Imagine a criminal statute commanded: “Don’t be a jerk.” Or: “Thou shalt not be unpleasant.” Or even: “Don’t do what’s bad.” What would be wrong with that?
Well, for one thing, the categories of “jerk,” “unpleasant,” and “bad” are so undefined that citizens could not upon reading the law know what it is they are not to do. In constitutional terms, that is a due process / fair notice problem—an egregious one when coupled with the peculiar oppressiveness that is the hallmark of criminal law. Another problem is that such statutes would give effective lawmaking power to police, prosecutors, and courts—the executive and judiciary, rather than the legislature—because the content of the law would depend so thoroughly on how the law’s enforcers and interpreters defined “jerk,” “unpleasant,” and “bad.” That is a separation of powers problem, constitutionally speaking. A third problem is that standardless law gives cover to police and prosecutors who engage in arbitrary forms of law enforcement—another due process problem, although one that can have equal protection and even First Amendment implications (if, for example, the open-ended law is used to target political dissenters or racial or religious minorities). I would add that highly undefined criminal prohibitions mean that the executive acts under pretext of law but is in practice unbound—a rule of law problem. And—another unconventional but I think interesting constitutional argument—since to prosecute criminally is necessarily to punish, all of these constitutional issues could be cognized as cruel and unusual punishment.
For all these reasons, vagueness doctrine has been part of constitutional law for at least a century. The most famous example is Papachristou v. City of Jacksonville, in which a local ordinance criminalized, among other things, “persons wandering or strolling around from place to place without any lawful purpose or object,” “habitual loafers,” and “disorderly persons.” Justice Douglas breathed fire defending Anglo-American liberty from that affront: “These unwritten amenities have been in part responsible for giving our people the feeling of independence and self-confidence, the feeling of creativity. These amenities have dignified the right of dissent and have honored the right to be nonconformists and the right to defy submissiveness. They have encouraged lives of high spirits rather than hushed, suffocating silence.”
Void-for-vagueness doctrine has been putt-putting along ever since. It is a sibling to the rule of lenity: the two together mean that, where a penal statute is a little unclear, it should be interpreted in favor of the defendant (lenity), and, where it is very unclear, it should be struck down as unconstitutional (vagueness). Or at least that’s the ideal. The reality has been that a lot of criminal statutes are enforced despite significant and even profound vagueness (just think of the Sherman’s Act’s felony prohibition on “[e]very contract, combination … or conspiracy, in restraint of trade”) and appellate courts routinely split 2-1 (or even 5-4) on matters of statutory interpretation in criminal law while denying that the statute that so divided professional judges is unclear enough to justify lenity. Criminal law professors tend to teach these doctrines with disclaimers.
Or so it has been. But the landscape may be changing, and changing in ways that say a lot about what this new Supreme Court—the Court of Kavanaugh, Gorsuch, Sotomayor, and Kagan—will mean for substantive criminal law. A trio of recent Supreme Court cases has given void-for-vagueness doctrine a giant push. For those of us who think American criminal justice has gone profoundly awry, constitutional vagueness deserves careful attention. It is one of the few tools of criminal justice reform that falls under judicial control, that seems likely to persuade enough judges to get real uptake in the courts, and that seems powerful enough to make a difference, though I am not sure the difference will be benign.
Johnson v. United States, 576 U.S. __ (2015), was about a provision of the Armed Career Criminal Act under which an offender’s sentence could be enhanced if he had been thrice previously convicted of “a violent felony.” What is a “violent felony”? It is a crime punishable by a term exceeding one year that “has as an element the use … of physical force,” or falls under a list of specified crimes (e.g., burglary), or “otherwise involves conduct that presents a serious potential risk of physical injury to another.” Truth be told, that definition doesn’t strike me as half-bad relative to many others I come across in teaching and writing about criminal law. But the Court took exception to what has come to be called the “residual clause” of the statute—the catchall phrase, “otherwise involves conduct that presents a serious potential risk of physical injury to another.” The only way to apply the residual clause, the Court reasoned, is to take the defendant’s crime (e.g., possession of a short-barreled shotgun, the crime at issue in the case), imagine a typical case of that crime (whatever that might be), and ask whether, in that typical case, the crime would present sufficiently serious risk of injury. But how is one to fix upon a typical case? And how much risk is enough? The whole line of thought is just too speculative to be the basis for major punishment, the Court reasoned. Justice Scalia wrote for the six justice majority—one of his last opinions. Justices Kennedy and Thomas concurred and Justice Alito dissented, but that is mostly because Kennedy and Thomas did not think possessing a short-barreled shotgun counts as a “violent felony” and Alito did; all three agreed on the key point that the residual clause is not unconstitutionally vague.
Sessions v. Dimaya, 584 U.S. _ (2018), involved quite similar statutory text, at least superficially. The Immigration and Nationality Act renders deportable an alien convicted of an “aggravated felony.” Part of the definition of “aggravated felony” is: “any other offense that is a felony and that, by its nature, involves a substantial risk that physical force against the person or property of another may be used in the course of committing the offense.” In other words, at issue was another “residual clause,” and the Court rejected it for the same reason they rejected the one in Johnson: the toxically vague combination of having to imagine a typical case of the relevant crime and having to gauge that hypothetical crime’s level of risk. Justices Kennedy, Thomas, and Alito again dissented. Uniquely, Chief Justice Roberts “switched sides,” as it were, arguing that the sentencing statute in Johnson and the immigration statute in Dimaya were only superficially similar, and that the differences made the latter statute constitutional where the former was not.
But surely Justice Gorsuch wrote the most noteworthy of the opinions in Dimaya: a long concurrence defending void-for-vagueness doctrine on originalist grounds, arguing that the doctrine should hold broadly in civil as well as criminal cases, and then critiquing the text at issue for a vagueness so “implacable” that “[i]t leaves the people to guess about what the law demands—and leaves judges to make it up.” His concluding language was forceful and passionate: “Choice, pure and raw, is required. Will, not judgment, dictates the result…. A government of laws and not of men can never tolerate that arbitrary power. And, in my judgment, that foundational principle dictates today’s result.” One is left with the impression that vagueness doctrine goes to Gorsuch’s core sense of what law should be, and he will hold legislatures to it in the future. The opinion is an announcement.
The issue in United States v. Davis, 588 U.S. _ (2019), was yet another residual clause connected to defining violence. The defendants were charged with a string of armed robberies. Besides convicting them of the robberies, the government sought to increase their sentences for carrying firearms in furtherance of a “crime of violence,” defined under the residual clause as a felony “that by its nature, involves a substantial risk that physical force against the person or property of another may be used in the course of committing the offense.” Writing for the majority, Justice Gorsuch concluded that the statute here required the same approach as the statutes in Johnson and Dimaya—and was thus unconstitutional for the same reasons. The newest member of the Court, Justice Kavanaugh, dissented, arguing first that the statute at issue in this case is functionally and textually distinct from those at issue in Johnson and Dimaya, saving it from the “typical case” approach that was so problematic in those two cases, and second that, even if the functional and textual arguments were not on the government’s side, the Court should still reject the “typical case” approach here as a matter of constitutional avoidance. Justice Kennedy and Thomas joined the dissent. Chief Justice Roberts joined as to all but the last part: constitutional avoidance did not in his view require saving the statute if the functional and textual arguments were not on its side.
So how do the individual justices line up? The Court that was 6-3 in Johnson (Scalia, Roberts, Ginsburg, Breyer, Sotomayor, and Kagan versus Kennedy, Thomas, and Alito) was 5-4 in Dimaya (Gorsuch, Ginsburg, Breyer, Sotomayor, and Kagan versus Kennedy, Thomas, Alito, and Roberts) and 5-4 again in Davis (Gorsuch, Ginsburg, Breyer, Sotomayor, and Kagan versus Kanavaugh, Thomas, Alito, and Roberts). True, Justice Gorsuch took Justice Scalia’s place between the first two cases and Justice Kavanaugh took Justice Kennedy’s place between the second two cases, but the balance of votes did not change because both justices voted as their predecessor had. Only Chief Justice Roberts seems to have “switched sides,” though really that characterization is unfair. Assuming we take him at face value, it would be fairer to say that only Chief Justice Roberts saw the relevant statutes as distinct enough to merit different votes.
What should we make of all this? I think there are a few lessons.
First, there is stable coalition of five to six justices willing to strike down major federal criminal statutes—statutes affecting tens of thousands of cases—on vagueness grounds: Justice Gorsuch, the Court’s four liberal justices, and, depending on the statute, Chief Justice Roberts. Furthermore, those five to six justices have produced a substantial and noticeable enough body of precedent to influence lower federal courts and state courts. We are going to see a boom in vagueness challenges to criminal statutes and, likely, certain categories of civil statutes as well.
Second, the Court’s lineup of justices totally explodes the simplistic idea that conservative justices are pro-state in criminal cases. That is simply not true of Justice Gorsuch and Chief Justice Roberts, to say nothing of whether it is fair to Justices Kavanaugh, Alito, and Thomas. My view, based on cases other than vagueness, is that both sides of the equation are wrong: when it comes to substantive criminal law, it is neither true that conservative justices are consistently pro-state nor that liberal justices are consistently pro-defendant. The evidence at hand admittedly does not go that far, but it does speak to half the point.
Third, Chief Justice Roberts’s statutory approach in the vagueness cases is consistent with his approach in other substantive criminal law cases (e.g., Bond (2014) and McFadden (2015)). One of the more peculiar hallmarks of contemporary textualism is the tendency, not just to attend closely to statutory text, but to focus on very short provisions of statutory text and apply those provisions in as rule-like a way as possible. Roberts resists that tendency, at least in substantive criminal law. He takes the whole statute rule very seriously, focuses on statutory context, and is more a splitter than a lumper. It is thus not surprising that, in these vagueness cases, he sees distinctions between bits of statutory text that strike other judges as isomorphic. He is not really a textualist, as that term is conventionally used. He may not be a purposivist either. One could call him a “contextualist.”
Finally, the vagueness revolution may not be benign—at least, not for those who, like me, think American criminal justice is too often unreasonably punitive. Vagueness sets the Constitution against law that is open to interpretation. It favors rules over standards—the more mechanical, the better. It takes no law off the table for being over-harsh: it only requires that the law’s harshness be clear, which can often be accomplished by making the law’s rules more sweeping. Consider, for example, how Congress might fix the residual clause in Johnson. Rather than “otherwise involves conduct that presents a serious potential risk of physical injury to another,” Congress could write, “otherwise involves conduct that, based on the facts underlying the offense, presented any risk of physical injury to another.” That would be a clearer but more punitive and, in my view, less reasonable law. What about the residual clause in Davis and Dimaya, defining a crime of violence as a felony offense that “by its nature, involves a substantial risk that physical force against the person or property of another may be used in the course of committing the offense”? Just drop “by its nature” and write that a crime of violence is “any felony offense that presented any risk of physical injury to another.” My own view is that mechanical rules are generally bad for criminal defendants and bad for criminal justice. It remains to be seen what will come from the Supreme Court’s recent vagueness revolution, but I am skeptical that it will produce the sort of change criminal justice reformers would hope for.