In a decision which emerged as an afterthought in the wake of its contemporaneous Obamacare ruling, the Supreme Court on June 28 addressed a significant First Amendment question in United States v. Alvarez. While the immediate effect of the Alvarez ruling — striking down the Stolen Valor Act’s prohibition on lying about military decorations — is speech-protective, the Court missed an opportunity to take a stronger stand on the First Amendment protections afforded even to false speech.
In Alvarez, the Court fractured into three blocs in its answer to the broader question touched on by the case: what is the First Amendment value of falsity? Justice Kennedy, writing for a four-justice plurality that also included Chief Justice Roberts and Justices Ginsburg and Sotomayor, started from the premise that speech receives First Amendment protection unless it falls into a defined unprotected category. This stance is fully in line with the Court’s previous rejection of the “startling and dangerous” concept of applying an ad hoc balancing test to every proposed restriction on speech. While noting that bans on some types of lies, such as perjury or impersonating a police officer, have been consistently upheld, the Court’s plurality here dismissed the contention that these relatively isolated instances suffice to establish “lies” as a historically unprotected category. “Absent from those few categories where the law allows content-based regulation of speech is any general exception to the First Amendment for false statements.”
In dissent, Justices Alito, Scalia, and Thomas contended that the plurality’s basic premise should be turned on its head — they would start with the assumption that falsehoods are unprotected unless they have particular value or fall within a category of speech for which a “breathing space for truth” is particularly crucial. On their view, unless the lie in question relates to “philosophy, religion, history, the social sciences, the arts, and other matters of public concern,” government regulation — no matter how “chilling” to expression — does not offend the First Amendment.
Between these polar philosophical starting points stood the two justices whose concurrence controlled the outcome. Justice Breyer, writing for himself and Justice Kagan, staked out a hybrid approach. While agreeing with the plurality (and the dissent) that untruth spokenon core political subjects requires some First Amendment protection, Justice Breyer would apply “intermediate scrutiny” to lies spoken outside these core areas. His test would take into account “the seriousness of the speech-related harm the provision will likely cause, the nature and importance of the provision’s countervailing objectives, the extent to which the provision will tend to achieve those objectives, and whether there are other, less restrictive ways of doing so.”
The Court ultimately mustered a six-vote majority to strike down the Stolen Valor Act in Alvarez because the concurring justices agreed with the plurality that the government had failed to show the absence of less-restrictive means of achieving the law’s objectives. In their controlling opinion, the two concurring justices noted that the “seriousness of the speech-related harm” posed by the law was greater because of its breadth. Unlike trademark protection statutes previously upheld by courts, the Stolen Valor Act prohibits lying about military decorations in all circumstances, ranging from fraudulent attempts to gain monetary or political advantage to mere “bar stool braggadocio.” The broad sweep of the statute makes a means-end fit especially important, and all six justices agree that less restrictive means are available. According to the justices, a narrower statute, focusing on more concrete harms such as lies motivated by material gain, would strike more cleanly at the government’s goal of preventing exploitation of the military honors system. A publicly-available database of military honorees would also prevent degradation of the system, leaving the government free to prosecute only the most egregious instances of abuse (the Pentagon has created just such a database since the decision).
The Court’s six-to-three holding that the Stolen Valor Act is unnecessarily and unconstitutionally overbroad is certainly a welcome result. However, the narrow logic dictated by the controlling concurrence’s stance amounts to a missed opportunity. As I discussed in March after Alvarez was argued before the Court, several Justices raised pointed questions about whether the government’s asserted objective in promulgating the Act is even coherent; perhaps more importantly, they questioned whether even a narrowly-tailored statute can avoid the speech-chilling pitfalls stemming from any government mandate to stamp out lies. The plurality opinion gestures towards these larger problems, but Justices Breyer and Kagan decided the case only by reference to the “proportionality test” shortcomings of the Stolen Valor Act itself.
The most fundamentally disappointing aspect of the decision is the concurring justices’ failure fully to embrace the speech-protective First Amendment framework with which the plurality begins its analysis. As Justice Kennedy explained, the Court should start every free speech case from the assumption that all expression receives some constitutional protection unless it falls within a well-defined unprotected category. The Court does not, and should not, create new unprotected categories; it can, at the most, recognize the existence of unprotected types of speech long ratified by historical practice. As the plurality demonstrates, “lies” are no such category. In fact, untruth is not a category of speech at all; valid content-based restriction targeting untruth have been upheld solely because the targeted speech occurs in a social context in which truth is particularly essential. Laws forbidding perjury, impersonating a government official, or a political candidate’s lying about his incumbent status all serve as responses to concrete threats to government functions that cannot operate without some means of guaranteeing participants’ truthfulness.
In the limited instances where content-based restrictions targeting lies are supported by the overwhelming need to maintain the integrity of government processes, they can stand up to the same “strict scrutiny” that is normally applied to content discrimination. Even if such a test is “strict in theory and fatal in fact,” existing precedent leaves room for new legislation addressing analogous government interests. For instance, Senator Scott Brown of Massachusetts has proposed a new, narrower Stolen Valor Act — one which criminalizes only those lies about military decorations which are motivated by material gain. Since this new Act addresses only one discrete (and less-protected) category of speech, it could be upheld as analogous to the type of trademark legislation discussed in the “Olympic Rings” case.
Because the concurring justices did not accept the plurality’s strongly speech-protective doctrinal framework, however, we are left with a decision which strikes down the overbroad Stolen Valor Act but does nothing to repudiate the government’s assertion that its interests can be jeopardized by any falsehood which it claims diminishes the value of its military honors. As Justice Sotomayor pointed out in oral arguments, the government never convincingly explained just how a lie threatens the military honors system, nor did it convincingly counter the argument that the dissemination of the truth is sufficient as an antidote to any damage caused by lies.
Taken together, the dissent and concurrence amounted to a majority of five justices supporting the proposition that lies as a category receive less protection from the First Amendment. This argument relies on the common-sense intuition that falsehood contributes nothing to the pursuit of truth and little to the functioning of the democratic process. However, it fails to account for the often-enormous difficulty of separating truth from lies. More importantly, it fails to give full weight to the crucial role that our tolerance of lies and exaggerations plays in ensuring open dialogue and full autonomy of expression — what the plurality referred to as “the common understanding that some false statements are inevitable if there is to be an open and vigorous expression of views in public and private conversation.”
The less rigorous test proposed by the concurring justices produced a desirable result in this case, but it seems to portend an approach that gives less than full scrutiny to asserted government interest in content-discrimination. As Eugene Volokh has pointed out, the case does little to address prohibitions on certain kinds of lie by political candidates — an issue which has produced an unresolved circuit split. If the Court relies on the Alvarez concurrence’s approach in future cases, it could well give short shrift to what should be its starting presumption: that content-based restrictions on “lies,” like any other form of speech, are justifiable only in extraordinary circumstances.
Matt, I know this will come as a shock to you because we are usually in exact agreement on the scope of the First Amendment, but I think the concurrence got it exactly right. The dissent’s position is unworkable, mainly because the areas they list as generally protected are in fact topics in which most discussions are matters of opinion or theory, and therefore the concept of a “lie” doesn’t even apply. We shouldn’t craft a jurisprudence in which speech is ever unprotected as a presumption. Even the longstanding permissible speech restrictions are justified by compelling interests, meaning there is no need to make speech presumptively unprotected in order to protect those interests.
A balancing approach leads to the conclusion that knowing falsehoods generally have little First Amendment value, but that the breadth of the Stolen Valor Act allowed it to reach speech that threatened little to no harm. I am fine with a presumption that when the balancing approach yields a close result the appropriate response is to protect the speech. I think if applied only to situations in which the person was lying about military honors to receive a particular tangible benefit (including and especially election to public office), the harm is present and outweighs the minimal value of the false speech.
As far as the plurality, I don’t think we need to be quite so afraid of expanding categories of unprotected speech. If anything, I think a balancing approach would be preferable to the current categorical approach largely because there wouldn’t be such a strong preference of protection of the speech even when significant harm is shown (see, animal cruelty videos). At some point, each of the categorically unprotected types of speech was subjected to a balancing test, and I don’t think we should refrain from applying similar tests to new types of speech, or from reevaluating whether the balancing was appropriately done in prior cases.
I don’t think this case needed to do anything about the “enormous difficulty of separating lies from truth” because it dealt only with a situation in which truth was easily veriable. The Volokh post points out others in the context of political campaigns, such as knowing falsehoods about when and where to vote, about endorsements, or about incumbency. Those kinds of falsehoods shouldn’t be given a presumption of protection. They undermine rather than support the search for truth, and their falsity is easily verifiable as they is no room for differences of opinion on such factual matters. In this case, maybe the balancing approach should have included a higher standard for a showing of the government interest to be served, but I think if it is applied to election cases, the goverment interest is substantial and there is little risk for dangerous impediments to true or evenly arguably true speech.
Thank you for this concise and illuminating analysis. This piece cuts to the heart of the difference between the plurality opinion and the concurrence in United States v. Alvarez, and explains why the difference matters.
There’s one point I would like to highlight, in fairness to the concurrence.
Despite its failure to use the categorical test endorsed by the plurality, Justice Breyer’s concurring opinion makes the stronger case for the positive value of false speech. This argument lends credence to the idea that false speech merits protection.
As you note, the plurality refers to “the common understanding that some false statements are inevitable if there is to be an open and vigorous expression of views in public and private conversation.” But this is a relatively weak defense of false statements; their protection is only incidental to protecting other forms of speech.
In contrast, the concurrence starts from the premise that “[f]alse factual statements can serve useful human objectives” and goes on to explain their social value in specific private, public, and scientific contexts. Justice Breyer writes that falsehoods can “prevent embarrassment, protect privacy, shield a person from prejudice, provide the sick with comfort . . . stop a panic . . . [and allow the] examination of a false statement (even if made deliberately to mislead) . . . [to] promote a form of thought that ultimately helps realize the truth.” Presumably, these considerations would have weight when applied in Justice Breyer’s proportionality test. But within the plurality’s framework, these reasons explain why it makes sense to have such a robust presumption against content-based restrictions on speech.
I think the language from the concurrence gets at why a balancing approach is preferable to a categorical approach: because speech is nuanced. The exact same words can mean so many different things and can affect listeners in so many different ways based on the context, the identity of the speaker, the identitity of the listener, the intent of the speaker, etc. A categorical approach creates false equivalency by forcing judges to only consider whether statements fall into pre-defined categories of protected or unprotected speech, rather than allowing them to consider the full context of the false statement and its intended and actual effects on listeners.
Both the plurality and concurrence approaches can start from a presumption of speech protection. The difference comes in how they choose to analyze whether the presumption should prevail. Under the plurality approach, the next step is determine what category the speech falls into, accepting as absolute truth that the categories that have been previously determined as unprotected are the only categories that should merit that status. The concurrence would instead take the next step of balancing, taking a more complete look at the speech in question and analyzing the full context in which it occurred. That makes so much more sense to me in a society in which the ways we speak are constantly changing and the balancing that occurred decades ago to decide on the appropriate categories of unprotected speech may no longer come out the same way.