Papers

The instinct to blame, resent, and punish norm violations is deeply engrained in human nature. While the norms vary widely, it is difficult to imagine a society that did not respond to violations of the normative order with some form of condemnatory response. Moreover, it is unclear whether such a society would amount to a utopia of unconditional empathy or a dystopia of selfish indifference. After all, the retributive sentiments are tightly bound up with recognizing each other as subject to normative expectations in the first place.

I have three objectives in this paper. First, I show how the characteristic moral psychology of the retributive sentiments—retrospective, non-calculating, and impersonal—emerges from their role in stabilizing cooperation in informal contexts. Second, I contrast their role as decentralized mechanisms of norm enforcement with that of centralized institutions. Because institutions do not rely on the retributive sentiments to prevent free riding, formal social orders are freer to adopt forward-looking, instrumental approaches to cooperation. This explains the perennial tension between the moral logic of retribution and the more utilitarian logic of criminal justice. Finally, I sketch some of the ways in which formal and informal social norms interact to constitute social order.

Arguably, the rule of law rests on a social background of broad interpersonal trust, even across substantial social distance. At the same time, a willingness to trust strangers itself arguably presupposes robust rule of law. Although the precise parameters of the relationship between law and interpersonal trust remain conceptually and empirically contested, in this paper we explore the jurisprudential presuppositions of an endogenous relationship between the rule of law and trust. What must be true of law if the rule of law is to both support, and be supported by, generalized trust?

We argue that law must be action-guiding and possess content-independent authority to support generalized trust. The central condition is divergence in private judgment. Law is otiose if private judgments uniformly converge, as the easiest way to predict what others will do is simply to consider what you would do. But cooperation at scale spans social distance and heterogeneity in private judgment. Under those conditions, cooperation can expand when people are disposed to respect the law, because it is law, and even when it clashes with their private judgments. We contrast an action-guiding conception of law with an expressivist and moralized conception of law which, we argue, supports particularized rather than generalized trust. The rule of law is surely not the only input into generalized trust (and vice versa) but law, we suggest, is an especially flexible and powerful means of securing cooperative gains across social distance.

We draw on two case studies, both pertaining to the regulation of sexual misconduct, to motivate the more general argument. The first case study concerns an informal dispute resolution forum designed to resolve allegations of sexual impropriety within a tightly bounded religious minority, whereas the second, online reputational sanctioning, strips away shared solidaristic bonds while seeking to retain a shared set of moral attitudes. The first case study illustrates the limits of informal dispute resolution, and its relationship to more impersonal, formally defined legal rules. The second case concerns online reputational sanctioning by survivors of sexual violence. The norms invoked by survivors are universal in ambition but enforced informally through appeal to shared conviction rather than formal rules. Taken together, the cases illustrate how expressive and action-guiding law seek to sustain trust, whether in a competitive or complementary fashion.

Perfect compliance with law has traditionally been regarded as undesirable because the law is inevitably imperfect. Imperfect compliance is often the right response to imperfect law. Rapid progress in AI capabilities invites us to reconsider both points. A technology capable of translating general legal norms into highly personalized guidance—what I call “intelligence at scale”—promises to bring perfect law within reach. This raises the question: does perfect law demand perfect compliance? After all, a directive that enjoins exactly what you have reason to do is one you have, by hypothesis, every reason to follow.

I suggest that an ideal of perfect compliance with perfect law is self-effacing. Perfect compliance requires gating conduct—foreclosing non-compliance outright—over pricing conduct. Perfect compliance also renders law’s normative guidance explanatorily inert. Taken to its limit, perfect law favours governance by infrastructure rather than norms. Governance by infrastructure poses a challenge to the trust, contestation, and respect for agency long associated with the rule of law. This creates a tension between the value of perfect compliance with perfect law, on the one hand, and the value of normative guidance on the other. In short, a less noticed challenge that AI poses to the rule of law is not that it will be poorly aligned with our values, but that it may become too well aligned.

The appropriateness of using reasonable hypotheticals to review sentencing legislation under section 12 of the Charter has become settled orthodoxy amongst Canadian courts and legal scholars, with the Supreme Court leading the way. This article argues that this conventional wisdom is wrong for four main reasons. Such use of reasonable hypotheticals: (i) rests on the mistaken assumption that the proportionality of punishment can largely be determined, in isolation, by applying general principles to the facts of a given case, real or imagined; (ii) is inconsistent with the Supreme Court’s otherwise individualized approach to sentencing; (iii) bypasses traditional public law standing requirements and implausibly assumes the infallibility of courts; and (iv) infringes the separation of powers by encroaching on both Parliament’s authority to set penal policy and the prosecutorial discretion traditionally afforded to the executive. The article concludes by proposing, as a way forward, an alternative holistic approach to sentencing and its constitutional review.

Are there functions that a state must do for itself instead of relying on others to perform on its behalf? Some argue that suppressing private violence and enforcing the law are powers that only the state may exercise, meaning the state is responsible not only for ensuring that they are discharged but for discharging them itself. In this article, I challenge the do-it-yourself principle that animates much of the recent philosophical literature on privatization. In its place, I defend a rights-based theory of delegated state power. A state’s consistent failure to adequately protect rights is a reason for the state to voluntarily delegate powers to a third party that is better placed to use them to protect rights. This account grounds the legitimacy of delegation on the protection of rights rather than on state consent. I illustrate the theory by showing how it could justify delegating policing and penal powers to charter cities, semiautonomous cities run by a guarantor country or private developer. I consider the conditions under which a charter city developer is likely to adequately safeguard the rights of its residents. Finally, I argue that neither sovereignty nor democratic legitimacy prohibits the state from delegating its coercive powers. When a state is incapable of protecting rights, delegation is not a betrayal of sovereignty and democracy but a step toward their realization.

‘Mass incarceration’, as conventionally understood, refers to an imprisoned population that is both excessive in size and racially skewed in its demographics. However, in contrast to racial skew, the appropriate size of a prison system has largely escaped analysis. This article contributes to analysis of the scale of a prison system in two ways. First, I show why non-controversial principles linking crime to punishment, such as guilt and proportionality, are insufficient. Because incarceration rates are driven more by social policy than by crime, an adequate analysis of scale presupposes an account of what we hope to get out of punishing people in the first place. Second, drawing on a generic crime-prevention account of incarceration, I sketch three increasingly resolving, but also increasingly contentious, conceptions of excess: the Pareto, social welfare, and utilitarian conceptions. Along the way, I briefly consider the trade-off between how committal a theory of incarceration is and its ability to explain what is wrong with mass incarceration, as well as the concern that the social welfare and, especially, utilitarian concepts are excessively paternalistic. The ultimate aim of the article is to contribute to our understanding of mass incarceration as a distinctive normative concept.

The rule of law’s value stems from fundamental features of human nature: our limited cognitive capacity, susceptibility to bias, and challenges in coordinating behavior at scale. Legal rules help us transcend these limitations by facilitating coordination, decision-making, and information sharing. However, as artificial intelligence systems surpass human capacities in these domains, even when scaffolded by legal rules, they may provide more direct and effective means of serving the law’s functions in facilitating information sharing, promoting sound decision-making, and enabling large-scale coordination. Artificial intelligence may emerge as an alternative social technology to law, and the rule of law. That is, the rule of law may diminish in value not because it has been displaced by authoritarianism but because we have invented better ways of accomplishing its purposes. This suggests the need to reconceptualize the rule of law for an AI-enabled future. Specifically, I argue that as the importance of pure procedure diminishes, the need to clearly articulate ultimate goals will become increasingly critical.

Legal philosophers often ask whether a person has a reason to obey the law simply because it is the law. We ask the contrary question: does a person have a reason to disobey the law simply because it is the law? Many philosophers who have considered the question of disobedience have focused on injustice; others have defended disobedience on libertarian or anarchist grounds. In contrast, we argue that there is a content-independent reason to disobey the law even when it is not unjust, illegitimate, or otherwise undesirable. Legal philosophers generally agree that law claims peremptory authority, but they also generally agree that any duty to obey the law is substantially more limited. We argue that insofar as the law makes inflated claims to authority, it generates a content-independent reason to disobey. This anti-authoritarian principle is grounded in the virtue of clearly communicating one’s political commitments to others within a democratic society. By disobeying, one communicates one’s conviction that the law makes inflated claims to authority. We show how our account of disobedience as such is distinct from more familiar theories of anarchism and civil disobedience and argue that it is applicable whether one lives under conditions of justice or injustice.

The rapidly increasing role of automation throughout the economy, culture and our personal lives has generated a large literature on the risks of algorithmic decision-making, particularly in high-stakes legal settings. Algorithmic tools are charged with bias, shrouded in secrecy, and frequently difficult to interpret. However, these criticisms have tended to focus on particular implementations, specific predictive techniques, and the idiosyncrasies of the American legal-regulatory regime. They do not address the more fundamental unease about the prospect that we might one day replace judges with algorithms, no matter how fair, transparent, and intelligible they become. The aim of this paper is to propose an account of the source of that unease, and to evaluate its plausibility. I trace foundational unease with algorithmic decision-making in the law to the powerful intuition that there is a basic moral and legal difference between showing that something is true of many people just like you and showing that it is true of you. Human judgment attends to the exception; automation insists on blindly applying the rule. I show how this intuitive thought is connected to both epistemological arguments about the value of statistical evidence, as well as to court-centered conceptions of the rule of law. Unease with algorithmic decision-making in the law thus draws on an intuitive principle that underpins a disparate range of views in legal philosophy. This suggests the principle is deeply ingrained. Nonetheless, I argue that the powerful intuition is not as decisive as it may seem, and indeed runs into significant epistemological and normative challenges. At an epistemological level, I show how concerns about statistical evidence’s ability to track the truth can be resolved by adopting a probabilistic, rather than modal, conception of truth-tracking. At a normative level, commitment to highly individualized decision-making co-exists with equally ingrained and competing principles, such as consistent application of law. This suggests that the “rule of law” may not identify a discrete set of institutional arrangements, as proponents of a court-centric conception would have it, but rather a more loosely defined set of values that could potentially be operationalized in multiple ways, including through some level of algorithmic adjudication. Although the prospect of replacing judges with algorithms is indeed unsettling, it does not necessarily entail unreasonable verdicts or an attack on the rule of law.

Recent years have witnessed proliferating calls for technology and social media companies to more aggressively police the speech of their users. Social media companies have drawn criticism both for being too aggressive and too lax in censoring their users’ speech. While this controversy is typically framed in terms of the extent and significance of principles of free speech, we re-frame the debate in terms of a contrast between ex ante prevention and ex post punishment. Ex ante prevention operates as a form of censorship, preventing objectionable speech from occurring, whereas ex post punishment operates by censuring objectionable behavior after it has materialized. Content moderation operates in an ex ante manner by preventing targeted speech from reaching an audience, whereas ex post remedies are more diffuse, ranging from informal disavowals and condemnatory statements by other users or the platform itself to “de-platforming” offenders to, in extreme cases, formal legal actions.

We identify four factors that bear on the choice between ex ante prevention and ex post punishment of online speech. These are the closeness or fit between a substantive type of wrong and its codification in a rule; the potentially asymmetric costs of precaution; deliberative transparency; and the value of normative adaptation. Since these factors do not necessarily point in the same direction, the choice between ex ante prevention and ex post punishment of speech requires a substantive value judgment. That said, we argue that, in general, ex ante prevention is most appropriate when applied to tightly specified, high-stakes expressive acts, subject to stable and widely agreed upon norms.

The ‘black box’ quality of contemporary algorithmic tools raises concerns related to their use in court because of the law’s emphasis on explanations, transparency, and public reasons. We argue that the problems of explainability associated with contemporary algorithmic tools are, from a legal perspective, neither sui generis nor irreconcilable with existing norms. We distinguish between the types of explanations required by fact-finders and those required from judges. We conclude that apparent tensions can be reconciled by attending to the epistemic division of labour between the legal and scientific communities, contextualizing expert evidence appropriately, and distinguishing between explanation as reconstruction and as justification.

While punishment is a longstanding object of philosophical scrutiny, other controversial aspects of the justice system, such as policing, have flown under the radar. In this paper, we consider possible reasons why philosophers interested in crime and punishment have neglected policing. We make the case for a broader account of the political morality of the justice system, with a particular emphasis on policing. We sketch the outlines of an egalitarian version of such a theory, highlighting parallels between policing and the welfare state. Finally, we turn to recent calls for police abolition. Evaluating such radical claims requires, we argue, going beyond the traditional confines of the philosophy of punishment and considering a more holistic and interdisciplinary theory of criminal justice.

In this paper, I defend a deflationary account of proportionality, which suggests that proportionality does not explain anything valuable about a system of punishment. Proportionality, rather, is a conventional means for presenting judgments about whether punishment fits the crime. A system of punishment is proportionate to the degree that it coheres with widely shared norms about punishment. There are many reasons such coherence could be valuable, not all of which are retributive. Hence, while on a deflationary view it may be important for a system of punishment to be proportionate, proportionality does not identify a uniquely important retributive value. I motivate the argument for a deflationary account of proportionality by canvassing some of the problems associated with both relative (“ordinal”) and absolute (“cardinal”) proportionality and examining how both legal theorists and courts have actually used the concept. I focus on the work of Doug Husak, and the jurisprudence of the Supreme Court of Canada.

A frequently heard criticism, particularly in high-stakes settings such as criminal sentencing, is that automated decision-making aids can be highly opaque. In this chapter, the author analyzes two conceptions of transparency: transparency as a norm of publicity, and transparency as a norm of intelligibility. Automated decision-making aids plausibly raise concerns along both of these dimensions. However, the relative strength of these ethical concerns is mitigated on an incrementalist, rather than idealized, ethical analysis. This is because existing modes of human decision-making also suffer from defects of secrecy and intelligibility. In some cases, this is by design, as when officials are prohibited from revealing the bases of their decisions, and in some cases this is a feature of human psychology, as decision makers possess limited and partial insight into the reasons that actually explain why they decide as they do. Consequently, the strength of ethical objections grounded in transparency as publicity, and transparency as intelligibility, remains ambiguous.

The significance of On Crimes and Punishments for the social contract and utilitarian traditions is well studied. In this paper I consider two less well appreciated ways of reading Beccaria’s treatise. The first is in juxtaposition to a certain strand of American legal realism. Like the scientific realists, Beccaria was sceptical that law could be treated as an internally closed system of concepts and insisted that penal law be evaluated in terms of its observable effects. In this respect, his approach has more in common with proponents of “evidence-based policy” than with contemporary efforts at rational reconstruction in the criminal law. Although Beccaria espoused a formalistic mode of legal reasoning, he did not regard law as an autonomous science. Beccaria’s formalism was motivated by a theory of liberty: a strict division of labour between sovereign and judge, and a concern to protect citizens from arbitrary power. Seen from that perspective, Beccaria’s legal formalism presages the realists’ commitment to law reform.

Second, I suggest that Beccaria was a prophet of the modern administrative state. His remarks on crime and punishment reveal an attitude toward law as instrumental and bureaucratic, sceptical of case-by-case discretion and optimistic about the power of scientific administration. His vision of a just and rational approach to crime and punishment required that laws be rationally calculated to achieve independently specified ends, and that officials be strictly bound by law. In Weberian terms, this is a vision of law as impersonal, instrumentally rational and rule-bound. At the same time, Beccaria arguably overstated the power of formal rules, and did not fully appreciate how far the bureaucracies he foresaw would come to challenge his own formalistic conception of law.

On a popular understanding, the rule of law is valuable because it enables people to plan their lives. However, planning conceptions of the rule of law are undermined by the sheer quantity of legal rules, regulations, and policies characteristic of modern administrative states. Under conditions of hyperlexis, people cannot reasonably be expected to reliably use the law as a guide to conduct. Rather than conclude that the rule of law is inimical to the administrative state, however, I defend an alternative conception of the rule of law. On what I term a contestatory conception, the rule of law requires an adequate opportunity to challenge decisions made by officials in the exercise of their legal powers. The animating idea of a contestatory conception of the rule of the law is that officials should relate to citizens in the space of reasons rather than merely through the exercise of power.

Which interpretive tasks should be delegated to machines? This question has become a focal point of “tech governance” debates. One familiar answer is that while machines are capable of implementing tasks whose ends are uncontroversial, machine delegation is inappropriate for tasks that elude human consensus. After all, if human experts cannot agree about the nature of a task, what hope is there for machines?

Here, we turn this position around. When humans disagree about the nature of a task, that should be prima facie grounds for machine delegation, not against it. The reason has to do with fairness: affected parties should be able to predict the outcomes of particular cases. Indeterminate decision-making environments—those in which humans disagree about ends—are inherently unpredictable in that, for any given case, the distribution of likely outcomes will depend on a specific decision maker’s view of the relevant end. This injects an irreducible dynamic of randomization into the decision-making process from the perspective of non-repeat players. To the extent machine decisions aggregate across disparate views of a task’s relevant ends, they promise improvement on this specific dimension of predictability. Whatever the other virtues and drawbacks of machine decision-making, this gain should be recognized and factored into governance.

The essay has two parts. In the first, we draw a distinction between determinacy and certainty as epistemic properties and fashion a taxonomy of decision types. In the second part, we bring the formal point alive through a case study of criminal sentencing.

In this article, I consider the degree to which criminal justice interventions may be expected to ameliorate systemic corruption. I distinguish between two ideal types of corrupt actors – conditional cooperators and autonomous defectors – and argue that the prospects of reform through criminal justice are greatly affected by the relative preponderance of each type. When conditional cooperators predominate, the criminal law serves primarily to provide assurance that a perceived social norm is effective, in that the norm is both widely adhered to, and adhered to because people endorse the propriety of that norm. When autonomous defectors predominate, the criminal law serves primarily to deter would-be cheaters by attaching costs, at least in expectation, to cheating. Because patterns of compliance based upon a social norm tend to be self-reinforcing, unlike patterns of compliance motivated by fear of sanction, I argue that the prospects of sustainable reform through criminal justice interventions is likely to depend to a substantial degree upon convincing people to trust social norms rather than rely upon their private judgments of what is in their interest – that is, to become conditional cooperators.

Entrapment has been a prominent, if rarely successful, defence in terrorism prosecutions. In this chapter, I sketch an egalitarian case for entrapment. On this account, the primary moral significance of entrapment is to prevent the police from generating crimes that would not otherwise have been perpetrated. In a context in which most people are, as Richard McAdams puts it, “probabilistic offenders,” the power of the authorities to control the nature, frequency, and timing of an inducement to crime is the power to make criminals out of ordinary, but fallible, people. Entrapment is a means of constraining this power. In this regard, entrapment stands to undercover policing roughly as abuse of process stands to prosecutorial discretion: as a constraint on how officials choose which individuals to investigate, prosecute and punish. However, since judgments as to when this line is crossed are likely to be contestable, and since what is at issue is typically extraordinary state power used to ensnare particular individuals, I argue that courts should do more to encourage Parliament to regulate undercover policing ex ante rather than rely solely on an entrapment defence applied ex post, for instance by strictly applying an “authorized by law” condition in prosecutions based on undercover investigations.

Although capital punishment has been gradually disappearing from liberal democracies, philosophers remain divided as to its permissibility. The first part of this chapter considers arguments in favor of retention and abolition, with particular attention to recent contractualist arguments. I then consider the United States Supreme Court’s incrementalist approach, under the rubric of “evolving standards of decency.” On this view, the Constitution is limited to sweeping up stragglers; like Minerva’s owl, the Constitution announces a philosophy of punishment only in hindsight. The final section draws on Feinberg’s suggestion that punishment reflects a society’s particular “symbols of infamy.” A society may come to rethink its symbols of infamy, such as capital punishment, but that is unlikely to be because philosophical argument has silenced all reasonable doubt.

Over the last few years, legal scholars, policy-makers, activists and others have generated a vast and rapidly expanding literature concerning the ethical ramifications of using artificial intelligence, machine learning, big data and predictive software in criminal justice contexts. These concerns can be clustered under the headings of fairness, accountability and transparency. First, can we trust technology to be fair, especially given that the data on which the technology is based are biased in various ways? Second, whom can we blame if the technology goes wrong, as it inevitably will on occasion? Finally, does it matter if we do not know how an algorithm works or, relatedly, cannot understand how it reached its decision? I argue that, while these are serious concerns, they are not irresolvable. More importantly, the very same concerns of fairness, accountability and transparency apply, with even greater urgency, to existing modes of decision-making in criminal justice. The question, hence, is comparative: can algorithmic modes of decision-making improve upon the status quo in criminal justice? There is unlikely to be a categorical answer to this question, although there are some reasons for cautious optimism.

A basic principle in sentencing offenders is proportionality. However, proportionality judgments are often left to the discretion of the judge, raising familiar concerns of arbitrariness and bias. This paper considers the case for systematizing judgments of proportionality in sentencing by means of an algorithm. The aim of such an algorithm would be to predict what a judge in that jurisdiction would regard as a proportionate sentence in a particular case. A predictive algorithm of this kind would not necessarily undermine justice in individual cases, is consistent with a particularistic account of moral judgment, and is attractive even in the face of uncertainty as to the legitimate purposes of punishment.

An influential strain in the literature on state punishment analyzes the permissibility of punishment in exclusively deontological terms, whether in terms of an individual’s rights, the state’s obligation to vindicate the law, or both. I argue that we should reject a deontological theory of punishment because it cannot explain what is unjust about mass incarceration, although mass incarceration is widely considered—including by proponents of deontological theories—to be unjust. The failure of deontological theories suggests a minimum criterion of adequacy for a theory of punishment: it must take aggregation sufficiently seriously that it returns plausible results when scaled up from individual cases to large public institutions. In this vein, I briefly sketch a prioritarian metric for evaluating the use of custodial sanctions as a means of creating and allocating social advantage.

The traditional distinction between retributive and distributive justice misconstrues the place of the criminal law in modern regulatory states. In the context of the regulatory state, the criminal law is a coercive rule-enforcing institution – regardless of whether it also serves the ends of retributive justice. As a rule-enforcing institution, the criminal law is deeply implicated in stabilizing the institutions and legal rules by means of which a state creates and allocates social advantage. As a coercive institution, the criminal law requires justification as an instance of legitimate state authority. The operation of criminal justice institutions should therefore not be evaluated by reference to a distinct set of criteria, but should be evaluated by the same criteria that apply to coercive public institutions generally.

Philip Pettit’s conception of freedom as nondomination is modally robust in that it requires not simply reducing the probability of uncontrolled interference by others but entirely eliminating that possibility. In this article, I consider whether freedom as nondomination provides an attractive analysis of official discretion, particularly in the context of the criminal law, an area of recurring interest for Pettit. I argue that not only does the modally robust character of freedom as nondomination have some rather unattractive implications in the criminal law, but that it sits poorly with Pettit’s more general ambitions to provide a consequentialist framework for the evaluation of social institutions. Drawing on recent work by Niko Kolodny, I develop an ‘anti-deference’ interpretation of nondomination and contrast it to Pettit’s modally robust conception.

In this paper, I articulate two competing conceptions of the aims and subject matter of the criminal law. A private right conception views the criminal law as, in the first instance, a set of institutions meant to vindicate independent moral rights, and punishment as the uniquely appropriate moral response to invasion of those rights. A public law conception, in contrast, views the criminal law as the harshest, and crudest, means for enforcing compliance with legally enacted rules. From a public law point of view, the political legitimacy of the criminal justice system is a central philosophical problem for the philosophy of the criminal law in a way that it is not from a private right point of view. I go on to explore the contrast between the private right and public law conceptions in the context of preventive policing, an early development in the emergence of the welfare state. I suggest that the function of preventive policing – meant to prevent crimes ex ante rather than simply respond to them ex post – is more easily accommodated on a public law conception than it is on a private right conception, i.e. that preventive policing is more easily understood as an attempt to find less coercive means of promoting rule compliance rather than as a means of vindicating the invasion of private right.

This chapter examines the concept of actus reus as a basic, essential component of criminal liability. It considers a range of recent scholarly interpretations of actus reus and the extent to which they are supported in the case law, with particular reference to the Canadian and U.S. jurisprudence. It discusses minimalist and maximalist interpretations of actus reus, the first of which conceives of actus reus on the basis of whatever the legislature has decided to criminalize and the second of which restricts criminal liability to positive acts. The chapter looks at approaches that interpret actus reus based on two factors: a person’s “control” over the prohibited outcome or conduct, proposed by Husak, and the person’s practical reasoning, proposed by Duff. The chapter argues that both minimalist and maximalist views of actus reus conflict with well-established features of the criminal law.

The criminal law has at least two goals: to provide a degree of protection to a variety of individual and collective interests, and to communicate to those to whom it applies that those interests are protected. The question I consider is whether the criminal law should be used to advance the second goal independently of its use in advancing the first. Drawing on what I refer to as non-comparative egalitarianism, I argue that it should not. After developing a general argument for this claim, I turn to considering its implications for the criminalization of hate speech, focusing specifically on a line of argument found both in the Supreme Court of Canada’s s.2 jurisprudence as well as Jeremy Waldron’s recent book, The Harm in Hate Speech. I also briefly consider a structurally similar, but broader argument – recently defended by Alon Harel – which suggests that there is a constitutional duty to criminalize conduct that would, if engaged in, interfere with a person’s dominion over how her life goes, regardless of whether criminalization would or would not drive down the actual incidence of the targeted conduct. I claim that egalitarians should not recognize any such duty.

The United States Supreme Court has repeatedly insisted that what distinguishes a criminal punishment from a civil penalty is the presence of a punitive legislative intent. Legislative intent has this role, in part, because court and commentators alike conceive of the criminal law as the body of law that administers punishment; and punishment, in turn, is conceived of in intention-sensitive terms. I argue that this understanding of the distinction between civil penalties and criminal punishments depends on a highly controversial proposition in moral theory – namely, that an agent’s intentions bear directly on what it is permissible for that agent to do, a view most closely associated with the doctrine of double effect. Therefore, legal theorists who are skeptical of granting intention this kind of significance owe us an alternative account of the distinctiveness of the criminal law. I sketch the broad outlines of just such an alternative account – one that focuses on the objective impact of legislation on a class of protected interests, regardless of the state’s motivations in enacting the legislation. In other words, even if the concept of punishment is unavoidably intention-sensitive, it does not follow that the boundaries of the criminal law are likewise intention-sensitive, because the boundaries of the criminal law may be drawn without reference to the concept of punishment. I conclude by illustrating the application of this view to a pair of well-known cases, and noting some of its ramifications.

In Furman v. Georgia, the United States Supreme Court announced that it would not tolerate a capital sentencing regime that imposed death sentences in a seriously arbitrary fashion. The question I ask in this paper is whether we should in fact object to arbitrariness in punishment. The answer I propose is that under plausibly adverse conditions, we might not object to arbitrary penal outcomes, because under those conditions a fair distribution of punishment would be one that equalizes chances across a class of similarly situated criminals. In particular, fairness may require no more than a rough equalization of ex ante chances under conditions of resource scarcity, an inability to rank claims reliably by comparative desert, and a pressing need for punishment to be imposed. I call this an ex ante theory of fairness. The central virtue of ex ante fairness is that it is capable of reconciling parsimony in punishment with equity in its distribution, even when claims about who deserves what are deeply contested. Adopting an ex ante standard of fairness means that a concern for fair treatment of the guilty need not blind us to the realities of the severe resource constraints faced by American criminal justice, and vice versa.

Anglo-American criminal law traditionally demands a criminal purpose for an attempt conviction, even when the crime attempted requires only foresight or recklessness. Some legal philosophers have defended this rule by appeal to an alleged difference in the “moral character” or “intentional structure” of intended versus non-intended harms. I argue that there are reasons to be skeptical of any such differences; and that even if conceded, it is only on the basis of an unworkable view of criminal responsibility that such a distinction would support a rule restricting attempts to criminal purpose. I defend instead the “continuity thesis,” according to which attempts are functionally continuous with endangerment offenses: both are legal efforts to regulate unreasonably dangerous conduct. The upshot of the continuity thesis is that there is little substantive difference between attempt and endangerment in principle, no matter how they are labeled in law.

This paper offers a critical reconsideration of the traditional doctrine that responsibility for a crime requires a voluntary act. I defend three general propositions: first, that orthodox Anglo-American criminal theory (as represented by Michael Moore’s updating of Austin’s volitionalism) fails to explain adequately why criminal responsibility requires an act. Second, when it comes to the just definition of crimes, the act requirement is at best a rough generalization rather than a substantive limiting principle. Third, that the intuition underlying the so-called “act requirement” is better explained by what I call the “practical-agency condition,” according to which punishment in a specific instance is unjust unless the crime charged was caused or constituted by the agent’s conduct (broadly understood) qua practically rational agent. The practical-agency condition is defended as a reconstruction of what is worth retaining in Anglo-American criminal law’s traditional notion of an “act requirement.”