Skip to Main Content

The First

How to Think About Hate Speech, Campus Speech, Religious Speech, Fake News, Post-Truth, and Donald Trump

“Fish’s points arrive in thoughtful, dense provocations.” —Kirkus Reviews

From celebrated public intellectual and New York Times bestselling author, Stanley Fish, comes an urgent and sharply observed look at one of the most hotly debated issues of our time: freedom of speech.

How does the First Amendment really work? Is it a principle or a value? What is hate speech and should it always be banned? Are we free to declare our religious beliefs in the public square? What role, if any, should companies like Facebook play in policing the exchange of thoughts, ideas, and opinions?

With clarity and power, Stanley Fish, “America’s most famous professor” (BookPage), explores these complex questions in The First. From the rise of fake news, to the role of tech companies in monitoring content (including the President’s tweets), to Colin Kaepernick’s kneeling protest, First Amendment controversies continue to dominate the news cycle. Across America, college campus administrators are being forced to balance free speech against demands for safe spaces and trigger warnings.

Ultimately, Fish argues, freedom of speech is a double-edged concept; it frees us from constraints, but it also frees us to say and do terrible things. Urgent and controversial, The First is sure to ruffle feathers, spark dialogue, and shine new light on one of America’s most cherished—and debated—constitutional rights.

Chapter 1: Why Censorship Is a Precondition of Free Speech CHAPTER 1 Why Censorship Is a Precondition of Free Speech “Speech Is Everywhere”
The one thing speech isn’t is free. There are costs to those who produce it and costs to those who are subjected to it. Of course, the term “free speech” does have a clear meaning under the Constitution: if you want to say something, you don’t have to ask the government’s permission and you won’t be punished by the government for saying it. But even this freedom from state interference with your speech has its limits, and it does not protect you in private life, where speaking out carries with it the risk of censorship and penalty. Freedom of speech, despite ritual celebrations of it, is not all it is cracked up to be, and it is difficult even to say what it is. As a concept it refuses to sit still and remains elusive to the grasp no matter how closely and rigorously it is examined. When one says “freedom of speech,” the referent seems solid enough, but, as we shall see, it melts at the analytical touch. The phrase “free speech” doesn’t refer to a single core doctrine that works itself out in different practices. Rather, different practices, with their in-place definitions and goals, determine in any instance what “free speech” means. Yet even if freedom of speech is not a thing but a promissory note that can never be cashed in, it is a large and inescapable component of our political rhetoric. We are invested in it, and almost everything of note that happens is attached, sometimes by the loosest of ligatures, to free-speech issues.

Consider, for example, the last ten days of October 2018. A man named Cesar Sayoc sent pipe bombs to CNN and a dozen prominent critics of President Donald Trump. Another loner, Robert Bowers, killed eleven Jews in Pittsburgh as they prayed. Almost as soon as these events occurred, they became attached to issues of free speech. Sayoc drove, and sometimes slept in, a van plastered with pro-Trump stickers and with images of Hillary Clinton and other progressives portrayed as the targets of gunfire. Was that the mere expression of political opinion or a “true threat,” a term of art that means a threat likely to be acted on? Bowers spewed anti-Semitic slogans as he went about his murderous work. Both men had online histories that identified them as haters, conspiracy theorists, and racists. Should the authorities have been alert to the danger they posed? Could they act only after Sayoc and Bowers had already acted? The digital trail was long, and the van with its disturbing images was visible on the streets of Miami for some time. Should President Trump be held to account for the anti-immigrant, anti-“other” rhetoric that, some said, made atrocities like these inevitable? Did his rantings against Hillary Clinton, George Soros, Cory Booker, Kamala Harris, Barack Obama, Eric Holder, Maxine Waters, and others make up a hit list Sayoc dutifully followed? Was Trump saying to Sayoc what Henry II said to those who promptly went out and murdered Thomas Becket: “Will no one rid me of this meddlesome priest?” Was his invective more than a dog whistle? Was it, in fact, a marching order? Was the order heeded by Coast Guard Lt. Christopher Paul Hasson when, in February 2019, he compiled a list of Democrats and anti-Trump journalists whose murder would announce the beginning of the race war he hoped to bring about? At least one prominent First Amendment theorist has posed the relevant question: “[W]e might ask when the State or political leaders may be held constitutionally responsible for encouraging private parties to punish critics. I suggest here that if the president or other officials direct, encourage, fund or covertly command attacks on their critics by private mobs of foreign powers, the First Amendment should be implicated.”1

Mainstream commentators clucked over the possible relationship between Trump’s words and these horrible deeds, but mostly they muttered “First Amendment” and said that however hateful someone’s speech might be, unless it amounted to a direct incitement of violence, it was protected by the Constitution; free speech must prevail. “Free speech above all” was also the mantra of Andrew Torba, the CEO of the website Gab, Robert Bowers’s favorite venue, characterized by the New York Times as the “last refuge for internet scoundrels.” Torba often wears a green hat embroidered with the message “Make Speech Free Again.” Is he a free-speech hero, bravely suffering the vituperation hurled at him by the New York Times? Or is he a prime example of how the promotion of unregulated free speech, proclaimed by First Amendment apostles as the cornerstone of democracy, can lead to a cascade of words that in time is corrosive of that same democracy? If Sayoc and Bowers didn’t have an internet community where their views could be parroted back at them and amplified to the point where every toxic thought they entertained seemed universally shared, would the seeds of hatred perhaps not have flowered in the actions they ultimately took?

You might be surprised that actions so undeniably physical (what could be more physical than bombs and assault rifles?) turn out to be imbued with elements of speech, but as Supreme Court Justice Elena Kagan observed recently, “Speech is everywhere—a part of every human activity.”2 Everything we do sends a message and everything we say has an effect. What this means is that freedom of speech is not a discrete value. You can’t carve speech out and pay homage to it in isolation from the actions from which it is inextricable. Free speech is not, despite Justice Robert Jackson’s memorable pronouncement, the “fixed star in our constitutional constellation,” the abiding light that will guide us through the kaleidoscope of circumstances if only we keep our eyes on it.3 In fact, there is nothing “fixed” about free-speech doctrine at all. It’s a grab bag of analogies, invented-for-the-occasion arguments, rhetorical slogans, shaky distinctions, and ad hoc exceptions to those distinctions, all combining to make it an artifact of the very politics it supposedly transcends. That’s a mouthful, but what it means is that the First Amendment is a participant in the partisan battle, a prize in the political wars, and not an apolitical oasis of principle. That’s the first thesis of this book. The second thesis is that there is nothing wrong with that. A First Amendment whose content and operations are through and through political is fully capable of doing the work we need it to do. Indeed, it may be more capable because, freed from the stringent demand of principle—the demand that freedom of speech be the rule without exception and no matter what the circumstance—the amendment can display the flexibility required to make useful sense of the many situations in which debates about speech emerge. The very malleability of the First Amendment—its lack of a hard center or of any center at all—may be its greatest recommendation.

The squishiness of the First Amendment infects it at the most basic level, the level of its prime purpose, which is, of course, to protect speech. But before we can begin to do that, we must clearly distinguish speech from action, and that turns out to be far from easy. Supreme Court First Amendment decisions have told us that flag burning is speech and that saying to a policeman, “You’re a fascist,” isn’t.4 If speech can be categorized as “symbolic action” (because it has an effect the state finds distressing) and action can be categorized as speech (because it sends a message the state wants to protect), isn’t the distinction infinitely manipulable? How do you draw the line, and who should be authorized to draw it?

Suppose we could answer those questions and come up with a generally accepted definition of what is speech and what isn’t. With that definition in place, is a citizen free to engage in speech without fear of repercussions? Not really, for there are “time, manner, and place” restrictions that limit your free-speech rights by telling you, for example, that you can’t deliver your message from a truck with a loudspeaker while driving through a residential neighborhood at two in the morning. Neither, if you are a nurse, can you lobby for higher wages and better working conditions in the middle of an operation. Just the fact of having a certain job constrains your exercise of free-speech rights. How many of these restrictions are there, and is there a point when their cumulative weight is so great that free-speech rights, at least in some circumstances, have been reduced almost to nothing? Let’s assume (as a hypothesis, not a fact) that those questions too have been answered and we have identified the scope of free-speech rights and a rough sense of the situations in which they can be invoked. Are we home free? Not quite, because inevitably someone will cite competing interests—the moral fabric of society or the mental health of individuals—and argue that they should outweigh the First Amendment in some contexts. In response, the strong First Amendment proponent will reaffirm the abstract principle and say that it should trump every time. But how do we think about those occasions when allowing speech to flow freely, no matter what its effects, causes harm to others? Is that harm just collateral damage, the necessary cost of doing First Amendment business? Are the eleven dead in Pittsburgh martyrs to the First Amendment? Is the First Amendment so basic a value that it must be upheld even when it appears to facilitate evil?
Must We Give Holocaust Deniers a Voice?
This last question is a perennial one and is the stuff of newspaper headlines. On July 19, 2018, the headline for the front-cover story of the New York Daily News was “Holocaust Deniers Deserve a Voice.” That statement was made by Facebook founder and CEO Mark Zuckerberg, who in an interview on the previous day had explained why his platform would not banish Holocaust deniers and other “conspiracy pushers and hoax peddlers”: “[A]s abhorrent as the content can be, I do think it comes down to the principle of giving people a voice.” Zuckerberg is saying (and here he echoes a remark incorrectly attributed to Voltaire: “I disapprove of what you say, but I will defend to the death your right to say it”) that however loathsome and reprehensible an idea or proposition may be, the core democratic principle of freedom of speech requires us to air it and forbids us to repress it. Even liars and defamers should be given a voice. That, as we shall see, is standard First Amendment doctrine. It goes along with a commonplace we’ve all heard and perhaps repeated: freedom of speech is meaningful only if its protection is extended to the worst speech imaginable.

Sounds good, but think about it. The argument that the more despicable the speech is, the more it merits protection makes sense only if we can be confident that when abhorrent views are given a place in the conversation, they will be exposed for what they are and rejected in favor of better views. Zuckerberg at first seemed to have that confidence. Reacting to some negative comments on the interview, he declared in a follow-up that “often the best way to fight offensive bad speech is with good speech.” Here again he is piggybacking on the words of others (nothing wrong in that), in this case on the declaration by Supreme Court Justice Louis Brandeis that “the remedy [for harmful speech] is more speech, not enforced silence,” and on the question posed eloquently by John Milton in 1644: “Who ever knew truth put to the worst in a free and open encounter?”5 (The answer is: a great many people.) Note how Zuckerberg’s phrase “offensive bad speech” lowers the stakes by diminishing the harms that speech might produce, for if the badness of a form of speech is that the speech is offensive—some people’s sensibilities are ruffled—it is surely an overreaction to regulate or censor it. If that’s all there is to it, what’s the fuss? A little unpleasantness is a small price to pay for so important a principle. But as Anti-Defamation League CEO Jonathan Greenblatt pointed out in his response to Zuckerberg’s remarks, Holocaust denial’s effects are not so mild: “Holocaust denial is a willful, deliberate, and longstanding deception tactic by anti-Semites that is incontrovertibly hateful, hurtful and threatening to Jews.”6 With the word “threatening,” Greenblatt challenges Zuckerberg’s deflationary “offensive.” These words, he is saying, do real damage; letting them out into the world and affording them space on the most powerful of social media platforms is not an innocent act whose consequence is merely someone’s hurt feelings.

Zuckerberg tries to cover his flank by saying that Facebook is moving toward taking down “misinformation that is aimed at or [is] going to induce violence.” But Greenblatt would no doubt reply that the anti-Semitism of which Holocaust denial is a version is always produced with a violent intention even if violence is not immediately and explicitly urged. It is hard to draw a bright line between aggressively hostile speech that is the mere voicing of opinion and aggressively hostile speech that bleeds into violence, although the effort to draw that line has been an ongoing project ever since 1859, when John Stuart Mill in On Liberty (the theoretical fountainhead of modern free-speech doctrine) distinguished between language vilifying corn-dealers as a class in the press—that’s just talk—and rehearsing that same language “to an excited mob assembled before the house of a corn-dealer”: that’s intimidation. One could say, however, that this is merely the difference between slow poison and fast poison, between, for example, generations of anti-Semitic rhetoric in Germany and the explosion of Kristallnacht (November 9–10, 1938), when Jewish homes, synagogues, businesses, and hospitals were destroyed by mobs. Wouldn’t it be better, perhaps, to eliminate the poison the moment you see traces of it? Bhikhu Parekh, a member of the British House of Lords, argues that discrete incidents of hate speech are “cumulatively capable of coarsening the community’s sensibility, poisoning the minds of the young, weakening the norms of civility and decency and creating a situation in which it becomes common practice to ridicule, mock, malign.”7 The next step, Parekh says, is inevitable: “If anything can be said about a group of persons with impunity, anything can also be done to it.” With that statement Parekh blurs the distinction between racist, anti-Semitic rhetoric and racist, anti-Semitic acts, and therefore blurs the distinction between speech and action on which the First Amendment depends. (If speech is not a distinct category, an amendment protecting it would have no object.) That would seem to be the choice: either be faithful to First Amendment principles by extending their protection to speech that leads to harm, or regulate harmful speech as if it were an action and become a First Amendment apostate.

Zuckerberg is caught in this dilemma. On the one hand, he doesn’t want to censor anyone’s speech, but on the other, he doesn’t want his company to be a vehicle of evil effects. “There are,” he explains, “two core principles at play here. There’s giving people a voice.… Then there’s keeping the community safe.” “Look,” he says almost plaintively, “I wanna make sure that our products are used for good.” But how can he make sure of that without setting himself up as the arbiter of the good, something every tech CEO says he doesn’t want to do? At times Zuckerberg puts his faith in the development of computer programs that will do the job by distinguishing mechanically between good speech and harmful speech. (“We took the time to build AI [artificial intelligence] tools.”) The idea is that if we can come up with the right algorithms, they will identify what is truly harmful speech all by themselves without any input from fallible human judgment. (I will come back to this techno-fantasy in chapter 5.) At other times, however, Zuckerberg is less optimistic and says what Aristotle said in his defense of rhetoric: “People use tools for good and bad.” In short, it’s human nature (Zuckerberg also says that), and there’s not much you can do about it. So in one moment Zuckerberg is putting his faith in technology, and in another he’s throwing up his hands. A few weeks later he had apparently migrated from the giving-everyone-a-voice side, where he had originally positioned himself, to the keeping-the-community-safe side. Facebook removed pages belonging to conspiracy theorist Alex Jones and his website InfoWars because, the company announced, they violated “Community Standards,” words capitalized in order, perhaps, to deflect attention from the obviously local and subjective nature of those standards. If it’s Alex Jones and InfoWars this time, who is going to be next? And what happened to the flinty First Amendment warrior who only a short while ago refused to silence Holocaust deniers even though, as he pointed out, he is himself Jewish? Maybe there will be a new announcement on that front soon.

It is easy to poke fun at Zuckerberg’s performance, but to be fair he is simply a walking emblem of the tensions, paradoxes, and contradictions that emerge whenever there is an attempt to draw a clear line between speech that is a genuine contribution to democratic deliberation and speech that threatens democracy’s foundations. Facebook and other social media platforms are forever trying to draw that line (if only because they are under so much pressure from politicians and the media), but no such line can be drawn because, as I said earlier, the doctrine of free speech is radically unstable and infinitely elastic. It is not so much a clear principle as a rhetoric—a collection of high-sounding phrases (like “the Marketplace of Ideas”) available for appropriation by anyone who can marshal them persuasively.

Here’s another piece of evidence, this time in the headline of the lead story in the July 1, 2018, Sunday edition of the New York Times: “How Conservatives Weaponized the First Amendment.”8 No doubt some readers were surprised to see free speech characterized as a weapon in a political struggle. Isn’t the right to free speech a protection against being silenced by one’s political opponents? Isn’t it the point of First Amendment doctrine to provide a space free of ideology and indoctrination? Don’t the guarantees provided by the First Amendment ensure the participation in civic life of individuals whose voices would otherwise be drowned out? The scholars quoted in the Times article answer no to all these questions. Maybe in the public’s mind free-speech doctrine serves these progressive purposes, but in recent decades, says one commentator, free speech “has become a sword for authoritarians, racists and misogynists, Nazis and Klansmen, pornographers and corporations buying elections.”9

These are strong words, and I quote them not to endorse them but to underline their implication: far from being a concept that stands to the side of the fray, free speech is right in the middle of the fray, where it is wielded as a sword by all parties to a controversy: You say you’re for free speech, but the truth is that we are, so take that! The party that succeeds in claiming the free speech label for its side is more than halfway to victory. So, for example, the moment the majority in Citizens United v. Federal Election Commission turned the question of regulating campaign contributions into a free-speech issue rather than a corruption issue, the case was over and the big money interests prevailed.10 “It chills free speech” is almost always a winner, but free speech is not what wins because free speech is not an independent value. Free-speech arguments are never made in the name of the abstraction itself but in the name of some agenda to which free-speech rhetoric has been successfully attached, and when the argument is won, the victor will not be free speech but that agenda. Although it is often invoked as a principle with its own shape, freedom of speech is given shape (and content) by the partisan agents skillful enough to appropriate its vocabulary for their preferred ends. If I am right about this, many of the so-called free-speech controversies—about hate speech, campus speech, fake news, and much else—will wear a different aspect than they do when they are framed in the high-flown vocabulary we have inherited from over a century of First Amendment jurisprudence.
When Can We Speak Freely?
Let’s begin by getting a few basic things straight. Freedom of speech is at once a cornerstone of democracy and a concept in law. The two often overlap, but they are not the same. Democracy means “rule by the people,” and if the people, and not some dictator or theocrat or central committee, are to rule, each citizen must have the right to freely express his or her views about proposed policies and the performance of elected leaders. The idea is that those obliged to live under the laws should have a part in making them. But some of those laws, although established by democratic procedures, institute limits on what those who made them can say. You can’t lie under oath (although you can lie to your spouse), and you can’t shout “Fire!” in a crowded theater when there is no fire. In these and other contexts your freedom of speech is abridged for reasons the courts specify. Other laws abridge the freedom of the state either to curtail your speech or to compel it. The state cannot stop you from affirming Jesus Christ as your savior and it cannot force you to affirm Jesus Christ as your savior.

These do’s and don’ts are the work of the First Amendment, which regulates speech interactions between you and your government. Insofar as there is a First Amendment right, it is a right you hold against the government’s efforts to curb it; it is not a right you hold against nongovernmental actors who may wish, for a variety of reasons, to silence you. More often than not, when nongovernmental actors restrict or censor your speech, the First Amendment cannot be invoked to stop them. And even when the amendment guarantees your right to say something, it does not guarantee that you will suffer no consequences for having said it. What the law cannot do—penalize, dismiss, or exile you for speaking—can be done by your parents or your partner or your friends or your boss. Many people misunderstand this and think that the answer to the question When can you exercise your free-speech rights without fear of penalty or retaliation? is “Most of the time.” They think of free speech as something the Constitution guarantees in almost all circumstances, as the default condition that holds except when an exception is allowed for a special reason. But in fact the reverse is true. Situations in which freedom of speech is a right unalloyed by any competing values or in-place restrictions are rare. The obvious (and perhaps the only) example is the Hyde Park corner or some equivalent “free-speech zone”—a place dedicated to the production of speech that is insulated from both seriousness and consequences: people get up on a soapbox or stand on a street corner emoting about anything or nothing; everyone has his or her say; no one is asked to do anything but listen, and even listening is not required.

It used to be that sports stadiums were also places where you could say anything that came into your head, but in recent years fans have been ushered out for using profanity or for taunting players with what amounts to “fighting words” (words intended to provoke a confrontation). In most contexts (ranting to yourself in the shower is another of the rare exceptions), speech is produced within an implicit understanding of the concerns it might be addressing, and the structure of those concerns—which could be political, economic, domestic, agricultural, educational, whatever—silently and without any fuss constrains which assertions are relevant and which irrelevant or frivolous or out of bounds. When you are at home, you refrain from saying what you’re thinking because you know that frankness will come with a price—such as the unraveling of your marriage—you don’t want to pay. When you are a student in a classroom, you can’t insist (although you can politely request) on the topics to be discussed; if you are an instructor, you can’t talk about anything you like (the syllabus and professional decorum constrain you) and you can’t ridicule or berate your students. Those constraints—marked out in advance by the context—are not added to or imposed on the scene of expression; they give the scene of expression its shape. Whatever the practice you engage in, protocols of speech are always and already in place and you know, without reflection, which utterances are okay and to the point and which are not. As long as there is something at stake, as long as speech is more than noise indifferently produced, there’s no such thing as free—that is, completely unfettered—speech. Speech is always attached and tied down to the pre-known situational context of utterance, and it is only because speech is attached and tied down that it has significance. Were speech to emerge from a void or be found in a bottle cast into the sea years ago, we could make something of it—attribute meaning to it—only by providing the context it lacked: we ask ourselves, “What kind of person might have produced this untethered utterance?,” and the answer we come up with or, more precisely, invent stabilizes and configures the communicative act.

It should now be clear how unusual is the Hyde Park corner or free-speech zone where anything and everything is freely said. In what we might call the “standard free-speech story,” that scene is thought to be the norm, and scenes where free speech has been chipped away by exceptions are special and therefore always in need of justification. But in fact it is the other way around: what is special is the Hyde Park corner or any other artificially designated space where it doesn’t matter what you say because when you enter it, the ordinary conditions of significance and consequence have been suspended. Absolutely free speech is the outlier case; constrained speech is the norm. Limitations on speech are part and parcel of any context in which speech is produced for a reason and not just for amusement. The standard story has it backward. Censorship, in the form of it-goes-without-saying restrictions on expression, is built into ordinary occasions of speech production. It might seem odd to say so, but censorship precedes free speech and is its precondition: if there were no censorship in the form of social or institutional purposes and goals that mark out what is appropriate and inappropriate to say, there would be no speech that was meaningful. All we would have is a succession of utterances without any frame telling us what we should pay attention to. One might say then, with only a slight metaphorical stretch, that meaningful speech—speech that says something and not everything—performs an act of censorship, every time. As the philosopher and cultural critic Judith Butler explains, “[N]o speech is permissible without some other speech becoming impermissible.… [C]ensorship is what permits speech” to have sense.11

By “censorship” Butler doesn’t mean what we usually understand it to mean, an action taken by the state limiting the speech rights of a citizen. Her sense of censorship is more diffuse and is in line with the definition offered by Sue Curry Jansen: “[C]ensorship encompasses all socially structured prescriptions which inhibit or prohibit the dissemination of ideas, information and other messages… whether those obstructions are secured by political, economic, religious, or other systems of authority.”12 In short, censorship is all around us. Now one might well question the usefulness of so expansive a definition of censorship, for if everything is censorship, the word loses its critical edge; if it is the general condition, nothing stands opposite to it, and no one can be blamed for engaging in it. First Amendment jurisprudence requires that we distinguish between the restrictions on speech built into every socially organized context and the restrictions on speech imposed by government. Silencing is the result in both cases, but only governmental restrictions implicate the First Amendment. Perhaps it should be otherwise; perhaps silencings of all kinds should trigger First Amendment concerns. Many who wish to regulate hate speech think so, but that conversation must wait for the next chapter.
Roseanne Barr and First Amendment Opportunism
The proposition that the production of speech is constrained even when no state censor is at work is easy to illustrate. Do you and I have free speech in the workplace? No. Can an employer discipline or dismiss us because he doesn’t like what we say or what our T-shirt says? Absolutely. Remember, freedom of speech is a right we have against government’s efforts to suppress it, not a right to speak freely on any occasion without fear of repercussions. An employer has every legal right to fire you if he judges your speech or your clothing to be disruptive of the workplace or if he doesn’t want his business to be associated with people who express your kind of views. You don’t even have to espouse those views in the office in order to be shown the door. Some of those who marched on the alt-right side in Charlottesville were terminated when their employers learned what they had been up to. They of course had the right to demonstrate and to carry placards and shout “Jews will not replace us,” but that right did not protect them from being penalized when they exercised it. A Google employee was let go in 2017 after he sent out an internal memo criticizing the company’s diversity policy and suggesting that the underrepresentation of women in some job areas is the result not of intentional discrimination but of biological differences.13 “[We don’t want] harmful gender stereotypes in our workplace,” said a company spokesperson defending the firing. She could say that because employers in the private sphere are generally free to determine which behaviors, including speech behaviors, they will countenance and which they will not.

Another example: It has been widely speculated that Colin Kaepernick, formerly a quarterback for the San Francisco 49ers, has been blackballed because NFL owners don’t want the baggage associated with someone who performs a political act (taking a knee rather than standing with hand on heart during the playing of the national anthem) on the field. If that’s true—and, given the scarcity of skilled quarterbacks, it likely is—the owners are fully within their rights and Kaepernick has no legal case to make unless it can be proven that two or more owners colluded in an effort to deprive him of his livelihood. To be sure, his gesture was political speech and the Constitution says that he cannot be prevented from making it, but any argument that he should not be penalized for doing something the Constitution protects will go nowhere. Your free-speech rights insulate you (in most cases) from government sanctions (“Congress shall make no law…”), not from sanctions issued by your employer.

That’s something the comedian Roseanne Barr found out when she tweeted that Valerie Jarrett, an Obama aide, was the product of a union between the Muslim Brotherhood and the Planet of the Apes. Within twelve hours Barr’s highly rated sitcom was canceled by ABC, and the CEO of Disney, ABC’s parent company, called Jarrett to apologize. Did Barr have a right to say what she said? Yes. Did ABC have the right to take away her platform and, in effect, shut her up (although not, it turned out, on Twitter)? Yes, and the decision to do so no doubt involved a quick calculation of the constituencies the network would offend if it didn’t and the constituencies it would offend if it did. Were there any free-speech/First Amendment issues involved? No. The issues were economic. Network broadcasting depends on advertising revenues, and advertisers are understandably wary of associating their brand with a controversial celebrity if only because of the prospect of boycotts. In 2017–18, as the #MeToo movement grew, some companies associated with a figure in the negative spotlight dithered while advertisers began to drop away; then they ended up pulling the plug anyway. Obviously ABC learned from the unhappy experience of its predecessors that it is usually better to cut your losses right away, for then you have a chance at least of retaining your corporate dignity and protecting your revenue.

When an incident like this one emerges—and they seem to emerge daily—there are always going to be those who cry “Free speech, free speech!” and ask “Can’t Roseanne say what she likes without being penalized for it?” The answer is no, and the penalties she has endured have nothing to do with her free-speech rights, which have not been infringed at all. She did something stupid and professionally inadvisable, and it took the form of speech, but the consequences she suffered have no constitutional implications (no more than do the consequences television commentator Megyn Kelly suffered when she said it was all right to dress up in blackface on Halloween if you did it respectfully). To think otherwise is to engage in what Frederick Schauer calls “First Amendment opportunism.” It occurs, he says, when some debated matter that appears “to have no special philosophical or historical affinity with the First Amendment” is nevertheless framed, at least by one of the parties, in First Amendment terms.14 You want to do something but are blocked from doing it, or something has been done to you and the law seems to offer no redress; you’re stymied, but what can you do? Well, Schauer explains, you can redescribe your cause as one touching First Amendment interests and hope that the redescription is persuasive to the courts.

Schauer cites as a successful example the argument made in 1976 by those who felt burdened by a Virginia law prohibiting the advertising of pharmaceutical prices. The prohibition was sought by independent pharmacists who feared that the aggressive advertising of deep-pocketed chains would drive them out of business. As Schauer observes, the case was really about whether the state should intervene in the free market in order to protect local interests. It became a First Amendment case because advertising (obviously an activity performed by speech) and not price controls happened to be the vehicle of the state’s design. Ruling for the megastores, the majority in Virginia State Board of Pharmacy v. Virginia Citizens Consumer Counsel, Inc. turned an attempt to sell more toothpaste into a noble exercise of First Amendment freedom by equating the familiar phrase “free flow of ideas” with the new phrase “the free flow of commercial information.”15 Advertisers aren’t trying to get your money; they’re just providing information that will enable you as a consumer to make well-informed choices, and therefore massive advertising campaigns like those mounted by large corporations are “indispensable to the formation of intelligent opinions.” (As if the formation of the public’s intelligence were the goal advertising firms set for themselves! Are you kidding?) In dissent, Justice William Rehnquist remarked that he had always understood the First Amendment to be “an instrument to enlighten public decision-making… rather than the decision of a particular individual as to whether to purchase one or another kind of shampoo.” Never mind; after the Court’s holding, the First Amendment’s protections extend to that too. And the way is open to decide, as the Court did thirty-four years later in Citizens United, that the “information” provided by campaign ads is just what the framers had in mind when they wrote the First Amendment.
Is There a Free-Speech Principle?
Obviously, then, First Amendment opportunism exists, but how should we regard it? Is it a practice courts should not engage in because it stretches the First Amendment beyond its proper shape? Or does the First Amendment have no proper shape and, in Schauer’s words, is it nothing more than “a particular set of social, political and ideological moves that are available at a particular point in time”? Is there a core to the amendment, a baseline value or set of values from which one can fall away, or is the amendment the sum of whatever actions taken under its name have been recognized by the courts? This question will become urgent when we turn to the topic of free speech on campus, for we will find that in every dispute each party flies the flag of free speech and accuses the other party of tearing it down. If free speech is an identifiable thing, the two sides can’t both be right: one party would be affirming it and the other would be undermining it. But if each party can make a plausible claim to be wearing the free-speech mantle, we must face the possibility that there is no general free-speech principle and that the label of free speech is applied by polemicists to speech practices that affirm the values they already hold: This is the speech that should be uttered freely without restrictions because it says things I agree with.

That is the conclusion reached by the law professor Larry Alexander, who for over three decades has been telling us that formal abstractions like free speech (formal because what it celebrates is not a particular assertion but assertion in general, no matter what its content) are empty vessels that acquire content when they are filled with preferences of the kind they claim to transcend. Any free-speech calculation will be made in relation to some pre-understanding of what speech is for. If you think that the point of free speech is to promote the free flow of ideas no matter what their content or effects, the fact that some forms of speech produce documentable harms will not be a sufficient reason for regulation. But if you believe that the point of free speech is to facilitate the search for truth, you will be open to the argument that lies and disinformation should be regulated because they frustrate that search. As Alexander explains, “[W]hat comes out of the process will be predetermined by what went into setting it up.” It follows, he adds, that “a pure freedom of expression process is a vacuous concept. There are just different background conditions for expression each of which will lead to different substantive outcomes.”16 By “background conditions,” Alexander means a definition of speech along with the specification of the purposes speech serves. Any such definition or specification will be local, context-specific, and challengeable by someone inhabiting another context who offers a different definition. This means that however speech and its purposes are characterized at a given moment, that characterization will be political and ideological through and through. The moral is one I drew earlier. The zone of free speech is not an oasis apart from political considerations; it is fashioned by political considerations and will always have a shape that is politically angled. As Yale Law professor Robert Post puts it, “[T]he search for any general free speech principle is bound to fail.”17
Principle or Balance?
The failure to find a general free-speech principle is acknowledged whenever First Amendment jurisprudence downgrades freedom of speech to the status of a value. Principles are often described as inviolable (that’s what we mean when we say “It’s the principle of the thing”); if they are to be infringed, it is only in the most extreme circumstances. A value, in contrast, is one among many; therefore values can be and often are in competition, and the resolution of a conflict between them will involve calculations of probability (If we do this, what risks do we incur?) and the weighing of the costs of choosing one over another (If we go with value X, how much of value Y will we sacrifice?). The name for this weighing is “balancing.” The great Judge Learned Hand tells us how it is to be done: first tote up the probable costs of allowing the troublesome speech to flourish, and then tote up the probable costs of regulating it, and go with the option that costs less.18 A balancing calculation might have led the Seventh Circuit Court to side with the city of Skokie, Illinois, in its effort to prevent a neo-Nazi group from marching through neighborhoods populated by Holocaust survivors.19 When the American Civil Liberties Union (ACLU) supported the marchers (who never in fact marched in Skokie), the organization lost some members, although many applauded what was termed its “principled stand.” The principle is articulated by journalist Glenn Greenwald, who writes that the doctrine of freedom of speech is intended to foster “exactly those political ideas that are the most offensive, most provocative, and most designed to inspire others to act in the name of its viewpoints.”20 In short, the worse the speech and the more dangerous its effects, the more it deserves protection. No balancing allowed.

The tug-of-war between balance and principle is a general feature of life in a democracy, but it has a particularly visible form in Supreme Court cases, where free speech is often celebrated as a, if not the, central principle of our democracy. But in many of those same cases freedom of speech is treated as a value, an especially honored value to be sure, but one that can be set aside in certain well-defined circumstances. Here’s the formula: when free-speech rights are in the mix, they can be outweighed if (1) limiting them serves a “compelling state interest” (lawyer talk for “really important”) and (2) the protection of that interest requires the limitation of speech which must be (more lawyer talk) “narrowly tailored” and the “least restrictive means” of achieving the desired goal. This is a balancing test with a high bar—free-speech interests can be slighted only if the interest they clash with is held in the highest regard—but one that can be cleared, as it has been, for example, when the courts have allowed some limitations on the ability of antiabortion advocates to directly confront women entering a clinic.21 In the balance, the vulnerability of women contemplating a legal act outweighed the unfettered exercise of free-speech rights.
What Is the First Amendment For?
If balancing goes on all the time despite the repeated invocation of principle, there is a disconnect between what we say about freedom of speech and what we do when freedom of speech issues arise. This disconnect is an inheritance from the history of free-speech jurisprudence, a history that unfolds in three stages. (The history I offer is a bit crude, but it will do for our purposes.) In the first stage, lasting from the beginning of the republic to the early decades of the twentieth century, speech could be regulated if it harbored within itself a “bad tendency” so that permitting it would constitute a threat to the public welfare. (“Bad tendency” has made a comeback in the form of demands by college students that certain words and phrases should not be said because they are inevitably imbued with racism or sexism or homophobia.) In stage 2, the bad tendency theory was succeeded by the “clear and present danger” test formulated by Supreme Court justices Oliver Wendell Holmes and Louis Brandeis. Where the bad tendency test assumes that the badness of some utterances is inherent—they are just bad no matter when or where they surface—the clear and present danger test is sensitive to the circumstances of utterance: the government must wait until it is apparent that a particular use of language is likely in the near future to bring about a “substantive evil” the state has a right and an obligation to prevent. The two tests measure the potential harm inflicted by both the content of speech and its likely effects. Speech considered deficient in these two ways—it says bad things and/or it does bad things—was thought not even to raise a constitutional issue; in a 1942 case, Chaplinsky v. New Hampshire, a famous paragraph lists the kinds of speech (obscene speech, insulting speech, fighting words, among others) the First Amendment will not protect because they are “no essential part of any exposition of ideas, and are of such slight social value as a step to truth that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality.”22 The opposition between liberty—free speech for all no matter what is said and what the harm—and the interest society has in maintaining good order will structure First Amendment jurisprudence for decades to come.

Within twenty-five years, almost all of the forms of speech labeled “low value” (another term of art) by Chaplinsky had been brought under the constitutional umbrella. (This is the third stage.) How did that happen? It was in part the work of New York Times v. Sullivan, a case that rejected both content and effect tests and declared that speech is valuable in and of itself because no matter what it says or what it does, it contributes to the “uninhibited, robust and wide-open” conversation that is the lifeblood of a democracy.23 Even false speech must be protected because falsehoods, as much as true statements, play a part in the ongoing work of the Marketplace of Ideas. Justice William Brennan declared, “Neither factual error nor defamatory content suffices to remove the constitutional shield.” Indeed, the distinction between fact and falsehood is blurred, and that blurring is made official when in Gertz v. Robert Welch the Court declares, “Under the Constitution there is no such thing as a false idea,” which doesn’t mean that there are no false ideas, but that the falsity of an idea does not disqualify it from participation in the public dialogue.24 (As I shall argue in chapter 5, it is this logic that paves the way for fake news and for the modern mantra “My opinion is as good as yours.”) It is with New York Times v. Sullivan that freedom of speech officially becomes a principle, at least in the legal world, although it is a principle often breached just after it is invoked; in practice there have been any number of competing values—public safety, national security, the welfare of children—that captured the attention of the courts and led to restrictions of speech.

Yet, even if the rhetoric of principle is belied by what courts actually do, that rhetoric remains socially and politically strong. Many American citizens reflexively affirm the value of free expression and don’t require that it be justified by reference to some other value. Strong free-speech advocates don’t ask “What is the First Amendment for?” Merely to pose that question would be to imply that the rationale for the amendment is to be located in a goal prior to it. Freedom of speech, on the strong view, is a freestanding value. Yet any celebration of that value typically includes a list of the benefits free speech provides. It facilitates the search for truth, or it provides the free flow of information necessary to an informed citizenry, or it opens a space for dissent and thereby provides a counterweight to the pronouncements of entrenched authority. But if these are the goals the First Amendment helps us to realize, there must be some forms of speech that impede rather than aid their realization—speech that blocks or undermines the search for truth, speech that corrupts rather than informs the minds of citizens, speech that shuts the democratic conversation down rather than keeping it open. It follows that regulating such speech, rather than violating the First Amendment, is an act of fidelity to it. If you have any answer at all to the question “What is the First Amendment for?,” you are logically committed to censorship somewhere down the line because your understanding of the amendment’s purpose will lead you to regulate or suppress speech which serves to undermine that purpose. We see once again that censorship is not a violation of the First Amendment but the necessary vehicle of its implementation. The choice is never between free speech or censorship but between different paths of censorship, each of which will follow logically from an answer to the question “What is the First Amendment for?”

Of course if you have no answer to that question, if you think it wrong even to ask because freedom of speech just is our guiding principle no matter what, no argument for state regulation of speech, even one made in the name of a cherished interest, will sway you. But you will then have to explain why we should revere something that doesn’t seem to be good for anything specific. Identifying freedom of speech as a self-justifying value unlinks it from any goal a society might think to promote. A value that admits no other measure than its own would seem to be at home only in a Hyde Park corner or some other place where talk is produced for its own sake and not for any result, good or bad, it might bring about. To be sure, strong free-speech proponents have a response to this line of reasoning: while the free exchange of ideas may not deliver an immediate payoff, in the long run it will lead to the ultimate clarification of issues that now seem muddied. (“Now we see in a glass darkly, but then face to face.”)25 That clarification, it is promised, will be the work of the Marketplace of Ideas, a concept often invoked by free-speech champions but one that is, as we shall see in the next chapter, more problematic than its proponents believe it to be.
The Two Stories of Speech
To this point the ruling assumption of the discussion has been that, all things considered, speech is a good thing. But there is another, darker story about speech, and it is told by Thomas Hobbes in his great work Leviathan (1651). When he first mentions speech, Hobbes praises it as the capacity that distinguishes us from the animals; it is “by the help of speech” that man’s faculties “may be improved to such a height as to distinguish men from all other living creatures.” While both men and animals will react to a loud noise, only men will devise plans to muffle it. But later, when Hobbes returns to the topic, he says that the same capacity of speech is what enables us to be duplicitous, deceptive, and destructive; it is by the “art of words” that some men “represent to others that which is good in the likeness of evil, and evil in the likeness of good.” Speech as a category of action is thus a leading character in two narratives. In one, originated by the Pre-Socratic Sophists and popularized by Cicero and other classical humanists, speech is the deliverer of civilization, the faculty that allows us to formulate plans, recommend policies, urge actions, rise to life’s challenges, and coordinate our efforts to better the human condition. The British historian Timothy Garton Ash provides a recent reformulation of this optimistic view: “?‘[F]reedom of information’ and ‘freedom of expression’ enable ‘us to get as close as humanly possible to the truth.… The more freely a wide range of alternatives is aired for any decision, the better chance we have in choosing the best course of action.’?”26 In the other, less happy narrative, speech is the medium through which we deceive our wives and husbands, manipulate our fellow citizens, betray our civic missions, and incite violence against our enemies. If those are the things the free flow of speech helps us to do, more free speech may not be what we want to encourage.

The thing about these two narratives is that they are both true, which is why we have two contrasting attitudes toward the production of speech enshrined in our laws. Some of our laws are obviously designed to increase the amount of speech available to the citizenry; others of our laws take note of the harms that speech engenders. The debates between the ACLU and the antipornography and anti–hate speech forces reflect this dual evaluation of speech which has been with us since by his word God created the world and by his words Cain tried to deny that he had killed Abel. Is there a way of purifying or distilling speech so that its good effects remain and the bad ones are filtered out? It has long been a project to do just that. It must be possible, many have thought, to fashion a mode of communication incapable of distortion and hostage to no parochial interests. If there were such a device and speakers tied themselves to it, they would be able to say only the thing that is true, for the linguistic resources necessary to falsehood would have been removed. If in speaking and writing you employ a language that cannot give voice to bad motives, you would yourself be truthful and honest, at least in your verbal behavior.27

This fantasy—and fantasy it is—has given rise to any number of projects. In the Middle Ages and early modern periods the plan was to re-create the language spoken by Adam and Eve in Eden, a language whose vocabulary, it was said, perfectly captured the essence of the things it named. In the seventeenth century John Wilkins, one of the founders of the British Royal Society, wrote An Essay towards a Real Character and a Philosophical Language (1668). The idea was to replace the ambiguity and redundancy of ordinary language with a finite universal set of symbols that would provide “elementary building blocks from which could be constructed the universe’s every possible thing and notion.” Such a language, Wilkins argued, would be free of redundancies (more than one word signifying a single thing), equivocals (words that refer to more than one thing), and metaphors (words that tell you not what a thing is but what it is like). Much later, in the twentieth century, the logical positivist Rudolf Carnap constructed a language in which “every primitive term is a physical term”;28 should you wish to speak of matters other than physical ones, you must define them in relation to those basic terms. Once again, no equivocals, no redundancies, no metaphors, and, also, no metaphysics. In the mid-twentieth century George Orwell offered a popular version of this perennial project when in “Politics and the English Language” (1946) he urged speakers and writers to purge foreign vocabularies from the English language and to limit themselves to good down-to-earth Anglo-Saxon words that refer precisely to things and not to abstractions. If we can thus purify our language, he declared, we will at the same time purify our thoughts and engage in a purified politics. (If it were only that easy!) The latest chapter in this story emerges with the internet and the hope that data freely circulated without filters or gatekeepers will eliminate misinformation and distortion because everything will be out in the open and deception will be impossible. In this utopian vision, the unregulated internet will be the true realization of the Marketplace of Ideas.

Projects like these represent an effort to erase the darker of the two stories told about speech, but the effort will inevitably fail. Eliminating the human stain from the human process of conversing and proposition-making is a losing game, even if the game is given honorific names like “artificial intelligence.” (More of this in chapter 5.) We’re stuck with what we are and with the glories and liabilities of being speaking creatures. Speech will always be a gift with two faces: it makes possible the civilizing arts and enables the growth of individuals, but it is also the vehicle of suffering in the form of lies, slanders, and deceptions. As we saw, Mark Zuckerberg tries to sit on both sides of this fence. He wants to give everyone a voice even when some voices spew poison, and he also wants his platform to be a force for good and do no harm. Can he have it both ways? Is there a formula that allows us to regulate speech whose effects we deplore without compromising the free-speech principle we honor? That is the question that informs debates about hate speech, speech that flouts the norms of civilized society and in its strongest form threatens to undo the ties that bind us. Must we afford such speech First Amendment protection even when its malign effects are documented? Do we follow the principle that every view must have its chance to be heard to the bitter end, or is there a point at which we say “Enough!” and crank up the machinery of regulation? For answers, read on.
Barney Cokeliss

Stanley Fish is the Davidson-Kahn Distinguished University Professor of Humanities and Law at Florida International University and a visiting professor of law at Cardozo University. He has previously taught at the University of California at Berkeley, Johns Hopkins University, Duke University, and the University of Illinois at Chicago where he was dean of the College of Liberal Arts and Sciences. He has received many honors and awards, including being named the Chicagoan of the Year for Culture. He is the author of many renowned books, including Winning Arguments and How to Write a Sentence: And How to Read One. Fish is a former weekly columnist for The New York Times. His essays and articles have appeared in The Chronicle of Higher Education, Harper’s Magazine, Esquire, and The Atlantic.

"Fish's points arrive in thoughtful, dense provocations that will require close attention from readers."
Kirkus Reviews

“Fish zeroes in on the First Amendment in this well-constructed analysis, offering his nonpartisan take on what it does and doesn’t protect and what kind of speech it should and shouldn’t regulate.” 
Publishers Weekly