Every morning in the United States 2.2 million people wake up in our nation’s prisons and jails, making us the world leader in incarceration. Another 4.4 million people currently live under some form of correctional supervision, and more than 12 million cases flood our courts, public defender offices, and probation offices each year. The result is a country where one out of every three adults possesses a criminal record—a staggering statistic. Of all the lives, families, and communities destroyed by this system, the poor, the underserved, and people of color bear the brunt. These grim realities are well known to many, but this summer’s events crystallized them yet again, as we watched COVID-19 rage through our overpopulated correctional facilities and recoiled in horror at the grotesque slow-motion asphyxiation of George Floyd at the hands of Minnesota police—yet another entry on the ever-growing list of unjustified killings of people of color by law enforcement.

Moments like these demand that we reexamine our approach to criminal justice reform. In our era of mass incarceration, most would say that we need fewer criminal convictions and less punishment. But exactly what conduct are we prepared to decriminalize, and which sentences are we ready to shorten? These are hard questions in part because low-level, non-violent offenders account for only a small percentage of the total number of incarcerated people; the vast majority of people in prison are there for serious offenses, including homicide, assault, and drug trafficking. But it’s also true that our most serious offenses are being applied in overly broad ways that conflict with our moral intuitions about guilt. To commit a crime, after all, is not just to do a bad thing. Conduct becomes criminal only when it is accompanied by a blameworthy state of mind. Or at least that’s the idea behind the legal principle of mens rea (Latin for “guilty mind”).

All too often, this principle is ignored by our criminal justice system—both in who it convicts of crimes and in the length of sentences it hands out. That should change. Good intentions may not be enough to shield someone who stumbles into harm’s way from civil liability, but they should keep individuals outside the reach of the criminal justice system. And even for those who act with a guilty mind, the criminal justice system should recognize important moral gradations between culpable mental states. Reforming our criminal codes in these ways won’t rid our system of all its problems, but doing so is an important part of a just vision for change.


Respect for the guilty mind is intuitive. As Justice Oliver Wendell Holmes, Jr., famously put it, “even a dog distinguishes between being stumbled over and being kicked.” Of course, humans do too: a wide array of studies has shown that our intuitive moral sense of how to respond to harm or wrongdoing is keenly sensitive to what is happening in the minds of others. And this appears to be true from a very young age. Psychologists have found that even kindergarteners “make their culpability and punishment decisions proportionately” based on morally relevant differences in mens rea.

To commit a crime is not just to do a bad thing.

In theory, the criminal law is like this too. U.S. legal culture generally accepts that the criminal justice system should not punish people who make reasonable mistakes or for accidents, nor should it punish those who cannot think or act morally (say, due to mental illness). It is also accepted that the criminal justice system should excuse those who make understandably bad decisions in excruciating and extenuating circumstances. No doubt, individuals in these situations may do something harmful—perhaps terribly so. But in the absence of a guilty mind, the law should—and typically does—view them as morally innocent, and therefore beyond the reach of criminal liability.

The law has also embraced a similar notion in sentencing. As I’ve highlighted elsewhere, the principle of “proportional mens rea” says that, all else being equal, punishment should track the guiltiness of a defendant’s state of mind. This is the animating idea behind our centuries-old homicide laws, which go to great lengths to differentiate between mental states, and, ultimately, to lessen sentences for those whose choices are less blameworthy than the paradigmatic case of a cold-blooded, premeditated murder. An intentional killing committed in the heat of passion (e.g., an aggrieved parent who kills her child’s bully in a fit of rage) is often mitigated down to manslaughter, as is one committed recklessly (e.g., a fatal decision to run a red light, in conscious disregard of a slowly approaching pedestrian). And where that killing is intentional but motivated by imperfect self-defense (e.g., someone who unreasonably mistakes a friend for a foe), or is just plain negligent (e.g., a thoughtless college student mishandling a gun accidentally fires it), the charge may be dropped even further to negligent homicide—or a lesser form of manslaughter, with labels varying from jurisdiction to jurisdiction.

All of these individuals have perpetrated the same wrongful act: killing a human being. But because of the important distinctions in mental state accompanying that act, criminal law provides them with increasingly lenient sentences.

Our nation’s lawmakers have sacrificed our intuitive sense of fairness in the name of “public safety.”

This is how the criminal justice system is supposed to operate: limiting convictions to guilty minds and apportioning punishment in accordance with their guiltiness. The U.S. Supreme Court has recognized this basic idea to be (1) “universal and persistent in mature systems of law,” (2) as fundamental as our belief in an individual’s ability to “choose between good and evil,” and (3) “essential if we are to retain ‘the relation between criminal liability and moral culpability’ on which criminal justice depends.” Similar statements are frequently made by courts and scholars around the country, who are drawn to the intuitively appealing vision of—as Douglas Husak puts it—“a world in which impositions of criminal liability and punishment correspond to our considered judgments of blame and desert.”


If legal rhetoric were an accurate gauge of legislative reality, our criminal justice system’s treatment of mens rea would be pristine. But this is simply not the case. Criminal statutes, the primary source of liability and punishment in this country, often fail to live up to this commitment to moral guilt. In a surprisingly large number of situations, our nation’s lawmakers have disregarded traditional mens rea principles in favor of a tough-on-crime approach that sacrifices our intuitive sense of fairness in pursuit of an amorphous idea of “public safety.”

The drug arena is a case in point, in part because it is where some of the most aggressive tough-on-crime campaigns have been waged. U.S. drug policies often focus on behavior to the near total exclusion of mental state considerations. Many drug possession statutes apply a “strict liability” approach, in which—as Markus Dubber explains—“you can be convicted . . . if you don’t know that you are ‘possessing’ a drug of any kind, what drug you are ‘possessing,’ how much of it you’ve got, or—in some states—even that you are possessing anything at all, drug or no drug.”

Just as strict is the ruthlessly quantitative approach to drug sentencing employed in most jurisdictions that bases punishment on what Doug Berman calls “fairly arbitrary questions about how the drugs involved in an offense are to be classified or quantified instead of on a defendant’s actual culpability.” What this means, in practice, is that small-time dealers can be punished just as severely as big-time traffickers (think decades, not years) based upon their tangential connection to massive quantities of narcotics they had no idea existed and had no intent to sell.

Similarly strict tendencies are reflected in the law of violence. Some jurisdictions have assault and homicide statutes that convict those who, absent clear moral fault, accidentally injure or kill another person in the course of daily life. (This situation is often the result of applying civil negligence standards in the criminal context.) And nearly all jurisdictions have homicide laws that treat unwitting fatal accidents that occur in the course of perpetrating a felony crime as murder. The consequences of these “felony murder” laws are often striking.

Consider the 1984 case of Forrest Perry Heacock, who supplied cocaine to a few friends. They all injected it together, at which point one of the partiers died from an overdose. No doubt Heacock lacked any intent to kill: indeed, he affirmatively hoped his friends would enjoy themselves. But because Heacock had committed the underlying felony of “drug distribution,” he was convicted of murder and sentenced to forty years of imprisonment. (Comparable prosecutions, under both felony murder and other drug-induced homicide statutes, appear to have substantially increased over the past decade.) Or consider the 1989 case of Allison Jenkins, whom an Illinois police officer chased at gunpoint based on suspicion of drug possession. As the officer caught him by the arm, Jenkins tried to shake free. The officer tackled Jenkins and fired the gun as they fell, killing his own partner. Jenkins, like Heacock, surely did not intend to kill anyone; but he, too, was convicted of felony murder, predicated on the underlying offense of battery of a police officer.

An even more jarring illustration of this “in for a penny, in for a pound” approach to criminal liability are felony murder cases involving unwitting accomplices who receive the same treatment based on their accidental facilitation. Consider the 2004 case of Ryan Holle, a young Florida man. One evening, as a party was wrapping up, Holle’s housemate spoke with a few guests about stealing a safe from a drug dealer’s home. The housemate then asked Holle, who had frequently loaned him his car, for the car keys. Tired, drunk, and unsure whether his housemate was even serious, Holle ultimately relented, giving up the keys and then going to sleep. The housemate and his friends proceeded to steal the safe. In the process of perpetrating the burglary, one of the friends killed a resisting resident. Having aided the underlying burglary, Holle was convicted of felony murder and sentenced to life without parole.

The punishment in a case like Holle’s is striking on its own terms, but perhaps doubly so once one considers how felony murder laws—as Rachel Barkow observes—“lump disparate categories of people together.” The application of felony murder in Holle’s case authorized punishing someone who unwittingly aided a homicide just as severely as someone who intentionally killed. Indeed, both Holle and the person who fatally struck the resisting resident received the same sentence: life without parole. Similar convictions may take place when the supporting actor is a juvenile, disabled, or suffering from mental illness, as a number of jurisdictions effectively preclude a defendant from raising defenses based on immaturity, insanity, or other cognitive deficiencies.

These moral equivocations are not limited to the law of homicide. Many jurisdictions extend them to all crimes, without regard to context and sometimes in even more egregious ways. This move is authorized by judicially created policies, such as the natural and probable consequences rule and the Pinkerton doctrine. These policies hold a person who purposely aids, or conspires in, the commission of one crime responsible for all reasonably foreseeable crimes committed by the principal actor. In effect, as Joshua Dressler has explained, this approach makes even a bit player’s assistance or agreement with any aspect of a criminal plan the basis for convicting that person of each and every offense associated with it, so long as it can be proven that this peripheral actor should have known the offense would or might occur (a low bar in both theory and practice). The result is a system where a girlfriend taking messages for or a parent offering a bed to a drug dealer can face a lifetime in prison for the multitude of crimes committed by every member of a massive criminal operation.


Outcomes like these shock our intuitive sense of justice by ignoring the central role that the minds of others play in our own personal evaluations of responsibility. But do these intuitions about guilty minds provide a sound basis for reforming the criminal law? Arguably they do, for at least two reasons.

The first is that criminal law’s formal judgments of responsibility, like the informal ones we make in private, are ultimately about the same thing: blame. A criminal conviction and sentence say something morally condemnable about the offender—that he or she failed to sufficiently concern him or herself with the interests of another person or society. But as scholars such as Kimberly Kessler Ferzan, Stephen Garvey, and Peter Westen have observed, whether a person actually failed to live up to our societal expectations, and therefore deserves to be condemned for it, depends upon his or her state of mind. For example, an actor’s blameworthiness hinges upon important questions of motivation and awareness: why did the wrongdoer do what she did, and did she know what might result when she did it? And it also depends upon important questions of rationality and free choice: did the wrongdoer have the capacities necessary to think and act morally, and under the circumstances was she able to exercise them freely?

It is these folk psychological differences that explain the moral difference between, say, the grieving mother’s accidental swiping of an unseen bicyclist on her way to her child’s funeral and the vengeful ex-husband’s premeditated decision to run over his former wife while she’s out for a bike ride. The outcome may be the same in both situations: grievous bodily injury. But only one of these minds clearly expresses the insufficient concern for physical security that aggravated assault laws seek to protect. So only one of these people deserves the formal expressions of blame—criminal liability and punishment—associated with violating them. (In this sense, respect for mens rea reflects what Kenneth Simons has described as the “classic liberal idea that moral culpability is, and criminal liability should be, based on a conscious choice to do wrong.”)

The injustice of blaming those who lack a guilty mind thus provides a good reason for mens rea reform. But the societal cost associated with that injustice provides another.

Money spent on incarcerating the morally blameless is money not spent on other things—like education, health care, or social services.

The most obvious cost is crudely economic: imprisoning the morally innocent is expensive and wasteful. Some have defended the punishment of innocent minds on the grounds that it deters future wrongdoers. Others have argued that dangerous people must be incapacitated, no matter their mental state. But we have very little reason to believe these are effective ways of promoting public safety. And we at least have some reason to believe that they do just the opposite. For one thing, there is the reality of opportunity costs: money spent on incarcerating the morally blameless is money not spent on other things, like education, health care, or social services. From the perspective of promoting public safety, one can question whether a harsh criminal justice response is a wise investment.

This is particularly dubious in light of the fact that incarceration can be criminogenic: given the harsh realities of prison life, people may end up leaving more likely to commit a crime than they were upon entry. And, just as important, strict liability policies may be criminogenic given that—as Tracey Meares explains—“the legitimacy of the criminal justice system is critical to the system’s proper functioning.” Indeed, in many situations, punishing people without regard to their state of mind can lead to outcomes that the community perceives as unjust, which undermines the legitimacy of the criminal justice system in the eyes of the public. That threatens public trust and social cohesion, as well as public safety. “As the criminal law’s moral credibility is incrementally reduced,” Paul H. Robinson has observed, “the system is incrementally more likely to inspire resistance and subversion rather than acquiescence and cooperation.”

Finally, there’s the unavoidable reality that the injustices and costs associated with strict liability policies are inevitably concentrated on those who disparately suffer from every other problem in the criminal justice system: the poor, the underserved, and people of color. This explains why David Patton, executive director and attorney-in-chief of Federal Defenders of New York, chose to speak out in support of a recent effort to limit strict liability in the federal criminal code. Although some Democratic lawmakers and reform organizations attacked the proposal as a giveaway to corporate executives, for whom mens rea protections could make white collar crime prosecutions more difficult, Patton pushed back:

As Federal Defenders, we are acutely aware of the need for mens rea reform. Over 80 percent of people charged with federal crimes are too poor to afford a lawyer, and nearly 80 percent of people charged with federal crimes are Black, Hispanic, or Native American. These are our clients, and too many of them are subject to laws that are neither fair nor consistent with traditional principles of criminal liability.

Other scholars echoed this point. Writing in support of the federal effort in the New York Times, Gideon Yaffe argued that mens rea reform can be “harnessed to aid some of those who are especially ill treated by the criminal justice system: the poor and racial minorities.” But ultimately, these arguments were not enough to assuage Democratic lawmakers, who remained focused on the bill’s impact on corporate prosecutions. (That is not to say that Democratic opinion was monolithic: as Vikrant Reddy observes, the federal mens rea reform effort was also supported by “prominent progressive voices, such as the National Association of Criminal Defense Lawyers and U.S. Representatives John Conyers and Bobby Scott.”)

This episode says more about the current state of criminal justice reform in the United States than it does about the law of guilty minds. Benjamin Levin observes in it the unyielding “temptation for mass incarceration critics to make exceptions and support harsh treatment for particularly unsympathetic defendants.” Rather than constrict liability in a way that would protect both people of color and corporate executives from being unfairly blamed or punished by the criminal justice system, opponents of federal mens rea reform effectively opted for what might be called a “level-up solution,” preferring to treat white collar defendants just as poorly as disadvantaged defendants of color.


What’s the solution to this widespread neglect of mens rea? To some extent, the answer already exists in a legal document called the Model Penal Code (MPC). Completed in 1962, the MPC was the culmination of a decades-long examination of criminal law conducted by a bipartisan group of judges, lawyers, and scholars for the American Law Institute (ALI). The group was initially tasked with summarizing the current state of criminal law for the ALI’s “Restatement of the Law” series, but the morass of judge-made criminal justice policies they encountered was deemed to be “too chaotic and irrational to merit restatement.” Instead, they embarked on drafting the nation’s first comprehensive criminal code, which offered a clear and precise legislative treatment of criminal justice policy issues intended to supplant the piecemeal, haphazard collection of criminal statutes in effect at the time.

Perhaps the MPC’s greatest achievement is its treatment of mens rea. At the time, criminal statutes rarely provided much concrete guidance as to mens rea requirements, and courts struggled to give a clear and consistent explanation of the state of mind necessary to secure a conviction for most offenses. This muddle ultimately led to what the Supreme Court famously described as “variety, disparity and confusion” around “definitions of the requisite but elusive mental element.” The MPC drafters believed that “clarification was essential,” which to their minds was something that only comprehensive mens rea legislation could achieve.

This commitment led the MPC drafters to several innovations. Most notably, they reduced the more than seventy amorphous mens rea terms in use at the time to just four precisely defined ones: purpose, knowledge, recklessness, and negligence. They also developed a nuanced statutory framework for communicating how these culpable mental states apply to the different elements of an offense. (For example, under the MPC’s framework, burglary prohibits recklessly entering an occupied structure, with the purpose to commit a crime inside). These conceptual innovations, among others, were intended to make the legislative work of drafting clear mental state requirements easier and more effective.

More significant to criminal defendants are the MPC’s substantive policy recommendations on mens rea issues. In an attempt to thwart the use of strict liability, the MPC includes a general provision that requires proof of a culpable mental state—whether purpose, knowledge, recklessness, or negligence—for just about every element of every offense in a criminal code. “Crime does and should mean condemnation,” the drafters reasoned, “and no court should have to pass that judgment unless it can declare that the defendant’s act was culpable.” From the drafters’ perspective, strict liability policies that risked convicting the morally blameless were “indefensible,” and they understood the criminal law’s commitment to the guilty mind to be “too fundamental to be compromised.”

The MPC makes a number of other recommendations to this end. For example, it strengthens affirmative defenses like insanity and duress, making it less likely that faultless individuals would face criminal liability or punishment. And it effectively abolishes felony murder, the natural and probable consequences rule, and the Pinkerton doctrine, driven by the drafters’ belief that the extreme sentences produced by these policies are “incongruous and unjust” in light of what an individualized assessment of a criminal defendant’s state of mind would otherwise call for.

Completion of the MPC in 1962 set off a cascade of comprehensive criminal code revision efforts across the country, and by 1989, thirty-five states had adopted new criminal codes on the basis of its recommendations. Although these MPC-based criminal codes are not identical, most incorporate the MPC’s four-part mental state framework. In this sense, the MPC approach to mens rea became the “representative modern American culpability scheme.”

So what’s the problem? Many of the MPC’s substantive policy recommendations—for example, its general abolition of strict liability and the specific jettisoning of felony murder—did not fare as well as the general conceptual framework. Some lawmakers watered down key MPC mens rea provisions, while others ignored them entirely. And courts have interpreted remaining protections in uncharitable ways, effectively placing a thumb on the scales in favor of strict liability in criminal law cases.

This is not surprising given political developments in the latter half of the twentieth century. During this era, as Darryl Brown has observed, criminal justice was increasingly driven by the tough-on-crime movement, which had been triggered by the rising crime rates and social disorder of the 1960s as well as the political transformations of the civil rights movement. So “when the MPC reform movement conflicted with the tough-on-crime movement, it was, unsurprisingly, the MPC’s reform efforts—the efforts of legal professionals and academics more than politicians—that lost.”

Despite these setbacks, the MPC’s treatment of mens rea offers an essential lens for understanding enduring problems in our approach to criminal justice. That is not to say the MPC does not suffer from flaws; indeed, it is far from perfect. Several aspects of the MPC have been subjected to close and withering criticism, including its abysmal approach to sex crimes, which (among other problems) recognizes and defends a marital exception to rape. Angela Harris and Cynthia Lee capture this dual legacy well when they write:

The Model Penal Code . . . encapsulated the aspirations of a generation of progressives who hoped to bring criminal law, and law generally, fully into the twentieth century by situating it within the social sciences. Today, however, the significance of the Model Penal Code has changed . . . In some areas, such as mens rea, it seduces our students with its promise of clear and unchanging definitions, but in other areas, such as sexual assault, it is beginning to look quite dusty and antiquated.

Of course, attracting law students is one thing; attracting lawmakers has been quite another.


There have been a few small but noteworthy mens rea reform successes in the last decade. Although the recent federal reform effort fell short, comparable efforts in Ohio and Michigan prevailed, with their state legislatures successfully enacting a default presumption of mens rea in 2013. (Still, it should be noted, each state’s law has important exclusions.) California passed a modest version of felony murder reform in 2018 that precludes liability for some accomplices (among other changes). But these are largely isolated and geographically limited projects. There hasn’t been anything akin to the centralized movements we’ve seen on other modern topics of criminal justice reform.

Mens rea reform won’t rid our system of all its problems, but it is an important part of a just vision for change.

In one sense, this is understandable. Enacting comprehensive mens rea reform, while of great moral significance, certainly would not cure everything that ails our deeply flawed criminal justice system. It would not dismantle mass incarceration on its own; even the most ambitious versions of mens rea reform would likely at most have a moderate impact on the number of people we introduce to prison on an annual basis and the length of time for which we send them there. Nor would it eliminate police and prosecutorial abuses (even if it could certainly help to diminish them). It is also clear that the notion of guilty minds may not intersect so neatly with other aspects of the way we talk about social justice. As Shaun Ossei-Owusu recently wrote in these pages, “intent can often be a sideshow, if not an outright distraction” when trying to address systemic social inequalities, including those produced by racism and sexism.

On the other hand, comparable limitations confront just about every other policy reform proposal on the table right now. Nor should we let the perfect be the enemy of good. The potentially broad political appeal of mens rea reform makes it an attractive target. Progress on this front would be a significant accomplishment, which in turn would help to advance efforts to rid our criminal justice system of its many other injustices. For all these reasons, comprehensive mens rea reform has a role to play in state and federal reform efforts. At the very least, in a time of both mass incarceration and overcriminalization, there’s no good justification for leaving the plight of the morally innocent out of our criminal justice reform conversation. Justice demands that criminal liability and punishment be limited to the guilty mind—and our criminal justice system can surely offer criminal defendants at least that much.