treatment of scrivener’s errors:

Justice Scalia’s main thesis with regard to statutory construction is that “[t]he text is the law, and it is the text that must be observed.” He asserts that what the legislature meant as opposed to what it actually stated in the language of a statute is immaterial. However, Justice Scalia acknowledges that one of the “sound principles of interpretation” is the interpretative doctrine of lapsus linguae (slip of the tongue) or “scrivener’s error,” where from the very face of the statute “it is clear to the reader that a mistake of expression (rather than of legislative wisdom) has been made.” . . . I agree that the scrivener’s error canon of construction is an appropriate tool in determining legislative intent. But intellectual honesty requires an acknowledgement that it involves a departure from the actual language used by the Legislature or by Congress.88

There is a sense in which then-Judge Kelly and those who share her skepticism are plainly correct and a sense in which they overreach. She is correct to suggest that the scrivener’s error doctrine shows that text is not law. But, like others who concur with Judge Kelly, presenting scrivener’s errors as decisive counterexamples to textualism goes too far.89 Indeed, as shown below, textualism is in principle divorceable from Scalia’s misleading slogan.

Why Textualists Need Not Claim Otherwise

The idea that text is law rests on a mistake. If, as Jonathan Siegel has argued, the conflation of text and law is the “fundamental axiom” of textualism,90 the preceding discussion would refute textualism. But, despite Jonathan Siegel’s claims to the contrary91—and, in fairness to Siegel, contrary to the claims of some of textualism’s leading proponents—textualists can stop embracing Scalia’s dictum without giving up textualism. Indeed, there seems to be considerable consensus among modern textualists (in theory) about the idea that statutory texts bear some relation to context-sensitive linguistic content, and that this content in turn bears some direct or unmediated relationship to legal norms.92 

As a first approximation, textualists articulate their project in terms of the “meaning” of a statutory text. Consider the following formulations from proponents of textualism, stated in terms of some relation between texts and their “meanings” as understood in context:

  • “An interpreter who believes that legislatures have authority only to pass statutes, not to form abstract ‘intentions,’ will describe statutory interpretation as a search for the meaning of statutory That interpreter can be called a ‘textualist.’”93
  • “[T]extualism does not admit of a simple definition, but in practice is associated with the basic proposition that judges must seek and abide by the public meaning of the enacted text, understood in context (as all texts must be).”94
  • “Textualism maintains that judges should seek statutory meaning in the semantic import of the enacted text and, in so doing, should reject the longstanding practice of using unenacted legislative history as authoritative evidence of legislative intent or purpose.”95
  • “In their full context, words mean what they conveyed to reasonable people at the time they were written—with the understanding that general terms may embrace later technological innovations.”96

These formulations differ from one another. But none of them conflate a statute’s text with law. They instead assert that the judge’s primary task is to ascertain a statutory text’s meaning, where “meaning” is understood as the context-sensitive linguistic meaning of the statutory text at issue.

Whether textualism can coherently identify a form of linguistic meaning while eschewing appeal to specific communicative intentions is an important issue.97 But for now, assume that statutory texts have ascertainable linguistic meanings without them. More importantly for our purposes, textualists must have or presuppose a relation between this meaning and the law’s content. Textualist and nontextualist writers alike often assume that the meaning of a statute, however understood, bears some sort of direct relation to a legal norm.98

Consider some recent formulations of textualism that explicitly recognize the distinction between linguistic meaning and legal content. Hrafn Asgeirsson claims that “textualists hold that judges should – in the first instance – determine the [legal] content of the relevant statute, which in their view amounts to determining the meaning of the statutory text.”99 John Perry construes textualism as “the view that the [legal] content of a statute—basically, what actions it mandates, forbids, or protects—is determined by the original meaning of the text of the statute.”100 Or consider Gideon Rosen’s formulation of “simple textualism”:

A legal sentence S, uttered in a determinate context C, expresses a proposition about the law in some jurisdiction. The proposition expressed by S depends on S’s meaning in the language—a meaning it has independently of any particular occasion of use—and perhaps on objective features of the context, such as the time and place of the utterance, but it does not depend on the private and potentially idiosyncratic beliefs or intentions of the speaker or his audience. Given suitable background conditions, the legal utterance is a performative whose effect is to add this very proposition to the law, and so to make this proposition true.101

In other words, as Asgeirsson, Perry, and Rosen claim, textualism does not simply connect textual sentences to linguistic meanings (i.e., the text does more than simply express those meanings or “propositions,” as philosophers of language say),102 textualists assume that those propositions or meanings count as legal norms.103

Returning to Bostock, we see that Justice Gorsuch himself recognizes that meaning, rather that text alone, is the operative concept in figuring out a statute’s legal content.104 That is, even though Justice Gorsuch’s majority opinion misleadingly suggests that text is law,105 the opinion also contains a somewhat more careful and by now familiar appeal to public meaning, remarking, “[t]his Court normally interprets a statute in accord with the ordinary public meaning of its terms at the time of its enactment.”106 Justice Gorsuch’s misleading turn of phrase—conflating text and law—seems little more than rhetorical window dressing, a small mistake at most, belied by a separate claim, a more widely held view already well understood by careful expounders of textualism: Linguistic meaning, as understood by a reasonable speaker in context, either counts as a legal norm or constrains the judge’s application of legal norms.107

With these examples in mind, textualists are well positioned to deny that statutory texts are themselves laws or legal norms. This is because they accept a different claim, which is that the context-sensitive meaning that those same texts express count as, or constrain, the legal norms that judges should apply in discharging their obligations as judges.108

Notice that this claim, or something in the vicinity, better resists counterexamples. Recall that a text changes depending on the language in which it is written, but the law need not.109 The fact that different texts written in different languages may communicate the same legal norm is no problem at all. A statutory text may express some legal norm. And the same norm may be described by a Spanish-language pamphlet despite containing a different text. Different texts can express the same meaning, and it is that meaning that in turn is supposed to constitute and constrain the relevant legal norm. This is indeed one of the functions of what philosophers of language call propositions; they are abstractions that explain, among other things, how sentences in different languages can express the same thing.110 Propositions are the common denominator.

Emphasizing the importance of context, moreover, helps textualists overcome scrivener’s errors.111 Appeals to ordinary conversation motivate the textualist response. For example, when someone utters sarcastically that he is having “so much fun,” they may communicate that they are not having fun at all. Or when a doctor tells a patient, “you’re not going to die,” this is strictly false but communicates something different and true, along the lines of you’re not going to die prematurely due to ill health.112 But if context includes the fact that the speaker spoke sarcastically or was a physician, and conventions that allow speakers to recognize sarcasm and clinical meaning, the assertion in context negates its literal meaning of the sentence uttered.

As with sarcasm and doctors, so too with scrivener’s errors and absurdity. The background context of a statutory text may make it clear that the written text expresses a meaning quite different than that which the text, literally construed in isolation, would communicate. Justice Scalia advocated this maneuver long ago, arguing that “it [is] not contrary to sound principles of interpretation . . . to give the totality of context precedence over a single word.”113 “[W]hen an appropriately informed reader would conclude that the statutory text contains a scrivener’s error,” adds Caleb Nelson, “textualists can assert that someone seeking the ‘objective’ meaning of the text would naturally correct the error.”114 Andrew Gold argues more specifically that shared communicative conventions between drafters and audiences help speakers and hearers recognize mistakes in communication and correct those mistakes in our understanding.115 The communications succeed despite the erroneous speech or, by extension, text. And the fact that an interpretation would produce absurd results, as Ryan Doerfler points out, may provide “evidence that Congress meant to say something else.”116

Appeals to context might explain why textualists seem so unconcerned by scrivener’s errors. This kind of claim—which suddenly retreats to some notion of reasonable corrections of obvious mistakes—puzzles some of textualism’s critics.117 But the claim shouldn’t be puzzling, at least not in all cases. Scurto illustrates, for example, that the Louisiana legislature meant to permit “lawful” rather than “unlawful” impeachment of witnesses.118 The audience, however plausibly specified, will have a shared background understanding that legally permitting unlawful conduct is incoherent. These background understandings form part of the relevant context—and in turn, serve to explain how text that appears to “say” one thing when taken out of context may assert something quite different when understood in context.119 So, although appealing to context won’t solve all of textualism’s problems, a textualist might reasonably point out that meaning actually asserted may depart from the literal meaning of the text taken alone, given context clues.120

Nothing here purports to exhaustively survey textualism’s commitments or arguments in favor of it. I offer instead only a list of ingredients—including statutory text, linguistic meaning, and legal norms—that can be used to formulate textualism’s commitments while consistently denying that text is law. To be sure, the meaning of “meaning” is fraught with ambiguity.121 As discussed in more detail below, textualists can and should be more precise in discussing what type of (presumably) linguistic meaning they are after.122 For now, the fact that a basic approach to textualism can be articulated without conflating text and law provides all the more reason to abandon the mistake.123

Why Conflating Text and Law Matters

This problem may seem much ado about nothing. After all, one might understand “text is law” as a slogan rather than expressing any serious conceptual thesis, a slogan that operates as an intellectual orientation, rallying cry, or pithy reminder. If so, conflating text with law understandably sacrifices nuance for the sake of simplicity. Even Mitch Berman remarks that, although the claim can “breed confusion,” it is probably “harmless” in most contexts.124 Part II might seem, ironically, to have committed the interpretive sin of taking certain textualist claims out of context and interpreting them too literally.

This Part argues, however, that conflating text and law—whether intended literally or not—likely has significant downsides. Section A explains why textualists themselves should reject the conflation, while Section B revisits Jonathan Siegel’s arguments that textualism becomes radicalized given its “fundamental axiom” that text is law. Although Siegel errs on this point, he is correct that certain pathologies come with thinking of text as law—e.g., a form of reasoning that Section B calls “opportunistic literalism.” Finally, Section C explains why the conflation perpetuates myths about appellate adjudication and props up overheated rhetoric that degrades public discourse. In short, conflating statutory text with law is not innocuous, even if not intended to be taken literally.

The Conflation Causes Confusion About Textualism

Textualists should deny that text is law. Apart from being false, the claim invites misunderstandings of their views. Before seeing how, notice first that textualists frequently lament having to battle uncharitable interpretations. Justice Scalia distinguished textualism from “strict constructionism,”125 denying that textualism is, as its critics suggest, “‘wooden,’ ‘unimaginative,’ or ‘pedestrian.’”126 John Manning has likewise cautioned against confusing textualism with literalism.127 Textualists have further assured that they pay close attention to historical context, denying that they privilege acontextual readings of statutory texts.128 Others have resisted claims that textualism serves primarily as a fig leaf covering reactionary political agendas.129 Textualists have even faced arguments that their position is fundamentally immoral.130 In short, textualists spend considerable time defending against what they regard to be mischaracterizations of their views.

If, however, textualists frequently lament that they are misunderstood, they would do well to acknowledge that they are at least partly to blame. Something about textualism invites mischaracterization. One reason is that textualism is sometimes a moving target—a natural state of affairs attributable to textualist theory working itself pure.131 Some arguments that initially loomed largely in favor of textualism have diminished in importance. We have already seen that a simple version of the bicameralism-and-presentment argument proved too much, and in response, a better but more limited version of the argument championed by Dean Manning took its place.132 So, to some extent, uncharitable interpretations of textualism reveal a failure to keep up to date with the most recent defenses of its core tenets. When targets move, critics miss.

But textualism is not simply a list of claims defended in scholarly journals. And it is not obvious that textualism always should be understood in terms of its best or most recent articulation.133 After all, we should also care—perhaps primarily—how legal concepts and doctrines of law operate in practice.134 Beyond a set of theoretical “theses,” self-proclaimed textualists do things in textualism’s name, even if mistaken according to the most sophisticated presentations of textualism’s doctrines. And as we have seen, textualist Supreme Court Justices and other judges routinely make basic mistakes in conflating text and law, while implying that those mistakes accurately reflect textualism’s commitments.135 So, textualists have good reason to correct the record; they themselves have good reason to give the best versions of their views room to breathe and develop, and in turn, to mitigate misunderstanding.

And the potential for misunderstanding runs deep. To illustrate, consider one of Jonathan Siegel’s criticisms of textualism,136 a possible textualist reply to Siegel, and a different textualist reply offered instead by Ilya Somin.137 The debate between Siegel and Somin will show that even thoughtful legal thinkers are led astray when grappling with the misleading idea that text is law.

Start with Siegel’s work. In The Inexorable Radicalization of Textualism, Siegel predicts that textualism will continue to “work itself pure,” but in so doing, it will get increasingly implausible—indeed, radical—in application.138 More specifically, Siegel makes conceptual, explanatory, normative, and predictive claims.

Citing Justice Scalia’s now-familiar dictum, Siegel’s conceptual claim is that, “[t]extualism’s fundamental philosophy—its prime directive—is that ‘[t]he text is the law, and it is the text that must be observed.’”139 The chief explanatory claim traces that directive to two arguments. The formalist one we have seen before. It is the familiar bicameralism-and-presentment argument that, because only a text is approved by the legislature and the president, and because the legislature thereby makes “law,” only the text counts as the law.140 The realist argument recognizes the fact that legislation reflects messy compromises, and honoring those compromises requires honoring the text that codifies them.141

Siegel’s predictive and normative claims walk hand in hand. Siegel claims that because the idea that text is law is foundational, and because many time-honored doctrines and judicial practices like the absurdity and scrivener’s error doctrines are inconsistent with this premise, textualist theory and practice will eventually shed them.142 Because these doctrines exert a moderating influence, however, textualism will radicalize.143 Specifically, textualism will become committed to enforcing absurdities and scrivener’s errors despite claims to the contrary.144 This is the radicalization thesis.145 Normatively, this radicalization will yield an “unworkable” and undesirable approach to statutory interpretation because: Those doctrines are sound techniques for construing statutes sensibly; the alternative is committed to enforcing absurdities and scrivener’s errors; and, given textualism’s tendency to ignore evident statutory purposes, legislation will less likely solve the problems it seeks to remedy.146

Siegel’s radicalization thesis gains traction because it assumes that textualism’s defining commitment is that text is law. But, as Jeremy Kessler and David Pozen point out, “[i]n focusing on the logic of textualism’s internal ‘axioms,’ Siegel misses all of the external dynamics that allow such axioms to be reconceptualized or reformulated in response to criticism—and that prevent mainstream prescriptive legal theories, more generally, from tending toward radicalism.”147 We have already encountered one family of reformulations, which allow textualists to simply reject Scalia’s conflation of text and law by focusing instead on how context-sensitive meaning, not text simpliciter, contributes to or constrains the law.148 Now, fully spelling out this reply incurs the cost of trying to explain in a positive way what textualism really is and how, if at all, it is distinctive from intentionalism or purposivism.149 This is no simple task.150 But for now, notice that rejecting Siegel’s “axiom” is the most obvious—and perhaps even most decisive—rejoinder available to the textualist.

But Ilya Somin, a self-described textualist, did not respond this way.151 Instead, Somin concedes Siegel’s characterization of textualism, granting that “[t]extualists are indeed committed to the proposition that the text is the law.”152 He then goes on to claim, implausibly, that those who embrace this conflation can nevertheless adopt the scrivener’s error doctrine.153 His argument answers Siegel by trying to minimize the scope of the problem rather than recognizing its force.154 Briefly, Somin construes scrivener’s errors as by definition departures from the version of a text that legislators actually voted on.155 In the alternative, Somin entertains the possibility that some legislation might be poorly drafted, self-defeating, yet still duly enacted156—i.e., statutes like the one encountered in Scurto v. LeBlanc.157 Somin says little about this latter kind of mistake other than remarking that judges sometimes disagree about whether legislation contains poor drafting (which presumably Somin would think should be enforced) versus scrivener’s errors (which, under Somin’s cramped definition, are texts that misstate what has actually survived bicameralism and presentment and thus have not truly been approved by a legislature).158 Somin appears, in other words, to provide no answer at all to Siegel for the subset of scrivener’s errors that were actually voted on and signed into “law.”159

This response concedes too much. Textualists need not engage with Siegel on his terms. This attempt to rescue rather than reject the premise that text is law is both unnecessary and feeds suspicion that textualism rests on implausible premises.160 It also implies that critics of textualism are not as uncharitable as they may seem. If textualist scholars like Somin agree with the literal claim that text is law, while providing inconclusive answers to Siegel’s objections, then they bolster Siegel’s objections. If text is law, then the scrivener’s error and absurdity doctrines, at a minimum, seem to be on the chopping block, no matter how entrenched and sensible they are. Textualism seems committed in principle to enforcing obvious mistakes and absurdities.

The Conflation Facilitates Opportunistic Literalism

Although textualists needn’t embrace the idea that text is law, the exchange between Siegel and Somin shows how that idea can mislead even sophisticated, good faith interlocuters. But conflating text with law does more than just cause confusion about textualism. To the extent that judges continue to embrace that conflation, a version of Siegel’s critique continues to gain a foothold. Indeed, this Section argues that, although embracing the notion that text is law will not likely result in textualism’s radicalization as Siegel predicts, accepting text-is-law-ism does facilitate a glib and dismissive form of literalism likely to be deployed in an ad hoc way—as Justice Gorsuch’s majority opinion in Bostock illustrates. This is not a welcome development, representing all the vices of literalism and none of the virtues that would at least follow from consistently applying it.

Revisiting the Radicalization Thesis

Revisiting Siegel’s radicalization thesis will be helpful. Recall that, according to that thesis, sound judicial doctrines and practices—like the scrivener’s error and absurdity doctrines—are logically incompatible with the thesis that text is law, since those doctrines license courts to ignore or depart from the text.161 As a result, textualist judges and scholars will increasingly abandon these time-honored doctrines and practices, yielding an uncompromising and radical form of textualism committed to enforcing absurdities.162

Ultimately, this Article argues for a modified version of Siegel’s worries, drawing on the Bostock opinion to illustrate. But it will be helpful to review how Siegel bolsters these points by highlighting an example that, in his view, already exhibits the kind of radicalization that he predicts.

The example is well known.163 The Class Action Fairness Act (“CAFA”) originally stated that appeals of certain lower court removal orders must be filed “no less than” seven days after a court had entered such an order.164 Taken literally, this language required an unhappy litigant to wait seven days before filing an appeal, then provided no deadline for doing so.165

Several courts rejected the literal approach.166 They reasoned, in effect, that this was an obvious typographical error that would yield an absurdity if taken literally.167 The error required courts to consider, among other things: that the statute in the broader statutory scheme plainly purported to provide a filing deadline, even though the text offered no deadline;168 that uncontradicted legislative history showed the statute aimed to discourage delay—not encourage it by creating a waiting period;169 and relatedly, that the evident purpose of the provision was to establish “tight deadlines,” not to prolong the appeals process surrounding class action suits.170 In short, most courts that addressed the issue saw a drafting error, whereby Congress intended to create a deadline giving an appealing party seven days to file an appeal—not a pointless seven-day cooling off period.171

Not everyone agreed. After a Ninth Circuit panel declined to follow the literalist route in Amalgamated Transit Union Local 1309 v. Laidlaw Transit Servs., Inc., several judges called sua sponte for a rehearing en banc.172 The active judges voted against rehearing the matter, inspiring dissents from textualists including Judge Jay Bybee and then-Judge Alex Kozinski.173 With Judge Bybee writing, the dissenting judges rejected construing “less” to mean more.174 Likewise, in Spivey v. Vertrue, Inc.,175 Judge Easterbrook, another textualist, also insisted on reading “less” to mean less.176 No textualist that considered the CAFA language concluded that it communicated, in context, a filing deadline rather than a waiting period—even though textualists nominally accept the absurdity and scrivener’s error doctrines. Indeed, Judge Bybee’s dissenting opinion in Amalgamated Transit viewed any other conclusion as an “abuse” of judicial power.177

Congress eventually fixed the mistake, rendering unnecessary the Supreme Court’s intervention.178 But for Siegel the unanimous textualist reaction to CAFA’s evident mistake was telling. Despite acknowledging the importance of context, consulting certain extratextual resources, and purporting to embrace the scrivener’s error and absurdity doctrines, the fundamental axiom of textualism—i.e., that text is law—rendered these acknowledgments and concessions incompatible with that theory and thus unavailable in practice.179 The textualist reaction to the CAFA filing deadline was, according to Siegel, a harbinger of things to come. “Soon enough,” he predicted, “textualist judges will be calling for the complete abolition of the absurd results and scrivener’s error exceptions.”180

As already noted, some of Siegel’s claims fall short. The conceptual claim is false. Textualism doesn’t necessarily embrace the claim that text is law because the relation between text and law is mediated, in some sense, by meaning.181 The same point shows why the bicameralism-and-presentment argument does not necessarily or perhaps even likely yield radicalization, since meaning supervenes on text and is just as plausibly validated by that process as the text itself. The possibility that nuanced attention to a text’s meaning in context can be used to avoid extreme literalism undercuts, in principle, the predictive and normative claims. A text’s context-dependent meaning leaves conceptual room, as we have seen, for the scrivener’s error and absurdity doctrines.182 Although the question of whether textualism has become “radicalized” may in part depend on the eye of the beholder, Siegel’s specific predictions that certain time-honored canons of construction would fall by the wayside have yet to materialize.183 Justice Gorsuch, for example, seems to acknowledge a place for the absurdity doctrine where it would inform linguistic understanding rather than involve reforming a statute’s substance.184 And even if textualism were radicalizing in theory or practice, other explanations—e.g., political ones—might more straightforwardly explain why.185

Still, dismissing Siegel’s insights entirely would be a mistake. After all, although textualists do not necessarily conflate text and law, we have seen that some of them—including at least two sitting Supreme Court Justices—continue to claim this idea as a fundamental principle.186 And he is also correct that conflating statutory text with law provides fertile ground for pathologies of legal reasoning, though perhaps not the radicalization that Siegel had in mind. One such pathology is described immediately below.

Opportunistic Literalism, Not Radicalization

Taking too seriously the idea that statutory text is itself law facilitates a form of argumentation that this Article calls “opportunistic literalism.” An interpretation is “literalist” to the extent that the judge seeks the text’s meaning acontextually—or at least to the extent that any text can be understood relatively independently of its context.187

There are some formalist reasons why judges might be attracted to literalist interpretations, especially if they embrace the conceit that text is law. First, conflating text with law makes it easier for judges to systematically dismiss or discount just about any extratextual guidance they dislike. Extratextual sources by definition stray beyond the text. But if text is law, systematically relying on those sources risks subverting the text and thus subverting, again, the law. Second, and relatedly, although textualists purport to rely on context, conflating text with law also stands in tension with that reliance. Establishing the relevant context in which to understand a text’s meaning, with certain exceptions,188 likewise involves relying on extratextual considerations. But, again, none of those considerations count as law. The text is the law. Better to strictly adhere to that which is purely law, the text, than risk contaminating it with the unwieldy process of assembling extratextual facts constitutive of the text’s context. Text is law, extratextual guidance and context are not.

These formalist considerations make literalism superficially appealing.189 They give generally applicable reasons to shun an ever-increasing number of extratextual factors—whether doctrinal or contextual—from consideration. As a result, judges who accept that text is law, to the extent that they are consistent, may find it difficult to help themselves to many extratextual considerations. These formalist arguments cast doubt on context, leaving acontextual interpretations—i.e., literalist ones—as presumably safer and more appealing. While Jonathan Siegel predicted the demise of specific doctrines like absurdity, scrivener’s errors, and interpretive practices that appealed to statutory purpose,190 his concern about radicalization—it seems—generalizes to all extratextual considerations. Concerns about “radicalization” seem fundamentally about literalism.

This is not to say that de facto literalism will always prevail, even for judges who embrace the notion that text is law. Legal realist considerations suggest otherwise. Judges advance literalist interpretations “opportunistically” in the sense that judges will advance literalist interpretations in some cases but not others, depending primarily on the judge’s conviction about what the right outcome should be.191 That is, sometimes judges deploy literalist interpretations as sufficient reason to reject alternative interpretations that draw heavily from extratextual considerations to inform context. Other times those same judges may dismiss literalist interpretations as insufficiently informed by context.192

Bostock itself bolters this point. As Anuj Desai, Cary Franklin, and Anita Krishnakumar have independently shown, both the “textualist” majority and dissenting opinions help themselves to extratextual—indeed, nontextualist—resources like appeals to legislative intent, precedent, and practical consequences.193 So even if the formalist arguments encourage judges who conflate text with law to eschew ever-greater swaths of extratextual sources of guidance, it is unlikely that judges—even adherents of the conflation—will consistently ignore those sources. Rather, the point is that judges who accept that text is law will more likely feel empowered to engage in de facto literalism in an opportunistic way.

Justice Gorsuch’s opinion in Bostock illustrates opportunistic literalism at work. Recall Gorsuch’s core argument, as succinctly recounted by Kavanaugh: “When a gay man is fired because he is gay, he is fired because he is attracted to men, even though a similarly situated woman would not be fired just because she is attracted to men.”194 Justice Kavanaugh and Justice Alito’s respective dissenting opinions share some common complaints against the majority opinion.195 But the most salient difference for present purposes is that Justice Kavanaugh—unlike Justice Alito—characterizes the majority opinion as the “literalist approach.”196

Justice Kavanaugh’s attempt to paint the majority opinion as a literalist one is unrelenting; his dissent uses the word “literal,” or some variation thereof, 39 times.197 If Kavanaugh’s characterization sticks, then Justice Gorsuch’s opinion is supposed to fail on its own terms, since textualism purports to reject literalism.198 After all, literalism’s sins include failing to read statutory language in context, and relatedly, deriving a statute’s linguistic meaning word-by-word rather than phrase-by-phrase.199 Time again we are reminded by Justice Kavanaugh that textualism, correctly applied, looks for reasonable meanings, not literal ones.200

Justice Gorsuch’s responses to the dissents are noteworthy because they are often framed as textualist ripostes to these textualist objections.201 For present purposes, however, Gorsuch’s most important response is to Justice Kavanaugh’s charge of literalism. There is a sense that Kavanaugh’s objection backfires, at least rhetorically. Justice Kavanaugh’s repeated attempts to paint Justice Gorsuch’s opinion as literalist entertains the possibility, if not outright concedes, that there is a literal sense in which discrimination “because of” homosexuality or transgender status counts as sex discrimination.202 And that concession is all that Justice Gorsuch needs—again, at least rhetorically. Seizing on the point, Justice Gorsuch remarks, “[y]ou can call the statute’s but-for causation test what you will—expansive, legalistic, the dissents even dismiss it as wooden or literal. But it is the law.”203 Tellingly, Gorsuch never outright rejects the charge of literalism, notwithstanding textualists’ repeated disavowal of literalism.204

Setting aside the dubious suggestion that the but-for test derives exclusively from Title VII’s thin “because of” language,205 why embrace literalism or the accusation of it in Bostock? One answer is that Gorsuch regards the ordinary meaning and the literal one to be the same in this case. But embracing allegations of literalism carries rhetorical advantages, at least once textualism is regarded by all as the baseline interpretive commitment.206 Consider the criticisms Justice Gorsuch’s opinion faces, which pull in opposite directions. Justice Alito recognizes and chastises Justice Gorsuch for updating the statute by reading present-day social developments into the text’s meaning, as opposed to considering “the social context” prevailing at the time of Title VII’s enactment.207 Justice Kavanaugh’s accusation of literalism takes in effect the opposite stance, suggesting that Justice Gorsuch errs by reading Title VII acontextually. After all, that is a hallmark of literalist interpretations: They are acontextual, providing a theoretical dividing line between textualism and literalism.

Although there is good reason to think that little about Gorsuch’s opinion is truly acontextual,208 it is easy to see why rhetorically Gorsuch would prefer Kavanaugh’s label of “wooden” or “literalist” to Alito’s accusation of rewriting the statute,209 like a legislator who updates old statutes given ever-changing social demands. After all, the literalist’s main sin is hewing too closely to the text as written—a sin of loving the text too much rather than too little. But if text is law, then the “sin” is tantamount to loving the law too much—hardly a judicial vice. By contrast, Justice Gorsuch’s dissenting colleagues help themselves to whatever extratextual facts, doctrines, and dictionaries they wish, provided they limit themselves to a certain time frame. But they are still exercising discretionary judgment unmoored from text, and as a result, more likely to discover meanings that cohere with their own preferred outcomes. Or as Justice Gorsuch puts it, “[i]f judges could add to, remodel, update, or detract from old statutory terms inspired only by extratextual sources and our own imaginations, we would risk amending statutes outside the legislative process reserved for the people’s representatives.”210 Better to be a literalist who sticks to the text—which, again, is the law—than a textualist who picks out his friends from the extratextual crowd.

To be sure, Justice Gorsuch does not explicitly embrace literalism in Bostock. The present point is to suggest how embracing the charge of literalism—even if only facetiously—proves useful for Gorsuch. Doing so turns vice into virtue, performing theoretical jiujitsu: Acontextual, “literalist” readings better honor the text qua law, while excessive reliance on an amorphous understanding of extratextual guidance or contextual evidence of meaning carries risks usually associated with purposivists and intentionalists. Literalism can prove useful when text is law.

A more general question emerges: Why should judges who conflate statutory text and law be inclined to embrace literalism or the charge of it? Certainly there is no necessary slide from claiming that text is law to advancing literalism. But taking literally, and seriously, the idea that text is law does grease the wheels. To see why, recall that literal interpretations are acontextual, almost by definition.211 Taken literally, the sentence, an appeal must be filed no less than seven days after the judge issues the order, does not allow the losing party to appeal before seven days after the judge issues the pertinent order. Nobody can plausibly deny that this sentence, taken alone and out of broader context, provides at least this one unambiguous understanding—perhaps the only one—that the text will bear.

But now consider a committed textualist who denies that text is law and embraces the notion that contextually enriched meaning is not identical to literal meaning. They will face an uphill battle in rejecting literalist interpretations. A textualist who wants to argue, for example, that “less” really communicates “more,” in context, will have to identify an evident mistake. This, in turn, requires leaning heavily on extratextual resources including canons of construction, dictionaries, legislative history (for some textualists, at least), prior interpretive precedent, other statutory provisions, linguistic databases, and contestable and usually empirically unsubstantiated appeals to what “most” readers of the English language would understand, and so on.

Now, textualists routinely rely on extratextual resources.212 But any extratextual resource is susceptible to objections perfectly at home in the textualist’s rhetorical arsenal. More specifically, most extratextual considerations are subject to textualist objections on a piecemeal basis (e.g., “Dictionaries may be consulted, but my dictionary is better!”), as a class (e.g., “We should never consult legislative history!”), or—and this is most disconcerting to anti-literalist textualists—on a wholesale basis (e.g., “Of course we may rely on some extratextual sources to resolve ambiguities, yet the text here is wholly unambiguous and we may not create ambiguities that do not exist or rewrite legislation that is plain as day!”). An anti-literalist project of responding to textualist literalism moves the anti-literalist ever farther away from the text itself, and ever closer to methodologies like purposivism by relying heavily on extratextual resources.

The notion that appeals to context bring textualism and its competitors closer together is familiar.213 What is less often noticed, however, is how literalism in the guise of textualism provides a tempting path of least resistance.214 After all, a literal interpretation will always have at least one advantage over competing ones: Literalist interpretations require no explicit appeal to extratextual resources. They are—again, almost by definition—interpretations that the text bears acontextually.215 This simple advantage makes literalism tempting for the textualist, especially if they already believe that text is law, anyway.

To see the advantage, recall again Justice Kavanaugh’s exchange with Justice Gorsuch. Kavanaugh complained that Justice Gorsuch’s opinion manifested precisely the objectionable “literalist approach” that under discussion now,216 reminding us at great length that the “the good textualist is not a literalist” and that the Supreme Court has consistently rejected literalism.217 Reasonable or ordinary meaning, not literal meaning, is the interpreter’s guiding light. And, according to Justice Kavanaugh, the ordinary meaning of discrimination “because of” sex would be understood, at the time of adoption, to exclude discrimination on the basis of sexual orientation.218 To bolster this empirical claim, Justice Kavanaugh cites, among other things, “common parlance, common usage by Congress, the practice in the Executive Branch, the laws in the States, and the decisions of this Court.”219 Justice Alito adds his own evidence, offering a litany of dictionary entries,220 state laws outlawing homosexual sex and excluding homosexuals from employment or security clearances,221 and the decisions of the lower appellate courts.222

But that is a lot of work! More to the point, the literalist advantage over a textualist that marshals mountains of extratextual evidence becomes apparent. Because none of that evidence—not failed or contemporaneous bills, not original expected applications, not time-travel speculations about how ordinary persons would understand statutes, not dictionaries, corpus databases—none of it is law, nor overrides it, any more so than legislative history can. The text is law, and nothing else, so nothing else requires much attention if the text will suffice to provide guidance. And on that basis Justice Gorsuch writes a comparably short “refutation” of dozens of pages of the dissenters’ extratextual evidence, detailed argumentation, and appendices. Purporting to rely solely on the text, given that only text is law, likewise allows the literalist to safely dismiss attempts to enrich the relevant context, leaving a relatively acontextual analysis—perhaps even a literalist one. So, one can see why literalism may seem appealing, even if only rhetorically.

Consider two objections. The first observes that Justice Kavanaugh disclaims literalism in Bostock, even though he too repeatedly asserts that text is law.223 This should undermine the idea that the conflation inclines judges toward literalism. The second objection is related. No judges are committed literalists,224 nor are any theoretical defenders of textualism.225 Justice Gorsuch, accused of literalism by Justice Kavanaugh, would deny that he is a literalist.226 So literalism seems unlikely to flow merely from an embrace of the text-is-law dictum.

Both objections miss the point. The problem isn’t that embracing the idea that text is law necessarily leads to consistent and explicit legal literalism of the kind decried universally. Nor is the problem, as Jonathan Siegel suggested, a relentless march towards textualism’s radicalization.227 The more likely problem is one of selective literalism, which goes unacknowledged and deployed on an ad hoc basis. Conflating statutory text with law allows textualist judges, and others who might endorse the idea, to rest content with superficial literalism and provides them a false moral and conceptual high ground that relieves them of having to answer serious attempts to discern a text’s public meaning. These attempts are often hard work—again, dozens of pages from the dissenting Justices in Bostock—and in principle empirically testable.228 But it is much easier to dismiss it all on the grounds that text is law, and to insist that any systematic efforts to discover public meaning by mining extralegal context presents the real risk of making rather than applying the law.

So, identifying statutory text with law will not likely radicalize textualism as Siegel suggests.229 Instead, the text-is-law idea facilitates a form of opportunistic literalism by licensing strict adherence to the only thing that counts as law: the statutory text, understood acontextually. Textualists now have a tool to eschew extratextual context on the grounds that it contaminates the law’s purity. And they can do this while criticizing appeals to extratextual resources—even sources that textualists may support—as making the same “mistakes” as purposivists or intentionalists by affording them the power to pick out their friends from the crowd. The result is opportunistic literalism, deployed to support acontextual readings, turning the virtues of contextual readings into vices, while simultaneously turning textualist orthodoxy on its head.

Against Opportunistic Literalism

Textualists traditionally distance themselves from literalism, opting for “fair” or “reasonable” readings of statutory text, understood in context.230 And with good reason. The worst caricatures of formalism—where judges mechanically apply acontextual readings of statutes to reach nonsensical results divorced from reality—are associated with literalism. Justice Kavanaugh adds that literalism fails to provide fair notice.231 If there is one thing that all disputants in the interpretation wars can agree on, it is that systematically applying literalist interpretations is a bad thing.

But the critique changes when the problem is opportunistic literalism rather than either a radicalized textualism or literalism consistently applied. A “radicalized” textualism—of the kind Siegel predicted—might at least carry the virtues of being relatively rule governed. Compare a textualism that categorically and explicitly excludes, for example, legislative history from judicial consideration,232 to versions that consider some types of legislative history.233 The more categorical, rule-like approach would yield greater predictability about the sources of guidance that the courts will or will not rely on explicitly, at least when compared to the textualist with a relaxed attitude toward legislative history. As for consistent literalism, Justice Kavanaugh’s objection about fair notice would have little bite if literalism were consistently applied over the mine-run of cases. Doing so might actually improve predictability and fair notice, since individuals would not have to engage in any fieldwork about which discretionary appeals to context courts will rely on to flesh out what statutes mean. Literally, “less” will always means less,234 even at the cost of occasional absurdities arising when “unlawful” is construed to mean unlawful.235

Opportunistic literalism, by contrast, carries no such minimal virtue of consistency or of being governed by categorical exclusionary rules. Whether a literal, acontextual reading carries the day will not be predictable. Literalist arguments, accepted notwithstanding considerable extratextual evidence against literalist interpretations, might sometimes be afforded decisive weight. Sometimes not. Sometimes the assertion that text is law—and its rhetorical corollaries, like refusing to “rewrite” statutes—will operate as conversation stoppers, rendering any other appeals to context merely attempts to legislate from the bench.236 Other times the person committed to the trope that text is law—like Justice Kavanaugh—will emphasize ordinary conversational contexts to dislodge us from literalism.237 Indeed, the very same judge may make both maneuvers in the same case. Opportunistic literalism promises all the vices of literalism, without any of the virtues of being predictably applied.

Facilitating Myths About Judicial Lawmaking

There are other reasons to reject the conflation of text and law. Most basically, neither scholars nor judges should be in the business of spreading falsehoods that breed confusion.238 Teachers do not need to be reminded of this. Highly regarded judges who engage with the public—e.g., through semi-autobiographical trade press publications239have a similar responsibility. Popular engagements by judges constitute judicial outreach, which is a public service.240 Such outreach, writes Judge Katherine Beaumont Kern, “educate[s] the public about the courts and, in a larger sense, aim[s] to erase the divide between those on the bench and those in front of it.”241 Easy slogans like “text is law” serve to misinform unnecessarily.242

Nor is the mistake benign. The conflation of text and law belongs in the same family as other misleading oversimplifications, like Chief Justice Roberts’s comparison of judges with umpires who call “balls and strikes.”243 Judges are not and cannot be umpires.244 Umpires make empirical determinations about whether a pitch falls within the strike zone.245 Appellate courts, by contrast, almost always create law when rendering opinions, especially when construing statutes.246 If a statute is ambiguous or vague, appellate courts will pick one of the plausible interpretations the text will bear, thereby settling going forward the question that was unsettled before (a process that counts as a form of “construction,”247 according to some writers, though it is doubtful whether appellate-level interpretation ever fully avoids involving itself in some manner of construction).248 The appellate judgments partially constitute what the law requires. Stare decisis requires that later courts adhere to this “precisification”249 or “law elaboration.”250 By contrast, decisions by umpires rarely fundamentally change how the game is played in a constitutive way. Decisions by appellate judges, and Supreme Court Justices in particular, do so routinely.

More important differences exist. As Charles Fried writes in reflecting (largely sympathetically) on Chief Justice Roberts’s only “partially apt” metaphor:

I point out that both Justices [Breyer and Scalia] claim to be calling balls and strikes according to the statute, so there must be a further game—a meta-game, as it were—according to which one or the other approach to statutory interpretation is correct. But that meta-game is nowhere set down. It is a product of legal and political reflection—and, in respect to that, the judge is rule maker, player, and rule applier. Nonetheless, both Breyer and Scalia believe that they are judging a meta-game: they speak with great certitude—like an umpire calling a beanball—when they call, say, a committee report a ball or a strike.251

Fried observes that although the words arranged in a statutory text might be fixed, methodological disagreement—i.e., clashes in Fried’s “meta-game”—persists and is not tightly constrained.252 Judges still retain significant discretion as to whether they hitch their wagons to purposivism, intentionalism, textualism, pluralism, or a certain conception of each of these methodologies—or whether they will eschew any methodological pre-commitment.253 This remains a particularly deep problem for the umpire metaphor, since a judge’s position on the meta-game impacts the outputs of the game itself.254 Of course Chief Justice Roberts knows all this, so it is possible that Roberts is—as one commentator puts the point—“describing an ideal that right-thinking people should honor in theory, and judges and justices should strive at least to approximate in practice.”255 But it is also possible to view Roberts’s folksy analogy “as a gesture of contempt,” one “designed to preserve the fraudulent mystique of the high court as detached and apolitical when anyone can see from a surfeit of narrowly decided cases that justices are fully immersed in questions of politics and power.”256

Setting aside whether the baseball analogy counts as contemptuous, the trope that statutory text is law is, at best, similarly a misleading oversimplification that does more harm than good. The notion reinforces the misguided view that any lawmaking activities in cases involving statutes—even if the creation arises from a routine application of general terms to concrete facts—necessarily usurps the legislator’s role. This view ignores that courts almost always make new law when rendering opinions, especially when resolving statutory disputes.257 Again, if a statutory text presents a genuine linguistic ambiguity, appellate courts will pick one of the plausible interpretations that the text will bear, thereby settling the question that was unsettled before. And once again, if an appellate court applies a vague term to a particular set of facts, it effectively “makes new law” by making it more precise, at least under a particular set of facts.258 By contrast, texts are fixed, so all of the aforementioned, routine judicial engagements with statutes involve going beyond statutory texts. Everyday activities of appellate judges engaging with those texts become, if only the text is law, presumptively illegitimate because they usurp the legislature’s statutory drafting functions.

And notice what this means. The myth that “text is law” facilitates these other well-known but misleading metaphors that fuel overheated rhetoric, and in turn, might undermine the legitimacy of appellate courts. Whenever an unhappy litigant, commentator, or even Supreme Court Justice disagrees with a statutory interpretation or construction, they complain that the court illicitly “rewrote the statute”259 or “legislated from the bench.”260 As we have seen, Justice Alito’s dissent in Bostock uses precisely this rhetoric in trying to discredit the majority opinion, claiming that “[t]here is only one word for what the Court has done today: legislation.”261 Not only does Justice Alito accuse the majority of legislating from the bench—a familiar insult that accuses the majority of judicial activism—he construes Justice Gorsuch’s opinion as “deceptive.”262 This rhetoric may seem overheated.263 But the notion that “text is law” feeds these tired clichés about judges illicitly legislating from the bench.264 In short, facts about adjudicative practices in the United States continue to get drowned out by simplistic popular rhetoric about the role of judges and courts, facilitated by and including the notion that only statutory text is law.265 “Text is law” belongs in the same hall of shame as “balls and strikes.”266

Why the Conflation Will Continue

This Article has taken aim at a common assertion among textualists: that text is law. This conflation reflects a category mistake because text cannot be law, causes confusion about textualism, facilitates opportunistic literalism, and undermines popular understanding about appellate judging.

But this raises a puzzle. Why have textualists embraced the slogan given the costs associated with it? This Part speculates that textualists find the conflation of text and law rhetorically useful. More specifically, conflating text and law allows textualists, first, to claim that they, and only they, are honoring legislative work product. Second, claiming that text is law allows textualists to avoid difficult issues about linguistic meaning, including identifying the limiting principles for what does and does not constitute the relevant interpretive context. Finally, even if textualists could agree on how to identify that context, they still face deeper jurisprudential questions about the relationship between linguistic content and legal content. Conflating text and law remains rhetorically appealing because it redirects attention away from these thorny questions by emphasizing that which is fixed and determinate: a text.

Honoring the Text

Conflating text and law is rhetorically useful, first, because doing so bolsters textualism’s narrative of determinacy and its alleged ability to constrain judges, at least as compared to other approaches like purposivism or pluralism.267 Textualism, it is claimed, disciplines judges and prevents them from engaging in policy advocacy by tethering them to the text and cabining judicial discretion in giving legal effect to it.268 For example, textualists argue that this prevents judges from cherry picking legislative history that supports their preferred conclusions.269 There is also the stock argument that only textualism respects the products of legislative compromise.270 By hewing closely to the text, as opposed to allowing perceptions of policy or purpose to override it, textualists assert that they are more likely to avoid straying from what that compromise requires.271

All of these purported reasons to endorse textualism are familiar. In advancing these types of arguments, emphasizing text above all reinforces the claims that textualism is comparatively determinate and cabins judicial discretion. After all, a statutory text is static. Its constituent words and phrases are known and fixed—just like its meaning is supposed to be, according to textualists. Nor does text change absent further legislative reform. The textualist rhetorically leverages these static and fixed features of statutory texts by claiming fidelity to those same features. Given their preferred label, “textualists” could hardly claim to do otherwise.

By now we know that fidelity to statutory text is, according to the textualists themselves, at best a misleading shorthand for fidelity to the text’s context-sensitive linguistic meaning.272 For appellate judges routinely facing questions of statutory interpretation, the text’s meaning in context is the very thing that is routinely contested, even though the words comprising the underlying text is not. So, to the extent that textualists continue to tout their approach(es) to statutory interpretation on the grounds that it better cabins judicial discretion, textualists will readily slide into emphasizing fidelity to the statutory text which is indisputable rather than that which is routinely disputed—namely, statutory meaning. So, the rhetorical conflation of text and law by textualists will likely continue.

Avoiding Hard Questions, Part I: Linguistic Meaning

Conflating text and law also allows textualists to sidestep some difficult questions related to the role of intent and context in ascertaining linguistic meaning.

The Problem of Communicative Intent

As discussed above, textualists need not conflate text and law to articulate their core ideas, given that they can rely on ideas like context-sensitive linguistic meaning or what has been called “communicative content.”273 But these terms paper over some important difficulties that emerge in understanding “communicative content.”274 Texts can “communicate” many things, not all of which have anything to do with linguistic meaning, strictly speaking. For example, regulations that racially segregate residential communities communicate impermissible and false judgments of racial inferiority, even if those judgments are not expressly encoded in the words of the statute.275 Textualists seek something less unwieldy than what a text “communicates” or “expresses” in full generality. They are after something like linguistic meaning.

But as Mark Greenberg has shown, it is not always obvious what type of linguistic meaning textualists have in mind because they reject one obvious candidate—semantic content.276 Roughly, semantic content is content “conventionally encoded in the words that constitute a sentence” and “is approximately literal meaning.”277 As already noted, textualists deny that they seek literal meaning,278 contrasting it with the “fair” meaning that a statutory text can bear.279 What counts as “fair” meaning? Texts must be understood in context, textualists insist.280 So, naturally, this suggests that the relevant linguistic content is pragmatic content, which Greenberg describes as “what a speaker or author, by uttering words on a particular occasion in a particular context, manages to convey beyond, or different from, the semantic content of the words.”281 Recall our previous example, which is also Greenberg’s: The semantic content conveyed when a doctor utters to his patient, “you are not going to die,” is false.282 But in that particular conversational context, the pragmatically conveyed content will be something like, you are not going to die prematurely as a result of illness presently under discussion.283 Textualists who express preferences for “reasonable meaning” over “literal meaning” seem to have something like pragmatic content in mind.284

This is problematic. Understanding a statutory text’s meaning as its pragmatic content relies on a model of ordinary conversation that does not straightforwardly apply in the legislative context.285 This is because, as Greenberg points out, in ordinary conversational contexts—the touchstone for most textualists—pragmatic content is a function of the speaker’s intent.286 Figuring out what the physician means when he utters, “you are not going to die,” depends on an understanding of what the physician intended to communicate.287 But textualists have eschewed any search for fine-grained communicative intentions, having argued that specific communicative intentions with respect to specific statutory language is a hopeless task.288

This reveals a tension. If, on the one hand, textualists appeal to a model of ordinary linguistic exchange between speakers and listeners to extract meaning from texts, one would expect them to seek the speaker’s communicative intentions to ascertain meaning because that is what we do in ordinary conversation.289 On the other hand, textualists have compelling arguments showing that seeking legislative intent is often a fool’s errand.290 So on pain of incoherence, textualists must pick their poison—either by acknowledging that it seeks legislative intent or by abandoning the model of ordinary conversation.291

To relieve the tension, textualists appeal to an idealized agent—a reasonable speaker of a language, for example.292 That is, rather than seeking actual communicative intentions, textualists ask what an ordinary speaker of the English language would take the legislature’s communicative intentions to be in the context at issue.293 But manufacturing an intended audience is not the textualist’s only idealization. Textualists must also posit an idealized speaker, as though the legislature speaks with one voice.294

Natural worries arise about these idealizations.295 If discovering the text’s meaning becomes the project of discovering what a “reasonable” speaker would understand an idealized single legislator would have intended to communicate, we seem to invite indeterminacy rather than the promised determinacy that initially makes textualism so appealing.296 So, as others have argued, textualists owe us limiting principles to help cabin discretion.297

Textualists, to be sure, have a bevy of canons that they rely on.298 They appeal to dictionary definitions.299 Sometimes they rely on thought experiments to elicit linguistic intuitions.300 But the extent to which one ought to rely on such methodologies, if at all, is not settled by mere linguistic considerations alone. They depend on normative arguments that go well beyond the text, as even Scalia and Garner admit in their treatise on statutory interpretation.301 The content of those limiting principles remains an ongoing part of the research agenda among textualists.302 And, as Tara Leigh Grove points out, Bostock illustrates that some of these in-house debates are playing out quite publicly now.303

Rather than leaning into these difficulties, which in fairness textualist scholars recognize and take seriously,304 textualist judges or other public advocates of textualism keen on promoting the alleged comparative determinacy of textualism will likely end up embracing slogans like “text is law” to paper over these difficult issues. After all, to the extent that those promoting textualism emphasize its alleged comparative determinacy, better to emphasize that which is determinate and uncontroversial—namely, the words constituting a statutory text.

The Problem of Context Gerrymandering

It is one thing to say that a statutory text’s meaning is the meaning that a reasonable reader would understand in context, but it is quite another thing to identify what that context includes and excludes—i.e., what set of facts that reader or listener will understand or presuppose in reading the text at issue. As Grove acknowledges, “textualists have not always been precise in their use of the term ‘context.’”305 By emphasizing that text is law, however, textualists thereby avoid the difficult problems associated with identifying what to include or exclude from that context.

William Eskridge, Jr., and Victoria Nourse have usefully coined the term “textual gerrymandering” to refer to all the ways that textualists can—and have—exercised their discretion to selectively emphasize or de-emphasize fragments of statutory text and background context to support a given statutory interpretation.306 Focusing on context, “context gerrymandering” refers to the practice whereby judges determine what type of facts they include or exclude as part of the relevant context, as well as the practice of treating aspects of that context as important or comparatively trivial.307 To illustrate, some textualists exclude legislative history entirely.308 Others will consult it, and those who do decide how much, if any, weight to place on it.309 As with legislative history, so too with other available evidence contemporaneous with the statute’s enactment.310

Tara Grove likewise observes that textualists disagree among themselves about what considerations are legitimately considered part of the interpretive context.311 Returning to Bostock, Grove argues that the case involves dueling conceptions of textualism driven in large part by how each conception understands the relevant context of interpretation.312 She distinguishes between “formalistic textualism” and “flexible textualism,” two poles on a possible continuum of views.313 The latter conception “authorizes interpreters to make sense of the statutory language by looking at social and policy context, normative values, and the practical consequences of a decision.”314 Grove urges, in the final analysis, that courts side with the formalistic version of textualism.315 Formalistic textualism also looks at context, but favors what she calls “semantic context,” that excludes from consideration social and policy context, and studiously avoids where possible substantive canons of construction that reflect policy prescriptions rather than rules of linguistic usage.316 She takes the majority opinion to reflect formalistic textualism, with Justice Alito’s dissent—with its appeal to the “social context” of the statute—to reflect more flexible textualism.317

Regardless of whether this distinction is sound, there is a deeper tension between textualism of both “formalistic” and “flexible” flavors on the one hand, and the textualist’s foundational commitment to the reasonable interpreter standard on the other. Insofar as textualists seek the meaning of a statute that a reasonable person would understand in context, and to the extent that they try to model interpretation in terms of ordinary conversation, it is difficult to understand how any relevant piece of information could be automatically excluded from consideration—e.g., likely policy implications
—even if we care only about what a statute communicates. After all, absurd practical implications also influence our understanding of communicative intent, and in turn, communicative content. If I follow directions while driving to your home and following them to the letter would make me drive off a cliff, I ought to understand your directions differently. Because if I regard your directions as intending to mean strictly what they say, then that entails imputing to you murderous intent.

As with practical consequences, so too with other potentially relevant sources of information. But we need not pursue that point here. Grove correctly observes an in-house dispute among textualists that centers on debates about how to fix context. And those debates turn on matters of principle and policy far removed from the task that textualists purport to focus on—namely, understanding the text as written, nothing more, nothing less. But because this context gerrymandering is an inevitably value-laden enterprise318—one that potentially divides textualists—judges are, as a rhetorical matter, less likely to emphasize how undertheorized their positions are on these questions. But by professing to adhere to text—and treating that adherence as synonymous with fidelity to law—judges can avoid the hard questions of what to exclude or include from context.319

Avoiding Hard Questions, Part II: General Jurisprudence

Assume, for a moment, that textualists could agree on context. As legal philosophers have recently observed, however, the communicative content of a statutory text does not automatically count as the legal content that that statutory text effectuates.320 By “legal content” we mean the content of the legal rights, powers, and obligations that the text in question creates or changes. More to the point: The meaning of a statutory text is one thing, what that meaning contributes to the law’s content is quite another. Just as Part II showed that it is a mistake to conflate text and law, Mitchell Berman and Mark Greenberg have shown that conflating the linguistic meaning of texts and law also makes a basic conceptual mistake.321 Hrafn Asgeirsson, Dale Smith, and Lawrence Solum have also accepted the distinction between linguistic meaning and legal content.322

Asgeirsson usefully summarizes the point, observing that while the communicative content of some statutory text:

[D]epend[s] on one’s view about linguistic content . . . . [T]he legal content of such a provision . . . is its contribution to the law – ie a legal obligation, power, permission, etc (or set thereof). Metaphysically, therefore, these two types of content appear to be quite distinct. If that is correct, it cannot be the case that the legal content of such a provision is identical with, or constituted by, its communicative content. Or so the worry goes.323

In other words, linguistic content and legal content are simply different sorts of things, and we need a theory explaining how the former relates to the latter. Identifying the two involves, Asgeirsson continues, a “category mistake.”324

To bolster this point, Greenberg has observed that courts have historically “read into” criminal statutes mens rea elements even though the statutory text’s linguistic meaning apparently contains no such element.325 Solum points out that much of U.S. constitutional law, like doctrines prohibiting prior restraint under the First Amendment, have the same constitutional status as the First Amendment but are not communicated by the First Amendment’s sparse text.326 Similarly, Smith points out that when courts get the linguistic meaning of a statute wrong in a particular case, that mistake continues to bind courts—at least in jurisdictions with strong traditions of stare decisis.327

These are examples where an underlying text’s linguistic meaning contributes something to the law’s content. Construing texts to include mens rea requirements is an activity that still depends on, in some way, an underlying text’s meaning.328 So too, arguably, is the judicial construction of constitutional doctrine and the continued adherence to mistaken interpretations.329 But sometimes the linguistic meaning of duly enacted statutory texts contributes nothing to the law’s content. For example, Texas still maintains a duly enacted statute “on the books” that makes it a class C misdemeanor “to engage[] in deviate sexual intercourse with another individual of the same sex,”330 even though the U.S. Supreme Court declared it unconstitutional.331 This is a case where not only is the text of the statute not law, the text’s meaning contributes nothing to the law’s content.

These examples are harder for the textualist to respond to than scrivener’s errors—though not impossible. As argued above, textualists can exploit an ordinary conversational phenomenon, whereby speakers who make mistakes nevertheless succeed in communicated intended content.332 Some scrivener’s errors may be explained away by relying on misfires of these kinds of conversational mistakes. But relying on ordinary conversational contexts to explain away all putative counterexamples seems implausible. As Greenberg observes, “[i]t would be a strain to argue that mens rea requirements are somehow part of the linguistic content of criminal statutes, whatever their wording and whatever the circumstances of their enactment.”333 Asgeirsson concurs: Repeatedly relying on an underspecified understanding of context to yield the “right results” is suspicious, and involves “getting rather – and perhaps unjustifiably – creative with the resources available to the communicative-
content theorist in order to get around the problem posed by Greenberg’s [and others’] example[s].”334 Under threat of implausibility, the proposed counterexamples succeed in showing that we should not necessarily conflate a sentence’s linguistic meaning and the law’s content.

This does not mean that textualists are incorrect. But they owe a substantive theory about the relation between a statutory text’s linguistic meaning and its contribution to the law.335

Judges—who are not legal theorists—generally do not have the time or inclination to delve deeply into these thorny debates. But even if they do, it remains much easier to continue to pledge allegiance to duly enacted text, the existence of which is far less theoretically contested. The fact that textualism raises such hard philosophical questions provides yet further incentive for textualist judges to rest content with the claim that text is law. Because if text is law, the challenging metaphysical questions about how legal content supervenes on linguistic meaning, if at all, melt away. Those questions are just stipulated out of existence. So, the rhetorical advantages of running together text and law remain appealing, ensuring (unfortunately) that the conflation will likely continue.


Justice Gorsuch’s Bostock opinion suggests that statutory texts, and the words that make up that text, count as law.336 This is a mistake. Text and law are different kinds of things.337 Text is not law. Saying so is not only false but also contributes to a legal culture that already misleads the public about the nature of adjudication, and falsely suggests that judicial decisions confronting novel legal issues are illegitimate because they necessarily usurp the legislature’s law-making functions. What’s more, textualists have good reason to abandon the slogan that text is law, because it invites uncharitable interpretations of their doctrines, and because the slogan is not necessary given that those doctrines can be formulated without the dictum.338 And the text-is-law doctrine facilitates precisely the kind of literalism that textualists rail against. Textualists have enough difficulties and disputes on their plate. They have little need to add an unforced error.

None of these remarks refute textualism. That was not this Article’s aim. Some of these points are familiar. Different conceptions of textualism must make many implicit or explicit decisions—what Cary Franklin has called “shadow decision points”339—en route to claiming fidelity to textual meaning. But, unfortunately, by insisting that text is law some textualists succeed in obscuring these decisions. The text is what it is, they say, a fixed and objectively ascertainable arrangement of written symbols. Answering questions surrounding the meaning of “meaning” and how to understand context is, however, far more difficult to pin down and indeterminate in application, undermining the veneer of determinacy and judicial constraint often touted as a chief, if not the chief, advantage of textualism.

The Article concluded by speculating about why the conflation has proven so influential. Apart from its simplicity, the slogan embodies a textualist aspiration, which is to root its methodology in something so seemingly trivial, so seemingly undeniably true, that nobody could reasonably contest it. Congress makes law, it seems, by voting on texts, which the President signs. So, it must follow, it also seems, that texts are law. The problem is that, even if the meanings of statutory texts are law (which we have seen is itself contestable),340 the texts themselves are not. And failure to consistently appreciate this fact has pernicious effects in understanding textualism, facilitates literalism, and undermines our legal discourse and the public’s understanding of how appellate courts work. Text is not law. And we should stop thinking otherwise.

  1. [1]. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1754 (2020).

  2. [2]. See Julie Moreau, Supreme Court’s LGBTQ Ruling Could Have ‘Broad Implications,’ Legal Experts Say, NBC News (June 23, 2020, 3:40 AM),
    /supreme-court-s-lgbtq-ruling-could-have-broad-implications-legal-n1231779 [
    /2STV-WJ5Y]; John L. Culhane, Jr. & Ballard CFS Grp., SCOTUS Decision on Title VII Sexual Orientation Discrimination Has Significant Implications for Credit Arena, Ballard Spahr LLP: Consumer Fin. Monitor (June 19, 2020),
    06/19/scotus-decision-on-title-vii-sexual-orientation-discrimination-has-significant-implications-for-credit-arena []; Nina Totenberg, Supreme Court Delivers Major Victory to LGBTQ Employees, NPR (June 15, 2020, 5:52 PM),
    /863498848/supreme-court-delivers-major-victory-to-lgbtq-employees [
    W-7HAQ]; Joanna Wuest, The Supreme Court, Capital, and Queer Equality, Nation (June 16, 2020), [

  3. [3]. Even dissenting, Justice Brett M. Kavanaugh recognized the significance of the victory. Bostock, 140 S. Ct. at 1837 (Kavanaugh, J., dissenting) (“Notwithstanding my concern about the Court’s transgression of the Constitution’s separation of powers, it is appropriate to acknowledge the important victory achieved today by gay and lesbian Americans.”).

  4. [4]. Ilya Shapiro, After Bostock, We’re All Textualists Now, Nat’l Rev. (June 15, 2020, 9:08 PM),
    nty-we-are-all-textualists-now (“We’re all textualists now!”); Joshua Matz & Robbie Kaplan, Opinion, The Supreme Court’s Ruling on LGBTQ Protections Is a Triumph for Textualism — and Dignity, Wash. Post (June 15, 2020),
    n-lgbtq-protections-is-a-triumph-for-textualism—and-dignity/2020/06/15/4c8268ca-af4e-11ea-856d-5054296735e5_story.html [] (describing Bostock as “a triumph of textualism”); Daniel Hemel, The Problem with that Big Gay Rights Decision? It’s Not Really About Gay Rights., Wash. Post (June 17, 2020),
    /17/problem-with-that-big-gay-rights-decision-its-not-really-about-gay-rights [
    W6LT-FSYA] (“This week’s decision [in Bostock] is, yes, a triumph for textualism.”). For Justice Kagan’s remark, see Harvard L. Sch., The 2015 Scalia Lecture: A Dialogue with Justice Elena Kagan on the Reading of Statutes, YouTube (Nov. 25, 2015),
    FT0Tg [].

  5. [5]. See Josh Blackman & Randy Barnett, Justice Gorsuch’s Halfway Textualism Surprises and Disappoints in the Title VII Cases, Nat’l Rev. (June 26, 2006, 6:30 AM), https://www.nationalreview

  6. [6]. See Matz & Kaplan, supra note 4; Tara Leigh Grove, Comment, Which Textualism?, 134 Harv. L. Rev. 265, 266–67 (2020); Cary Franklin, Living Textualism, 2020 Sup. Ct. Rev. 119, 123–24 (explaining a compelling but dissenting view, which argues that textualism alone fails to fully explain both the majority opinion and dissenting opinions in Bostock); see also Anuj C. Desai, Text Is Not Enough, 93 U. Colo. L. Rev. 1, 13–28 (2022) (analyzing the arguments in Bostock).

  7. [7]. See, e.g., Melvin Aron Eisenberg, Strict Textualism, 29 Loy. L.A. L. Rev. 13, 36 (1995); Daniel S. Goldberg, I Do Not Think It Means What You Think It Means: How Kripke and Wittgenstein’s Analysis on Rule Following Undermines Justice Scalia’s Textualism and Originalism, 54 Clev. St. L. Rev. 273, 305 n.178 (2006) (“Textualism and originalism are incoherent in terms of their avowed purposes—they simply cannot constrain interpretation in the formal way that its proponents assert it does.”).

  8. [8]. Eisenberg, supra note 7, at 14 (arguing that “judges have an obligation to be faithful servants of the legislature, and the use of strict textualism violates this obligation”).

  9. [9]. Larry Alexander & Saikrishna Prakash, “Is That English You’re Speaking?” Why Intention Free Interpretation Is an Impossibility, 41 San Diego L. Rev. 967, 974–78 (2004) (arguing that textualism makes sense only as a form of intentionalism); Caleb Nelson, What Is Textualism?, 91 Va. L. Rev. 347, 353 (2005) (making a similar argument as Alexander and Prakash); Jonathan T. Molot, The Rise and Fall of Textualism, 106 Colum. L. Rev. 1, 36 (2006) (arguing “that which unites textualists and purposivists seems to outweigh that which divides them”); Anita S. Krishnakumar, Backdoor Purposivism, 69 Duke L.J. 1275, 1280 (2020) (arguing that “textualist Justices regularly speculate about legislative purpose and intent through other interpretive tools”). But see generally John F. Manning, What Divides Textualists from Purposivists?, 106 Colum. L. Rev. 70 (2006) (arguing that important distinctions between textualism and purposivism remain).

  10. [10]. Aharon Barak, Purposive Interpretation in Law 281-82 (Sari Bashi trans., 2005) (arguing that textualism harms democracy by treating legislation as “a decision made with no goal”); Nancy Staudt, Lee Epstein, Peter Wiedenbeck, René Lindstädt & Ryan J. Vander Wielen, Judging Statutes: Interpretive Regimes, 38 Loy. L.A. L. Rev. 1909, 1938 (2005) (“The textualist approach is allegedly anti-democratic because it ignores results the legislature intended and, thus, privileges the justices’ own idiosyncratic views regarding the meaning of words—a meaning that may well differ from the underlying legislative intent and purpose.”); Franklin, supra note 6, at 168-72 (arguing that textualism’s unacknowledged and unexplained “shadow decision points” creates a problem for democratic accountability).

  11. [11]. William N. Eskridge, Jr., Textualism, the Unknown Ideal?, 96 Mich. L. Rev. 1509, 1535 (1998) (reviewing Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law (1997)) (denying that Justice Scalia’s textualism “narrows the options of Scalia’s bête noire, the willful judge”); James J. Brudney & Corey Ditslear, Canons of Construction and the Elusive Quest for Neutral Reasoning, 58 Vand. L. Rev. 1, 6 (2005) (finding “little evidence to support legal process scholars’ claims that the canons serve as consistent or predictable guides to statutory meaning”); William N. Eskridge, Jr., The New Textualism and Normative Canons, 113 Colum. L. Rev. 531, 534 (2013) (reviewing Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts (2012)) (“[C]herry-picking among interpretive sources is a problem for all methodologies, and the new textualism offers no solution to this ancient dilemma.”); Anita S. Krishnakumar, Dueling Canons, 65 Duke L.J. 909, 959 (2016) (arguing, based on original empirical work, that “the canons and interpretive tools do not seem to be constraining the Justices to vote against their ideological preferences”); Franklin, supra note 6, at 154-55 (“This kind of cherry-picking of textualist tools and definitions offers significant reason to doubt the objectivity of the textualist enterprise, and the Justices offer no arguments for their methodological choices that would in any way allay those doubts.”).

  12. [12]. See generally Mark Seidenfeld, Textualism’s Theoretical Bankruptcy and Its Implication for Statutory Interpretation, 100 B.U. L. Rev. 1817 (2020) (defending a conception of legislative intent that escapes textualism’s criticisms).

  13. [13]. See Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1754 (2020) (Alito, J., dissenting); id. at 1822 (Kavanaugh, J., dissenting).

  14. [14]. The present author does, however, have serious doubts about textualism. See, e.g., Andrei Marmor, The Language of Law 107–09 (2014) (arguing that textualism is best understood as claiming that a statue’s legal content is its pragmatic content, while denying that textualism is useful for appellate judges); Mark Greenberg, Legal Interpretation and Natural Law, 89 Fordham L. Rev. 109, 113-24 (2020) (arguing that the linguistic meaning that textualism seeks is probably pragmatic content, and that this creates problems, among other things, because pragmatic content normally depends on communicative intent, which textualists foreswear); see also Franklin, supra note 6, at 169–72 (arguing that textualist reasoning is fraught with “shadow decision points” that obscure value-laden choices that textualism purports to eschew).

  15. [15]. See Bostock, 140 S. Ct. at 1737.

  16. [16]. Id.

  17. [17]. Id. at 1738.

  18. [18]. Neil M. Gorsuch, A Republic, If You Can Keep It 132 (2019) (emphasis added).

  19. [19]. In addition to the citations below, see Shapiro, supra note 4 (approving Justice Gorsuch’s assertion that “[o]nly the written word is the law”).

  20. [20]. Mitchell N. Berman, Judge Posner’s Simple Law, 113 Mich. L. Rev. 777, 804 n.113 (2015) [hereinafter Berman, Posner’s Simple Law](reviewing Richard A. Posner, Reflections on Judging (2013)); Mitchell N. Berman, The Tragedy of Justice Scalia, 115 Mich. L. Rev. 783, 786–87 (2017) [hereinafter Berman, Tragedy of Justice Scalia](reviewing Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law (1997)).

  21. [21]. See Berman, Posner’s Simple Law, supra note 20, at 804 n.113; Berman, Tragedy of Justice Scalia, supra note 20, at 786–87.

  22. [22]. Unless otherwise specified, this Article will use “linguistic meaning” and “communicative content” interchangeably. This may sound misleading to philosophers of language, who might regard linguistic meaning as synonymous with “semantic meaning,” where the latter term is associated (roughly) with the meaning encoded words and sentences as understood acontextually. See Hrafn Asgeirsson, The Nature and Value of Vagueness in the Law 131 (2020) (“[S]emantics is in the business of describing those features of expressions that are invariant between contexts of use.”). I use “linguistic meaning,” as well as “communicative content” for ease of exposition, since many textualists write in terms of “meaning” but appear to conflate it with “communicative content.” Where necessary, this Article tries to disambiguate.

  23. [23]. Brett M. Kavanaugh, Fixing Statutory Interpretation, 129 Harv. L. Rev. 2118, 2118 (2016) (reviewing Robert A. Katzmann, Judging Statutes (2014)); see also Brett M. Kavanaugh, Keynote Address: Two Challenges for the Judge as Umpire: Statutory Ambiguity and Constitutional Exceptions, 92 Notre Dame L. Rev. 1907, 1910 (2017) (“The text of a law is the law.”).

  24. [24]. Cont’l Can Co. v. Chi. Truck Drivers, Helpers & Warehouse Workers Union (Indep.) Pension Fund, 916 F.2d 1154, 1157 (7th Cir. 1990).

  25. [25]. Id. (emphasis added).

  26. [26]. Frank H. Easterbrook, What Does Legislative History Tell Us?, 66 Chi.-Kent L. Rev. 441, 444 (1990) (“The text of the statute—and not the intent of those who voted for or signed it—is the law.”); id. at 445 (“Because the text is the law ... .”).

  27. [27]. 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: Federal Courts and the Law 3, 22 (Amy Gutmann ed., 1997).

  28. [28]. Church of the Holy Trinity v. United States, 143 U.S. 457, 459 (1892).

  29. [29]. Scalia, supra note 27, at 22.

  30. [30]. Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 397 (2012) (emphasis added).

  31. [31]. Kofa v. U.S. I.N.S., 60 F.3d 1084, 1088 (4th Cir. 1995); Larry J. Soldinger Assocs., Ltd. v. Aston Martin Lagonda of N. Am., Inc., No. 97 C 7792, 1999 WL 756174, at *4 (N.D. Ill. Sept. 13, 1999); Anderson v. Hancock, 820 F.3d 670, 674 (4th Cir. 2016); Leggett v. EQT Prod. Co., No. 13CV4, 2016 WL 297714, at *10 (N.D. W. Va. Jan. 22, 2016); Young v. Schmucker, 409 B.R. 477, 481 (N.D. Ind. 2008).

  32. [32]. United States v. Evans, 148 F.3d 477, 483 (5th Cir. 1998); Appoloni v. United States, 450 F.3d 185, 199 (6th Cir. 2006) (Griffin, J., concurring in part); Solomon v. United States, 467 F.3d 928, 937 (6th Cir. 2006) (Griffin, J., dissenting); Freeman v. Wainwright, 959 F.3d 226, 232 (6th Cir. 2020); id. at 235 (Donald, J., dissenting) (“The majority is right: the text is the law.”); United States ex rel. Feingold v. AdminaStar Fed., Inc., 324 F.3d 492, 495 (7th Cir. 2003); Carmichael v. The Payment Ctr., Inc., 336 F.3d 636, 639 (7th Cir. 2003).

  33. [33]. Key v. Grayson, 163 F. Supp. 2d 697, 703 (E.D. Mich. 2001); United States v. Lentz, 275 F. Supp. 2d 723, 739 n.16 (E.D. Va. 2003), rev’d, 383 F.3d 191 (4th Cir. 2004); DeOtte v. Azar, 393 F. Supp. 3d 490, 509 n.9 (N.D. Tex. 2019), vacated and remanded sub nom., DeOtte v. State, 20 F.4th 1055 (5th Cir. 2021).

  34. [34]. O.P-G. v. State, 290 So. 3d 950, 956 (Fla. Dist. Ct. App. 2019); Singletary v. State, 713 S.E.2d 698, 701 n.13 (Ga. Ct. App. 2011); State v. Quintanilla, 276 So. 3d 941, 948 (Fla. Dist. Ct. App. 2019); In re L.R., 729 S.E.2d 520, 521 n.6 (Ga. Ct. App. 2012); People v. Lint, 629 N.W.2d 924, 925 (Mich. 2001) (Corrigan, C.J., dissenting); Ga. Lottery Corp. v. Tabletop Media, LLC, 816 S.E.2d 438, 445 n.4 (Ga. Ct. App. 2018) (Dillard, C.J., concurring) (“The traditional view is that an enacted text is itself the law.” (quoting Scalia & Garner, supra note 30, at 397)); Alamo Heights Indep. Sch. Dist. v. Clark, 544 S.W.3d 755, 763 (Tex. 2018).

  35. [35]. United States v. Hall, 617 F.3d 1161, 1167 (9th Cir. 2010), aff’d, 566 U.S. 506 (2012) (“[I]t is our duty to follow the text because the text is the law.”); Owen v. Univ. of Ky., 486 S.W.3d 266, 272 (Ky. 2016) (“The text is law.”); In re Smith, 447 B.R. 435, 446 (W.D. Pa. 2011), as amended (Mar. 14, 2011) (“[I]t is our duty to follow the text because the text is the law.” (quoting Hall, 617 F.3d at 1167)); Hernandez v. State, 861 S.W.2d 908, 909 (Tex. Crim. App. 1993) (en banc).

  36. [36]. Saikrishna B. Prakash, Unoriginalism’s Law Without Meaning, 15 Const. Comment. 529, 536–37 (1998) (reviewing Jack N. Rakove, Original Meanings: Politics and Ideas in the Making of the Constitution (1996)) (“When the text is law, the desires, fears, and floor statements of the legislators are not.”). Prakash later softened his view, allowing that intentions play an inescapable role in legal interpretation. See Alexander & Prakash, supra note 9, at 979–82.

  37. [37]. Ilya Somin, Response, Is Textualism Doomed?, 158 U. Pa. L. Rev. PENNumbra 235, 237 (2010).

  38. [38]. See A. Benjamin Spencer, Class Actions, Heightened Commonality, and Declining Access to Justice, 93 B.U. L. Rev. 441, 488 (2013).

  39. [39]. Cass R. Sunstein, Must Formalism Be Defended Empirically?, 66 U. Chi. L. Rev. 636, 662 (1999) (“To be sure, the text is the law, and we can agree that policy judgments, and intentions standing alone, do not represent the law.”).

  40. [40]. This “realist” critique of the concept of legislative intent originates in Max Radin, Statutory Interpretation, 43 Harv. L. Rev. 863, 870 (1930) (“A legislature certainly has no intention whatever in connection with words which some two or three men drafted, which a considerable number rejected, and in regard to which many of the approving majority might have had, and often demonstrably did have, different ideas and beliefs.”). Textualists have endorsed the critique to the extent that it construes intentionalism as the search for some legislative agent’s subjective intent. See John F. Manning, The Absurdity Doctrine, 116 Harv. L. Rev. 2387, 2390 (2003) [hereinafter, Absurdity Doctrine] (“[T]extualists argue that the (often unseen) complexities of the legislative process make it meaningless to speak of ‘legislative intent’ as distinct from the meaning conveyed by a clearly expressed statutory command.”); see also John F. Manning, Textualism and Legislative Intent, 91 Va. L. Rev. 419, 430 (2005) [hereinafter, Manning, Textualism and Legislative Intent] (“[T]extualists deny that a legislature has any shared intention that lies behind but differs from the reasonable import of the words adopted; that is, they think it impossible to tell how the body as a whole actually intended (or, more accurately, would have intended) to resolve a policy question not clearly or satisfactorily settled by the text.”).

  41. [41]. This version of the textualist complaints against competing methodologies has been simplified for ease of exposition. For a discussion of both the common ground between textualism and purposivism and remaining differences, see generally Manning, supra note 9 (examining in depth textualism and purposivism).

  42. [42]. Amy Coney Barrett, Substantive Canons and Faithful Agency, 90 B.U. L. Rev. 109, 113 (2010). Elsewhere Barrett is more careful to avoid suggesting that text is law, observing that “when we refer to the originalist commitment to ‘text,’ we mean text as originalists interpret it—i.e., in accordance with its original public meaning.” Amy Coney Barrett & John Copeland Nagle, Congressional Originalism, 19 U. Pa. J. Const. L. 1, 5 (2016). I thank Brian Pérez-Daple for bringing this point to my attention.

  43. [43]. See, e.g., infra notes 47–51 and accompanying text.

  44. [44]. Jonathan R. Siegel, The Inexorable Radicalization of Textualism, 158 U. Pa. L. Rev. 117, 134 (2009) (“Textualists believe that the passage of text through the constitutional process of enactment imbues that text with legal force, regardless of anyone’s intent or purpose.”).

  45. [45]. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1738 (2020). Here, Gorsuch echoes his pre-Bostock book, which asserted that “[t]extualism honors only what’s survived bicameralism and presentment—and not what hasn’t. The text of the statute and only the text becomes law.” Gorsuch, supra note 18, at 132 (emphasis added).

  46. [46]. Cont’l Can Co. v. Chi. Truck Drivers, Helpers & Warehouse Workers Union (Indep.) Pension Fund, 916 F.2d 1154, 1157 (7th Cir. 1990).

  47. [47]. See generally John F. Manning, Textualism as a Nondelegation Doctrine, 97 Colum. L. Rev. 673 (1997) (arguing that relying on judicial reliance on “legislative history as an ‘authoritative’ source of legislative intent” incentivizes Congress “to bypass bicameralism and presentment” and that the judiciary “should not impute a committee’s ... declaration of intent to Congress as a whole”).

  48. [48]. Cf. William N. Eskridge, Jr., The New Textualism, 37 UCLA L. Rev. 621, 673 (1990) (observing that the Federal Rules of Civil Procedure, although promulgated consistent with the Rules Enabling Act, were not themselves approved by congress or signed by the president). John Manning describes legal norms arising without bicameral approval as examples of “law elaboration,” which is either administratively necessary to effectuate broad Congressional mandates (in the case of agencies) or inherent in the roles of the official (in the case of judges). See Manning, supra note 47, at 691–96.

  49. [49]. See Eskridge, Jr., supra note 48, at 623 (“Legislative history should not even be consulted to confirm the apparent meaning of a statutory text.”).

  50. [50]. Compare Gorsuch, supra note 18, at 131–32 (2019) (Textualism “starts with dictionary definitions, rules of grammar, and the historical context in which a law was adopted to see what its language meant to those who adopted the law”), and John F. Manning, Inside Congress’s Mind, 115 Colum. L. Rev. 1911, 1926 (2015) (“Textualists tend to move briskly to dictionary definitions, rules of grammar and syntax, and (since statutes are, after all, legal instruments) canons of interpretation or terms of art peculiar to the legal community.”(emphasis omitted)), with Eskridge, Jr., supra note 48, at 672 (“Consulting [committee reports] does not violate bicameralism or presentment any more than would consulting a dictionary.”).

  51. [51]. Victoria Nourse, Misreading Law, Misreading Democracy 166–67 (2016).

  52. [52]. More specifically, John Manning argues that the reason judges should avoid treating legislative history as a decisive source of authority flows from constitutional norms against Congressional self-delegation, which promises to explain why legislative history counts as a uniquely problematic source of authority. See Manning, supra note 47, at 706–31.

  53. [53]. Gorsuch, supra note 18, at 132.

  54. [54]. For example, if x is an amount of snow, then x is an amount of H2O. See Scott Soames, Beyond Rigidity: The Unfinished Semantic Agenda of Naming and Necessity 287–88, 294
    –306 (2002). The motivation for this conditional formulation of the identity relation is to avoid the absurdities that allow us to “prove,” for example, that some amount of ice is some amount of steam, given that both are water. See id.

  55. [55]. Ofra Magidor, Category Mistakes, Stan. Encyc. of Phil. (July 5, 2019), https://plato.stan [].

  56. [56]. Id.

  57. [57]. This example comes from Gilbert Ryle, who is credited with coining the term “category mistake.” Gilbert Ryle, The Concept of Mind 16–19, 203–06 (1949).

  58. [58]. See generally Brian Epstein, The Ant Trap: Rebuilding the Foundations of the Social Sciences (2015) (arguing against methodological individualism in the social sciences—i.e., arguing that social phenomena like institutions must be explained by reference to more than merely aggregations of individuals).

  59. [59]. Jonathan Siegel has argued at length that (a) the claim that text is law is a distinctive and essential component of textualism, and (b) this shows how textualism inevitably leads to implausible implications. See Siegel, supra note 44, at 120–22. I do not think (a) is correct, and if textualists wish to avoid (b), they can and should reject (a). See infra Part III.

  60. [60]. Berman, Posner’s Simple Law, supra note 20, at 778–80; Mitchell N. Berman, Our Principled Constitution, 166 U. Pa. L. Rev. 1325, 1384–85 (2018) [hereinafter Berman, Our Principled Constitution].

  61. [61]. Berman, Our Principled Constitution, supra note 60, at 1384–85 (footnotes omitted); see also Berman, The Tragedy of Justice Scalia, supra note 20, at 787 (offering “proof that text is not the same as either meaning or law”).

  62. [62]. Scalia, supra note 27, at 22.

  63. [63]. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1737 (2020).

  64. [64]. Berman, Posner’s Simple Law, supra note 20, at 804­–05.

  65. [65]. Id. at 804­–05 n.113.

  66. [66]. Id.

  67. [67]. ProCD, Inc. v. Zeidenberg, 86 F.3d 1447, 1452 (7th Cir. 1996).

  68. [68]. Here is another illustration: Most statutes are codified in English, but the law’s requirements are widely translated into different languages. English texts—linguistic symbols in certain syntactical structures—differ from Spanish texts, which have their own conventional symbols and syntactic structure. These are different texts. But they purport to articulate the same legal requirements or permissions. As it turns out, this move directly from meanings of legal text to legal content is itself controversial. See infra Part IV.

  69. [69]. Cf. Ryan D. Doerfler, The Scrivener’s Error, 110 Nw. U. L. Rev. 811, 816 (2016) (“[A] ‘scrivener’s error’ is a case in which the words of a legislative text diverge from what Congress meant to say.”).

  70. [70]. Scurto v. LeBlanc, 184 So. 567, 574 (La. 1938) (emphasis added).

  71. [71]. Id. (emphasis added).

  72. [72]. See id.

  73. [73]. See, e.g., Lexington Ins. Co. v. Precision Drilling Co., L.P., 830 F.3d 1219, 1223 (10th Cir. 2016) (Gorsuch, J., writing for himself alone in this portion of the opinion) (citing Scalia & Garner, supra note 30, at 235–38 (“Take the scrivener’s error ... . In cases like these, the error in the statute is so ‘unthinkable’ that any reasonable reader would know immediately both (1) that it contains a ‘technical or ministerial’ mistake, and (2) the correct meaning of the text.”)); Pharm. Mfg. Rsch Servs., Inc. v. Food & Drug Admin., 957 F.3d 254, 262 (D.C. Cir. 2020) (“True scrivener’s errors are unusual and we should not lightly assume that Congress has made one.”); Michael S. Fried, A Theory of Scrivener’s Error, 52 Rutgers L. Rev. 589, 590 (2000) (“Even staunch textualists typically concede the necessity of permitting judges to reform statutes that exhibit plain typographical mistakes.”). For an illuminating and recent discussion of scrivener’s errors doctrine, see generally Doerfler, supra note 69 (discussing the effect of scrivener’s doctrine).

  74. [74]. Scalia, supra note 27, at 20–21.

  75. [75]. See e.g., Scalia & Garner, supra note 30, at 234 (“No one would contend that the mistake cannot be corrected if it is of the sort sometimes described as a ‘scrivener’s error.’”).

  76. [76]. See sources cited supra note 73.

  77. [77]. See William N. Eskridge, Jr., Dynamic Statutory Interpretation 45–47 (1994); Alexander & Prakash, supra note 9, at 979–82.

  78. [78]. Siegel, supra note 44, at 146.

  79. [79]. Cameron v. Auto Club Ins. Ass’n, 718 N.W.2d 784, 787–89 (Mich. 2006), overruled by Regents of Univ. of Mich. v. Titan Ins. Co., 791 N.W.2d 897 (Mich. 2010).

  80. [80]. See id. at 787.

  81. [81]. Id. at 788 (quoting Mich. Comp. Laws § 600.5851(1) (1994)).

  82. [82]. Id. at 789.

  83. [83]. Id.

  84. [84]. Id.

  85. [85]. Manning, supra note 9, at 99; Barrett, supra note 42, at 113.

  86. [86]. Cameron, 718 N.W.2d at 791 (“The reason is that there are several conceivable explanations, as we have pointed out, why the Legislature could have intended the result the plain language of the statute requires.”).

  87. [87]. Id. at 822 (Kelly, J., dissenting) (describing the majority’s opinion as “rooted in the personal views of Associate Justice Antonin Scalia of the United States Supreme Court”).

  88. [88]. Id. (citations omitted).

  89. [89]. See Eskridge, Jr., supra note 77, at 45–47; Veronica M. Dougherty, Absurdity and the Limits of Literalism: Defining the Absurd Result Principle in Statutory Interpretation, 44 Am. U. L. Rev. 127, 158–59 (1994); Eisenberg, supra note 7, at 29; Jonathan R. Siegel, Textualism and Contextualism in Administrative Law, 78 B.U. L. Rev. 1023, 1100 (1998); Jonathan R. Siegel, What Statutory Drafting Errors Teach Us About Statutory Interpretation, 69 Geo. Wash. L. Rev. 309, 333–35 (2001); Siegel, supra note 44, at 146–47; Linda D. Jellum, But That Is Absurd!: Why Specific Absurdity Undermines Textualism, 76 Brook. L. Rev. 917, 939 (2011) (“Even narrow versions of these [absurdity and Scrivener’s error] doctrines undercut textualist principles to the extent courts are permitted to consider policy or legislative intent that runs contrary to unambiguous statutory language.” (quoting Andrew S. Gold, Absurd Results, Scrivener’s Errors, and Statutory Interpretation, 75 U. Cin. L. Rev. 25, 60 (2006))).

  90. [90]. Siegel, supra note 44, at 145.

  91. [91]. Id.

  92. [92]. See infra text accompanying notes 93–96.

  93. [93]. See Adrian Vermeule, Interpretive Choice, 75 N.Y.U. L. Rev. 74, 83 (2000) (footnote omitted).

  94. [94]. Manning, Textualism and Legislative Intent, supra note 40, at 420. For a version of this claim articulated in terms of “objective” meaning, see Gold, supra note 89, at 29–30.

  95. [95]. John F. Manning, Second-Generation Textualism, 98 Calif. L. Rev. 1287, 1288 (2010).

  96. [96]. Scalia & Garner, supra note 30, at 16 (describing the above-quoted language as the key principle of textualism).

  97. [97]. See infra Section IV.B.

  98. [98]. This assumption is what Mark Greenberg has called the “standard picture.” See Mark Greenberg, The Standard Picture and Its Discontents, in 1 Oxford Studies in Philosophy of Law 39, 39–42 (Leslie Green & Brian Leiter eds., 2011).

  99. [99]. Asgeirsson, supra note 22, at 126.

  100. [100]. John Perry, Textualism and the Discovery of Rights, in Philosophical Foundations of Language in the Law 105, 105 (Andrei Marmor & Scott Soames eds., 2011).

  101. [101]. Gideon Rosen, Textualism, Intentionalism, and the Law of the Contract, in Philosophical Foundations of Language in the Law, supra note 100, at 130, 131–32 (emphasis added) (footnotes omitted).

  102. [102]. See, e.g., Scott Soames, What is Meaning? 9 (2010).

  103. [103]. Asgeirsson, supra note 22, at 126; Rosen, supra note 101, at 131–32; Perry, supra note 100, at 105–07. Lawrence Solum’s account of the “constraint principle” proposes that the “communicative content”—i.e., the content identified via the process of interpretation—ought to constrain the judge’s construction of legal texts. See Lawrence B. Solum, Communicative Content and Legal Content, 89 Notre Dame L. Rev. 479, 482 (2013) [hereinafter, Solum, Communicative Content and Legal Content]. That is, Solum offers conceptions of textualism as involving the search for communicative content capable of constraining the choice of legal rule that governs cases. Id.; see also Lawrence B. Solum, Contractual Communication, 133 Harv. L. Rev. F. 23, 41 (2019) [hereinafter, Solum, Contractual Communication] (“Textualism affirms that the plain meaning (communicative content) of a statutory text should constrain the legal content of statutory constructions.”). What Solum means by “constraint” is not altogether clear. “Constraint” suggests normativity, e.g., an authority obligating that adjudicators not rule in a manner that is inconsistent with original public meaning. But that too is a fraught question. For a helpful discussion of Solum’s work on the idea of constraint, see generally Ethan J. Ranis, Note, Loose Constraints: The Bare Minimum for Solum’s Originalism, 93 Tex. L. Rev. 765 (2015) (discussing what Solum has coined “the Constraint Principle”). For a more recent attempt by Solum to formulate a “simplified” version of textualism’s core positive commitment as requiring that legal content be “consistent with” and “fairly traceable to” a text’s communicative content, see generally Lawrence B. Solum, Essay, Deferentialism: Soames on Legal Interpretation, Phil. Stud. (Jan. 23, 2022) (book review), [https://].

  104. [104]. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1749 (2020).

  105. [105]. See id.

  106. [106]. Id. at 1738 (emphasis added).

  107. [107]. See supra notes 93–101 and accompanying text.

  108. [108]. For the notion that a sentence’s meaning counts as legal content, see Rosen, supra note 101, at 131–33. For the notion that meaning—or as Solum prefers to put it, “communicative content”—constrains legal content, see Solum, Communicative Content and Legal Content, supra note 103, at 482; Solum, Contractual Communication, supra note 103, at 41.

  109. [109]. See supra note 101 and accompanying text.

  110. [110]. See, e.g., Soames, supra note 102, at 9–14 (discussing the theoretical role of propositions in modern philosophy of language).

  111. [111]. See, e.g., In re Sinclair, 870 F.2d 1340, 1342 (7th Cir. 1989) (allowing that interpretation must attend to a text’s “contexts—linguistic, structural, functional, social, historical”); John F. Manning, Textualism and the Equity of the Statute, 101 Colum. L. Rev. 1, 108 (2001); Gorsuch, supra note 18, at 10.

  112. [112]. Greenberg, supra note 14, at 113–14.

  113. [113]. Scalia, supra note 27, at 20–21.

  114. [114]. Nelson, supra note 9, at 356.

  115. [115]. Gold, supra note 89, at 56. For a criticism of Gold, see John David Ohlendorf, Textualism and the Problem of Scrivener’s Error, 64 Me. L. Rev. 119, 139–42 (2011).

  116. [116]. Doerfler,supra note 69, at 833 (emphasis omitted).

  117. [117]. See, e.g., Siegel, supra note 44, at 146–47.

  118. [118]. Scurto v. LeBlanc, 184 So. 567, 574 (La. 1938).

  119. [119]. This way of putting things helps itself to terminology at home in speech act theory. Marmor, supra note 14, at 12 (“The simple, or ‘standard’ view that I strive to defend here can be stated as follows: the collective action of the legislators enacting a law is a collective speech act, whereby some content is communicated that is, essentially, the content of the law voted on.”); Scott Soames, Deferentialism: A Post-Originalist Theory of Legal Interpretation, 82 Fordham L. Rev. 597, 597 (2013) (“The first question in interpretation is: what does the law say, assert, or stipulate? Saying, asserting, and stipulating are speech acts—or, in more technical philosophical terminology, illocutionary acts—as are confirming, denying, ordering, and promising.”). This approach likewise appeals to context, though with the added benefit of allowing us to distinguish in more fine-grained ways how context matters. For example, context might help to identify the deontic status of the speech act—e.g., does the illocutionary force of the legislative act count as a prohibition, assignment of rights, delegation of duties, or empowerment? See J.L. Austin, How To Do Things with Words 109 (J.O. Urmson & Marina Sbisà eds., 2d ed. 1975) (explaining the “force” of illocutionary speech acts, roughly, in terms of what they aim to accomplish in virtue of uttering them). But context might also help to identify the descriptive content of a given prohibition, assignment, delegation, or empowerment. For our purposes, however, a more coarse-grained appeal to “context” will suffice.

  120. [120]. This is not to say that these textualist rationales for departing from text will always be convincing, despite the theoretical possibility that communicated content may depart from a text’s literal meaning. Nor does the present author necessarily endorse these rationales. Instead, the point is to show how textualists have the theoretical resources to mount such a defense, at least once they reject the conflation of text and legal content. I thank Jonathan Siegel for pressing me on this point.

  121. [121]. Richard H. Fallon, Jr., The Meaning of Legal “Meaning” and Its Implications for Theories of Legal Interpretation, 82 U. Chi. L. Rev. 1235, 1244–45 (2015).

  122. [122]. See infra Section IV.A.

  123. [123]. Lawrence Solum’s Legal Theory Lexicon shows a careful, nuanced range of interpretations of textualism’s commitments, none of which presupposes or depends on the false claim that text is law. See Lawrence B. Solum, Legal Theory Lexicon: Textualism, Legal Theory Blog (Oct. 27, 2019, 1:50 PM),
    lism.html [].

  124. [124]. Berman, Our Principled Constitution, supra note 60, at 1384–85.

  125. [125]. Scalia, supra note 27, at 23 (“Textualism should not be confused with so-called strict constructionism, a degraded form of textualism that brings the whole philosophy into disrepute ... . A text should not be construed strictly, and it should not be construed leniently; it should be construed reasonably, to contain all that it fairly means.”).

  126. [126]. Id.

  127. [127]. Manning, supra note 111, at 108 (“Modern textualists ... are not literalists.”).

  128. [128]. See, e.g., In re Sinclair, 870 F.2d 1340, 1342 (7th Cir. 1989) (allowing that interpretation must attend to a text’s “contexts—linguistic, structural, functional, social, historical”); Manning, supra note 111, at 108–15; Gorsuch, supra note 18, at 10.

  129. [129]. See Eskridge, Jr., supra note 48, at 668 (“[S]ome criticize Justice Scalia and the new textualists for having a ‘hidden agenda,’ to wit: By deferring to Republican-controlled agencies and narrowly construing the liberal laws of the Democrat-controlled Congress, the new textualists (mainly conservative Republicans) seek to reduce the power of government to do good in our society.”); Jerry L. Mashaw, Textualism, Constitutionalism, and the Interpretation of Federal Statutes, 32 Wm. & Mary L. Rev. 827, 834 (1991) (“Is this retreat to the text merely a conservative plot to undermine liberal statutes?”); Andrei Marmor, The Immorality of Textualism, 38 Loy. L.A. L. Rev. 2063, 2065 (2005); Jeremy K. Kessler & David E. Pozen, Working Themselves Impure: A Life Cycle Theory of Legal Theories, 83 U. Chi. L. Rev. 1819, 1850 (2016) (“The theory of new textualism was suspected from the start of being a political project aimed at restraining judicial and legislative efforts to create a more liberal administrative state.”).

  130. [130]. Marmor, supra note 129, at 2065. Marmor has since softened his critique, observing that, to the extent that textualism asserts that a statute’s legal content consists of its communicated content, textualism is unobjectionable. He nevertheless argues that the thesis performs little useful work in contested cases of statutory interpretation that are the mainstay of appellate adjudication. See Marmor, supra note 14, at 107–10.

  131. [131]. Or “impure,” depending on one’s perspective. See Kessler & Pozen, supra note 129, at 1850 (treating the evolution of textualist theory as a case study in the development and change of legal theories).

  132. [132]. See supra Part I.

  133. [133]. But see generally Ronald Dworkin, Law’s Empire (1986) (defending constructive interpretation as involving imputing to practices a purpose that casts that practice in its best light).

  134. [134]. See, e.g., Nourse, supra note 51, at 103–08 (urging greater attention to textualism as applied in practice).

  135. [135]. See supra Part I.

  136. [136]. See generally Siegel, supra note 44 (criticizing what he sees as the increasing radicalization of textualism).

  137. [137]. See generally Somin, supra note 37 (responding to Siegel’s claims by arguing “textualism’s adherence to text is compatible with eliminating scrivener’s errors” and that “Siegel understates the importance of textual ambiguity”).

  138. [138]. Siegel, supra note 44, at 144–68.

  139. [139]. Id. at 120 (quoting Scalia, supra note 27, at 22). Siegel followed up in a series of articles insisting that the “textualist ideal” consists simply in the strict identification of duly enacted text and the law’s content. Jonathan R. Siegel, Symmetries—and Asymmetries—Between Theories of Statutory Interpretation, 99 Cornell L. Rev. Online 182, 186 (2014) (“The core distinction is simple: textualists believe that ‘the text is the law.’” (quoting Scalia, supra note 27, at 22)); Jonathan R. Siegel, The Legacy of Justice Scalia and His Textualist Ideal, 85 Geo. Wash. L. Rev. 857, 906 (2017) (asserting that the textualist ideal holds that “statutory text simply is the law by virtue of being legislatively adopted”).

  140. [140]. Siegel, supra note 44, at 169–78.

  141. [141]. This is an oversimplification. For a more detailed discussion of the various realist attacks on intentionalism and purposivism, see id. at 131–32.

  142. [142]. Id. at 148–53.

  143. [143]. Id. at 144–45.

  144. [144]. Id. at 148–53.

  145. [145]. Siegel also discusses the radicalizing effects of ignoring statutory purposes. Id. at 153–57. But for the sake of simplicity, we will set aside that helpful discussion.

  146. [146]. Id. at 170–78 (criticizing a radicalized textualism as “unworkable”).

  147. [147]. Kessler & Pozen, supra note 129, at 1853 n.122.

  148. [148]. See supra Section III.B.

  149. [149]. For an argument there is little practical difference between textualism and other theories of statutory interpretation, see Molot, supra note 9, at 36–43; see also supra Part II (discussing how textualists conflate statutory text and the law in detail).

  150. [150]. See Molot, supra note 9, at 36–43.

  151. [151]. See generally Somin, supra note 37 (discussing how textualism is compatible with textual ambiguity and eliminating scrivener’s errors, as well as the future of textualism).

  152. [152]. Id. at 237.

  153. [153]. Id.

  154. [154]. See id. at 237–38.

  155. [155]. Id.

  156. [156]. Id. at 238.

  157. [157]. Scurto v. LeBlanc, 184 So. 567, 574 (La. 1938).

  158. [158]. See Somin, supra note 37, at 237–41.

  159. [159]. Id. In a similar vein, John Manning tries to reduce the scope of scrivener’s errors to a vanishingly small set of cases involving typos, where those typos were not the result of legislative compromise. Manning, Absurdity Doctrine, supra note 40, at 2460 n.265 (“Or a court might notice a common mistake of grammar or punctuation that makes linguistic nonsense of an otherwise comprehensible sentence.” (citation omitted)). I don’t see how this is responsive. Typos are text too, text no less ratified through the bicameral process and presented to the President for signature. And typos may involve no linguistic nonsense, as the Scurto case shows. See Ohlendorf, supra note 115, at 139 (“Imposing minimum conditions of meaningfulness on statutory texts can at most justify the correction of deviant texts. It cannot be used to justify correcting non-deviant errors, like those in ... Scurto, which authorized any ‘unlawful’ impeachment of testimony ... .” (footnote omitted)).

  160. [160]. This might be true, anyway. But my point above is that there is no need to make an unforced error on the matter.

  161. [161]. See supra Section III.A.

  162. [162]. See supra Section III.A.

  163. [163]. For a helpful summary, see generally Adam N. Steinman, “Less” is “More”? Textualism, Intentionalism, and a Better Solution to the Class Action Fairness Act’s Appellate Deadline Riddle, 92 Iowa L. Rev. 1183 (2007) (discussing textualism and intentionalism in the context of the Class Action Fairness Act).

  164. [164]. Id. at 1187 (“When a class action is removed from state court to federal court, CAFA authorizes discretionary appeals of district court decisions on whether removal is proper, but only ‘if application is made to the court of appeals not less than 7 days after entry of the order.’” (citation omitted)).

  165. [165]. See, e.g., Miedema v. Maytag Corp., 450 F.3d 1322, 1326 (11th Cir. 2006) (“We now reaffirm that construction of § 1453(c)(1), for to read it literally would produce an absurd result: there would be a front-end waiting period (an application filed 6 days after entry of a remand order would be premature), but there would be no back-end limit (an application filed 600 days after entry of a remand order would not be untimely).”).

  166. [166]. See id.; Morgan v. Gay, 466 F.3d 276, 278–79 (3d Cir. 2006); Amalgamated Transit Union Local 1309, v. Laidlaw Transit Servs., Inc., 435 F.3d 1140, 1146 (9th Cir. 2006); Pritchett v. Office Depot, Inc., 420 F.3d 1090, 1093 n.2 (10th Cir. 2005).

  167. [167]. Miedema, 450 F.3d at 1326; Morgan, 466 F.3d at 277 (“Because the uncontested legislative intent behind § 1453(c) was to impose a seven-day deadline for appeals, we conclude that the statute as written contains a typographical error and should be read to mean ‘not more than 7 days.’”); Amalgamated Transit Union, 435 F.3d at 1146; Pritchett, 420 F.3d at 1093 n.2 (“Given Congress’ stated intent to impose time limits on appeals of class action remand orders and the limited availability of appeals prior to the statute’s enactment, we can think of no plausible reason why the text of [the] Act would instead impose a seven-day waiting period followed by a limitless window for appeal.”).

  168. [168]. See Miedema, 450 F.3d at 1326.

  169. [169]. See S. Rep. No. 109-14, at 49 (2005).

  170. [170]. See 151 Cong. Rec. 1,638 (2005).

  171. [171]. See cases cited supra note 167.

  172. [172]. See Amalgamated Transit Union, 435 F.3d at 1146; Amalgamated Transit Union Local 1309 v. Laidlaw Transit Servs., Inc., 448 F.3d 1092, 1093–94 (9th Cir. 2006).

  173. [173]. Amalgamated Transit Union, 448 F.3d at 1093–94 (denying rehearing en banc).

  174. [174]. Id. at 1094–95 (Bybee, J., dissenting).

  175. [175]. Spivey v. Vertrue, Inc., 528 F.3d 982, 983–85 (7th Cir. 2008).

  176. [176]. Id.

  177. [177]. Amalgamated Transit Union, 435 F.3d at 1095 (Bybee, J., dissenting).

  178. [178]. See Statutory Time-Periods Technical Amendments Act of 2009, Pub. L. No. 111-16, § 6(2), 123 Stat. 1607, 1608 (2009) (codified as amended at 28 U.S.C. § 1453 (c)(1)).

  179. [179]. Siegel, supra note 44, at 142.

  180. [180]. Id. at 152.

  181. [181]. See supra Section III.B.

  182. [182]. See supra Section III.B.

  183. [183]. Siegel, supra note 44, at 168–78.

  184. [184]. Yellen v. Confederated Tribes of Chehalis Rsrv., 141 S. Ct. 2434, 2460 n.3 (2021) (Gorsuch, J., dissenting) (“At most, [the absurdity doctrine] may serve a linguistic function
    —capturing circumstances in which a statute’s apparent meaning is so ‘unthinkable’ that any reasonable reader would immediately (1) know that it contains a ‘technical or ministerial’ mistake, and (2) understand the correct meaning of the text.”).

  185. [185]. Although characterizing “late-stage textualism” as “wooden” and “embarrassing” rather than “radical,” Ryan Doerfler argues that textualism as practiced on the Supreme Court has become mired in dueling linguistic canons of construction, making Karl Llewellyn’s famous critique of the interpretive canons relevant again. He explains the Court’s move toward linguistic pedantry on political factors (i.e., the rise of textualist personnel on the Supreme Court), as well as a legal environment unable and unwilling to acknowledge linguistic or legal underdeterminacy. See generally Ryan D. Doerfler, Late-Stage Textualism, 2022 Sup. Ct. Rev. (forthcoming 2022), [] (discussing the development of textualism).

  186. [186]. See supra Section III.A.

  187. [187]. Somewhat more precisely, and following Mark Greenberg, a sentence’s “literal meaning” is to be roughly synonymous with its “semantic content,” which is (also) roughly the content expressed, if at all, independent of context when words are arranged in a certain grammatical form. See Greenberg, supra note 14, at 113–24; see also Asgeirsson, supra note 22, at 131 (“[S]emantics is in the business of describing those features of expressions that are invariant between contexts of use.”). For a helpful discussion of literalism, see generally Bill Watson, Literalism in Statutory Interpretation, 2021 U. Ill. L. Rev. Online 218.

  188. [188]. If “context” is understood as statutory text surrounding the text under immediately evaluation, then presumably this type of context would not stand in tension with the notion that text is law. The law, not extratextual considerations, would count as the relevant context for the text under examination.

  189. [189]. The superficial appeal is this Article’s primary concern, though I should be clear that the formalist considerations discussed above are not compelling, and I don’t endorse them. As Joseph Raz argued long ago, the fact that judges systematically rely on extralegal norms doesn’t necessarily convert them to legal ones. See Joseph Raz, The Authority of Law: Essays on Law and Morality 46 (2d. ed. 2009) (using conflicts of laws to illustrate the point that the judicial practice of consulting a normative system does not automatically convert those norms into legal norms of the judge’s jurisdiction). A fortiori systematically relying on a given extratextual source does not necessarily suggest unlawful adjudicative practice. The point in the discussion above is that considerations that lead judges to regard skeptically extratextual sources will also make literalism more appealing.

  190. [190]. See infra Section IV.B.3 (discussing opportunistic literalism).

  191. [191]. Brian Leiter, Explaining Theoretical Disagreement, 76 U. Chi. L. Rev. 1215, 1229 (2009) (arguing that judges “tend to be opportunistic . . . to their approach to the meaning of authoritative sources in contested cases” (emphasis omitted)).

  192. [192]. Opportunistic literalism need not be pursued in bad faith; judges may inconsistently offer literalist arguments without realizing precisely what they are doing, and without acknowledging that their arguments are literalist at all.

  193. [193]. See Krishnakumar, supra note 9, at 1280; Franklin, supra note 6, at 124–25; Desai, supra note 6, at 3.

  194. [194]. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1824 (2020) (Kavanaugh, J., dissenting).

  195. [195]. Both lecture the majority for writing their own policy preferences into the statute or legislating from the bench. Id. at 1754 (Alito, J., dissenting) (“There is only one word for what the Court has done today: legislation.”); id. at 1836 (Kavanaugh, J., dissenting) (“[W]hen this Court usurps the role of Congress, as it does today, the public understandably becomes confused about who the policymakers really are in our system of separated powers, and inevitably becomes cynical about the oft-repeated aspiration that judges base their decisions on law rather than on personal preference.”). Both stress that Title VII’s meaning must be understood in context and further specify that the relevant temporal context was the time of enactment. Id. at 1755 (Alito, J., dissenting); id. at 1828 (Kavanaugh, J., dissenting). Both deny that the ordinary meaning of discrimination “because of ... sex” could possibly include, at that time, discrimination on the basis of sexual orientation or transgender status. Id. at 1771–72 (Alito, J., dissenting); id. at 1828
    –33 (Kavanaugh, J., dissenting). And both Justices appealed to failed efforts to amend Title VII to explicitly bar discrimination against LGBTQ+ persons as evidence that the ordinary understanding of discrimination “because of” sex excludes discrimination because of homosexuality or transgender status. Id. at 1755 (Alito, J., dissenting); id. at 1824 (Kavanaugh, J., dissenting).

  196. [196]. Id. at 1824–25 (Kavanaugh, J., dissenting).

  197. [197]. Id. at 1822.

  198. [198]. See supra Section IV.A.

  199. [199]. Bostock, 140 S. Ct. at 1824–25 (Kavanaugh, J., dissenting).

  200. [200]. Id. at 1822.

  201. [201]. For example, he dismisses as inconclusive evidence that congressional bills attempted to amend Title VII, as well as evidence that nobody would have understood Title VII to encompass discrimination because of homosexuality or transgender status. See id. at 1747 (majority opinion). The former is just another type of legislative history, inferences about which could cut either way, while the latter is just a roundabout way of appealing to subjective intentions or expected applications, rather than the meaning of the statute as written. Id. at 1750 (majority opinion) (“Rather than suggesting that the statutory language bears some other meaning, the employers and dissents merely suggest that, because few in 1964 expected today’s result, we should not dare to admit that it follows ineluctably from the statutory text.”).

  202. [202]. See, e.g., Id. at 1824 (Kavanaugh, J., dissenting) (“According to this theory, it follows that the man has been fired, at least as a literal matter, because of his sex.”); id. (“For the sake of argument, I will assume that firing someone because of their sexual orientation may, as a very literal matter, entail making a distinction based on sex.”); id. (“The plaintiffs must establish that courts, when interpreting a statute, adhere to literal meaning rather than ordinary meaning.”).

  203. [203]. Id. at 1745 (majority opinion).

  204. [204]. Id.

  205. [205]. See Nelson Lund, Unleashed and Unbound: Living Textualism in Bostock v. Clayton County, 21 Fed. Soc’y Rev. 158, 162 (2020) (“Taking ‘because of’ to mean ‘motivated by’ is every bit as literal as Gorsuch’s but-for causation interpretation.”); see also James A. Macleod, Ordinary Causation: A Study in Experimental Statutory Interpretation, 94 Ind. L.J. 957, 962 (2019) (“As it turns out, most people do not interpret common causal phrases to imply but-for causation, even in the linguistic and factual contexts at issue in these and similar cases.” (emphasis omitted)).

  206. [206]. More rhetorical advantages will be discussed infra Part V.

  207. [207]. Bostock, 140 S. Ct. at 1767, 1773 (Alito, J., dissenting).

  208. [208]. See generally Franklin, supra note 6 (arguing that both the majority and dissenting opinions presuppose a set of “shadow decision points” about, among other things, the context in which the text is understood).

  209. [209]. Id. at 176; see also Lund, supra note 205, at 161 (discussing the opinions in Bostock).

  210. [210]. Bostock, 140 S. Ct. at 1738 (emphasis added).

  211. [211]. See Greenberg, supra note 14, at 113.

  212. [212]. See, e.g., S. Austin Coal. Cmty. Council v. SBC Commc’ns Inc., 274 F.3d 1168, 1172 (7th Cir. 2001) (approving of reliance on certain forms of legislative history under certain conditions); Bostock, 140 S. Ct. at 1739 (citing Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 350 (2013)).

  213. [213]. Molot, supra note 9, at 3; see also generally Nelson, supra note 9 (canvassing ways that differences between textualism and other methodologies are overstated).

  214. [214]. Victoria Nourse has argued that “petty textualism” involves boiling complex legal issues down to a handful of words or phrases, taking them out of context, and then re-contextualizing those words or phrases in a way that yields a desired interpretive result. See Nourse, supra note 51, at 107–08. Her analysis is compelling. But Nourse seems more skeptical than the present author about the possibility of acontextual interpretations. See id. at 107 (arguing that words taken out of context have “so many meanings” so as to have “none”). And petty textualism seems predicated on isolating small snippets of text from a “lengthy statute,” whereas opportunistic literalism need not. See id. Whatever the differences, there is likely considerable overlap in cases involving petty textualism and opportunistic literalism. Both practices, to the extent they are engaged in, are anathema to textualism in theory.

  215. [215]. Greenberg, supra note 14, at 113.

  216. [216]. Bostock, 140 S. Ct. at 1824 (Kavanaugh, J., dissenting).

  217. [217]. Id. at 1824–28 (quoting Scalia, supra note 27, at 24).

  218. [218]. Id. at 1833–34.

  219. [219]. Id. at 1833.

  220. [220]. Id. at 1765–66 (Alito, J., dissenting).

  221. [221]. Id. at 1772 (“We must ask whether Americans at that time would have thought that Title VII banned discrimination against an employee for engaging in conduct that Congress had made a felony and a ground for civil commitment.”).

  222. [222]. Id. at 1760–61.

  223. [223]. See supra Part II.

  224. [224]. Nelson, supra note 9, at 376 (“[N]o mainstream judge is interested solely in the literal definitions of a statute’s words ... .” (footnote omitted)).

  225. [225]. Scalia, supra note 27, at 23–24; Manning, supra note 111, at 108.

  226. [226]. See Bostock, 140 S. Ct. at 1750 (“[W]e must be attuned to the possibility that a statutory phrase ordinarily bears a different meaning than the terms do when viewed individually or literally.”).

  227. [227]. See generally Siegel, supra note 44 (arguing that textualism is different from intentionalism and purposivism and that this difference is growing over time).

  228. [228]. Corpus linguistics approaches to interpretation take seriously the empirical dimension of “ordinary” meaning. See Thomas R. Lee & Stephen C. Mouritsen, Judging Ordinary Meaning, 127 Yale L.J. 788, 821–45 (2018) (arguing that empirical methods exist to test claims about a text’s ordinary meaning, and that they should be more widely adopted). For critical assessments, see John S. Ehrett, Against Corpus Linguistics, 108 Geo. L.J. Online 50, 61–70 (2019) (arguing that judges should refrain from corpus linguistic assessments); Evan C. Zoldan, Corpus Linguistics and the Dream of Objectivity, 50 Seton Hall L. Rev. 401, 409–41 (2019) (arguing that corpus linguistics do not make legal interpretation more objective); Kevin P. Tobia, Testing Ordinary Meaning, 134 Harv. L. Rev. 726, 747–78 (2020) (pointing out limitations of corpus linguistics on conceptual and empirical grounds); Franklin, supra note 6, at 145 (criticizing the use of corpus linguistics databases in Bostock); Matthew Jennejohn, Samuel Nelson & D. Carolina Núñez, Hidden Bias in Empirical Textualism, 109 Geo. L.J. 767, 782–802 (2021) (arguing that corpus linguistics incorporates invisible biases while presenting a method for mitigating them); and Anya Bernstein, Legal Corpus Linguistics and the Half-Empirical Attitude, 106 Cornell L. Rev. 1397, 1418
    –29 (2021) (criticizing as overstated corpus linguistics, inter alia, as a technique for ascertaining meaning insofar as it ignores peculiarities of legal language as it appears in legal contexts).

  229. [229]. See Siegel, supra note 44, at 120–22.

  230. [230]. See, e.g., Scalia, supra note 27, at 23–24.

  231. [231]. Bostock, 140 S. Ct. at 1828 (Kavanaugh, J., dissenting).

  232. [232]. See Scalia, supra note 27, at 31–35.

  233. [233]. See S. Austin Coal. Cmty. Council v. SBC Commc’ns Inc., 274 F.3d 1168, 1172 (7th Cir. 2001) (“[T]he legislative history—the enactment history, not the fog of words generated by legislators—shows that ‘common carrier’ [in § 7 of the Clayton Act of 1914] means all common carriers. The version of § 7 that passed by the House used the word ‘railroad’; the Senate amended this to ‘common carrier’, a broader designation; the House acceded to the Senate’s amendment.”); Manning, supra note 47, at 737 n.272 (1997) (“[T]extualist judges ... do not categorically exclude a statute’s drafting evolution from their consideration of statutory context.”).

  234. [234]. Steinman, supra note 163, at 1187.

  235. [235]. See Scurto v. LeBlanc, 184 So. 567, 574 (La. 1938).

  236. [236]. Again, witness Justice Gorsuch all but conceding the characterizations of his opinion as wooden and literalist yet insisting that the blame lies with the text. See supra Section IV.B.2.

  237. [237]. See supra Section IV.B.2.

  238. [238]. See Berman, Our Principled Constitution, supra note 60, at 1384–85; Berman, Posner’s Simple Law, supra note 20, at 804 n.113. I qualify “unnecessarily” to make room for well-understood legal fictions, like legal personhood for corporations.

  239. [239]. See generally, e.g., Gorsuch, supra note 18 (discussing the author’s career as a lawyer and Supreme Court justice and reflecting on the role of the judiciary in society).

  240. [240]. Katherine Beaumont Kern, Be the Change, Judges’ J., Fall 2019, at 15, 15.

  241. [241]. Id.; see also Shirley S. Abrahamson, Judicial Independence as a Campaign Platform, Bench & Bar Minn., November 2004, at 28, 28 (“Lawyers and judges must educate the public on judicial roles and duties.”).

  242. [242]. Some oversimplifications are useful first steps en route to deeper understanding. A model of the atom that compares the nucleus to the sun and electrons to circling planets may help new learners gain a foothold on the concepts of electrons and the atomic nucleus, even though the laws governing electrons and planets differ dramatically. But construing the conflation of text with law as an educational oversimplification assumes that more carefully specifying that statutory texts express meanings, which in turn count as or constrain law, is significantly more difficult to understand than asserting that that text is law simpliciter. Understanding quantum mechanics is not on par with understanding the idea that texts have meanings, at least in terms of justifying pedagogical oversimplification. The educational function of the conflation is limited. At a minimum, were education the goal, then those who peddle the conflation should pause to note the oversimplification. Cf. Paul G. Hewitt, The Bohr Model of the Atom, Sci. Tchr., Jan.–Feb. 2021, at 14, 16 (observing that physicist Niels Bohr “was quick to stress that his model was to be interpreted as a crude beginning, and the picture of electrons whirling about the nucleus like planets about the Sun was not to be taken literally,” and that his “sharply defined orbits were conceptual representations of an atom whose later description involved waves—quantum mechanics”). I thank Adrienne Erickcek and Thomas S. Jackson for helpful discussion.

  243. [243]. Confirmation Hearing on the Nomination of John G. Roberts, Jr. to be Chief Justice of the United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 56 (2005) (statement of John G. Roberts, Jr., Nominee to be Chief Justice of the United States).

  244. [244]. Smith v. Farley, 59 F.3d 659, 665 (7th Cir. 1995) (“Judges are not umpires, calling balls and strikes; or judges of a moot court, awarding victory to the side that argues better; least of all is that their disposition in a death case.”); Charles Fried, Balls and Strikes, 61 Emory L.J. 641, 641 (2012) (“Critics balked because the metaphor suggests that there is always, at least in principle, an objectively correct call, the umpire being only a necessarily imperfect human approximation of what an accurate electronic monitor could settle beyond possibility of dispute—as is done in determining the order of finish in a horse race (a photo finish).”).

  245. [245]. Strike Zone, Major League Baseball, [] (“The official strike zone is the area over home plate from the midpoint between a batter’s shoulders and the top of the uniform pants—when the batter is in his stance and prepared to swing at a pitched ball—and a point just below the kneecap. In order to get a strike call, part of the ball must cross over part of home plate while in the aforementioned area. Strikes and balls are called by the home-plate umpire after every pitch has passed the batter, unless the batter makes contact with the baseball (in which case the pitch is automatically a strike).”).

  246. [246]. Vaughn R. Walker, Moving the Strike Zone: How Judges Sometimes Make Law, 2012 U. Ill. L. Rev. 1207, 1207 (arguing that “judges not only make law but cannot avoid doing so”).

  247. [247]. Lawrence B. Solum, The Interpretation-Construction Distinction, 27 Const. Comment. 95, 96–108 (2010).

  248. [248]. The difficulty in strictly separating in practice the oft-asserted distinction between interpretation and construction is one of the overarching lessons of Franklin, supra note 6, at 154–55.

  249. [249]. Soames, supra note 119, at 604 (describing the process of stipulating answers to vague legal questions in a way that coheres with statutory rationales).

  250. [250]. Manning, supra note 47, at 699 (“Consistent with that traditional understanding, textualists also recognize that ‘no statute can be entirely precise,’ and that the elaboration of statutory detail inevitably takes place outside the formal confines of bicameralism and presentment.” (footnote omitted) (quoting Mistretta v. United States, 488 U.S. 361, 415 (1989) (Scalia, J., dissenting))).

  251. [251]. Fried, supra note 244, at 643 (emphasis added).

  252. [252]. See id.

  253. [253]. See Michael C. Dorf, Will Liberal Justices Pay a Price for Signing onto Justice Gorsuch’s Textualist Opinions?, Dorf on L. (July 22, 2020, 7:00 AM),
    al-justices-pay-price-for.html [] (“The very fact that [the liberal Justices] were willing to join the textualist majority opinions in Bostock and McGirt even though they have also written and joined purposivist (or occasionally intentionalist) opinions shows that they regard methodology as secondary to results.”).

  254. [254]. Former Judge Richard Posner makes a similar point. “Neither [Chief Justice Roberts] nor any other knowledgeable person actually believed or believes,” scolds Judge Posner, “that the rules that judges in our system apply, particularly appellate judges and most particularly the Justices of the U.S. Supreme Court, are given to them the way the rules of baseball are given to umpires.” Richard A. Posner, How Judges Think 78 (2008). To make the umpire metaphor work, Judge Posner continues, “[w]e must [also] imagine that umpires, in addition to calling balls and strikes, made the rules of baseball and changed them at will.” Id. at 78–79. Although Fried strikes a more conciliatory tone than Judge Posner, and would probably deny that all the rules are changeable at will, surely the methodologies reflect discretionary judgment calls for which there are few binding rules laid down ex ante.

  255. [255]. John F. Harris, Take the Hint, John Roberts—It’s Time for You to Retire, Too, Politico (Feb. 3, 2022, 4:30 AM), [].

  256. [256]. Id.

  257. [257]. See Walker, supra note 246, at 1207.

  258. [258]. Soames, supra note 119, at 604 (emphasis omitted) (describing the process of stipulating answers to vague legal questions in a way that coheres with statutory rationales).

  259. [259]. See, e.g., Waddell v. USS Agri-Chemicals, 523 So. 2d 683, 684 (Fla. Dist. Ct. App. 1988) (Barfield, J., dissenting) (“This court in Bonner rewrote the statute to provide that benefits were payable unless their entitlement were somehow disproved by the amount of income received by the worker.”).

  260. [260]. See, e.g., Avallone Mech. Co. v. City of Las Cruces, No. CV 12-0677, 2013 WL 12333496, at *2 (D.N.M. Jan. 14, 2013). Plaintiffs assert that: ... the state district court improperly legislated from the bench by imposing the code of evidence upon the city’s ordinance in the January 2012 order ... .” Id.

  261. [261]. Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1754 (2020) (Alito, J., dissenting).

  262. [262]. Id.

  263. [263]. For a more sympathetic take on Justice Alito’s rhetoric, see Mitchell N. Berman & Guha Krishnamurthi, Bostock was Bogus: Textualism, Pluralism, and Title VII, 97 Notre Dame L. Rev. 67, 120–25 (2021).

  264. [264]. None of this denies that other myths may loom larger. A deeper misunderstanding holds that courts should never make new law, a mistake that depends on a highly simplistic understanding of separation of powers. But the notion that only statutory text is law is a natural corollary to that myth, one that helps to prop it up.

  265. [265]. Cf. Erik Encarnacion, Response, On Raphael Demos’s Legal Fictions,” 125 Ethics 540, 541–42 (2015) (arguing that legal fictions may “mislead[] the public about the nature of appellate adjudication and create[] false expectations about judicial roles”).

  266. [266]. Confirmation Hearing on the Nomination of John G. Roberts, Jr. to be Chief Justice of the United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 56 (2005) (statement of John G. Roberts, Jr., Nominee to be Chief Justice of the United States).

  267. [267]. For a concise summary of textualism’s broadside against purposivism, see Grove, supra note 6, at 271–74.

  268. [268]. Id. at 271 (“Textualists claimed to offer an approach that would be more faithful to the words actually used by the legislature and also better constrain the federal judiciary.”).

  269. [269]. Id. at 274.

  270. [270]. See Manning, supra note 9, at 77.

  271. [271]. See id.

  272. [272]. See supra Section III.B.

  273. [273]. Id.

  274. [274]. The meaning of “meaning” as used in law is famously ambiguous. See, e.g., Fallon, Jr., supra note 121, at 1244–45 (distinguishing among, “(1) semantic or literal meaning; (2) contextual meaning as framed by shared presuppositions of speakers and listeners, including shared presuppositions about application and nonapplication; (3) real conceptual meaning; (4) intended meaning; (5) reasonable meaning; and (6) interpreted meaning”).

  275. [275]. See generally Plessy v. Ferguson, 163 U.S. 537 (1896) (upholding racial segregation in the face of constitutional challenges); Brown v. Bd. of Educ., 347 U.S. 483 (1954) (overruling Plessy).

  276. [276]. Greenberg, supra note 14, at 113.

  277. [277]. Id.

  278. [278]. Scalia, supra note 27, at 23–24 (“[T]he good textualist is not a literalist ... .”).

  279. [279]. Id. at 23 (“A text should not be construed strictly, and it should not be construed leniently; it should be construed reasonably, to contain all that it fairly means.”).

  280. [280]. See Greenberg, supra note 14, at 110–13 nn.5–19 (collecting sources).

  281. [281]. Id. at 113.

  282. [282]. Id. at 114.

  283. [283]. Id.

  284. [284]. Scalia, supra note 27, at 23 (“A text should not be construed strictly, and it should not be construed leniently; it should be construed reasonably, to contain all that it fairly means.”).

  285. [285]. Greenberg, supra note 14, at 114.

  286. [286]. Id. at 119–22.

  287. [287]. Id. at 113–14.

  288. [288]. Id.

  289. [289]. Id.

  290. [290]. Id. at 121.

  291. [291]. For more extensive criticism of attempts to model statutory interpretation on the interpretation of ordinary conversation, see Evan C. Zoldan, The Conversation Canon, Ky. L.J. (forthcoming), [

  292. [292]. Greenberg, supra note 14, at 121.

  293. [293]. See Solum, Communicative Content and Legal Content, supra note 103, at 482.

  294. [294]. See Greenberg, supra note 14, at 120–21. For a sophisticated defense of legislative intent as a fiction, see generally Ryan D. Doerfler, Who Cares How Congress Really Works?, 66 Duke L.J. 979 (2017).

  295. [295]. Doerfler, supra note 294, at 981–82.

  296. [296]. See id. at 1023 nn.226–27.

  297. [297]. See, e.g., id. at 986 (proposing a few suggestions for managing interpretations of intent).

  298. [298]. Scalia & Garner, supra note 30, at 397–99.

  299. [299]. Ellen P. Aprill, The Law of the Word: Dictionary Shopping in the Supreme Court, 30 Ariz. St. L.J. 275, 280–82 (1998) (criticizing the practice of “dictionary shopping”).

  300. [300]. See Smith v. United States, 508 U.S. 223, 242 (1993) (Scalia, J., dissenting) (positing that nobody would understand the phrase “us[ing] a cane” to mean displaying it on a wall, and analogizing that usage with the phrase “us[e] a firearm”).

  301. [301]. See Scalia & Garner, supra note 30, at 8–11, 53–69, 397­–410 (advocating a “normative” defense of certain canons of construction, while disputing others that courts have recognized).

  302. [302]. See id.

  303. [303]. Grove, supra note 6, at 271–74.

  304. [304]. See id.

  305. [305]. Id. at 281.

  306. [306]. See 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, 1720–38 (2021).

  307. [307]. See id. at 1743.

  308. [308]. See Scalia, supra note 27, at 31–35.

  309. [309]. See Manning, supra note 47, at 731–38.

  310. [310]. Richard Fallon has argued that textualists are not alone in owing us an explanation of what generally will be included or excluded from the context in which statutory text will be interpreted, and why. Richard H. Fallon, Jr., Three Symmetries Between Textualist and Purposivist Theories of Statutory Interpretation – and the Irreducible Roles of Values and Judgment Within Both, 99 Cornell L. Rev. 685, 719 (2013).

  311. [311]. Grove, supra note 6, at 271–74.

  312. [312]. See id.

  313. [313]. Id. at 285–86.

  314. [314]. Id. at 286.

  315. [315]. See id.

  316. [316]. See id.

  317. [317]. See id.

  318. [318]. See Fallon, Jr., supra note 310, at 713–24 (discussing value-laden judgements inherent in the practice of statutory interpretation); Franklin, supra note 6, at 153–55 (arguing that background “shadow decision points” are outcome determinative even on textualist analyses of Title VII, rendering textualism unlikely to provide a value-free approach to statutory analysis). See generally Grove, supra note 6 (exploring two types of textualism and arguing that the more formalistic type protects the legitimacy of the judicial branch).

  319. [319]. After asserting that “[c]ontext is the ‘mutually salient information’ that an author exploits to make evident to her audience what she means[,]” and that common knowledge determines context, Ryan Doerfler argues that this in turn bolsters the ordinary conversation model of statutory interpretation. Doerfler, supra note 294, at 1031–32 (footnote omitted). Although there is much to be said, many of the debates within and about textualism boil down to normative debates about what should and should not be considered common knowledge for the purposes of statutory interpretation, as well as whether we ought to be empowered to serve as epistemic gatekeeper even when certain information would ordinarily count as common knowledge. Doerfler does not deny this, I gather, but might encourage normative appeals to conversational norms governing context to constrain normative theorizing about context grounding in political or constitutional theory, rather than the other way around. See id. at 1033.

  320. [320]. See supra Section IV.B.1.

  321. [321]. No one has done more to draw attention to this issue than Mark Greenberg. See Greenberg, supra note 98, at 39; Mark Greenberg, The Moral Impact Theory of Law, 123 Yale L.J. 1288, 1296–99 (2014); Mark Greenberg, Legislation as Communication? Legal Interpretation and the Study of Linguistic Communication, in Philosophical Foundations of Language in the Law, supra note 100, at 217, 223–24. See generally Mark Greenberg, Response, What Makes a Method of Legal Interpretation Correct? Legal Standards vs. Fundamental Determinants, 130 Harv. L. Rev. F. 105 (2017) (emphasizing, for example, “the importance of distinguishing between, on the one hand, the linguistic meaning of legal texts and, on the other, the content of the law” (footnote omitted)). For Mitchell Berman’s views on legal content, see generally Berman, Our Principled Constitution, supra note 60 (introducing the theory of principled positivism, which posits that constitutional rules derive from constitutional principles, which are rooted in social norms).

  322. [322]. See Asgeirsson, supra note 22, at 8.

  323. [323]. Id. (emphasis added).

  324. [324]. Id.

  325. [325]. Greenberg, supra note 98, at 76–77.

  326. [326]. Solum, Communicative Content and Legal Content, supra note 103, at 507–08.

  327. [327]. See Dale Smith, Is the High Court Mistaken About the Aim of Statutory Interpretation, 44 Fed. L. Rev. 227, 229–30 (2016) (critiquing Australia’s High Court’s Approach to statutory interpretation); Solum, Communicative Content and Legal Content, supra note 103, at 501.

  328. [328]. See Greenberg, supra note 98, at 77.

  329. [329]. See Smith, supra note 327, at 229–30.

  330. [330]. Tex. Penal Code Ann. § 21.06 (West 2020), declared unconstitutional by Lawrence v. Texas, 539 U.S. 558 (2003).

  331. [331]. Lawrence, 539 U.S. at 578–79.

  332. [332]. See supra Section IV.B.

  333. [333]. Greenberg, supra note 98, at 72.

  334. [334]. Asgeirsson, supra note 22, at 11.

  335. [335]. Perhaps the most sophisticated one available is offered by Hrafn Asgeirsson, who argues that communicative content provides pro tanto legal content that is defeasible by other legal content. See id. at 6–33, 131. Fully engaging with his theory goes far beyond the present Article. In presenting these ongoing debates about the nature of law, however, I aim merely to point out that textualists owe an explanation of how linguistic content—even to the extent that it is ascertainable—contributes in a fairly direct and unmediated way to legal content, and to point out that this is a difficult and contested task about which legal philosophers have different theories.

  336. [336]. See Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1750 (2020) (“[L]egislative history can never defeat unambiguous statutory text ... .”).

  337. [337]. See supra Section III.A.

  338. [338]. See supra Section III.B.

  339. [339]. Franklin, supra note 6, at 169–72.

  340. [340]. See supra Section IV.C.


Assistant Professor of Law, The University of Texas School of Law.

For comments, conversations, and email exchanges, thanks to Mitch Berman, Ryan Doerfler, Adrienne Erickcek, Joey Fishkin, Willy Forbath, Cary Franklin, Thomas S. Jackson, Guha Krishnamurthi, Brian Pérez-Daple, and Jonathan Siegel. For editorial assistance, I thank Scott Vdoviak. For thoughtful research assistance, I thank Zach Werrell.