We live in a time of intense anxiety about the problem of lies. It is an “Age of Disinformation”; lies, it is often said, are destroying our democracy; they are tearing us apart; they are creating an “epistemic crisis” of truth. And yet, it is often also said, the First Amendment prevents the government from doing much about it because lies are protected speech. Except for the narrowest exceptions, the claim goes, laws that limit the dissemination of false speech are unconstitutional. Many critics of the regulation of disinformation accept that lies are dangerous and something that we all would be better off without. The problem, alas, is that once you give the government the power to regulate lies, you’re one step away from creating an Orwellian Ministry of Truth.

These blanket claims find some support in the language of the Supreme Court’s 2012 opinion in United States v. Alvarez. Alvarez is the most recent and most important Supreme Court opinion about the regulation of lies, and Justice Anthony Kennedy’s plurality opinion in that case did indeed invoke the dystopia of Oceania described in George Orwell’s 1984 when striking down a law that made it a misdemeanor for a person to lie about whether they had received a military medal of honor.

But Alvarez, like many of the other Supreme Court opinions on the topic of lies, is a rather more nuanced opinion than is commonly assumed, and one whose precise reach remains far from clear. It is also not the steadiest precedent. There is good reason to think that it articulates a view of the First Amendment and lies that many members of the Supreme Court are no longer terribly sympathetic to. Given the fact that lies, their ubiquity, and their regulation, are perhaps the most prominent topic in debates about freedom of speech today, it’s a good time to reread Alvarez to examine what it does—and doesn’t—say.

Alvarez isn’t Clear

In defending the Stolen Valor Act in Alvarez, the government argued that lies—because they do not contribute to the search for truth, or democratic self-government, or any of the other good things the First Amendment protects—have no constitutional value. A plurality of the Court disagreed. Because lies are so pervasive in social life, the Alvarez plurality argued that categorically excluding lies from the protection of the First Amendment would give the government too much power—power that could be abused.

But the Alvarez plurality did not issue a sweeping holding that all regulations of lies violate the First Amendment. Justice Kennedy’s opinion in Alvarez made clear that lies could be regulated when they create “legally cognizable harms.” The obvious next question is what the term “legally cognizable harm” means. Frustratingly (but perhaps not surprisingly), Justice Kennedy does not define the term. But he does give some hints of what it might mean by listing some kinds of lies that threaten it, including defamatory falsehoods, lies designed to cheat their listener out of money or something else of material value, lies to government officials about matters within their jurisdiction, and perjuries. Reasoning analogically, these examples suggest that legally cognizable harms include harms to an individual’s reputation, economic or personal well-being, harms to the efficient operation of government processes or the legitimacy of government institutions, and harms to the “function and province of the law” and to, as Justice Kennedy put it, “the integrity of [legal] judgments.”

There’s no reason to think that this list is comprehensive. The plurality opinion at no point suggested that it was. But even if we limit ourselves to these kinds of harms, it should be immediately obvious that there are many kinds of lies that the government may constitutionally regulate under the rule in Alvarez.

And really, this should be no surprise. We all readily accept government regulation of lies all the time. Federal laws make it unlawful to lie when applying for visas, passports, or federal benefits; to lie in connection to the Social Security program; to lie about the value of securities; to lie about the location of polling places; to lie to a federal officer about a matter under his or her jurisdiction; to lie about a product that is being advertised; to lie when purchasing a handgun; to lie in order to con someone else out of money or some other valuable good; and to lie in many other contexts. Many state laws do the same. Lies are commonly regulated, in other words, when they threaten harm to government programs, or undermine the government’s ability to enforce its laws, or pose a particular threat to consumers. The Stolen Valor Act that was struck down in Alvarez was itself repassed by Congress to criminalize the same lies, but with the added requirement that the lie be told with intent to benefit.

But, again, there’s no reason to think that the government has exhausted its capacity to punish falsehoods. Surely lies about electoral procedures or outcomes threaten a similar kind of harm to the integrity of government processes as do lies about the location of polling places. Certain kinds of medical misinformation—misinformation designed to prevent listeners from accessing effective health care, perhaps, or that try to profit from harmful “miracle cures”—threaten personal well-being in the same way that lies aimed to con their victims out of money or valuable goods do. Other examples could be imagined.

This is not to say that the government has carte blanche. There are other guardrails on truth policing. Mens rea is important: Alvarez strongly suggests that the government can only regulate lies when they are intentionally or recklessly made because to do otherwise would chill too much well-intentioned speech. Laws regulating lies should also sweep no more broadly than necessary: Part of the constitutional infirmity of the Stolen Valor Act was that it applied “in almost limitless times and settings” so that it criminalized even lies that were made inadvertently, that no one believed, and that were not publicly disseminated—lies that did not cause, to any significant extent at least, the harms the act was intended to prevent. And there is no question that the First Amendment prevents the government from restricting “false” opinions—to claim, for example, that communism is a good system of government, or that President Trump has great hair and made America great again. For almost 100 years, the Supreme Court has consistently recognized that such laws violate the fundamental democratic right for citizens, not their government, to decide what to believe about matters of opinion.

These qualifications are important. It should go without saying that a ministry of truth would be, um, bad (although often anyone that suggests regulating any falsehoods at all is accused of implicitly thinking otherwise). Shutting down or polluting free public discourse is the first recourse of would-be authoritarians both at home and abroad. Recent history and our current moment should be enough to counsel caution in asking the government to police political speech—and to show that almost any kind of lie (including, for example, health misinformation) can be political.

Government regulation of truth is fraught and often dangerous. But our point is that, contrary to what a lot of public rhetoric might suggest, it’s not completely off the table in any and all cases. To give just one example: Colorado recently proposed a law that included a provision that would have prohibited election officials from recklessly or knowingly lying about election procedures. But that provision was dropped because it was warned by an apparent expert that it would be an unconstitutional speech regulation. But this reflexive conclusion is not justified. The law very likely would have satisfied all of Alvarez’s limiting conditions—it punished lies that pose a legally cognizable harm, it applied only to knowing or reckless lies, and it didn’t sweep broadly beyond the circumstances in which telling such lies threatens the legitimacy of government processes. Such a law would therefore have been on strong constitutional ground. To read Alvarez as standing for the proposition that there is no room for this kind of carefully crafted legislation is a mistake, and a costly one.

Alvarez isn’t Stable 

Even if that is what Alvarez meant, it’s unlikely to be the Court’s last word on the subject. Alvarez was, after all, a plurality opinion that only four members of the Court joined. One of the other justices in the case—Justice Elena Kagan—signed on to a concurring opinion that argued that lies possess less constitutional value than other kinds of speech, and so laws that regulate them should be subject only to intermediate scrutiny, rather than the strict scrutiny that the plurality applied to the Stolen Valor Act. As this suggests, it would be a mistake to remember Alvarez as an easy case in which the outcome was foreordained. Indeed, Don Verrilli Jr., solicitor general at the time, has described the case as one of the two he most regrets losing (Shelby County being the other). Alvarez is often thrown around now as if it represents an obvious truth (ha!) practically baked into the Constitution from the founding, but that is, in fact, false. It was not self-evident in 2012 what the Court would find then, and it’s not self-evident what the new Court would hold now.

It is perfectly possible that the current Court would reach a different conclusion than the Alvarez Court about how lies should be treated under the First Amendment. Only two of the justices who joined the Alvarez plurality remain on the Court today. And recent cases suggest that at least some of the Court’s members are getting increasingly concerned about the threat that certain kinds of lies pose to the health of the democratic public sphere. Justice Clarence Thomas, for example, has argued in multiple recent opinions that the limits the First Amendment imposes on the use of libel laws to punish falsely defamatory speech should be loosened to better protect against the “proliferation of falsehoods” in the United States. Justice Neil Gorsuch has similarly argued that changes to the libel laws may be necessary given a changing media environment in which “virtually anyone in this country can publish virtually anything” that “facilitates the spread of disinformation.” There is no reason why Justices Thomas’ and Gorsuch’s concern with the proliferation of falsehoods in the United States should affect only their rulings in libel cases. The basis of their dissatisfaction with current law is the changed technological, social, and political realities of the current information environment and its far-reaching consequences.

These are profound shifts in the composition and disposition of the Court. We should therefore not assume that the existing cases give us all the answers. What constraints the First Amendment imposes on regulators seeking to limit lies will not be determined by a close parsing of this or any other recent case but will require engaging with the principles that motivate those opinions, the purpose of the First Amendment, and the nature of our changing information ecosystem.

It certainly would be easier if Alvarez provided a simple answer to the constitutional questions raised by the problem of lies. But in fact, doctrine provides no escape from the messy and difficult business of figuring out, under existing conditions, when the danger of not regulating lies exceeds the danger of regulating them. The opportunities—both salutary and not—that this leaves open are in fact vast and remain to us to work out.