Crime: Definition

views updated


A crime is an act proscribed by law and subject to punishment. It can also be an omission instead of an act, namely a failure to act where the law imposes a duty to act. Traditionally, crimes have been restricted to acts and omissions that harm the interests of others. Sometimes, however, a legislature will criminalize an act or omission because it is harmful to the perpetrator himself, or because the conduct is morally reprehensible. Such criminal provisions are known as "victimless" crimes. The possibility of a victimless crime underscores the central difference between criminal and civil law: a crime is an offense against public welfare, whereas a civil wrong is an offense against private interests. While civil damages are awarded to compensate a victim for harm he has suffered at the injurer's hands, criminal punishment is inflicted to allow the state to vindicate its interest in the common good.

In our history, the concept of the public wrong emerged after the Norman Conquest, replacing what was essentially a system of private plea-bargaining under the Anglo-Saxons. Prior to the conquest, an injurer would pay his victim a sum of money in order to buy off the latter's right to revenge. These payments, known as wer, wíte, and bót, were not determined by law, but instead depended on what injurer and victim could negotiate. We can already discern the concept of a public harm at this time, however, in the fact that some injuries were bóteás, or beyond monetary redemption, and for these a man might be put to death. A later possible source of the offense against the state may be the jurisdictional concept of the "king's peace." Under this concept, the Crown reserved the right to control for violent acts that might occur along any route on which the king traveled. Finally, while an even later development, the advent of a public police force made the concept of a public wrong institutionally feasible.

Civil and criminal divide

In recent years, the distinction between civil and criminal wrongs has become somewhat blurred. On the civil side, for example, there is the institution of "punitive damages," by which an individual is punished for the intentional infliction of an injury or a malicious breach of contract. Punitive damages are intended as punishment for the injurer, unlike the ordinary civil remedy of compensatory damages that cannot exceed the amount required to make the victim whole. On the criminal side, there is the increasingly common use of monetary penalties in lieu of incarceration. Such penalties are often paid as compensation to the victim in the form of restitution. There is also increasing use of the criminal sanction against corporations. Since a corporation can only be punished with monetary sanctions, and since punitive damages are increasingly awarded in civil suits, the distinction between civil and criminal in such cases is a nominal one. It would appear to consist mostly of procedural differences, such as the different standards of proof and different rules of evidence. Finally, there is a recent movement to enhance the role of the victim in criminal proceedings, stemming from the belief that crime victims have a right to representation in the prosecution of their attackers. The idea of victim's rights most strongly suggests a shift away from the conception of crime as a public offense. It suggests that the punishment of the offender serves, at least in part, to satisfy the victim's need for vengeance. This trend toward the "privatization" of crime finds expression in various proposed institutional reforms as well, such as the proposal to convert prisons to private ownership.

The acceptability of these various modifications of the traditional notion of crime depends partly on what we take a crime to be. Is a crime simply a prohibition that appears in one of the state or federal penal codes under the heading "criminal"? Or is the criminal category a deeper one, one that does not derive its meaning from any particular use to which the notion of crime is put? The first would be what we might call a "positivistic" stance toward the notion of an offense. It treats crime entirely as a legislative concept. The second would be a normative stance toward the notion of an offense identifying the criminal category by a theory of justified prohibition. On a positive approach, there can be no objection to punishing corporations or enhancing the role of the victim, since there is no obligatory content to the notion of an offense. On a normative approach, by contrast, there may be grounds for objecting to these modifications to the traditional treatment of crime. For it may turn out that punishing an offender at the behest of the victim, especially if associated with the payment of restitution, is not legitimate according to our best theory of justified punishment.

The positivistic approach

The prevailing approach of the American legal system toward crime is positivistic. As Henry Hart once wrote facetiously: "a crime is anything which is called a crime, and a criminal penalty is simply the penalty provided for doing anything which has been given that name" (p. 404). By refusing to recognize constitutional boundaries on the notion of an offense, this is precisely the position the U.S. Supreme Court has articulated over the course of the last fifty or so years. The Court has held, for example, that a legislature may criminalize conduct without including a mental state element (mens rea ) in the definition of the offense (U.S. v. Dotterweich ; U.S. v. Balint ). It has also found it a matter of legislative discretion whether to treat exonerating conditions like insanity as part of the definition of the offense to which they apply or as so-called affirmative defenses. The former approach would place the burden on the prosecution to prove, for example, that the defendant was not insane at the time he performed the criminal act, whereas the latter would place the burden on the defendant to prove he was. The Court famously articulated its commitment to the positivistic approach to crime in a case involving the defense of extreme emotional disturbance where it upheld a New York provision that shifted the burden to the defendant to prove the defense, instead of requiring the prosecution to prove the absence of the defense beyond a reasonable doubt (Patterson v. New York ). Given its premise, the Court's reasoning was flawless: It argued that because a state has the power to eliminate the defense altogether, it must also have the power to shift the burden to the defendant to prove it, since "the greater power implies the lesser power" (p. 211). The same argument has been found applicable to other defenses as well, even one as fundamental as self-defense. Recently, however, the Supreme Court has indicated a renewed willingness to place limits on state burden-shifting. The case concerned a New Jersey hate-crime statute that authorized substantially increased penalties for any defendant whose crime was committed from the motive of racial animus. The Court found the statute unconstitutional on the grounds that it obviated the state's duty to prove mental state by treating racial bias as a sentencing factor instead of as an element of the offense. The implication of such a decision is that legislatures do not have unfettered discretion to decide how and whether to criminalize, even outside the area of fundamental rights. For if it is constitutionally impermissible for a state to shift the burden on a mental state element, it would seem to follow that it does not have unfettered discretion to decide whether to include such mental state elements in its offense definitions in the first instance. The question, then, is whether the Court's recent holding in the area of burden of proof signals a fundamental shift away from the positivist approach to crime, or whether its influence will be confined to the area of burden of proof. Is the Court embarking on a new constitutional jurisprudence of substantive criminal law or will it continue to shy away from any real attempt to place limits on the substantive criminal provisions legislatures can pass?

While the positivistic approach to crime has prevailed, there are some isolated areas in which the Supreme Court has traditionally attempted to place limitations on offense definition. For the most part, these limitations have consisted of a set of formal restrictions on how legislatures may draft offenses, stemming from the due process clauses of the Fifth and Fourteenth Amendments. While these restrictions purport to speak only to how conduct is criminalized, rather than what is criminalized, they often turn out to impose substantive conditions on offense definition as well. Consider, for example, the following four important limitations on the notion of an offense.

First, the doctrine of vagueness requires that criminal statutes define the prohibited conduct with sufficient specificity to place potential defendants on notice of their vulnerability to criminal prosecution. This doctrine has most notably been applied to loitering ordinances, many of which are thought to leave too much discretion to police officers to arrest individuals on grounds of physical appearance or demeanor. In many cases, the objection to such statutes would not be eliminated by more precise drafting. As the Court made clear in a recent case involving a Chicago loitering ordinance, sometimes a statute cuts too deeply into the ordinary activities of everyday life, with too little justification, to be constitutionally acceptable (City of Chicago v. Morales ). A second, related doctrine is that of overbreadth, which forbids a legislature from drafting criminal statutes in a way that risks prosecution and conviction for ordinary, noncriminal behavior. The Court will strike down criminal statutes on overbreadth grounds mostly where the prohibition risks infringing freedom of speech and expression (R.A.V. v. City of St. Paul ). A third doctrine is also articulated under the heading of "due process," namely the doctrine of legality. Criminal statutes must provide clear notice of a citizen's potential subjection to criminal punishment in order to afford ordinary citizens a fair opportunity to conform their behavior to the law. For example, punishment must not be retroactive, and it must be certain and definite. Finally, the Eighth Amendment ban on "cruel and unusual punishment" has been interpreted as containing a doctrine of proportionality that serves to restrict the punishment selected for a given offense (Solem v. Helm ; see Harmelin v. Michigan ). While this doctrine retains its force mostly in the death penalty area, it has served in the past to ensure that the sanction authorized for a given offense is roughly on a part with the sanction for the same offense in other jurisdictions, and that it is appropriate given the sanction authorized for other offenses in the same jurisdiction.

Nonpositivist approaches

The foregoing limitations on the notion of an offense suggest that while the positivistic approach to offense definition may be the prevailing one in our constitutional jurisprudence, there is reason to question the depth of our commitment to it. We do not in fact accept that any conduct a legislature wishes to make criminal is rightly punished, and the restrictions we impose on the use of the criminal sanction cannot be entirely accounted for as restrictions imposed by the first eight amendments to the Constitution. Some conduct seems so unsuitable as an object of criminal prohibition that we feel it stretches the concept of crime to apply it to those cases. In extreme cases the point would be clear: Statutes that made criminal punishment retroactive rather than prospective, that punished for thoughts without any accompanying deeds, that enacted a separate set of prohibitions for each separate member of the community, that established a separate count of theft for each thirty-second period that a thief withheld the stolen item from its owner, or that adopted an arbitrary class of subjects to whom the prohibition would apply, would be so out of keeping with the way we think of crime that we might be inclined to reject the suggestion that the statutes made the conduct (or thoughts) crimes. In what sense would they be crimes? Simply arresting a person and subjecting him to incarceration or other harsh treatment does not by itself make the conduct for which he was arrested criminal. It does not even do so when the legislature has authorized the behavior in the form of a law. While one might hope to limit the use of the criminal sanction in such cases by the sorts of ancillary constitutional restrictions on legislative discretion discussed above, these will prove insufficient to capture our current understanding of crime. It may be, therefore, that it is the concept of crime itself that limits what a legislature may prohibit and how it may ensure adherence to those limits.

At least to some extent, then, our understanding of crime is normative as well as descriptive. In particular, there may be conditions of justification that are themselves part of the notion of crime. If this is correct, then part of what we mean when we speak of a criminal offense is that the infringement of liberty the statute authorizes is justified by the importance of inducing conformity with the criminal prohibition. This approach would suggest not only that punishing an individual for something he had no reason to know was forbidden is not, properly speaking, punishment, but that the conduct thus penalized could not be correctly called "criminal," even if the legislature has called it a crime and has attached the kinds of penalties to it that typically accompany so-called criminal conduct. The normative approach to crime would thus provide a way of evaluating legislative uses of the power to criminalize by establishing criteria that are internal to the notion of crime itself. Such criteria would make it possible to say quite directly that the legislature erred in prohibiting a certain kind of conduct and providing stringent penalties for its occurrence, on the grounds that the prohibited conduct is not an appropriate object of criminal prohibition. And while legislatures might have significant latitude in determining the acceptable objects of criminal prohibition, under a normative approach to crime, their decisionmaking would operate within certain broadly defined limits.

Legal moralism. Unlike their judicial counterparts, criminal law scholars tend to favor some sort of normative approach to the notion of an offense. There is, however, no nonpositivistic definition of crime that would command uniform assent among them. One school of thought about crime is called "legal moralism." The legal moralist maintains that a crime is an immoral act, and accordingly that all and only immoral acts ought to be punished. Thus the legal moralist not only believes that every crime is in some way an immoral act, or that it tends to produce an immoral act, but also that there are no immoral acts that should go unpunished. One class of crime appears to pose a problem for the legal moralist, namely the crimes often referred to as mala prohibita. Mala prohibita crimes identify acts that are bad only because the legislature has forbidden them. By contrast, mala in se crimes prohibit acts that are bad in and of themselves. The legal moralist has difficulty with this distinction, because he seems to regard all crimes as mala in se, to the extent that he thinks it is the underlying immorality of an act that justifies prohibiting it under the criminal law. Legal moralists sometimes seek to solve the problem of mala prohibita crimes by saying that the acts they prohibit are instrumentally related to an act or state of affairs that is mala in se. While it is not immoral to drive on the left rather than on the right, it is immoral to impose grave risk of injury on one's fellows. In this way, the legal moralist explains the law mandating driving on the left, in the United States, or on the right, in Britain, as a necessary prohibition in order to avoid the truly immoral act of plowing into cars coming in the opposite direction.

Social practice view. A second nonpositivistic view of the notion of an offense sees crimes as prohibited acts, where the explanation for these prohibitions is that they are forbidden by certain social practices, or by those possessing authority to make criminalization decisions in light of a social practice allocating the power to do so. H. L. A. Hart, for example, thought of criminal law as a set of "primary rules" designed to regulate conduct. But the primary rules, he argued, are law only because they are made by officials whose authority rests on a social practice that identifies when a rule counts as law. The rule that men must remove their hats in church, he wrote, identifies a social practice. But not all social practices have the force of law. Unlike customs and ordinary, quotidian conventions, the social rules that are law are ones that are identified in a special way within the practice as having the force of law. Only those rules possessing a certain "pedigree," namely those created by individuals authorized by "secondary rules" to create, interpret, and apply primary rules, will be so recognized. The social practice view of crime may seem similar to the positivistic approach, given that both approaches treat crime as a set of prohibitions created by those authorized to do so. It might thus be thought simply a different brand of positivism. But unlike the Supreme Court's brand of positivism about crime, Hart's account would allow for evaluative judgments about a legislature's criminalization decisions, based on their fidelity to an underlying notion of crime. A legislature that created draconian criminal prohibitions under a social practice view could be found to be exceeding its authority as established by the relevant secondary rules. As such, its dictates would not have the force of law.

Economic account. A third prominent nonpositivistic alternative is the economic account of crime. According to some theorists, a crime is an inefficient actinefficient because it bypasses a voluntary market. Criminal sanctions are necessary to give individuals sufficient incentive to obtain what they want through the market, rather than to take what they want by force. In this, criminal sanctions are slightly different from civil penalties. While the legal economist sees rules of civil and criminal liability as serving the same purpose, namely to provide incentives for efficient behavior, the incentive structures needed to promote efficiency for the two kinds of acts diverge. According to the economic account of crime, the criminal sanction ought to apply to acts that are always inefficient. The criminal law must threaten potential defendants with sufficiently stringent punishment to ensure that criminal acts are never worthwhile. Sometimes, by contrast, the acts that violate civil law are in fact efficient, despite the fact that they are prohibited. It is thus sometimes efficient to allow individuals to break a contract or to run a risk of injuring another person. Unlike criminal sanctions, which must always induce conformity, the penalty for civil wrongs need only be equal to the damage caused in order to provide the incentives for efficient behavior. By forcing injurers or those wishing to breach a contract to "internalize" the cost of the damage they cause, they will injure or breach only when it is efficient to do so. Criminal penalties are just like civil penalties, with the exception that civil sanctions must contain a "kicker" added to the damage caused, in order to ensure that it is never sufficiently advantageous to violate the prohibitory norm. Indeed, the decreasing distance between tort law and criminal law in recent years may itself be testimony to the influence of law and economics on judicial and legislative methodology.

While the positivistic view of crime enjoys a rhetorical advantage in our system, the actual understanding of crime our legal system presupposes seems rather to display an admixture of descriptive and normative facts. We look to legislative pronouncement to learn the content of those prohibitions we call "crimes," but we also make normative judgments about criminal statutes based on an implicit sense of what constitutes a correct application of the notion of crime. It is perhaps, moreover, because the conceptual limits of "crime" are reasonably well ensconced in our public use of the term that states do not attempt to eliminate the defense of self-defense or, for the most part, make chatting on a street corner a crime.

Harm-based theory. Jeremy Bentham is often thought of as the father of legal positivism. But even Bentham recognized that the notion of crime must incorporate normative elements. Bentham took the standard positivist line that laws, and criminal laws in particular, are commands of the sovereign. Whatever is commanded has the force of law. But Bentham also argued interestingly that a command does not count as law if it is not "complete." In order for a law to be complete, it has to identify a discrete harm or evil at which the legal prohibition aims. Thus even for Bentham, the notion of crime rests on a pre-legislative concept, namely the notion of harm. Building an account of crime on the idea of harm represents a fourth nonpositivistic approach. The beginnings of such an account were suggested by John Stuart Mill, who articulated what has come to be knows as the "harm principle." In On Liberty Mill wrote: "The only purpose for which power can rightfully be exercised over any member of a civilized community against his will is to prevent harm to others" (pp. 1011). More recently, Joel Feinberg has developed Mill's basic approach in greater detail. He has argued, however, that harm may not provide the only legitimate grounds for making criminal sanction. Even if Feinberg is right that we do not adhere to the harm principle without exception, the harm principle may nevertheless lie at the heart of American criminal law's approach to the notion of an offense.

Claire Finkelstein

See also Actus Reus; Burden of Proof; Causation; Civil and Criminal Divide; Guilt; Mens Rea; Punishment; Strict Liability; Victimless Crime.


Bentham, Jeremy. Of Laws in General. New York: Oxford University Press, 1970.

Coleman, Jules L. "Crimes, Kickers and Transaction Structures." Repr. in Markets, Morals and the Law. New York: Cambridge University Press, 1988. Pages 153166.

Feinberg, Joel. The Moral Limits of the Criminal Law. Vols. 14. New York: Oxford University Press, 19841998.

Finkelstein, Claire. "Positivism and the Notion of an Offense." California Law Review 88: (2000): 335394.

Fletcher, George P. With Justice For Some: Protecting Victims' Rights in Criminal Trials. Reading, Mass.: Addison-Wesley Publishing Co., 1996.

Goebel, Julius, Jr. Felony and Misdemeanor: A Study in the History of English Criminal Procedure. University of Pennsylvania Press, 1937. Reprint, 1976.

Hart, H. L. A. Punishment and Responsibility. Oxford, U.K.: Clarendon Press, 1967.

Hart, Henry M. "The Aims of the Criminal Law." Law and Contemporary Problems 23 (1958): 401.

Mill, John Stuart. On Liberty (1859). Reprint. New York: W. W. Norton, 1975.

Moore, Michael. Placing Blame. New York: Oxford University Press, 1997.

Pollock, Sir Frederick, and Maitland, Frederic William. The History of English Law before the Time of Edward I. Lawyer's Literary Club Edition, 1959.

Posner, Richard. "An Economic Theory of the Criminal Law." Columbia Law Review 85 (1985): 1193.


Apprendi v. New Jersey, 120 S. Ct. 2348 (2000).

City of Chicago v. Morales, 119 S. Ct. 1849 (1999).

R.A.V. v. City of St. Paul, 505 U.S. 377 (1992).

Harmelin v. Michigan, 501 U.S. 957 (1991).

Martin v. Ohio, 480 U.S. 228 (1987).

Solem v. Helm, 463 U.S. 277 (1983).

Patterson v. New York, 432 U.S. 197 (1977).

United States v. Dotterweich, 320 U.S. 277 (1943).

Lambert v. California, 355 U.S. 225 (1957).

Leland v. Oregon, 343 U.S. 790 (1952).

United States v. Balint, 258 U.S. 250 (1922).