Introduction
Law students almost always encounter criminal law in their first year, frequently in the first semester. The traditional course in criminal law introduces students to the way the law approaches deep moral questions. What justifies the state’s deliberate imposition of suffering on offenders? What conduct should be criminalized? The criminal law does not (and should not) punish everything that is wrong or harmful — so what are the limits? What are the general conditions of criminal liability? Before anyone can be punished, we need an account of what must be proved: a voluntary act, a culpable state of mind, and the absence of justification or excuse.
These questions correspond to the three main divisions of the field: the theory of punishment, the theory of criminalization, and the theory of what criminal lawyers call the “general part.” This Lexicon entry provides an introduction to the central questions of criminal law theory. As always, the Lexicon is aimed at law students, especially first-year law students, with an interest in legal theory.
Theories of Punishment
The justification of punishment is the oldest and deepest question in criminal law theory, and the debate is organized around a fundamental divide: does punishment look backward or forward? This divide is a special case of the distinction between ex post and ex ante perspectives, explored in a prior Lexicon entry (Legal Theory Lexicon 001: Ex Ante & Ex Post). Backward-looking theories such as retributivism justify punishment by reference to the offender’s past wrongdoing — the offender deserves to be punished because of what she did. Forward-looking theories such as deterrence justify punishment by reference to its future consequences — punishment is justified because it prevents crime and thereby produces good outcomes. This divide corresponds to the two great traditions in moral philosophy: deontology on the backward-looking side and consequentialism on the forward-looking side. Both traditions are explored in prior Lexicon entries (Legal Theory Lexicon 008: Utilitarianism and Legal Theory Lexicon 010: Deontology).
The most prominent backward-looking theory is called “retributivism.” The core retributivist idea is that punishment is justified because offenders deserve it: wrongdoing creates a moral desert basis for punishment, and the state acts permissibly (perhaps even obligatorily) when it gives offenders what they deserve. Immanuel Kant is the classic historical source — Kant famously insisted that even a society about to dissolve itself must first execute its last murderer, so that everyone receives what their deeds deserve. The leading contemporary retributivist is Michael Moore, whose book Placing Blame offers a sustained defense of the view that desert is not merely a necessary condition for punishment but a sufficient one: the state should punish culpable wrongdoers because they deserve it, full stop. Notice what retributivism does not say. Retributivism is not the view that punishment satisfies the victim’s desire for revenge, and it is not committed to harsh punishment — the retributivist principle of proportionality (punishment must fit the crime) can condemn excessive sentences just as forcefully as it condemns lenient ones.
The forward-looking alternatives to retributivism are consequentialist theories, and they usually focus on the idea of deterrence. Bentham’s classic formulation captures the structure: punishment is itself an evil — it inflicts suffering — and so it can be justified only if it prevents greater evils. The most familiar preventive mechanism is deterrence: the threat of punishment gives potential offenders a prudential reason to comply with the law, and punishing actual offenders makes the threat credible. But deterrence is not the only forward-looking rationale. Incapacitation prevents crime by physically restraining offenders — a burglar in prison cannot burgle. And rehabilitation aims to change offenders so that they no longer want to offend. All three rationales share the consequentialist structure: punishment is justified by the good consequences it produces. And that structure generates the standard objection. If good consequences justify punishment, then the theory seems to permit punishing the innocent whenever doing so would deter others — framing a scapegoat, for example, to quell a crime wave. Consequentialists have responses, but the scapegoat objection has convinced many theorists that consequences alone cannot be the whole story.
The scapegoat objection helps to explain the enduring appeal of mixed theories, which combine backward-looking and forward-looking elements. The most influential mixed theory belongs to H.L.A. Hart. In his famous essay “Prolegomenon to the Principles of Punishment,” Hart distinguished the “general justifying aim” of punishment from the principles governing its “distribution.” The general justifying aim of the institution of punishment, Hart argued, is forward-looking: we have systems of criminal punishment because they prevent crime. But the distribution of punishment is constrained by backward-looking principles: only the guilty may be punished, and only in proportion to their culpability. Desert operates as a side constraint on the pursuit of good consequences. Hart’s move — asking different questions about the institution as a whole and about its application to individuals — remains the standard framework for organizing the punishment debate.
A fourth family of theories emphasizes the meaning of punishment rather than its consequences or the offender’s desert. Joel Feinberg’s classic essay “The Expressive Function of Punishment” observed that punishment differs from a mere penalty (a parking fine, a tax) because punishment communicates a message of condemnation. Expressive theories build on this insight, arguing that condemnation is part of what justifies punishment. The most developed version is R.A. Duff’s communicative theory. For Duff, punishment is a communicative enterprise between the political community and the offender: it aims to communicate to offenders the censure they deserve and to persuade them to repent, reform, and reconcile with the community. Duff’s theory is backward-looking in its foundation (censure must be deserved) but aspirational in its aims — and it treats offenders as responsible members of the community, not as objects to be managed.
Finally, some theorists challenge the entire framework. Restorative justice theorists argue that the central question should not be “how do we justify punishment?” but “how do we repair the harm?” — and they favor practices (victim-offender mediation, restitution, community conferencing) that address crime without the deliberate infliction of suffering. Prison abolitionists press further, arguing that the institution of imprisonment is so unjust in practice that it cannot be reformed and should be dismantled. These views remain minority positions in the theoretical literature, but they have grown in influence, and they perform an important function even for those who reject them: they force defenders of punishment to shoulder the justificatory burden rather than resting on the familiarity of existing institutions.
Theories of Criminalization
Suppose we have a justification for punishment. A second question immediately follows: what conduct may the state punish? Murder, obviously. But what about drug possession? Gambling? Blasphemy? Failing to rescue a drowning stranger? The theory of criminalization asks what the limits of the criminal law are — and whether there is any conduct that the state simply may not criminalize, no matter how many people want it criminalized.
The classic starting point is John Stuart Mill’s harm principle. In On Liberty, Mill argued that the only purpose for which power can rightfully be exercised over a member of a civilized community, against his will, is to prevent harm to others. On this view, the fact that conduct is immoral, offensive, or harmful to the actor herself is simply not a reason for criminalization. The most influential development of this liberal tradition is Joel Feinberg’s four-volume work, The Moral Limits of the Criminal Law. Feinberg distinguished four “liberty-limiting principles” — four kinds of reasons that might be offered in favor of criminalization. The first is the harm principle: conduct may be criminalized because it harms others. The second is the offense principle: conduct may be criminalized because it seriously offends others — think of public indecency, which may harm no one but is nonetheless prohibited. The third is legal paternalism: conduct may be criminalized because it harms the actor herself — seatbelt laws and drug prohibitions are often defended this way. The fourth is legal moralism: conduct may be criminalized simply because it is immoral, even if it harms and offends no one. Feinberg accepted the first two principles (with important qualifications) and rejected the last two. His four-part framework is enormously useful: almost any debate about criminalization — of drugs, of sex work, of hate speech — can be organized by asking which liberty-limiting principle is doing the work.
Legal moralism, however, has its defenders. Michael Moore — whose retributivism we have already encountered — argues that the criminal law’s function is to punish moral wrongdoing as such; for Moore, the immorality of conduct is always a reason (though not necessarily a conclusive reason) for criminalizing it, and countervailing considerations like privacy and liberty do the work of limiting the criminal law’s reach. R.A. Duff defends a more moderate version: on Duff’s view, criminalization is appropriate only for “public wrongs” — wrongs that properly concern the political community as a whole, not merely the individuals involved. The debate between liberal and moralist theories of criminalization remains one of the liveliest in the field.
Whatever one’s theory, there is widespread agreement that contemporary American law suffers from overcriminalization. Douglas Husak’s book of that title documents the phenomenon: thousands of federal crimes (no one knows the exact number), vast bodies of regulatory offenses carrying criminal penalties, and criminal statutes so broad that prosecutors effectively decide who goes to prison. Husak argues that criminalization requires justification because punishment does — every criminal statute is a standing threat to deprive citizens of liberty, and the state needs a good reason for each one. Overcriminalization also connects the theory of criminalization back to the theory of punishment: if most criminal cases are resolved by plea bargaining in the shadow of overlapping statutes, the theorist must ask whether the practice of American criminal justice can be justified by any of the theories of punishment on offer.
Critical theories of criminal law approach the criminalization question from a different direction. Rather than asking which liberty-limiting principles justify criminalization in the abstract, critical theorists ask how criminalization decisions actually operate — and their answer is that the criminal law’s reach has been shaped by race and class hierarchy. The historical record supplies the evidence: vagrancy laws, the differential treatment of crack and powder cocaine, and the concentration of drug enforcement in poor communities of color are standard examples. Abolitionist theorists draw a further conclusion: the question is not which conduct deserves criminalization but whether criminalization should be the state’s default response to social problems at all — and they argue for shifting resources from policing and prisons toward housing, health care, and education as the primary means of producing safety. For the theory of criminalization, the critical challenge is important even for those who reject its conclusions, because it insists that a theory of what may be criminalized in principle must reckon with how criminalization works in practice.
The General Part
Criminal lawyers distinguish the “special part” of the criminal law — the definitions of particular offenses like murder, burglary, and theft — from the “general part”: the doctrines that apply across all offenses. The general part includes the requirement of a voluntary act, the requirement of a culpable mental state, and the defenses of justification and excuse. Each of these doctrines raises deep theoretical questions, because each embodies a view about the conditions under which persons are responsible for wrongdoing.
Start with the act requirement — what criminal lawyers call the “actus reus.” The criminal law does not punish thoughts, character, or status; it punishes acts. Why? One answer is practical — thoughts are hard to prove. But the deeper answers are theoretical: punishing thoughts would be an intolerable intrusion on liberty, and mere thoughts, unlike acts, do not wrong anyone. Michael Moore’s book Act and Crime offers the most sustained philosophical treatment, defending the view that the act requirement rests on a theory of human action: acts are willed bodily movements, and only through action does an agent’s practical reason engage the world in a way that can constitute wrongdoing. The standard complication is liability for omissions. The criminal law generally imposes no duty to rescue — the passerby who watches a stranger drown commits no crime — but it does punish omissions when there is a legal duty to act, as when a parent fails to feed a child. Whether the act requirement and omission liability can be reconciled within a single theory of criminal conduct remains a contested question.
Next, the mens rea requirement — the actus reus must be accompanied by a culpable mental state. The common law’s mens rea vocabulary was notoriously chaotic, and the Model Penal Code’s great achievement was to reduce it to four levels of culpability: purpose, knowledge, recklessness, and negligence. Most students encounter this hierarchy as doctrine, but it rests on a theory: culpability tracks the actor’s practical attitude toward the interests of others. The purposeful wrongdoer aims at harm; the knowing wrongdoer accepts it as a certainty; the reckless wrongdoer consciously disregards a substantial risk; the negligent wrongdoer fails to perceive a risk that a reasonable person would have perceived. The theoretical controversies cluster at the bottom of the hierarchy. Negligence liability is contested because the negligent actor, by definition, was unaware of the risk — and some theorists, most prominently Larry Alexander and Kimberly Kessler Ferzan, argue that culpability requires a conscious choice: on their view, recklessness is the core of culpability, and the negligent actor, who chose nothing, deserves no punishment at all. Strict liability — criminal liability without any culpable mental state at all — is more contested still: on most theories of punishment, and certainly on retributivist theories, punishing the blameless is simply unjust. Yet strict liability offenses persist, especially in the regulatory sphere, and their persistence is one of the standing puzzles of criminal law theory.
The general part also includes the defenses, and here the key theoretical distinction is between justification and excuse. A justified actor did the right thing (or at least a permissible thing) in the circumstances: the classic example is self-defense, where the defender’s use of force is not wrongful at all. An excused actor did something wrongful but is not blameworthy for doing it: insanity and duress are the standard examples. The distinction matters theoretically because it separates two different questions — was the conduct wrong? and is the actor responsible? — that the verdict “not guilty” otherwise runs together. The distinction can also matter practically: third parties may assist justified conduct but not excused conduct, and justifications generalize to anyone in the same circumstances while excuses are personal to the actor.
Finally, consider a puzzle that connects the general part back to the theory of punishment: the problem of moral luck. Suppose two drivers text behind the wheel; one arrives home safely, the other kills a pedestrian. Their conduct and culpability are identical — the difference between them is pure luck. Yet the law punishes the killer severely, while the lucky driver might be charged with a minor offense for texting or get off scot free. The same puzzle appears in the law of attempts: the assassin whose shot misses is typically punished less than the assassin whose shot hits, though the two are equally culpable. Should outcomes matter to punishment, or only culpability? Subjectivists — Alexander and Ferzan again — say culpability is everything and the law’s harm-based grading is an indefensible concession to primitive intuitions; objectivists reply that results matter morally — that causing harm is a different (and worse) thing than risking it. Michael Moore, in Causation and Responsibility, defends the objectivist position: causing harm increases blameworthiness, and the law’s differential treatment of completed crimes and attempts is therefore no embarrassment. The moral luck debate is ongoing and complex, and it is a fine example of how a doctrinal detail — the grading of attempts — opens onto some of the hardest questions in moral philosophy.
Conclusion
Criminal law theory begins with a question that the first-year course mostly takes for granted: what entitles the state to punish? From that question, the field radiates outward — to the limits of criminalization, and to the general conditions of responsibility that the doctrines of actus reus, mens rea, justification, and excuse embody. The three bodies of theory are connected. A retributivist about punishment will be drawn toward moralism about criminalization and toward culpability-centered views of the general part; a consequentialist will see deterrence, and not desert, doing the work at every level. R.A. Duff’s work displays the connections especially clearly: his communicative theory of punishment, his account of crimes as public wrongs, and his theory of criminal responsibility as answerability to the political community are all pieces of a single, unified vision of criminal law. One of the pleasures of criminal law theory is discovering these connections — and finding that a position taken on the first day of class, about why we punish at all, has consequences that reach into every corner of the course. I hope this entry provides a useful starting point for your own exploration of these questions.
Related Lexicon Entries
Legal Theory Lexicon 001: Ex Ante & Ex Post
Legal Theory Lexicon 008: Utilitarianism
Legal Theory Lexicon 010: Deontology
Legal Theory Lexicon 018: Justice
Legal Theory Lexicon 020: Causation
Legal Theory Lexicon 022: Intention
Legal Theory Lexicon 113: Property Theory
Legal Theory Lexicon 114: Contract Theory
Legal Theory Lexicon 115: Tort Theory
Legal Theory Lexicon 116: Civil Procedure Theory
Legal Theory Lexicon 117: Constitutional Theory
Bibliography
Larry Alexander & Kimberly Kessler Ferzan (with Stephen J. Morse), Crime and Culpability: A Theory of Criminal Law (Cambridge University Press 2009).
Jeremy Bentham, An Introduction to the Principles of Morals and Legislation (1789) (Dover ed. 2007).
R.A. Duff, Answering for Crime: Responsibility and Liability in the Criminal Law (Hart Publishing 2007).
R.A. Duff, Punishment, Communication, and Community (Oxford University Press 2001).
Joel Feinberg, The Expressive Function of Punishment, 49 The Monist 397 (1965).
Joel Feinberg, Harm to Others: The Moral Limits of the Criminal Law, Volume One (Oxford University Press 1984).
H.L.A. Hart, Punishment and Responsibility: Essays in the Philosophy of Law (2d ed., Oxford University Press 2008).
Douglas Husak, Overcriminalization: The Limits of the Criminal Law (Oxford University Press 2008).
Allegra M. McLeod, Prison Abolition and Grounded Justice, 62 UCLA L. Rev. 1156 (2015).
John Stuart Mill, On Liberty (1859) (Hackett ed. 1978).
Michael S. Moore, Act and Crime: The Philosophy of Action and Its Implications for Criminal Law (Oxford University Press 1993).
Michael S. Moore, Causation and Responsibility: An Essay in Law, Morals, and Metaphysics (Oxford University Press 2009).
Michael S. Moore, Placing Blame: A Theory of the Criminal Law (Oxford University Press 1997).
Thomas Nagel, “Moral Luck,” in Mortal Questions (Cambridge University Press 1979).
Dorothy E. Roberts, Foreword: Abolition Constitutionalism, 133 Harv. L. Rev. 1 (2019).
(This entry was created on July 5, 2026.)
Link to the Most Recent Version of this Lexicon Entry
Legal Theory Lexicon 118: Criminal Law Theory
Link to the Legal Theory Stack
The Legal Theory Stack informs readers of a new or revised Legal Theory Lexicon entry every Sunday along with summaries of the Download of the Week and the Legal Theory Bookworm recommendation. Subscribe to the Legal Theory Stack here.