ONE

ACCOUNTABILITY IN CRIMINAL LAW

ALMOST 2.3 MILLION PEOPLE currently sit in prison or jail in the United States. This number is equivalent to the size of the combined populations of the cities of Boston, San Francisco, Atlanta, and Miami. The size of the incarcerated population is larger than that of fifteen U.S. states—from New Mexico, with a population of just over 2 million, to Wyoming, population 576,000. It is hard to imagine how many buildings, cells, and employees it would take to incarcerate the entire population of New Mexico, or all the people who live in Boston, San Francisco, Atlanta, and Miami.
By some estimates, over half of the incarcerated population is mentally ill.1 The prison population also includes intellectually disabled inmates, probably at a higher rate than they are represented in general population.2 Clearly, the law puts into practice a notion of individual responsibility that permits us to criminalize people who are mentally ill, mentally disturbed, and intellectually disabled. The law maintains that, despite their illnesses or disabilities, people who have been convicted had a reasonable opportunity to refrain from their criminal actions and thus to avoid criminal sanctions.
There are some good reasons for the law to be cautious about excusing less than fully rational people from eligibility for criminal punishment. When such a person commits a serious crime, we have reason to use the criminal justice system to reject his wrongful conduct and to protect ourselves from the threat he poses to our rights. It is morally permissible for us to do this, even when a criminal wrongdoer is not fully rational and lacks a normal measure of control over his own conduct. Though we should be reluctant to impose incarceration on anyone, and we should expand the menu of consequences the state is entitled to impose on people who have committed crimes to include, for example, treatment for mental illness and drug addiction, we are permitted, within reason, to use the criminal justice system to protect our rights from people who have demonstrated a willingness to violate them. Of course, retributivists are not satisfied with a rights-protection rationale for criminal punishment. Instead, they maintain that the state should distribute punishment according to moral desert: The state should ensure that criminals get the punishment they deserve.
Standard criminal-law conceptions of accountability, though, complicate efforts to rely on a retributive theory of punishment to explain or to justify current legal practices. Retributivists aim to ground punishment in blameworthiness and desert, yet in criminal law, the criteria of criminal liability call for the punishment of actors who may be neither blameworthy nor deserving of punishment. This implies that our system is not genuinely retributive. To be so, it must set much higher standards of accountability—standards that would not permit the punishment of individuals whose moral blameworthiness is seriously in doubt. It would need to establish, case by case, that each defendant is blameworthy enough to deserve the punishment he gets, whether it be a monetary fine, community service, or a prison sentence.
In the United States, many crimes, including nonviolent and even trivial offenses, as well as violent crimes, trigger severe punishment. Although the U.S. Constitution prohibits the imposition of “excessive fines” and the infliction of “cruel and unusual punishments,” these limitations are almost toothless. As a result, exceedingly harsh sentences are permitted and routinely imposed by the courts. For example, in Ewing v. California, the U.S. Supreme Court rejected Ewing’s appeal of his twenty-five-years-to-life sentence for stealing three golf clubs, a theft that constituted his “third strike” under the California law that imposes a sentence of twenty-five years to life in prison on repeat offenders whose third conviction is for a “serious or violent” felony. The Court argued that granting Ewing’s appeal “would fail to accord proper deference to the policy judgments that find expression in the legislature’s choice of sanctions.”3 The Court attested that, short of a sentence such as life imprisonment for a parking violation, legislation mandating extremely severe punishment is permitted within our legal system.4
Excessive punishment is unjust. It is a disgrace to the United States, and any reasonable retributive theory would reject it.5 The problem for the retributive theory does not come down to the practice of over-punishment. Even apart from that serious problem, the criteria of legal guilt and punishment in our system are not well calibrated to judgments of blameworthiness and desert. This does not necessarily mean that less-blameworthy and less-deserving criminal lawbreakers should be excused from punishment. Perhaps many should not be. But the mismatch between punishment and moral blameworthiness should prompt us to understand the basis for punishment, when it can be justified, in nonretributive terms. The retributive theory is out of joint with the law’s principles for determining guilt, even when those principles are reasonable. The legal criteria it makes sense to accept, including the law’s verdict that a person who commits a crime is normally eligible for punishment, are not best supported by the retributive principle. This is because legal guilt does not imply moral blameworthiness.
Later I will argue that we should take a “just harm reduction” approach to criminal justice. Thinking in terms of harm reduction would lead us to correct the problem of over-punishment, and to ensure that the punishments we impose are both beneficial to society and not unfair to the people who are punished. It would also lead us to acknowledge that a criminal justice system cannot, without the support of a broader set of just institutions, achieve justice. Problems like socioeconomic inequality and racial injustice interfere with the state’s capacity to deliver criminal justice. Yet even in an unjust society, a criminal justice system might be used, within reason, to reduce the harms associated with crimes, provided that it treats the convicted fairly. Aiming for just harm reduction would, however, require us to change the way we regard people who have been convicted of crimes as well as how we justify and organize the practice of criminal justice. It would require us to correct our tendency toward maximal punishment, and it would direct us to refrain from using the criminal justice system to morally condemn people, some of whom cannot reasonably be said to deserve punishment.

Retributive Justice

Advocates of retribution hold that wrongdoers merit a punitive response proportional to their blameworthy wrongdoing and that therefore wrongdoers should suffer corresponding harms.6 Retributivists disagree about whether wrongdoers deserve to suffer “pain and anguish” or that they deserve only that their lives go less well.7 In cases of serious wrongdoing, however, most retributivists believe that wrongdoers deserve physical or psychological pain; as philosopher Douglas Husak puts it, a culpable wrongdoer “should feel bad about what he has done.”8 Papers advocating retribution drive home this point with thought experiments directing us to imagine scenarios in which a person commits a serious crime, yet we lack reasons to punish on grounds of deterrence, incapacitation, rehabilitation, or other consequence-sensitive considerations. In such cases, retributivists argue, we would still insist that retributive punishment is a requirement of justice.9 Something is amiss when we fail to make a wrongdoer feel bad.
Philosophical attempts to justify the retributive ideal rely heavily on purportedly intuitive assertions about the foundations of justice, or morality, or the essence of human decency, and they are unsuccessful. One of the most influential defenders of retributive justice is legal scholar Michael S. Moore. The crux of his argument is an introspective appeal to the guilt he would feel if he had committed a horrific act of violence against another person. Moore proclaims, “My own response, I hope, would be that I would feel guilty unto death. I couldn’t imagine any suffering that could be imposed upon me that would be unfair because it exceeded what I deserved.”10 While one might admire Moore for his stated willingness to own up to any grievous wrongs he might commit, his expression of a hypothetical willingness to endorse whatever form of suffering might be inflicted upon him hardly offers a compelling rationale for a measured and fair criminal justice system. By his own testimony, his retributive sentiments are unbounded.
Of course, intuitions play an important role in justifying conceptions of morality—a role that, generally speaking, is not always suspect. John Rawls described the process of formulating and justifying a theory of justice as a process of reaching “reflective equilibrium.”11 The political philosopher aims to capture common beliefs about justice in a coherent set of moral principles. This involves adjusting familiar moral intuitions so as to formulate them as principles and revising the principles to make them consistent with one another, while taking care that the principles retain the support of an array of important intuitive judgments. Rawls begins with a set of intuitions about justice: a just society is a fair system of social cooperation, all people should have the material means needed to attain acceptable life prospects, all are entitled to due process of law, slavery and other forms of servitude and oppression are wrong, citizens should be treated as free and equal, everyone should be secure in their personal property, have decent health care, opportunities for education, and more. He organizes this collection of beliefs about the requirements of justice into a set of principles that would protect equal basic rights and liberties for all citizens, including rights associated with due process and the rule of law, assure that everyone’s political participation has fair value, protect fair opportunities for education and employment, and guarantee a distribution of material goods that is acceptable to even the least well-off members of society. The principles he proposes can, he believes, guide us in establishing and assessing a political state’s institutions on the basis of “public reasons,” reasons that could be shared by all reasonable members of society, whatever social positions people happen to occupy.12
By showing how intuitions yield insight when considered systematically, critically scrutinized, and extended to new questions, Rawls demonstrates that the work of philosophy can produce a theory of our intuitions without resting on any Archimedean point. There is no single grounding value from which the theory begins or on which it ultimately rests. Unlike the approach taken by many retributive theorists, basic notions of criminal justice are considered together with broader questions of social justice.13 Criminal sanctions are understood as part of a system of law that should “have the consequence, in the long run, of furthering the interests of society.”14 These interests, Rawls argues, should be those that could be endorsed by all reasonable members of society on the basis of shared reasons. It is when and because intuitions about justice plausibly establish a standard of public justification that they are worth taking seriously.
A Rawlsian approach to criminal justice, which is not oriented fundamentally around the value of retribution, seems to retributivists to be missing something crucial. The elusive element, however, proves to be impossible to justify with “public reasons.” In a compelling survey of philosophical attempts to justify retribution, philosopher John Mackie has formulated what he refers to as “the paradox of retribution.” The paradox, he says, is that “on the one hand, a retributive principle of punishment cannot be explained or developed within a reasonable system of moral thought, while, on the other hand, such a principle cannot be eliminated from our moral thinking.”15 The paradox of retribution follows from granting a foundational status to the retributive principle, for if a fundamental commitment to retribution were revoked, the value of retribution could be eliminated from our principles of justice, and the paradox would dissolve. Though retributive theorists do not see how, I believe it is both possible and urgent to retract commitment to retribution from a theory of criminal justice.
Mackie shows that popular attempts to justify retribution rationally are not convincing. Criminal wrongdoing does not imply the necessity of punishment (or, as Mackie puts it, the relation between crime and punishment is not analytic). Consequences, including the satisfaction retribution offers to some crime victims and the educative effects of bringing criminals to justice, are not relevant to justifying an entirely backward-looking principle. (According to retributivists, punishment is justified according to what wrongdoers deserve, not by the consequences of punishing them.) The idea promoted by some professed retributivists, that through punishment a criminal discharges a debt to society, also has a consequentialist ring, and seems closer to the notion of reparation than retribution.16 The Hegelian idea that punishment somehow annuls the crime is not promising, Mackie argues: “The punishment may trample on the criminal, but it does not do away with the crime.”17 And the related idea that punishment “absorbs and wipes out guilt” presupposes rather than explains the retributive thesis.
Mackie considers the proposal that what is negated by punishment is the unfair advantage a criminal gains by committing a crime.18 He finds this formulation of the value of retribution equally implausible. The proposal focuses on the wrong thing—advantage rather than wrongness or guilt—and it supposes something that is not always true: that the criminal gains something through criminal wrongdoing. Wrongful benefits are most plausible in the case of property crimes, but do not characterize relationship-destroying actions, like domestic violence, or inchoate crimes, like attempted murder. Furthermore, the notion that people who do not commit crimes suffer the burden of restraining themselves seems wildly off. Surely, for most people, refraining from murder or kidnapping does not demand genuine self-restraint.19
Still, as Mackie correctly notes, retributive justice holds great attraction for many people. He explains this by reference to natural selection and sociological theory. He stresses the role retributive conventions play in stabilizing social cooperation, and he hypothesizes that our moral sentiments have evolved as responses to a collective recognition that certain types of behavior trigger resentment and opposition. Specifically, he connects our attachment to the notion of just deserts with the emergence of the concept of moral wrongness, a concept he dissects into three components: the notion of an action’s being harmful, its being forbidden, and its calling for a hostile response.20 According to Mackie, the harmful and forbidden nature of wrongful acts is conceptually connected with the judgment that generally harmful and...