The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Legal Doctrine and Conventional Criminal Law

Legislatures and courts should assist the development and enforcement of beneficial legal conventions.


In previous posts, I have argued that legal conventions form a de facto common law and that this de facto common law remedies many defects in our statutory criminal law system. Despite these criminal law conventions, however, significant overcriminalization problems remain. Broadly, overcriminalization problems occur in areas of criminal law in which (1) there do not exist any legal conventions, (2) there exist legal conventions, but we have difficulty controlling isolated defections from them, and (3) traditional methods of enforcing legal conventions and controlling prosecutorial discretion substantially break down.

Let me start with the first. Much of the worst overcriminalization problems involve excessive harshness against defendants who are, in fact, guilty of violating some community norm. But we do not have many legal conventions governing plea bargaining and sentencing. Individual judges and prosecutors may have "going rates" for plea bargains. But these rates are often arbitrarily set, not public, and sometimes primarily known only by insiders of that particular courthouse. "Going rates" are not true legal conventions because, even if they are special customs, they are not generally accepted as binding and they are not enforceable through nonlegal sanctions.

Recidivist statutes compound the problems with plea bargaining. Prosecutors often use these provisions to extort guilty pleas, and they follow through with their threats to charge these provisions when defendants will not plead. Moreover, based on their views of what is just, different prosecutors may (or may not) charge such provisions, creating arbitrariness across jurisdictions and, in some cases, within any individual jurisdiction. The breadth of these recidivist provisions often ensnares those for whom the mandatory minimum penalty seems disproportionate, such as defendants who have stolen golf clubs or possessed a single round of ammunition.

A second problem with conventional criminal law is that criminal conventions, unlike constitutional ones, are easily susceptible to isolated deviations. Individual deviations are often low-information events that affect marginalized members of the community. The community may care a great deal if a town's police officers routinely ticket residents who exceed the speed limit by any margin. The community probably does not care if its officers use de minimis speeding as a pretext to detain drivers suspected of drug violations or other crimes. Overbroad statutory law facilitates arbitrary enforcement against isolated individuals in a manner that would not be tolerated if the law were applied in the same way to the wider community.

Isolated deviations create other serious rule-of-law problems. Rudy Giuliani infamously instituted "federal day" for drug crimes in New York City. Ordinarily conventional rules determine which drug crimes are prosecuted in state court (e.g., possession crimes related to personal use) and which are handled in federal court (e.g., interstate and foreign trafficking). But on federal day, which was a random day rotated from week to week, Giuliani had local drug cases diverted to federal court, which imposed more serious penalties. General principles of distributive justice require treating like cases alike, and it is extraordinarily arbitrary to increase someone's penalty because he was caught on Monday rather than on Tuesday.

In a conventional system, the harm from this kind of arbitrary enforcement runs even deeper than just violating norms of distributive justice. Unusual prosecutions unsettle the community's shared understanding of the law and, correlatively, they interfere with the basic requirements of notice, clarity, and prospectivity that underlie the rule of law. An unwritten common law or conventional system can still respect these rule of law values—but only if the law is actually enforced in ways that track widely understood social norms and customs.

A third problem is that, in many cases, traditional legal checks that curb prosecutorial discretion break down. In many misdemeanor cases, defendants lack the right to counsel and to trial by jury. Many poor defendants, moreover, lack adequate money to post bail, leading them to languish in pretrial detention. Prosecuting weak or trivial cases "in the shadow of the jury" looks quite different when the prosecutor knows that the defendant cannot afford to sit in jail for months while he awaits a jury trial.

Because penalties are often small, misdemeanors and traffic offenses may not seem like a huge deal; but they are. Convictions can have lingering effects as defendants face burdens of paying off fines, complying with probation, and getting transportation in the face of suspended driver's licenses. Additionally, more arrests and detentions for small matters lead to more felony arrests, as police use such offenses as pretexts to fish for more serious wrongdoing. Uneven enforcement of misdemeanors and traffic laws leads to uneven enforcement of felonies.

In the article, I offer a series of suggested policy proposals in problematic areas of criminal law. Few are novel. But I want to suggest a different theoretical grounding for them. To continue fixing the defects in statutory criminal law, we need to develop legal conventions. And to do that, we need to provide mechanisms to sanction violations of those conventions.

For example, one option to cabin prosecutorial discretion over the sentencing of guilty defendants is to further empower juries. Judges often recoil at informing juries about a defendant's sentencing exposure. And under Almendarez-Torres, juries do not have to find the fact of a previous conviction, even if it leads to extraordinary high sentencing enhancements (e.g., the Armed Career Criminal Act, which elevates a felon-in-possession gun crime from a maximum of ten years' imprisonment to a minimum of 15 years and a maximum of life for those with three or more qualifying convictions).

Faced with a defendant who unlawfully possessed a single bullet and with an understanding of the sentencing consequences, a jury might return a conviction on simple felon in possession, while acquitting on the recidivist charge. Judges often argue that such proposals invite lawless juries. But this may be the wrong way to look at it. In these cases, the community, through the jury, polices whether the prosecutor has acted lawlessly by overcharging individual cases—that is, by charging them outside of widely understood norms about the cases to which such laws should apply.

Another sentencing reform is to have greater transparency in plea bargains. Prosecutors should have to publish their "going rates" for pleas. In a conventional system, information transparency confers several advantages. Among them, publication of information will deter prosecutors from deviating in isolated cases. Disclosure also subjects the information to public scrutiny, including by voters. This facilitates political sanctions against prosecutors who act unreasonably as defined by community norms.

More broadly, increasing transparency in law enforcement may help curtail arbitrary deviations from widely understood criminal law customs. For example, it would be helpful to know the details of the 40 people ticketed in Virginia in 2018 for going fewer than five miles per hour over the limit. Were the drivers engaged in significant speeding but the tickets reduced by the officer? Were drivers going at an unsafe speed through a school zone? Or were these pretextual stops or individuals whom the police arbitrarily selected? Having this information facilitates the application of political pressure against improper enforcement practices.

I am more skeptical of the ability of legal conventions to fix all the problems that overbroad criminal law has on criminal procedure and civil redress. Overbroad statutory law narrows the effective scope of the exclusionary rule because officers may justify detentions and arrests based on objective probable cause that any offense has been committed, no matter how minor or how customary it is for individuals to violate it. Relatedly, overbroad statutory law limits civil redress because officers who have probable cause of any legal violation will have a defense for false imprisonment or malicious prosecution either under state law or when raised as an unreasonable search or seizure under 42 U.S.C. § 1983.

But judges may ameliorate some of these problems by acknowledging that criminal law conventions exist and by facilitating their enforcement. For example, judges may require officers to have subjective good faith when they enforce laws that customarily go unenforced. In Whren v. United States, the Supreme Court said that they were "aware of no principle that would allow [them] to decide at what point a code of law becomes so expansive and so commonly violated that infraction itself can no longer be the ordinary measure of the lawfulness of enforcement." But judges, no less than other members of the community, have general knowledge about customary norms of how laws are enforced. Officers who pull drivers going one mile per hour over the speed limit as a pretext to search are not enforcing the law in accordance with traditionally understood community norms.

The development and recognition of criminal law conventions will not fix every problem of statutory overcriminalization. But in a system like ours in which legislating takes place under nonideal conditions, we primarily improve our criminal law by developing and improving our unwritten customs and traditions.