Textualists insist that judges should follow the ordinary meaning of a legal text, and sometimes texts do have an ordinary meaning that judges can follow. But sometimes texts have no such thing, in the sense that they are reasonably susceptible to two or more interpretations. Some textualists fall victim to something like the duck-rabbit illusion. They genuinely see a duck; they insist that a duck is the only thing that reasonable people can see. Their perception is automatic, even though it might have been primed, or a product of preconceptions. But reasonable people might well see a rabbit. Various approaches are possible to determine whether we have a duck or a rabbit; most of them do not turn on the text at all.
Have a look, if you would, at the following image:
Is it a duck, or is it a rabbit? Many people see it as a duck; many others see it as a rabbit. You might see it as one and then as the other, perhaps with some effort. Look closely. As if by magic, the rabbit’s ears become a duck’s beak, and vice versa. Look once more. Again, as if by magic, the rabbit’s mouth becomes the back of the duck’s head, and vice versa.
You might try simultaneously to see the image as both a duck and a rabbit, but that is not possible. It is one or the other; it is not both. If you can easily see it as one and then the other, congratulations. You might be especially creative.
Wittgenstein used the duck-rabbit figure to make a point about perception. His short, dense discussion distinguishes between “seeing that” and “seeing as.” In his words:
I may, then, have seen the duck-rabbit simply as a picture-rabbit from the first. That is to say, if asked “What’s that?” or “What do you see here?” I should have replied: “A picture-rabbit.” If I had further been asked what that was, I should have explained by pointing to all sorts of pictures of rabbits, should perhaps have pointed to real rabbits, talked about their habits, or given an imitation of them. I should not have answered the question “What do you see here?” by saying: “Now I am seeing it as a picture-rabbit.” I should simply have described my perception: just as if I had said “I see a red circle over there.”—
Nevertheless someone else could have said of me: “He is seeing the figure as a picture-rabbit.”
If you see a table, you see that it is a table. If you see a duck in the duck-rabbit image, you are seeing the image as a duck. Seeing it in that way is not mandated or foreordained by the image itself. You are doing the relevant work, even if you are doing it automatically or unconsciously. A central lesson of the image, emphasized in psychology as early as 1899, is that perception is not merely a result of the stimulus; it is also a product of mental activity. Hence an observer could rightly say: “He is seeing the figure as a picture-rabbit.”
The duck-rabbit image has a long history. It appeared for the first time in the German humor magazine Fliegende Blatter in October of 1892, without any attribution (Figure 2A). It was captioned, ‘‘Welche thiere gleichen einander am meisten?’’ (“Which animals are most like each other?’’). Underneath, it offered an answer: “Kaninchen und Ente” (“Rabbit and Duck”). Shortly thereafter, a slightly more horizontal version was featured in the Harper’s Weekly journal (Figure 2B). All of these versions involved intentional visual ambiguity: Someone designed an image that could be interpreted in more than one way. Seven years later, the psychologist Joseph Jastrow called the image “ingenious” and briefly explored its implications for perception and interpretation.
Although widely known as an “illusion” (as in my title), the duck-rabbit image is actually a bistable, reversible, or ambiguous figure. There are many such images; one is the horse-seal (Figure 2C). Over the years, the duck-rabbit image has gained in notoriety, and it is now well-known. It is even a book for preschoolers and kindergarteners, with over 100,000 copies sold.  More than a dozen variations have appeared in print, and the figure is featured in most introductory psychology textbooks. It is also a common staple in popular science. Some of the images are quite vivid and even startling (Figure 3).
Although the duck-rabbit figure has intrigued people for more than 100 years, it remains to be fully understood, and people do not agree about how to explain why different people see it differently. It is clear that people’s perception of the figure as a duck or a rabbit is rapid and automatic. In the terms of modern behavioral science, the immediate perception involves the cognitive operations of “System 1.” If you see it as a duck, you might have to invoke the more reflective and effortful “System 2” to see it as a rabbit. Indeed, some people cannot see the figure as anything but a duck, while others cannot see it as anything but a rabbit. But why?
Some early papers hypothesized that everything depended on whether people were right-handed or left-handed. Right-handed people, the theory went, scanned the figure from left to right, thus seeing a duck. By contrast, left-handed people scanned the figure from right to left, thus seeing a rabbit. But studies failed to find any correlation between the initial response and whether people were right-handed or left-handed.
Figure 2. A) The original duck-rabbit figure from Fliegende Blatter; B) the slightly more horizontal replica from Harper’s Weekly; C) another example of a bistable figure: the horse-seal figure.
Figure 3: A vivid duck-rabbit.
Many people now endorse a different theory: preconditioning or priming. On this view, whether people see the image as a duck or a rabbit depends on which they find more familiar. For example, Brugger found that people of all ages, and particularly those between the ages of two and ten, were significantly more likely to see the image as a rabbit on Easter Sunday. In contrast, participants polled on a randomly picked Sunday in October of the same year were more likely to see the duck, or another long-beaked bird, first. Indeed, the duck is generally spotted faster, prompting the speculation that most people encounter ducks or images of ducks more often than rabbits or images of rabbits in day-to-day life.
In a similar vein, Mathewson showed that semantic framing could bias the viewer’s perception. When presented with two duck-rabbit figures side by side (Figure 3), the overwhelming majority of respondents (97.8 percent) were unable to see both a duck and a rabbit at the same time. (Note that we are not speaking here of seeing both a duck and a rabbit in the same image, which is impossible. We are speaking of seeing a duck and a rabbit in two images placed side by side.) But when asked to “[i]magine the duck eating the rabbit,” 86.6 percent of the respondents indicated that they could see both animals next to each other. (Can you see them?)
Figure 4. Mathewson’s two duck-rabbit figures.
In support of this general hypothesis emphasizing preconceptions and priming, consider this image:
Figure 5: My Wife and My Mother-in-Law
Do you see an old woman with her nose to the left, or a young woman looking over her shoulder? Can you see one and then the other? If so, you might be quite startled. Younger people have been found more likely to see a young woman; older people are more likely to see an old woman. As the researchers put it, “[H]igh-level social group processes have a subconscious effect on the early stages of face processing.” That finding fits well with the idea that preconceptions help determine what people see. Indeed, some young people cannot see an old woman in the picture, and some old people are unable to see a young woman.
Textualists insist on following the ordinary meaning of legal texts when there is an ordinary meaning to follow. Suppose, for example, that a statute forbids people from going over 65 miles per hour. Suppose that someone is stopped for going 75 miles per hour, but strenuously objects that he had an exceedingly important meeting and that he could not have made that meeting if he had obeyed the law. Or perhaps the driver was rushing to the hospital; perhaps he was evading a madman, who was chasing him with his own vehicle. The law might allow excuses or justifications in such cases. Even if it does not, courts might invoke some kind of canon, such as the absurdity canon, to allow an exemption. But most textualists would refuse to pay attention to these purported grounds for failing to follow the statute’s ordinary meaning.
Textualism may or may not be the right approach to interpretation. Those who emphasize legislative intentions or purposes, or who approve of the absurdity canon, have some trouble with it. Nonetheless, textualism is certainly an honorable approach to statutory interpretation in the sense that it can be justified by reference to reasons, and is not deceptive or dishonest in any way.
But some texts do not have a single, ordinary meaning, certainly as a matter of semantics or communicative content. Consider the word “magazine.” It might mean a kind of periodical; it might mean an ammunition storage device inside a gun. Context is usually sufficient to sort out ambiguities, ensuring that words have ordinary meanings in context, but sometimes context itself is not enough. Consider the use of the word “source” in the nonattainment provisions of the Clean Air Act, which sets federal standards for stationary and mobile sources of air pollution. Does the term “source” refer to particular smokestacks, or does it refer to whole plants? In cases of this kind, the textual ambiguity might be intentional: It might be the linguistic equivalent of an intentionally ambiguous image. Or it might be inadvertent: Those who wrote it, or enacted it, might not have seen the ambiguity. As the Court famously put it, with respect to the word “source”:
Perhaps [Congress] consciously desired the Administrator to strike the balance at this level, thinking that those with great expertise and charged with responsibility for administering the provision would be in a better position to do so; perhaps it simply did not consider the question at this level; and perhaps Congress was unable to forge a coalition on either side of the question, and those on each side decided to take their chances with the scheme devised by the agency.
What makes these words so helpful for present purposes is that they isolate different reasons for ambiguity. Congress may have left ambiguity on purpose (as in the duck-rabbit image), or might have hoped some other institution would solve the problem, or might not have focused on—or even understood—the possibility that a statutory term could mean more than one thing.
Often textualists insist that legal texts have an ordinary meaning when they actually do. A law might contain specified a speed limit, and a driver might violate it; someone might run for president at the age of 21, when the Constitution requires that presidents be at least 35; someone might seek disability benefits in the face of a statute whose text plainly forbids them to receive such benefits. In such cases, judges really are “seeing that.” But in other cases, textualists proceed as if legal texts have an ordinary meaning even when they do not. Judges see a rabbit, or a duck, when other reasonable readers see a duck, or a rabbit. Such judges are “seeing as.” Nonetheless, they insist that they are “seeing that.” They do not think, do not know, and might not even believe, that “someone else could have said of [them]: ‘He is seeing the figure as a picture-rabbit.’”
It is true that in sone cases, textualists fully recognize that texts have more than one plausible reading. They insist that they are simply searching for the best one, with the goal of ensuring against the injection of their own will into the interpretive process. If so, a great deal is buried into the idea of “simply searching for the best one”; I will have something to say about that shortly. The basic point is that texts, as such, sometimes fail to resolve hard questions even though some textualists sincerely insist that they do— and that something like the duck-rabbit image helps to illuminate their mistake. In constitutional law and statutory interpretation, there are countless examples. Consider just a few, beginning with the most straightforward.
The text of the First Amendment states in relevant part, “Congress shall make no law . . . abridging the freedom of speech.” Suppose that Congress enacts a law, based on the common law of defamation, that allows people to recover damages for certain kinds of libelous falsehoods. Is that a violation of the First Amendment? You might insist that the answer is as plain as day. Such an enactment is a law “abridging the freedom of speech.” But is it? We need to know the meaning of the word “abridging,” and also of the phrase “the freedom of speech.” We might have a rabbit, not a duck.
It is generally understood that taken in the abstract and without reference to history or something else, the Free Speech Clause presents a kind of duck-rabbit image. Consider another issue not always understood to present the same kind of ambiguity: whether affirmative action programs, containing explicitly racial preferences, violate the Equal Protection Clause. You might insist that treating people differently because of the color of their skin cannot possibly be “equal.” You might wonder: Why is that question even difficult? But to reach a conclusion about whether Equal Protection Clause forbids affirmative action programs, a language lesson is not enough. An English speaker who is committed to following the text of the Equal Protection Clause might think that an affirmative action program is (1) permitted by the clause, (2) forbidden by the clause, or (3) required by the clause. None of those thoughts is foreclosed by the ordinary meaning of the text, which is to say that the text simply does not have a single such meaning. (I will get to originalism in due course; it might well be seen as a way of overcoming the duck-rabbit dilemma.)
Consider Article II of the Constitution, which vests “the executive power” in “a president of the United States.” Does Article II forbid Congress from creating independent agencies, immunized from plenary presidential removal authority? It might seem plain that because Article II creates a unitary executive, independent agencies are necessarily unconstitutional. But that is not so plain. An English speaker could say: “I agree that the executive power is vested in a president of the United States. But I do not agree that Congress lacks the authority to immunize Cabinet officials from plenary presidential removal authority.” Alexander Hamilton spoke English, and that is exactly what he thought.
Or consider Article I, Section 1, which says, “All legislative Powers herein granted shall be vested in a Congress of the United States.” Does that provision forbid Congress from granting open-ended discretionary authority to the executive branch, and thus require some version of the nondelegation doctrine in its modern form? A reasonable reader might think so, on the ground that such authority is “legislative” and thus must be exercised by Congress alone. But as a linguistic matter, the vesting of legislative powers in Congress can mean several different things. It might authorize the grant of open-ended discretionary authority to the executive branch so long as it is Congress that is doing the granting. (It’s a rabbit, not a duck.)
One of the most famous problems in statutory interpretation in American law can be seen in similar terms. Church of the Holy Trinity v. United States involved the proper understanding of a statute that prohibited an employer to “prepay the transportation, or in any way assist or encourage the importation or migration, of any alien or aliens, any foreigner or foreigners, into the United States . . . to perform labor or service of any kind in the United States.” The particular question was whether the statute prohibited a New York church from prepaying the transportation of an English rector who was to serve as the church’s pastor. On a widespread understanding, the text clearly applied to the church (“labor or service of any kind”), and the only question was whether the legislature’s intention, or some sort of background principle involving religious liberty, could trump the ordinary meaning of the statutory language. But perhaps that is not the right way to understand the case. “Labor or service of any kind,” to be sure—but what counts as “labor,” and what counts as “service”? Perhaps an English speaker in the late nineteenth century might have understood those words to apply only to manual labor, in which case the statutory language was not plain at all. And in fact, contemporaneous sources make it plausible to think that the statutory text at issue in Holy Trinity created a kind of duck-rabbit figure.
Turn in this light to section 706 of the Administrative Procedure Act, which states that “court[s] shall decide all relevant questions of law, interpret constitutional and statutory provisions, and determine the meaning or applicability of the terms of an agency action.” What does that provision mean? Is it inconsistent with Chevron v. NRDC, which held that courts should defer to reasonable agency interpretations of ambiguous statutory provisions? Many people, including some members of the Supreme Court, think that the answer is plain. After all, courts are instructed to decide “all relevant questions of law.” If judges are required to respect statutory enactments, Chevron might seem to be in serious trouble, at least if no other provision of law supersedes the APA. As Justice Gorsuch put it in a closely related context involving the question whether courts should defer to an agency’s interpretations of its own regulation:
The APA thus requires a reviewing court to resolve for itself any dispute over the proper interpretation of an agency regulation. A court that, in deference to an agency, adopts something other than the best reading of a regulation isn’t “decid[ing]” the relevant “questio[n] of law” or “determin[ing] the meaning” of the regulation. Instead, it’s allowing the agency to dictate the answer to that question. In doing so, the court is abdicating the duty Congress assigned to it in the APA.
It’s a duck! What could be more obvious?
But the text can be read very differently. Adhering to section 706, courts might decide that the right answer to the relevant question of law depends on the agency’s interpretation. In deciding that question, courts might ask what the agency thinks. Rejecting Justice Gorsuch’s view, the Court put it this way: “That provision does not specify the standard of review a court should use in ‘determin[ing] the meaning’ of an ambiguous rule. One possibility . . . is to review the issue de novo. But another is to review the agency’s reading for reasonableness.” It might be that the best way for a court to “determine the meaning” of law is “by deferring to an agency’s reasonable reading.”
Maybe it is not a duck after all. It might be a rabbit. By itself, the text does not resolve the Chevron question; it does not say that Justice Gorsuch was right or wrong. To answer that question, we need to do something other than stare at the APA and pound the table.
In the context of environmental law, the best example is Massachusetts v. EPA, which posed the question whether the EPA was obligated to treat greenhouse gases as “air pollutants” under the Clean Air Act. The relevant provision defines that term to include “any air pollution agent or combination of such agents, including any physical, chemical, biological, radioactive . . . substance or matter which is emitted into or otherwise enters the ambient air.” The Court did not find the question to be at all difficult: “On its face, the definition embraces all airborne compounds of whatever stripe, and underscores that intent through the repeated use of the word ‘any.’ Carbon dioxide, methane, nitrous oxide, and hydrofluorocarbons are without a doubt “‘physical [and] chemical . . . substance[s] which [are] emitted into . . . the ambient air.’ The statute is unambiguous.” In short, the Court saw a duck.
But is the statute really unambiguous? In a sense, the case could be seen as a rerun of Holy Trinity, in which a seemingly plain statutory term is not plain at all. “Any air pollution agent,” to be sure—but what is an air pollution agent? In dissent, Justice Scalia offered a series of arguments against the claim that the term is unambiguous. As he put it, “EPA’s conception of ‘air pollution’—focusing on impurities in the ambient air ‘at ground level or near the surface of the earth’—is perfectly consistent with the natural meaning of that term.” My goal is not to say that Massachusetts v. EPA was wrong. It is only to say that the governing text, taken by itself, could be seen as a rabbit.
Now turn to the question in Bostock v. Clayton County. Title VII of the Civil Rights Act of 1964 makes it “an unlawful employment practice” for an employer “to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s . . . sex.” A female employee is fired because she married a woman. She claims that if she were a man and married a woman, she would not have been fired, and hence she has been discriminated against “because of” her sex. On one view, accepted by the Supreme Court, her claim is clearly correct. The text forbids discrimination “because of such individual’s . . . sex,” and if a female employee is fired because she is married to a woman, the textual prohibition is clearly triggered. As the Court put it:
Today, we must decide whether an employer can fire someone simply for being homosexual or transgender. The answer is clear. An employer who fires an individual for being homosexual or transgender fires that person for traits or actions it would not have questioned in members of a different sex. Sex plays a necessary and undisguisable role in the decision, exactly what Title VII forbids.
As the Court saw it, the statutory text was plain. An employment who discriminates on the basis of sexual orientation necessarily and simultaneously discriminates on the basis of sex. But imagine if an English speaker, now or in 1964, says the following: “I am opposed to discrimination because of sex. I am also opposed to discrimination because of sexual orientation.” Does the speaker not understand the English language? Is she being redundant? (The answer to both questions is “no.”) Or suppose that an English speaker, now or in 1964, says the following: “I am opposed to discrimination because of sex. But I am not opposed to discrimination because of sexual orientation.” Is the speaker contradicting himself? Is he making some sort of logical error? Does he not understand the language? (The answer to all three questions is “no.”)
To be sure, a prohibition on discrimination “because of sex” could be meant and understood as a prohibition on discrimination because of sexual orientation, at least in cases in which people in same-sex relationships are treated worse than they would be if they were men rather than women, or women rather than men. A statute that bans discrimination “because of sex” might be universally taken to forbid discrimination against such people. But if it is, it is not because that is how the English language works. It is because that was the universal understanding of the particular text, taken in context.
Or perhaps we think that to interpret legal texts, we should attempt to place them in the best constructive light. That is, we are obliged to respect those texts by being faithful to them (duck or perhaps rabbit, but not elephant or tiger), but also make the best possible sense out of them (perhaps rabbits are better than ducks). This is Dworkin’s notion of law as integrity, which urges that it is impossible to avoid making some kind of moral conclusion when interpreting an ambiguous text. In a sense, Dworkin is saying that judges need to bite the duck-rabbit bullet. If he is right, judges might conclude that a prohibition on discrimination “because of sex” is best read to forbid discrimination on the basis of sexual orientation, where “best read” means that this reading makes Title VII the best it can be. Or judges might reach the opposite conclusion, again on moral grounds. If this is how we approach an ambiguous text, we are not (simply) textualists. Whether it is a duck or a rabbit depends on how we would most like to see it.
Recall in this light that priming or preconditioning might determine whether people see a duck or instead a rabbit. For some textualists, I suggest, something similar is at work, in the form of their own legal, political, or moral commitments. That is why they see a duck and are puzzled (and perhaps aghast) that anyone can see a rabbit. If you want to believe that affirmative action programs are unconstitutional, you might find it obvious that the Equal Protection Clause forbids them. If you think that the idea of a unitary executive is appealing, perhaps because it promotes accountability and liberty, you might find Article II, Section 1 to be unambiguous. If you favor independent judicial review of questions of law, you might find it clear that section 706 of the APA requires it. Some textualists see a duck rather than a rabbit because a duck is what they want to see. And because they want to see a duck, it does not take a great deal of work in order for them to see it. It happens quickly and automatically. They think that they are “seeing that,” even though they are actually “seeing as.”
Sometimes, of course, judges are aware that they are “seeing as” and that others see things differently, and that some separate argument must be made on behalf of what they are seeing. They might advert to purpose. They might invoke some canon of construction. They might do something else; recall the idea of law as integrity. I will turn to these possibilities shortly.
Some people believe in following the “original public meaning” of legal texts. Originalists would understand the First Amendment by asking about how its text was publicly understood at the time of ratification. We might approach section 706 by asking the same question. If we do that, we might know whether we have a duck. A word that might mean two or three things in the abstract might mean only one thing if it is taken in accordance with its original public meaning.
How often will that happen? To know, we need to settle on a specific understanding of “original public meaning.” That is not the simplest matter on which to settle. Do we ask about the original semantic meaning of the text? Do we ask about contemporaneous understandings of how the text would settle concrete cases? If the ratifiers had expectations about the reach of a clause, are those expectations decisive? Are they relevant if not decisive? Do we ask about the understandings, at the time, about whether the text set out a general principle whose meaning was not fixed?
On one approach, for example, we should ask: Did those who ratified the First Amendment understand their text to forbid libel actions? If there is an authoritative answer to this question, we would resolve the problem of interpretation not on the basis of anyone’s moral convictions, but on the basis of history; we might learn that we have a duck. Of course, it is possible that the First Amendment’s ratifiers understood their text to set out a general principle whose meaning with respect to particular issues would not be frozen in time. The only point is that if the original public meaning is authoritative, it is authoritative with respect to everything, including the question whether a text sets out a general principle or not. Interpretation becomes a historical inquiry.
In some cases, some forms of originalism will certainly settle otherwise difficult interpretive disputes. At the same time, the original public meaning might leave gaps or ambiguities; judges might be in “the construction zone,” in which they might be permitted to see a duck or a rabbit, and (it is profoundly to be hoped) justify their choice in an appropriate way (bracketing exactly what they means). In addition, the original understanding might indeed be a general principle, one that can be specified in multiple ways.
It is tempting to think that in the kinds of cases that are of concern here, textualism is a kind of lie. It might be. But it might also be an honest mistake, a matter of sincerely thinking that you are “seeing that” when you are actually “seeing as.” Still, it is a serious problem if a judge does not know that she is seeing as. If she is, in fact, seeing as, she should explain why that is the right way to see, and if she thinks that she is seeing that, she might see no need to offer an explanation. Return to section 706 of the APA. The key phrase—”court[s] shall decide all relevant questions of law”—might well seem unambiguous. It might take a great deal of immersion in the context to see that it can be understood in more than one way. With that immersion, readers will reach that conclusion at a minimum. (Recall that if they are committed to the original public meaning, they might settle on a particular view, not as a matter of the English language, but as an effort to capture how the phase was understood at the time.)
Once judges learn that they are “seeing as,” they have many paths forward. Originalism is a possibility, and as noted, it might turn out to be decisive, if it is specified in a certain way. In criminal law, judges might use the rule of lenity and select the approach that benefits the defendant. If a constitutional issue is raised, they might use the avoidance canon and so avoid the issue. If Native American tribes are involved, judges might select the interpretation that benefits them. In some cases, they might use some version of the “major questions” doctrine and forbid agencies from extending their authority to new areas unless Congress has clearly given them that authority. They might introduce background principles of various sorts, perhaps connected with the rule of law, to insist on a particular view of an ambiguous statute. They might try to make the statute the best it can be by their own lights, not because they are arrogant, but because they believe that they have no alternative.
The general point is that too often, readers of legal texts, including judges, act as if a term or phrase necessarily has a unitary meaning when it might have two, three, or more. A reader might “simply have described my perception: just as if I had said ‘I see a red circle over there.’— Nevertheless someone else could have said . . . : ‘He is seeing the figure as a picture-rabbit.’” To know what they are seeing, readers might be able to rely on history or context. They might need a canon of some kind. They might even need a moral principle. They are often using one or more of these, even if they insist, quite sincerely, that anyone can see it is a rabbit.
Cass R. Sunstein: Robert Walmsley University Professor, Harvard University. I am grateful to Richard Fallon, Daphna Renan, Lawrence Solum, and Adrian Vermeule for valuable comments on a previous draft and to Dinis Cheian for valuable research assistance.
. Ludwig Wittgenstein, Philosophical Investigations 165 (1953). For useful discussion, see generally William G. Lycan, Gombrich, Wittgenstein, and the Duck-Rabbit, 30 J. Aesthetics & Art Criticism 229 (1971).
. For various examples, see Alison Gopnick & Andrea Rosati, Duck or Rabbit? Reversing Ambiguous Figures and Understanding Ambiguous Representations, 4 Dev. Sci. 175 (2001); see also Mariana A. Preciado, Is There Truth in What We See? Motivation and Visual Perception, Psych. in Action (Feb. 25, 2011), https://www.psychologyinaction.org/psychology-in-action-1/2011/02/25/is-there-truth-in-what-we-see-motivation-and-visual-perception.
. For indirect evidence, see generally Gopnick, supra note 8. Especially intriguing is the suggestion that the capacity to experience reversal—first duck, then rabbit, or vice-versa—”may depend on a more abstract understanding of ambiguous representation.” Id. at 175. More specifically: “Not only do we need the specific information about the alternative interpretations, we also need the general conceptual understanding that multiple interpretations of ambiguous figures are possible.” Id. at 182. Judges, please take note.
. Note also the finding in one study that “younger viewers were more likely to see the figure as a duck,” which “might reflect the popularity of the cartoon figure Donald Duck, which was at its zenith in the 1940s and early 1950s—an effect similar to the increased popularity of the rabbit interpretation that occurs before Easter.” McManus, supra note 11, at 183.
. This is the thrust of the analysis in Seila Law LLC v. Consumer Fin. Prot. Bureau, 140 S. Ct. 2183 (2020), as summarized by this suggestion: “The President’s power to remove—and thus supervise— those who wield executive power on his behalf follows from the text of Article II. . . .” Id. at 2191.
. A similar view is defended in Tara Leigh Grove, Which Textualism?, 134 Harv. L. Rev. (forthcoming 2020) (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3674050). Grove urges that the majority embraced “formalistic textualism” and that the dissents favored “flexible textualism.” Id. at 2. In my view, the question in the case could not be resolved by reference to “formalistic textualism.”
. See Kent v. Dulles, 357 U.S. 116 (1958); Philip P. Frickey, Getting from Joe to Gene (McCarthy): The Avoidance Canon, Legal Process Theory, and Narrowing Statutory Interpretation in the Early Warren Court, 93 Calif. L. Rev. 397, 430 (2005) (“Kent v. Dulles . . . fit the established pattern of deploying the avoidance canon to protect alleged subversives.”).
. See Util. Air Regulatory Grp. v. EPA, 573 U.S. 302, 324 (2014) (“We expect Congress to speak clearly if it wishes to assign to an agency decisions of vast ‘economic and political significance.’”); Int’l Refugee Assistance Project v. Trump, 883 F.3d 233, 291 (4th Cir.) (Gregory, C.J., concurring) (“Courts require a clear statement of congressional intent before finding that Congress has ceded decisions of great economic and political significance. . . .”), vacated, 138 S. Ct. 2710 (2018).