The purpose of criminal law is to identify conduct that threatens the fabric of civil society and to make clear that such conduct will not be tolerated. The kind of conduct that makes civil society impossible falls into several categories that are familiar to most libertarians: (a) using unjustified force against others, (b) taking other people’s property by force or fraud, and (c) interfering with other people’s ability to engage in conduct that does not violate anyone else’s rights. In essence, a sound criminal code seeks to express and implement John Stuart Mill’s harm principle, which provides, “The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.”¹

A well-functioning criminal justice system seeks to discourage people from violating the relatively few prohibitions that are truly necessary to protect the fabric of civil society. But instead of focusing on the relatively few, genuinely anti-social criminals, our system outlaws a vast array of nonwrongful conduct and thus saddles itself with the responsibility of dealing with an enormous number of “transgressors,” only a fraction of whom are actual criminals.

Furthermore, by outlawing activity that is not truly wrongful, the government itself actually creates crime, both by designating people who continue engaging in the proscribed conduct as criminals and oftentimes, as with drug prohibition, by ensuring the existence of one of the most powerful generators of criminal activity known to man: black markets. As a result of these misguided and illiberal policies, the government ends up creating a great amount of crime that would likely not otherwise occur and designating as “criminals” a huge number of decent, peaceful, and otherwise law-abiding people. Faced with the ensuing deluge of transgressors, the government responds by ditching many of our system’s most important protective mechanisms and replacing them with more efficient—but often wildly unjust—workarounds.

To appreciate just how thoroughly hacked and unjust our system has become, imagine yourself in the following situation.

You are a young black man driving back to Denver from a friend’s farmhouse just outside Boulder, Colorado. Besides a small vegetable garden, your friend also grows several hundred marijuana plants for a nearby medical cannabis dispensary, which is perfectly legal under state law. For a belated birthday gift, your friend offers you a dozen marijuana plants, which you accept and place in the trunk of your car. On the way home, you pass an aggressively nondescript gray sedan with tinted windows and multiple antennae. The sedan pulls out and follows you down the highway for several miles. Looking back in the rearview mirror, you see blue lights flashing from behind the sedan’s grill. You are being pulled over, but you have no idea why. Here’s what can happen next—and every bit of it is considered perfectly legal by the U.S. Supreme Court.

The police officer approaches the driver’s side window and asks if you know why he pulled you over. You don’t, but you’re smart enough not to guess. The officer says he decided to pull you over for “driving while black,” so he followed you until he saw you commit a traffic violation—namely, driving one mile per hour over the speed limit. The officer asks for consent to search your car; you decline. He calls for a canine unit, which arrives just as the officer is handing you a warning for exceeding the speed limit. The officer walks the drug-sniffing dog around your car and it “alerts” near your trunk. The officer takes you out of your car, handcuffs your arms behind your back, and places you in the back seat of his unmarked police car. He pops your trunk, sees the marijuana plants, and advises you that you are now under arrest.

“But marijuana’s legal in Colorado,” you protest. “Furthermore, you just admitted you decided to stop me for ‘driving while black,’ which is unconstitutional, so any evidence you found as a result of the stop is fruit of the poisonous tree and therefore inadmissible.” (It turns out you’ve watched a lot of Law & Order.)

The officer responds, “Well, I’m part of a federal task force, so I’m authorized to enforce federal laws, and marijuana possession is still a federal crime. As for pulling you over for ‘driving while black,’ the Supreme Court has held that purely pretextual traffic stops do not violate the Fourth Amendment, which means it’s irrelevant why I actually pulled you over. The fact that I could have pulled you over for driving one mile an hour over the speed limit means the stop is valid under the Fourth Amendment and the evidence is admissible. Nice try, though.”

At this point, you play your last card. “Ah, but this marijuana was neither bought nor sold—it was given to me as a gift—and it was grown right here in Colorado and never crossed a state line. So the federal government lacks the authority to criminalize my possession of these plants under the Commerce Clause—or any other constitutionally enumerated power.”

“Very impressive,” the officer says. “But you see, the Supreme Court’s got my back there too. A 1995 case called Gonzales v. Raich held that Congress may criminalize the purely intrastate, noncommercial distribution of marijuana using its power to regulate commerce among the states. Now I’ll grant you that’s not the most intuitive reading of that particular passage, but let me ask you a question: Are you starting to notice a pattern here at all? Because I’ll let you in on a little secret, partner—this is how the system works and this is how it’s going to be from now until you decide what charges you want to plead guilty to. Trust me, things will go a lot smoother for you if you just sit back and accept the inevitable. All right, my friend—let’s go ahead and get you processed.”

Of course, this is a stylized and somewhat absurd hypothetical. But it captures the essence of the system from a defendant’s standpoint. The constitutional deck is stacked against you, the playing field is persistently tilted in favor of the government, and, for most defendants, the outcome is all but inevitable. At the risk of stating the obvious, ours is not remotely a libertarian criminal justice system. But is there any such thing, or are those two concepts—libertarian and criminal justice—mutually exclusive?

As explained below, most people—including most libertarians—believe some kind of criminal justice system is necessary to protect peaceful members of society from persistent rights violators. But the goals of a truly just and well-functioning system would be much more modest than those of our present system, and such a system would restore several features that we have largely eliminated, including: (a) robust protection for substantive and procedural rights, (b) strong accountability for those charged with enforcing the laws, and (c) high levels of citizen participation in the administration of criminal justice. After describing several of our current system’s most serious pathologies, I will offer some ideas for how we could change the system to more faithfully reflect our historical commitment to justice, liberty, and limited government.

The first problem to consider is the proliferation of illegitimate legal prohibitions, or what we might call “unconstitutional overcriminalization.” The scope and substance of criminal law are of course not set in stone. No Platonic ideal criminal code outlaws precisely the right sorts of conduct while leaving people free to do everything not specifically proscribed. Thus, creating a criminal code—that is, a set of laws dictating what people may and may not do and providing punishments for those who transgress—is an inherently political endeavor. But because we live in a constitutional republic rather than a pure democracy, the government is limited in what it may proscribe.

Some of those limits—such as laws censoring speech—are familiar and fairly well-theorized by the courts, whereas other constitutional limits on government power are either ignored outright by the courts or so undertheorized that they are effectively rendered meaningless. In addition to the judiciary’s failure to develop a philosophically serious theory of criminal sanction, strong incentives drive legislators to create far more prohibitions than are desirable or defensible, resulting in a recipe for overcriminalization.

Clearly, from a purely philosophical standpoint, one must have a good reason to interfere with another person’s freedom of action, and to initiate force against another person without sufficient justification is immoral. What is less clear—at least as far as the Supreme Court and many leading constitutional theorists are concerned—is whether this basic moral obligation applies equally to government. In other words, must government have a good reason to lock people up in cages, or may it do so more or less on a whim? Unfortunately, the Supreme Court has answered this vital question with what amounts to an institutional shrug. Accordingly, the courts have come to treat the arrest and incarceration of citizens as a fairly trivial imposition on their liberty that government may—except in a few settings—impose for no particular reason at all.

As already noted, powerful dynamics cause government officials to criminalize far more conduct than is socially optimal or morally defensible.² That’s because when voters feel anxious or outraged about something—whether it’s the opioid epidemic or a bubble-driven financial crisis—legislators have two basic choices. Either they can do the hard work of studying the problem to figure out what actually caused it and determine what, if anything, government can realistically do to make things better. Or they can simply pass a law making it illegal to do whatever it is that people think may have caused the problem and start locking people up as a means of assuring constituents that the problem is being addressed—whether or not it really is.

Take marijuana, for example. For most of our history, the cultivation or consumption of cannabis was not illegal; in fact, cannabis was historically used in various medicines, such as cough syrup. But, eventually, marijuana use became associated with certain countercultures, and the white establishment came to fear, then demonize, it. But now the pendulum is swinging back, and after a half century or so of severely punishing people for possessing and distributing marijuana, most states have either partly or wholly decriminalized it. The federal government has not followed suit, however, and the penalties under federal law for cultivating or distributing marijuana are severe.³

But as more than 60 percent of Americans now realize, those laws exist for no good reason; in fact, the decision to criminalize marijuana—but not far deadlier substances such as alcohol and tobacco—is essentially arbitrary. Yet judges not only have upheld the constitutionality of criminalizing marijuana—even when prescribed by doctors to desperately ill people for whom no other treatment has worked—they also continue to participate in the enforcement of those laws by imposing morally indefensible prison sentences on people who violate them. That, in a nutshell, is the face of unconstitutional overcriminalization.

America’s criminal justice system was designed to deal with the relatively small number of genuinely anti-social people who exist in a generally healthy and well-functioning society at any given time. But because our policymakers have chosen to criminalize vast amounts of nonwrongful conduct that many perfectly decent, otherwise law-abiding citizens wish to engage in—and will continue to engage in despite the prohibition—our criminal justice system finds itself completely overwhelmed with a task it was never designed for, namely, identifying, arresting, prosecuting, and punishing a vast number of mere transgressors, only a fraction of whom are actual criminals.

It would be impossible to document in a single chapter the myriad ways in which our system cuts corners and abandons painstakingly prescribed constitutional procedures for adjudicating criminal charges. But the essence of the issue is reflected in a single, astonishing statistic: some 95 percent of criminal convictions today are obtained by guilty pleas instead of by jury trials. As Orwell, Dostoevsky, and countless others have observed, many regimes throughout history have excelled in extracting confessions of guilt from their citizens for crimes real or imagined. Unfortunately, America has become a leading member of that ignominious club. We call this process “plea bargaining,” but that is really a misnomer—what we really have is a system of coerced confessions that has almost completely supplanted the criminal jury trial and transformed America’s criminal justice system into little more than a conviction machine.

Article III of the U.S. Constitution provides that the trial of all crimes shall be by jury. So central was the jury trial to the Founders’ vision of just government that it is the only right mentioned in both the original Constitution and the Bill of Rights. Not coincidentally, the Bill of Rights devotes more words to the subject of juries than any other, because the one thing that virtually every leading thinker of the American Founding appears to have agreed on was the critical role of juries in limiting the power of government.

Alexander Hamilton noted that the Federalists and the Anti-Federalists, “if they agree in nothing else, concur at least in the value they set upon the trial by jury; or if there is any difference between them it consists in this: the former regard it as a valuable safeguard to liberty; the latter represent it as the very palladium of free government.”⁴ Likewise, Thomas Jefferson considered “trial by jury as the only anchor ever yet imagined by man, by which a government can be held to the principles of its constitution.”⁵

Notwithstanding the ineluctably clear command of the Constitution, the jury trial is now practically extinct in America. As noted, more than 95 percent (and more than 97 percent in the federal system!) of criminal convictions today are obtained through plea bargains, using a process that would certainly have alarmed and astonished the Founders.

The story of how a practice that was unknown at the Founding and viewed with great suspicion and even disdain by 19th-century American and English jurists came to supplant the constitutionally prescribed method for adjudicating criminal charges is a fascinating one.⁶ But for present purposes, it is enough to briefly describe the two dynamics that transformed American criminal justice from a primarily adversarial system into a primarily transactional one. The first dynamic is coercive plea bargaining; the second is a radical devaluation of the jury trial compared with its role at the time of the Founding. These will be discussed in turn.

One of the most powerful levers prosecutors have is pretrial detention. Simply put, those who are locked up awaiting trial are cut off from family, friends, and other sources of support. They will have a much more difficult time securing and working with counsel to prepare their defense. And given the dirty, dangerous, and generally miserable conditions of so many of the nation’s jails, defendants will likely entertain just about any offer that promises either an immediate release or a transfer to a more sanitary and professionally run prison, which is where people who have been convicted of crimes typically serve their sentences.

Another powerful lever is prosecutors’ ability to stack charges against a defendant—including charges that carry a mandatory minimum prison sentence—and then offer to drop some of those charges or recommend leniency to the judge at sentencing. Other countries, such as England, impose strict limits on the differential between the length of sentence prosecutors may threaten defendants with if they go to trial and lose versus the discounted sentence prosecutors may offer if defendants confess. However, American prosecutors have unfettered discretion to make draconian threats while offering a comparatively trivial punishment if the defendant will simply agree to plead guilty.⁷

To appreciate just how much pressure prosecutors can bring to bear, consider the case of young computer genius and internet pioneer Aaron Swartz. Swartz was arrested for breaking into a computer closet at the Massachusetts Institute of Technology and connecting a laptop carrying a program he had designed to download articles from the JSTOR academic database. Despite the fact that Swartz caused no permanent damage to any of the systems, and despite JSTOR’s disavowal of any “interest in this becoming an ongoing legal matter” once its data were secured, federal prosecutors hit Swartz with a 13-count federal indictment that threatened him with 35 years in prison and a $1 million fine. They offered him six months if he agreed to plead guilty.

Tragically, the pressure was so great and his mental state so fragile that Swartz killed himself during plea negotiations. Outrageous and inhumane as they may seem to the casual observer, clearly nothing was particularly unusual about the tactics used in Swartz’s case: U.S. Attorney Carmen Ortiz defended the conduct of her prosecutors as “reasonable” and fully consistent with Department of Justice policy.

The third and fourth major levers of coercive plea bargaining are the complexity of the system and the persistent inadequacy of public defender services. On the one hand, criminal procedure has become so complex that ordinary laypersons cannot simply represent themselves effectively in a criminal trial as was the customary practice at the time of the Founding. That means all criminal defendants, and certainly defendants in felony cases, are well advised to retain counsel. But lawyers—especially good ones—are notoriously expensive, and the vast majority of criminal defendants simply cannot afford one. Accordingly, some 80 percent of criminal defendants are represented by government-furnished counsel—either a public defender or a court-appointed private practitioner who will typically receive a fixed fee, often one that is set at some archaic and shockingly low amount, for his or her services.

On the other hand, although some public defender offices are quite good, many are not, and virtually all are desperately overworked.⁸ The problems with public defenders are well-known and well-documented, and they include a persistent lack of funding that translates into crushing caseloads and inadequate support services, such as investigators and expert witnesses.

Another problem with public defenders is the fact that they are repeat players who carry dozens or even hundreds of cases simultaneously, often against the same small set of prosecutors. Because taking all of those cases to trial would be impossible, a public defender knows that most of the cases will have to plead out. This approach results in a potential conflict of interest, because public defenders have a strong incentive to maintain sufficiently good relations with the prosecutors to ensure favorable plea offers for all of their clients, even if such an approach may result in less than fully zealous advocacy for particular defendants who might benefit from a more aggressive defense.⁹

As suggested earlier, another dynamic is at work here besides coercion: a radical devaluation of the criminal jury trial as an institution. To understand what happened, one must appreciate that there are two distinct ways to conceive the role of a criminal jury. The first is to think of it as a purely fact-finding body whose sole purpose is to increase the likelihood of a factually accurate verdict: Did this defendant sell a prohibited substance to that undercover officer—yes or no? The second is to think of the jury as playing at least one other, perhaps more important, role in our system of criminal justice: to check the abuse of government power by refusing to allow unjust convictions. The former conception is a relatively recent invention and is essentially mythical; the latter reflects the true role of juries in our system and the consistent understanding of their function both at the American Founding and through more than 1,000 years of our common-law history.

Perhaps not surprisingly, the government transformed both the understanding and the role of juries in our system. That fascinating history is presented in compelling detail in the book titled Jury Nullification: The Evolution of a Doctrine.¹⁰ That transformation boils down to a successful campaign by prosecutors and judges to undermine the time-honored institution of conscientious acquittal (often referred to pejoratively and less accurately as “jury nullification”) and to deprive jurors of information that might incline them to embrace it in a particular case. Such information might include the punishment the defendant will receive if convicted or the substance of any plea offers made to the defendant, like the six months prosecutors offered Aaron Swartz after threatening him with 35 years in prison.

As a result, whereas Founding-era juries generally knew what punishment the defendant would receive if convicted and understood their time-honored right—even duty—to acquit factually guilty defendants to prevent injustices, modern juries are carefully vetted by prosecutors and supervised by judges to ensure that they remain ignorant of both points. Consequently, “the jury today is, in the words of [Yale law professor] Akhil Amar, a ‘shadow of its former self.’ ”¹¹

Thus, the profoundly pathological combination of unconstitutional overcriminalization, coercive plea bargaining, inadequate representation, and neutered juries has radically transformed our criminal justice system. Instead of focusing on genuine criminals and proving their guilt beyond a reasonable doubt in an open and adversarial system, prosecutors spend most of their time racking up convictions by extracting confessions from the guilty and the innocent alike.¹²

So what would a well-functioning and truly just criminal justice system look like and how would it differ from ours?

First and foremost, it would recognize the criminal sanction for what it is: a public policy tool of last resort that should be deployed only when the failure to do so would threaten the very fabric of civil society and prompt private citizens to take matters into their own hands in the form of vigilante justice. Second, it would restore citizen participation to its rightful place at the heart of the criminal justice system, both by ensuring that all or most convictions are obtained through jury trials rather than guilty pleas and by enabling civil juries to hold police and prosecutors liable for abusing their power. Third, it would seek to level the playing field between citizens and law enforcement to ensure that the adjudication of criminal charges is a genuinely fair and adversarial process. And finally, it would impose appropriate constraints on specially empowered actors within the system, particularly police officers and prosecutors, to ensure that they are properly incentivized to avoid abusing their authority and properly accountable to their victims when they do. Here’s how this might look in practice.

The first line of defense against overcriminalization would be a judiciary that properly interprets the Constitution as requiring the government to have a very good reason for putting people in cages. As discussed, that would be a radical departure from current jurisprudence, which requires the government to provide neither a good nor even an honest explanation for enforcing most criminal laws. A properly robust standard of judicial review would strike down a substantial number of existing criminal laws, including many, if not all, drug prohibitions. Ending the so-called war on drugs would reduce crime and enable police and prosecutors to devote their attention to actual, rights-violating crimes such as homicide, rape, robbery, and theft. The second major reform would be to vastly reduce the prevalence of plea bargaining by eliminating (or at least substantially constraining) several of the most coercive levers in the prosecutorial toolbox. As previously discussed, those levers include pretrial detention, inadequate legal representation, and the so-called trial penalty, that is, the often massive differential between what punishment prosecutors can threaten defendants with and what they can offer to induce a plea.

Such reform would result in pretrial detention becoming the exception, not the rule. Inadequate legal representation would be addressed by requiring roughly equal funding for defense counsel as for prosecutors and also by allowing defendants to hire counsel of their own choosing using government-funded vouchers instead of simply assigning defendants a government-employed public defender. The trial penalty would be addressed by imposing a hard cap of, say, 10 to 20 percent on the discount between the sentence threatened in an indictment and the sentence offered in plea bargaining.

Another key reform in this area would be a prohibition on bribing witnesses. Bribing witnesses is, of course, already illegal, and indeed a federal law makes it a felony to offer “anything of value to any person for or because of testimony” to be given by that person at any trial or hearing.¹³ Incredibly, however, that statute has been interpreted by the courts as inapplicable to a single group of lawyers (prosecutors) in a single type of judicial proceeding (a criminal prosecution).¹⁴ As a result, prosecutors may—and routinely do—pay for witness testimony using not just material inducements but also the one thing more valuable than money: freedom in the form of reduced or dismissed charges. The government’s use of “snitches” has had a predictable and well-documented corrupting influence on criminal prosecutions and has led to a considerable number of false convictions. Compelling reasons exist for making it illegal to bribe witnesses, and those reasons apply equally to prosecutors.

The third major reform we should implement is to restore citizen participation in the administration of criminal justice. That reform would in turn require two distinct changes to current policy. The first change involves restoring the criminal jury to its Founding-era role as not merely a finder of fact but also a preventer of injustice. The second policy change would be to eliminate various immunity doctrines that enable government agents—including particularly police and prosecutors—to avoid civil liability for abusing their authority.

To reiterate, a crucial distinction between Founding-era juries and modern juries is that the former typically understood what punishment defendants would receive if convicted, whereas the latter generally do not. Indeed, prosecutors now go to extraordinary lengths to prevent jurors from learning what will happen to defendants if they are convicted, including not just the length of the prison sentence, but also collateral consequences, such as loss of various civil rights or deportation for noncitizens.

Another important distinction is that Founding-era jurors would likely have been both familiar with and supportive of “jury nullification,” more accurately referred to as “jury independence” or “conscientious acquittal.” All of those terms refer to a jury’s decision not to convict a defendant whose factual guilt has been proved beyond a reasonable doubt when the jury believes it would be unjust or otherwise inappropriate to do so. This situation might include defendants who have been charged with conduct that jurors believe should not be illegal in the first place, such as simple marijuana possession; defendants who have been charged with a crime that should be punished, but not as harshly as the government proposes; and defendants who appear to have been unfairly singled out or otherwise targeted by the government for inappropriate reasons, such as retaliation for political dissent.¹⁵

But while conscientious acquittal has a long and storied history in the Anglo-American legal tradition that predates even the Magna Carta, it is subject to various misconceptions and caricatures, including the idea that it empowers jurors to “nullify” democratically enacted laws. In fact, jurors have no such power, and a moment’s reflection shows how nonsensical the concept is. Unlike judges, jurors are not called upon to assess the constitutionality of laws and there is no mechanism by which they can strike down, or “nullify,” any piece of legislation. Instead, the only power a jury has is to acquit a particular defendant of particular charges under the circumstances of a particular case. If a mother is charged with marijuana possession and child endangerment for administering cannabinoids to her epileptic son at the direction of his pediatrician and a jury acquits her, has that jury “nullified” laws against child endangerment or marijuana possession? Plainly not, and it is preposterous to suggest otherwise. Indeed, some or all of those same jurors might well vote to convict a different mother who fed her healthy child a marijuana-laced brownie “just for the fun of it.” Simply put, conscientious acquittal (aka “jury nullification,” a misnomer) is not about undermining duly enacted laws; rather, it is about checking the government’s abuse of power when the government seeks to enforce those laws in a manifestly unjust way.

Unfortunately, the current status of conscientious acquittal in American courts remains unclear. On the one hand, the Supreme Court has held that criminal defendants are not entitled to have jurors instructed that they are permitted to acquit the defendant for any reason, including their belief that it would be unjust to convict a defendant whom they believe to be factually guilty.¹⁶ On the other hand, it remains an open question in most jurisdictions whether the trial judge has the discretion to advise jurors about their power to engage in conscientious acquittal if the defendant requests such an instruction.

A related question is whether individuals who are not involved in the case may provide jurors with general information about conscientious acquittal as well as specific information about the particular case on which they are serving. Federal and state laws prohibit jury tampering, but those laws are not models of precision, and their wording leaves significant doubt about which communications are covered and which are not. As a result, people have been prosecuted for activity as prosaic as simply handing out jury nullification literature outside courthouses. And most prosecutors believe it is illegal to share case-specific information, such as probable sentence length or the substance of any plea offers made to the defendant by the prosecution before trial.

But even assuming such information is covered by existing jury-tampering statutes, the question remains whether communicating it to an empaneled juror (or anyone else for that matter) might be protected by the First Amendment. The short answer is yes, because dictating to people what they may say to empaneled jurors (cautioning them to arrive early at the courthouse to secure parking, for example, which is fine, versus advising them that the defendant in their case is facing a 25-year mandatory minimum sentence for having a gun in his car during a drug deal, which is not fine) is a classic content-based restriction on expression and as such must meet the highest level of judicial review: strict scrutiny.¹⁷ To satisfy strict scrutiny, the government must demonstrate that it has a compelling interest in restricting the speech in question, and that the means chosen to restrict the speech are narrowly tailored to that interest.

That poses a fascinating question: Does the government have a compelling interest in ensuring that no one communicates to jurors regarding such public—and publicly available—information as what punishment the defendant will receive (or what the likely range of punishments will be) if they convict and the fact that American courts have repeatedly acknowledged the legitimacy of conscientious acquittal even if they have not actually embraced it as something they wish to see jurors actually contemplate in specific cases? To ask that question is to answer it: the government has no interest in keeping jurors ignorant of most publicly available information about the cases they are participating in, and certainly not when the only reason for doing so is to prop up an essentially mythical vision of the criminal jury as a purely fact-finding body. And although the legal doctrine in this area is quite thin—not to mention undertheorized and even glib—the Cato Institute has launched a strategic amicus campaign designed to vindicate people’s right to educate jurors about their historic role in checking the abuse of government power—including the pervasive unconstitutional overcriminalization and coercive plea bargaining that have largely stripped our criminal justice system of both its efficacy and its legitimacy—and to equip them with information relevant to that duty.

A key question is how prosecutorial behavior might change with the advent of properly informed juries. The short answer is that we can’t say for sure because it is a practice that has not been tried in living memory. But we can surmise that prosecutors would have to exercise far greater caution, both in their charging decisions and in their conduct of plea negotiations, lest they incur the ire of a jury that has come to believe the prosecutor has exercised her authority in bad faith in a transparent effort to coerce the defendant into pleading guilty. Some prosecutors are fond of saying, “If you can’t stand the time, don’t do the crime,” to which the defendants who enjoy the protections of a properly informed jury might respond, “If you can’t toe the line, don’t threaten the time.”

The second jury-related policy change is to restore the ability of citizens to impose civil liability on public officials who abuse their power by violating people’s rights. In fact, a federal law on the books (42 U.S.C. § 1983, colloquially referred to as “Section 1983”) seeks to do precisely that by creating a cause of action for people who suffer the “deprivation of any rights” at the hands of any state or local official. Unfortunately, the Supreme Court has largely gutted Section 1983 by inventing a variety of immunity doctrines that make it difficult—and in some cases impossible—to hold government actors liable for their misconduct.

The three cornerstones of this near-zero-accountability policy for law enforcement are qualified immunity, absolute prosecutorial immunity, and the elimination of employer liability for government agencies such as police departments and district attorneys’ offices.

The first of those policies, qualified immunity, radically narrows the scope of Section 1983 liability by effectively inserting the words “clearly established” before “any right.” That means that to bring a lawsuit against a public official, would-be plaintiffs must show not only that the conduct in question violated their rights (as the literal language of the statute provides), but also that a preexisting case in the relevant jurisdiction had sufficiently similar facts that no police officer or other public official could have had any doubt that the particular act in question—whether it be ramming a car off the highway, throwing an unarmed woman to the ground for walking away from a voluntary interview with police, or conducting a strip search of a child without parental consent—is impermissible. And if there is no case directly on point, then the citizen cannot sue, and, perhaps equally important, a jury will have no opportunity to convey to the relevant official the community’s sentiment about the propriety or culpability of the act in question.

The next accountability-destroying doctrine is called absolute prosecutorial immunity. Like qualified immunity, it was invented out of whole cloth by the Supreme Court and was effectively appended to the text of Section 1983, which makes no special provision or exemption for prosecutors (or any other government official for that matter). Absolute prosecutorial immunity is precisely what it sounds like: a complete bar on suing prosecutors for anything they do within the scope of their job as a prosecutor. Incredibly, that includes deliberate acts of malfeasance, such as knowingly and even vindictively prosecuting an innocent person and even suborning perjury to help secure an unjust conviction.

The Supreme Court’s twin rationales for absolute prosecutorial immunity are both risible and empirically unfounded: first, the Court asserts, in the face of voluminous evidence to the contrary, that other mechanisms besides civil liability suffice to hold prosecutors accountable for their misconduct. And second, the Court fears, again without evidence, that prosecutors would be unduly distracted if they could be sued by people whose rights they are claimed to have violated in the course of a criminal proceeding—even to the point that some people might choose not to become prosecutors if, like every other professional, they could be sued for injuries they cause through their negligent or willful misconduct.

Finally, the Supreme Court has decided that the standard common-law doctrine of employer liability, or respondeat superior, should not apply to police departments and other public employers. Thus, if a Domino’s Pizza driver smashes into your car in an act of road rage while delivering a pizza, you can sue Domino’s for any injuries its driver causes you while acting in the course and scope of his employment—in this case, delivering a pizza. By contrast, if a police officer runs you off the road for no good reason, you cannot sue the officer’s employer, the police department he works for. Again, the text of Section 1983 contains no basis for this so-called Monell doctrine¹⁸—it is simply another special privilege invented out of whole cloth by the Supreme Court to help shield government officials and agencies from accountability.

A libertarian criminal justice system would eliminate all of these immunity doctrines because they are unjust and counterproductive. Instead of immunizing and indemnifying police for their misconduct, as we do now, a libertarian system would emphasize personal responsibility and require police to purchase professional liability insurance, just like doctors, lawyers, and engineers must. Besides providing better redress to victims, this approach would harness the power of the market to reward rights-respecting police who manage to do their jobs without violating people’s rights with lower premium costs while forcing rights-violating officers to internalize the costs of their misconduct in the form of higher premium costs—just like unsafe drivers and malpractice-prone doctors are penalized for their behavior. Indeed, because insurance companies are so adept at identifying risk, there is every reason to believe the worst offenders would eventually render themselves uninsurable and thus unemployable, which is precisely what we want. These positive dynamics would, of course, be reinforced by eliminating the Monell doctrine and imposing employer liability on police departments and district attorney offices. Among other things, this would give supervisors a strong incentive to weed out marginal applicants for those positions, maintain high standards of training and conduct for those who are hired, and exhibit a much lower tolerance for “bad apples.”

There is a less obvious benefit to substantially increasing police and prosecutors’ exposure to civil liability that bears further consideration. In our current system, police spend a remarkable amount of time ravaging low-hanging fruit. Thus, there were more arrests in 2015, 2016, and 2017 for marijuana possession than for all violent index crimes (murder and nonnegligent manslaughter, rape, robbery, and aggravated assault) combined.¹⁹ This is plainly a misallocation of law enforcement resources that is presumably occasioned in part by officers’ relative indifference to the quality of a given arrest as long as they have a sufficient quantity of arrests. Given that officers receive roughly the same credit for a marijuana bust as for taking down an armed robber or closing a cold rape case, we should expect police to focus on the easiest arrests, not the most socially beneficial ones that are harder and more time-consuming to solve.

But what if every interaction with a citizen carried the very real risk of personal liability for the police officer? And what if every prosecution carried a similar risk of liability for the prosecutor? It seems plausible to suppose that police and prosecutors would shift their focus to genuinely bad people—true criminals whose behavior actually hurts other people or otherwise threatens the very fabric of civil society—both because the value of getting those people off the street is worth the financial risk and because genuinely bad people have a much lower chance of winning a damages award against a police officer or prosecutor who may have engaged in some borderline conduct to make the arrest or secure the conviction. By contrast, people who were not hurting anyone and were simply minding their own business or engaged in peaceful commerce with others when law enforcement chose to involve itself—they would make much more sympathetic civil-rights plaintiffs, and the kind of people police and prosecutors might choose to steer clear of unless absolutely necessary. Can we be sure the incentives would align that way? No, we can’t be sure. But given the system we have now, it would be worth experimenting to see if they might. Most people—including most libertarians—agree that operating a criminal justice system is a legitimate and indeed vital function of government. To be legitimate, however, a criminal justice system must punish only those people who truly deserve punishment, and it must do so in a way that is proportionate to the magnitude of their wrongdoing. Importantly, it must do those things in a substantively fair way, one that also has the appearance of fairness. Across all of those axes, America’s criminal justice system is failing miserably. Fortunately, we know how to do better—all it takes is the will to do so

1. John Stuart Mill, On Liberty (New Haven, CT: Yale University Press, 2003), p. 80. ↩
2. See, for example, Paul J. Larkin, “Public Choice Theory and Overcriminalization,” Harvard Journal of Law and Public Policy 36, no. 2 (2013): 715–93. ↩
3. For example, the federal mandatory minimum sentence for cultivating 1,000 or more marijuana plants—even in a state that has legalized marijuana—is 10 years. ↩
4. Alexander Hamilton, The Federalist Papers, No. 83, in The Federalist, ed. George Carey and James McClellan (Indianapolis: Liberty Fund, 2001). ↩
5. H. A. Washington, ed., Writings of Thomas Jefferson, vol. 3 (Washington: Taylor & Maury, 1853), p. 71. ↩
6. See, for example, William R. Kelly with Robert Pitman, Confronting Underground Justice: Reinventing Plea Bargaining for Effective Criminal Justice Reform (Lanham, MD: Rowman & Littlefield, 2018); and Stephanos Bibas, The Machinery of Criminal Justice (Oxford, UK, and New York: Oxford University Press, 2012). See also Jed S. Rakoff, “Why Innocent People Plead Guilty,” New York Review of Books, November 20, 2014; and H. Mitchell Caldwell, “Coercive Plea Bargaining: The Unrecognized Scourge of the Justice System,” Catholic University Law Review 61, no. 1 (2011): 63. ↩
7. See “The Trial Penalty: The Sixth Amendment Right to Trial on the Verge of Extinction and How to Save It,” National Association of Criminal Defense Lawyers, 2018. ↩
8. See, for example, Richard A. Oppel Jr. and Jugal K. Patel, “One Lawyer, 194 Felony Cases, and No Time,” New York Times, January 31, 2019. ↩
9. See Albert Alschuler, “The Defense Attorney’s Role in Plea Bargaining,” Yale Law Journal 84, no. 6 (1975): 1179, 1210, 1222. “With a grinding, overwhelming caseload, a public defender may ‘play for the average’ and fail to represent individual clients with the vigor demanded by … professional responsibility… . The practical consequences of ‘mutual trust’ between a prosecutor and a defender … may become indistinguishable from those of a massive trade-out. An advantage to one client arises because the defender does not make the same effort for the others. Although never expressed in terms of explicit bargaining, this implicit trade-out may supply a basic governing principle in most negotiations between public defenders and prosecutors.” ↩
10. Clay Conrad, Jury Nullification: The Evolution of a Doctrine (Durham, NC: Carolina Academic Press, 1998). ↩
11. Aliza Plener Cover, “Supermajoritarian Criminal Justice,” George Washington Law Review 87, no. 4 (July 2019). ↩
12. See Rakoff, “Why Innocent People Plead Guilty.” ↩
13. 18 U.S.C. § 201(c)(2). ↩
14. For example, United States v. Singleton, 165 F.3d 1297 (10th Cir. 1999). ↩
15. See, for example, Lozman v. City of Riviera Beach, Fla., 138 S. Ct. 1945 (2018) (refusing to dismiss claim for retaliatory arrest of protester at city council meeting). ↩
16. Sparf v. United States, 156 U.S. 51 (1895). ↩
17. Reed v. Town of Gilbert, 135 S. Ct. 2218 (2015). ↩
18. Monell v. New York City Dept. of Soc. Svcs., 436 U.S. 658 (1978). ↩
19. “More Arrests in the US for Marijuana Possession Than Violent Crimes, 2020 Hopeful Cory Booker Says,” Politifact, July 11, 2019. ↩