Skip to main content Skip to Search Box

Definition: punishment from The Penguin English Dictionary

the act or an instance of punishing.


something done in order to punish; a judicial penalty

No punishment has ever possessed enough power of deterrence to prevent the commission of crimes

Hannah Arendt.


informal rough or damaging treatment

The contenders both took plenty of punishment in the last round


Summary Article: Punishment from Encyclopedia of Community Corrections

Punishment can be defined as a penalty imposed under the authority of law, inflicted on a person by other persons, for the commission of a crime. This definition of punishment, however, raises a host of attendant legal and moral questions. First, can the imposition of state punishment be morally justified? Second, under what specific circumstances is it warranted? Third, what kinds of punishment should be imposed and how much of it? Philosophers have debated these questions for centuries, and as the United States grapples with serious problems of prison overcrowding and mass incarceration, punishment looms as a—controversial—public policy issue of great urgency. Changing social attitudes and rapidly evolving legal practices mean that many questions about punishment remain unanswered, despite the serious financial and social consequences of those answers.

Consensus Definition of Punishment

The term punishment is used in a variety of contexts. People reference the state's “punishment” of criminals, but they also reference the “punishment” of children by parents and teachers. The term is also used to denote severe handling or treatment, such as that endured by boxers and soldiers. Even extreme heat or cold can be described as “punishing.” Within the context of criminal justice, however, the term is used to signify a particular sort of state behavior. Most commentators have agreed that punishment is both aversive and state-sanctioned. Thomas Hobbes described it as an evil imposed by public authority; Cesare Beccaria noted that punishment is not an act of violence of one individual against another but is inherently public, necessary, proportionate to the crime, and imposed under law. In his “Prolegomenon to the Principles of Punishment,” H. L. A. Hart outlined a jurisprudential definition of punishment by identifying five characteristics: (1) It must involve pain or other consequences normally considered unpleasant; (2) it must be for an offense against legal rules; (3) it must be of an actual or supposed offender for his offense; (4) it must be intentionally administered by human beings other than the offender; and (5) it must be imposed and administered by an authority constituted by a legal system against which the offense is committed.

Many scholars also agree that stigma is another essential aspect of punishment. While a hospital patient may be quarantined for the same duration as the prisoner who serves a criminal sentence, there is condemnation inherent in the detention of the criminal.

The Moral Justification of Punishment

Although a handful of scholars have argued for the wholesale abolition of punishment, most systems of law incorporate mechanisms of its administration. What most philosophers, jurists, and legislators have debated is not whether punishment is justifiable, but on what basis it is justified. Two fundamentally different, and often conflicting, rationales have been advanced: consequentialist theories (that is, utilitarian bases of punishment) and nonconsequentialist theories (that is, retributivist bases of punishment). Consequentialist theories are prospective and instrumentalist, whereas nonconsequentialist theories are retrospective and intrinsicalist.

Consequentialist Theories of Punishment

Consequentialist theories of punishment focus on future crimes. As far as consequentialists are concerned, a crime, once committed, is finished. While steps can be taken to ameliorate its effects, the crime itself cannot be undone. Thus, the proper aim of punishment is the reduction of future crimes. In the Protagoras, Plato argued that when punishing, authorities do not punish a man for past wrongs unless they are wreaking blind vengeance. Instead, he argued, rational men inflict punishment to prevent future offending. Cesare Beccaria shared this view. In On Crimes and Punishments, Beccaria argued that the end of punishment is to prevent the criminal from inflicting additional harm and to prevent others from committing like offenses. Perhaps the best known of the consequentialist theorists, however, was Jeremy Bentham, who in The Principles of Morals and Legislation devised an elaborate hedonic calculus of pleasure and pain. Bentham aspired to create a system of laws that maximized the greatest good for the greatest number of people. Crimes were evils that caused suffering to their victims and rational laws were a mechanism to minimize them. However, Bentham also cautioned that punishment is also a pain, and he urged legislators to weigh the benefits of reduced crime rates against the costs of increased punishment. It is possible, he warned, to overpunish and to create unnecessary suffering in the world. Excess or useless punishment is an evil to be avoided.

While Bentham's hedonic calculus may prove too cumbersome for practical application, the relation of punishment to measurable outcomes (such as levels of pleasure and pain, crime rates, or recidivism) can help to gauge whether a particular punishment achieves its measurable social end. Policy initiatives, in this sense, are falsifiable. For example, lawmakers can know whether three-strikes penalties empirically reduce rates of violent crime.

Of course, while the consequentialist approach is philosophically attractive in many ways, it does raise its own set of difficult questions. For example, if the general justifying aim of punishment is the reduction of future crime, why is the imposition of punishment conditioned upon a past violation of the criminal law? Could punishment be imposed with no crime to trigger it? Should, for example, an innocent man be executed if his death will produce the greatest amount of pleasure for the greatest number of people? Some utilitarians (“rule utilitarians”) avoid the problem by asserting that it is not an individual act that should be assessed in terms of social utility, but the system of rules that would produce the greatest net good. Other utilitarians (“act utilitarians”), however, do not have recourse to this argument. Instead, they might suggest that the punishment of innocent parties would lead to widespread skepticism about the legitimacy of the punishing authority (thereby making the punishment of the innocent a nonoptimal social action), while others might suggest that the innocent defendant should be executed and that doing so offends moral intuitions only because people do not often confront such unusual circumstances.

Consequentialists typically justify punishment by identifying specific means of reducing or ameliorating crime. Most commonly, they suggest that future crime can be reduced through deterrence (in both its specific and its general forms), incapacitation, and rehabilitation, although consequentialists also describe the utilitarian values of norm reinforcement, vengeance, restitution, and restoration.

Specific Deterrence

Most consequentialists believe that humans, as rational actors, choose behaviors that maximize pleasure and minimize pain. To induce people to forgo the pleasures of crime (such as the taking of goods or services without payment, the euphoria of illegal drugs, and the killing of people who make us angry), the state must credibly enforce the penalties attached to crimes. One way that people can know the state will enforce its threats is if it imposes punishment. If an offender commits a crime and is punished for it, that offender will be less likely to reoffend. In this way, the theory of specific deterrence closely resembles the psychological theory of operant conditioning, in which the introduction of rewarding stimuli or the withdrawal of aversive stimuli increases the frequency of a behavior, or vice versa.

Psychologists of learning would recognize many of the principles of punishment advocated by early penologists. For example, Beccaria's suggestion that punishment should follow the crime without delay is supported by research showing that immediate feedback is more effective than less immediate feedback. His claim that the certainty of punishment, not its severity, leads to lawfulness is supported by research showing that consistent reinforcement and punishment lead to faster learning. His suggestion that penalties should be proportional to crimes is supported by research showing that costs and benefits are weighed against each other in choosing behavior. Criminologists have reported strong support for a deterrent effect.

General Deterrence

Of course, as suggested by social learning psychologists, people learn from the rewards and punishments visited upon them not only directly but also vicariously. Thus, punishment imposed upon an offender may deter future crime not just for him but also for others. It has been argued that it is not justice to punish one man in order to influence the behavior of others. Georg Hegel claimed that justifying punishment in this way was like raising a stick at a dog, treating people as animals instead of with the dignity they should be afforded. Criminologists have found empirical support for general deterrence. In some cases, however, specific findings can be very contentious. For example, Isaac Erlich's 1975 study, concluding that each execution carried out deterred approximately eight murders, has been the object of fierce criticism.


The theory behind incapacitation is simple: Punishment can make it difficult (or impossible) for an offender to commit further crimes. Offenders can be incapacitated by spatially removing them from society (such as imprisonment or banishment) or by physically eliminating the capacity to offend (such as cutting off the hand of a thief or chemically castrating sex offenders). Some forms of punishment are more incapacitating than others: The death penalty is the ultimate incapacitant, whereas the incapacitative effects of incarceration are less certain (prisoners can escape or commit crimes against fellow inmates or corrections staff) and those associated with financial penalties and community supervision are less certain still.

Selective incapacitation is the practice of severely punishing persistent and dangerous offenders, incarcerating a modest number of people to achieve dramatic reductions in crime rates. Because criminologists have sometimes reported that a handful of chronic offenders are responsible for a disproportionate amount of crime (6 percent of the population may be responsible for 52 percent of reported crime), Peter Greenwood proposed an approach whereby chronic offenders could be identified and sentenced to extended terms of incarceration. The promise of reducing crime at minimal financial and social cost was attractive to many policy makers, but limitations and high error rates in the prediction of risk led others to challenge this approach. Today, most selective incapacitation operates in a more rudimentary way, such as the enhancement of sentences for repeat offenders under sentencing guidelines and three-strikes regimes.


In 1876, Cesare Lombroso wrote Criminal Man, in which he suggested that criminals are throwbacks to an earlier stage of evolution and that crime is the product of an attenuated form of epilepsy. The concept of rehabilitation is rooted in Lombroso's model of crime-as-disease. If people commit crimes because of inherent defects, one straightforward way to reduce future crime is to simply correct the defect, regardless of whether the defect is physical (such as a chemical imbalance), psychological (such as criminal thinking patterns), or social (such as association with criminal peers). During the 1960s and 1970s, many experts believed that crime was a social problem that could be diagnosed, treated, and cured. In 1974, however, catalyzed by Robert Martinson's pessimistic article “What Works?” existing dissatisfaction about the state of U.S. punishment fomented a wholesale rejection of rehabilitation as a legitimate basis of punishment. Only in recent years has genuine interest in rehabilitation reemerged within mainstream policy circles.

Norm Reinforcement

Although norm reinforcement is not one of the cornerstone foundations of punishment, it is another important consequentialist basis for punishment. Imposing punishment can build solidarity among members of society, permitting people to internalize the sense that crime will not be tolerated. If crimes appear to go unpunished, it may seem unfair to law-abiding citizens. They may be tempted to violate the laws as well, and to enjoy the advantages of crime. However, if transgressors are punished for their violations, law-abiding citizens may be more likely to view their acquiescence to the law as just and fair.


Vengeance is another consequentialist basis for punishment. When criminals steal from or injure victims, those victims, their families and friends, and members of their communities may feel resentful and angry. The imposition of punishment may prove cathartic for them (increasing their pleasure or eliminating their pain, under the hedonic calculus) and may discourage vigilantism and private acts of revenge. Vengeance is not synonymous with retribution (described below) and is not premised upon a theory of intrinsic moral rights; rather, it is based on the view that the punishment of the criminal may bring relief to the victim and the public, enhancing respect for the law and discouraging vigilantism.

Restitution and Restoration

Restitution, or the making whole of crime victims, is yet another consequentialist basis for punishment. When criminals commit offenses, their victims are harmed (financially, physically, emotionally, and/or socially). Punishments that include restitution may help to ameliorate those harms by requiring the criminal to compensate the victim (usually financially). In addition to repaying victims for their losses, restitution may foster respect for the law (since the victim is acknowledged as relevant to the imposition of punishment, which often seems more equitable than excluding victims from the process).

A popular idea in recent years, restoration consists of requiring offenders to perform compensatory service for the victim and/or the community, and it often incorporates processes through which the offender is reconciled with victims and the broader community. Restorative justice can help to make victims whole, but it can also help to knit neighborhoods together. If offenders are contrite and demonstrate genuine repentance, the community may be willing to forgive and welcome them back. In this way, restoration encompasses many of the qualities associated with norm reinforcement, vengeance, and restitution.

Nonconsequentialist Theories of Punishment

Instead of justifying punishment through the advantageous effects it produces for society, the non-consequentialist justifies punishment because it is deserved; instead of attempting to reduce prospective crimes, the nonconsequentialist adopts a retrospective view and punishes the criminal for the crime that has been committed.

In effect, the nonconsequentialist justification of punishment (sometimes called retributivism or “just deserts”) argues that criminals should be punished because they deserve it. Herbert Morris articulated this view of fairness theory in 1968. He suggested that the criminal has taken advantage of society and stands in breach of the social contract, but the imposition of punishment on the criminal annuls the unfair advantage and levels the playing field. Using an elegant logical twist, Georg Hegel explains this view by suggesting that when the criminal violates the law, his crime is the negation of the right of society. Punishment is the negation of this negation, and thus an affirmation of right, solicited and brought upon the criminal by himself. This notion of inflicting harm to redress an unfair advantage is central to nonconsequentialist theories of punishment and underlies the very ancient doctrine of lex talionis, the principle that the punishment should be identical to the offense (“an eye for an eye”). Lex talionis featured prominently in the Code of Hammurabi (law number 196 states, “If a man puts out the eye of an equal, his eye shall be put out”) and ancient Roman and Islamic law (similar assertions are made in Leviticus 24:19-21 and Qur'an 5:45).

There are two varieties of retributivism: Negative retributivism requires that only the guilty may be punished, and then only to the extent they deserve, but it does not require that punishments must be imposed; positive retributivism, on the other hand, requires that guilty must be punished, to the full extent of their desert. The view of positive retributivism has been forcefully articulated by Immanuel Kant, who argued that even if an island society was about to disband, the last murderers lying in its prisons should be put to death. Society not only has a right to punish them; it has a duty to do so. Otherwise, Kant claims, the islanders who refuse to execute the killers are complicit in their crimes.

Unlike the utilitarian, who is concerned with maximizing the good for the greatest number of people, the retributivist might insist upon the imposition of an appropriate punishment even if it produced undesirable social consequences: more crime, less happiness, and diminished overall social welfare. This view is expressed in the Latin maxim, Fiat justitia ruat caelum (May justice be done though the heavens fall).

The nonconsequentialist approach to crime is certainly attractive in some philosophical ways. By punishing past crimes, it avoids the utilitarian conundrum of relying upon antecedent crimes to influence subsequent conduct. By asserting that only the guilty may be punished to the extent they deserve, it avoids the punishment-of-the-innocent problem. The idea that people should be punished because they are morally blameworthy may be intuitively appealing. However, nonconsequentialists must face their own questions. First, there is the problem of free will. If crime is a product of causes that we do not choose, it is difficult to understand how the criminal can be punished as a morally blameworthy actor. It may be the case that, as Herbert Packer has written, the law treats people as autonomous not because they are, but because it is useful to proceed as if they were. A second problem for retributivists is this: Even if one accepts fairness theory (punishment redresses unfair advantage), it is unclear why punishment should consist of hard treatment (such as physical pain, imprisonment, or execution). Third, why is punishment a matter for the state? Presumably, when a criminal assumes an unfair advantage by committing a crime, it is the victim—not the government— who should see the balance restored. Fourth, while the utilitarian can measure the outcomes of punishment (such as crime rates) to know if punishment is achieving its intended social ends, it is not clear how the retributivist can know how much punishment to impose. Andrew von Hirsch has studied this problem at length and has suggested that the principle of proportionality can provide meaningful guidance. If crimes of like gravity are punished by penalties of like severity, parity can be maintained. If crimes can be rank-ordered in their seriousness (for example, murder is more serious than armed robbery, which is more serious than burglary), a schedule of retributive penalties can be devised. This approach, however, does not answer fundamental questions about where the top and bottom of the penalty range should be.

Other Theories of Punishment

Consequentialist and nonconsequentialist theories of punishment frequently conflict. For example, the very risk factors that make a criminal dangerous (such as poverty, drug addiction, and a hot temper) are often factors that make the criminal seem less blameworthy. However, neither conse-quentialist nor nonconsequentialist theories seem wholly satisfying, and for this reason many scholars have suggested hybrid theories of punishment. Norval Morris's concept of limiting retributivism is one well-known example. Under Morris's theory, retributivist concepts set upper (and occasionally lower) limits on sentencing severity. Within these outer limits, however, utilitarian principles (such as deterrence, incapacitation, rehabilitation, uniformity in sentencing, and imposing the least severe sentence that will achieve the goals of punishment) provide the necessary fine-tuning of the sentence imposed. Joel Feinberg's expressive theory of punishment is neither entirely consequentialist nor nonconsequentialist. Rather, it suggests that punishments communicate moral opprobrium to offenders, relating social attitudes about criminal behavior. Other theorists have posited variations on Feinberg's theory, justifying punishment in terms of its communicative and educational function.

Modes of Punishment

Legal punishment is as old as the law, but it has changed dramatically over time. For most of human history, punishment was corporal in nature. Serious crimes were punished by execution (such as hanging, crucifixion, and drawing-and-quartering), and less serious crimes were punished with nonlethal physical penalties (such as amputation, whipping, and the use of devices like ducking stools and pillories). Prisons existed, but they served principally as places of detention for those who could not post bail or pay their fines. Historically, it was said that carcer enim ad continendos homines non ad punien-dos haberi debet (prisons exist only in order to keep men, not to punish them).

Between about 1750 and 1850, however, punishment underwent a remarkable transformation in Western societies. Prompted by humanitarian reformers in England and America, the prison was changed from where criminals awaited punishment to the manner of their punishment. In Discipline and Punish, Michel Foucault traces this evolution from a system of corporal punishments to one of carceral punishments, using the prison as a metaphor for society at large (which also relies on order, discipline, and “docile bodies” to maintain existing economic and political systems).

However, the shift from corporal to carceral punishments was not the only rapid transformation in the history of punishment. During the 1980s, another sudden shift radically affected public attitudes. When Robert Martinson published an article dismissing rehabilitation programs as ineffective, underlying dissatisfaction with the state of U.S. crime and punishment coalesced into a unified attack on correctional treatment from both conservatives and progressives, and within a year or two, the justice system was dominated by the retributivist view of just deserts. Determinate sentencing replaced indeterminate systems of parole, sentence lengths increased, and prison populations soared. The U.S. imprisonment rate, which had steadily averaged about 109 persons per 100,000 between 1925 and 1980, began to grow exponentially in the 1980s, rising from a level of 139 per 100,000 citizens in 1980 to a level of 504 per 100,000 in 2008. The Pew Center on the States has reported that when populations from jails, probation, and parole are included, the number of people under U.S. correctional control increases to more than 7.3 million (or 1 in 31 people).

Understanding the implications of mass incarceration on this scale is but one of the challenges that scholars of punishment face. Given that most of the 2.3 million people in jail and prison will eventually come out, and given that recidivism rates in many jurisdictions exceed 50 percent within a three-year period after release from prison, experts are also struggling to understand how to facilitate reentry into communities in order to break the cycle of returning to prison. Understanding the ongoing evolution of the death penalty is another challenge. The United States is unique among Western democracies in its continuing reliance upon capital punishment, but since the death penalty was reintroduced after a four-year period of de facto abolition in 1976, its scope has been limited by a number of Supreme Court decisions. Death may not be imposed on insane persons, juveniles, or the mentally retarded, and as a general matter, death may not be imposed in nonhomicide cases. Social attitudes about capital punishment continue to change, as do attitudes about punishment generally, reflecting Winston Churchill's remark to the House of Commons that one of the most unfailing tests of civilization is the mood and temper of the public in regard to the treatment of crime and criminals.

See also

Attitudes and Myths About Punishment, Continuum of Sanctions, Prison Overcrowding, Public Shaming as Punishment, Punishment Units, Restitution

Further Readings
  • Beccaria, Cesare On Crimes and Punishments, Paolucci, Henry, trans. Englewood Cliffs, NJ: Prentice Hall, 1963.
  • Bentham, Jeremy The Principles of Morals and Legislation. New York: Hafner Press, 1948.
  • Ehrlich, Isaac “The Deterrent Effect of Capital Punishment: A Question of Life and Death.” American Economic Review, v.65.
  • Feinberg, Joel “The Expressive Theory of Punishment.” The Monist, v.49.
  • Foucault, Michel Discipline and Punish, Sheridan, Alan, trans. New York: Pantheon Books, 1977.
  • Greenwood, Peter W. Selective Incapacitation. Santa Monica, CA: Rand, 1982.
  • Hart, H. L. A. “Prolegomenon to the Principles of Punishment.” In Punishment and Responsibility: Essays in the Philosophy of Law. New York: Oxford University Press, 1968.
  • Kant, Immanuel The Philosophy of Law, Hastie, W., trans. Edinburgh, UK: T. and T. Clark, 1887.
  • Lombroso, Cesare Criminal Man, Rafter, Nicole, ed. Durham, NC: Duke University Press, 2006.
  • Martinson, Robert “What Works? Questions and Answers About Prison Reform.” The Public Interest, v.35.
  • Morris, Herbert “Persons and Punishment.” The Monist, v.52.
  • Morris, Norval The Future of Imprisonment. Chicago: University of Chicago Press, 1974.
  • Murphy, Jeffrie G. Punishment and Rehabilitation, 2nd ed. Belmont, CA: Thomson Wadsworth, 1985.
  • Packer, Herbert L. The Limits of the Criminal Sanction. Stanford, CA: Stanford University Press, 1968.
  • Pew Center on the States. One in 31: The Long Reach of American Corrections. Washington, DC: Pew Charitable Trusts, 2009.
  • Robinson, Paul H. Distributive Principles of Criminal Law: Who Should Be Punished, How Much? New York: Oxford University Press, 2008.
  • Tyler, Tom Why People Obey the Law. New Haven, CT: Yale University Press, 1990.
  • von Hirsch, Andrew Censure and Sanctions. New York: Oxford University Press, 1993.
  • Zimring, Franklin E. and Hawkins, Gordon Incapacitation: Penal Confinement and the Restraint of Crime. New York: Oxford University Press, 1995.
  • Oleson, James C.
    Copyright © 2012 by SAGE Publications, Inc.

    Related Credo Articles

    Full text Article Punishment
    Encyclopedia of Community Corrections

    Punishment can be defined as a penalty imposed under the authority of law, inflicted on a person by other persons, for the commission of a...

    Full text Article PUNISHMENT
    Encyclopedia of Behavior Modification and Cognitive Behavior Therapy

    Burrhus Frederic Skinner (1971 p. 75) “Nature, if not God, has created man in such a way that he can be controlled punitively. . . . The need...

    Full text Article PUNISHMENT
    Encyclopedia of Behavior Modification and Cognitive Behavior Therapy

    DESCRIPTION OF THE STRATEGY Parents, teachers, and others have used punishment throughout history to manage behavior. Nonetheless, in spite of...

    See more from Credo