Criminalization and Decriminalization

It is useful at the outset to distinguish between three fundamentally different types of arguments for and against the use of the criminal law. Arguments of principle assert that as a matter of political or moral philosophy, it is proper (or improper) to prohibit certain conduct. The second category, that of arguments of practicality, is logically relevant only if it has been determined that society may legitimately prohibit the conduct in question; practical arguments assert that, although legitimate, certain prohibitions are unwise because in practice they cause more harm than good. A third approach rejects the feasibility of devising a workable standard based on the substance of criminal prohibitions, and argues instead for additional procedural limitations on criminalization decisions and criminal law enforcement.

Arguments of principle. The broad question of whether, and to what extent, the law may enforce morality represents one of the classic debates in philosophical and criminal law literature. This debate has tended to focus on the use of formal criminal sanctions, and most of the arguments appear to make no distinction between criminal and noncriminal measures. Of course, if the law may not legitimately interfere at all with certain behavior, then neither criminal nor civil sanctions may be used. On the other hand, if the law may legitimately interfere, there is the further question of whether it is legitimate, necessary, or desirable to use criminal sanctions.

Arguments in favor of prohibition. Some writers have argued that society is permitted, and perhaps even obligated, to enforce morality by means of criminal or other legal sanctions. There are at least two distinct variations of this argument. What H. L. A. Hart calls the "conservative thesis" asserts that the majority in society have the right not only to follow their own moral convictions but also to preserve their "moral environment" as a thing of value and to insist that all members of society abide by their moral convictions (p. 2; see also Stephen). What H. L. A. Hart calls the "disintegration thesis" asserts that public morality is the "cement of society," which must be maintained to prevent social disintegration (p. 1). A major proponent of this thesis was Patrick Devlin, who argued that the law should protect society's political and moral institutions and the "community of ideas" necessary for people to live together. Devlin wrote, "Society cannot ignore the morality of the individual any more than it can his loyalty; it flourishes on both and without either it dies" (p. 22).

Even if it is conceded that the legal enforcement of morality is legitimate, however, several practical difficulties arise: whose "morality" is to be enforced, and how much of it? Although the United States has a highly "moralistic" criminal law (Morris and Hawkins, 1970), many types of behavior that would generally be considered "immoral" have never been considered criminal—for example, most breaches of contract (Packer). The nineteenth-century jurist James Fitzjames Stephen suggested that the criminal law should be limited to "extreme cases . . . [of] gross acts of vice," that public opinion and common practice must "strenuously and unequivocally condemn" the conduct, and that "a moral majority must be overwhelming" (1967 ed., pp. 159, 162). Although he saw no possibility of setting theoretical or a priori limits on the power of the law to enforce morality, Patrick Devlin conceded the need for "toleration of the maximum individual freedom that is consistent with the integrity of society"; only if the majority has "a real feeling of reprobation, intolerance, indignation and disgust" for the conduct may it be prohibited (pp. 16–17). But how are such judgments to be made, and by whom? Devlin answered that the standard should be that of the "reasonable" person, or the typical juror, because the "moral judgment of society must be something about which any twelve men or women drawn at random might after discussion be expected to be unanimous" (p. 15).

Although these formulations do suggest some limiting principles, they provide little concrete guidance to legislators. Moreover, in an increasingly secular, pluralist society there is less and less consensus about fundamental moral principles, and some doubt whether twelve persons "drawn at random" would be unanimous about anything. (However, as discussed more fully below, Stephen's and Devlin's overwhelming-moral-consensus standards do suggest potential procedural limits on criminalization decisions.) Finally, in deciding how much of morality to enforce with the law, there is no inconsistency in also considering the practical advantages and disadvantages of attempting to prohibit certain conduct. Indeed, Devlin accepted this, citing such practical considerations as the extent to which enforcement would be ineffective or would inevitably violate rights of privacy (pp. 18–22).

Principled arguments against criminalization. In sharp contrast to Devlin, other writers have argued that the law may not legitimately prohibit certain behavior. The classic statement of this position was made by the nineteenth-century English philosopher John Stuart Mill, who argued that society may interfere with the individual's freedom of action only "to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant" (1946 ed., p. 15). Mill believed that the individual must be accorded the maximum degree of liberty and autonomy that is consistent with the rights of others. Although writers such as Devlin concede the importance of protecting individual liberty, Mill and his followers appear to give this factor much greater weight. They view the individual's freedom and self-determination as preeminent rights, which outweigh mere utilitarian considerations of the greatest good for the greatest number (Richards, 1979a, pp. 1222–1223).

To some extent, the Supreme Court and some state courts have adopted this approach, holding that individual rights of privacy and free expression, implicit in the First, Fourth, and Ninth Amendments, prevent the state from prohibiting certain acts that cause no direct injury to any other person (Griswold v. Connecticut, 381 U.S. 479 (1965) (use of contraceptives); Stanley v. Georgia, 394 U. S. 557 (1969) (possession of obscene material in the home); Roe v. Wade, 410 U.S. 113 (1973) (abortion during the first three months of pregnancy); Ravin v. State, 537 P.2d 494 (Alaska 1975) (possession of marijuana in the home); Commonwealth v. Bonadio, 490 Pa. 91, 415 A.2d 47 (1980) (sodomy between consenting adults in private)). The Supreme Court has also held that, under certain limited circumstances, the ban on cruel and unusual punishments, and principles of due process or equal protection, preclude criminal liability for morals or regulatory offenses (Robinson v. California, 370 U.S. 660 (1962) ("status" crime of being an addict); Lambert v. California, 355 U.S. 225 (1957) (crime of failing to register as a convicted person, with no showing of reasonable opportunity to become aware of the duty to act); Loving v. Virginia, 388 U.S. 1 (1967) (interracial marriage)). On the other hand, the Court has also stated that broad deference should be given to legislative judgments about the wisdom of prohibiting certain conduct, even if those judgments are based on moral assessments. Thus, for example, the Court has upheld punishment of obscenity in a private theater to which access was limited to consenting adults (Paris Adult Theater I v. Slaton, 413 U.S. 49 (1973)), and has also upheld criminalization of private acts of sodomy between consenting adults (Bowers v. Hardwick, 478 U.S. 186 (1986)).

Analysis of the earlier quotation from Mill reveals several distinguishable justifications for punishment: individuals might be punished for their own "moral" benefit, for their own "physical good," or to prevent "harm to others." Modern writers have tended to agree with Mill that the first purpose is clearly improper: as far as the law is concerned, the individual has "an inalienable right to go to hell in his own fashion, provided he does not directly injure the person or property of another on the way" (Morris and Hawkins, 1970, p. 2).

However, most modern writers do not share Mill's total opposition to prohibitions aimed at protecting the defendant's physical well-being. Mill apparently felt that paternalism was a rationale too easily abused (Richards, 1979b, p. 1424), but modern authors seem more willing to recognize some limited version of this rationale. Mill did approve of laws to protect children from their own lack of judgment, and some modern writers have broadened this justification to include protection against exploitation and corruption of other especially vulnerable groups, including those "weak in body or mind, inexperienced, or in a state of special physical, official or economic dependence" (The Wolfenden Report, para. 13; Morris and Hawkins, 1970; H. L. A. Hart; Feinberg, vol. 3). Other writers maintain that individuals must be given the maximum freedom to make choices that they may later regret; these writers would narrowly limit paternalism to cases of extreme irrationality or nonrationality "likely to harm irreparably serious human interests," such as rationality, freedom, or life itself (Richards, 1979b, p. 1424).

As for Mill's third category, "harm to others," it is unclear to what extent he would have distinguished between tangible and intangible harms. If intangible harms include the weakening of "public morality," there is obviously little point to Mill's other limitations. A related problem involves the use of the criminal law for "verbal vindication of our morals" (Kadish, 1967, p. 162), with no serious effort actually to enforce the law. Since such symbolic legislation provides intangible benefits to "others," and does not directly inhibit the individual's freedom of action or punish him for his own moral benefit, how does it violate either the letter or the spirit of Mill's philosophy? Modern followers of Mill usually respond with practical arguments about the collateral disadvantages of unenforced law (for example, diminished respect for law), rather than with arguments of principle (Kadish, 1972).

Also problematic is Mill's support for laws punishing offenses against public "decency" (Mill; H. L. A. Hart; Morris and Hawkins, 1970; Richards, 1979a). The offense caused to the "victims" of public indecency seems different in degree, but not in principle, from the revulsion that such victims feel toward similar acts performed in private. However, one distinction is that prohibition of public solicitation, nudity, and other "indecencies" does not totally eliminate the freedom of the individual to engage in such behavior (in private), or in the presence of those who are not offended. Nevertheless, the result is to limit individual freedom for the sake of preventing an intangible harm to others.

Another intangible harm to others, recognized by at least one modern follower of Mill as a proper justification for punishment, is the breach of promises of marital fidelity (Richards, 1979a). Traditional laws prohibiting adultery, bigamy, and prostitution would seem to be justified on this basis, yet the same author opposes such laws, apparently on the ground that they are over- or under-inclusive. That is, these laws purportedly cover some conduct that poses no threat to fidelity, yet they fail to punish other conduct that clearly does pose such a threat.

A third category of harm that might be viewed as intangible, at least from the point of view of human beings, is that of cruelty to animals. Several modern followers of Mill's philosophy appear to support prohibitions against such cruelty (H. L. A. Hart; Morris and Hawkins, 1970), without considering whether these laws are consistent with the harm-to-others criterion.

As for more tangible harm to others, such as physical injury or property loss, there is widespread agreement with Mill's position that this is a proper basis for punishment. However, as Herbert Packer has pointed out, it is almost always possible to argue that a given form of conduct involves some risk of harm to the interests of others; the harm-to-others criterion is thus a matter of degree—"a prudential criterion rather than a hard and fast distinction of principle" (p. 266). Packer goes on to argue that the risk of harm to others must be "substantial" and unjustified by reasons of social utility. He also asserts that the harm should not be trivial in two senses: it should not be so minor that the imposition of any criminal punishment would be disproportionate to the social harm caused, nor should it be so minor that law enforcement and sentencing authorities are unwilling to enforce the law or to make regular use of "real criminal sanctions," such as imprisonment (pp. 271–273).

Other modern followers of Mill appear to recognize a much broader authority to "protect the citizen's person and property" (Morris and Hawkins, 1970, p. 4). Difficult problems of remoteness of harms are also posed by conduct that, if widely practiced, might cause serious social disorganization—for example, drug addiction. Thus, one court, while upholding the right to use marijuana in the home, implied that such use could be punished if it ever became so widespread that it might "significantly debilitate the fabric of our society" (Ravin, supra, p. 509).

A strict application of the Mill philosophy thus poses a number of difficulties. As noted above, it is difficult to find a "pure case" of behavior that harms no one but the actor (Dripps). Clearly, however, the extent of harm to others, both in seriousness and probability, is an important factor to be considered in the criminalization decision. Second, depending on how broadly "harm to others" is defined and on how willing we are to recognize paternalistic legislation, the Mill principle may not be all that limiting, particularly if it is recognized that the legislature has at least some discretion to be both over- and under-inclusive in pursuit of its goals. Third, to the extent that a strict interpretation of Mill's philosophy is based on an elaborate theory of the moral or human rights of the person (Richards, 1979b), this approach to criminalization may prove too vague or too subjective to command broad consensus and application. Fourth, Mill and his followers offer little guidance in the choice of various noncriminal sanctions; indeed, a strict reading of the Mill philosophy would invalidate not only civil and criminal penalties but any interference with individual "liberty of action," including, for example, steep excise taxes on tobacco or alcoholic beverages (Greenawalt, p. 719). Finally, there is no reason why "moral" arguments against criminalization should preclude consideration of practical disadvantages as well, particularly since most of the moral arguments do not lend themselves to definitive, "bright line" distinctions.

Arguments of practicality. Although modern advocates of decriminalization have often cited Mill's philosophy in support of their arguments, they usually go on to argue that even if prohibition might be legitimate, it is unwise. One group of writers argues that all "victimless" crimes should be repealed. A less simplistic approach seeks to catalog the specific advantages and disadvantages of trying to prohibit certain behavior, and argues that, on balance, the total "costs" of criminalization often outweigh the benefits.

Victimless crime. The concept of victimless crime is frequently suggested as a basis for decriminalization (Schur). The term itself is somewhat misleading, since it has been applied to offenses such as public drunkenness and adultery, which often have direct, readily identifiable "victims." Furthermore, it is arguable that most other so-called victimless crimes, such as drug offenses and prostitution, do have at least potential victims: the participants themselves, relatives, taxpayers, or society at large. The users of this term tend to stress the practical disadvantages of trying to enforce victimless crimes—for example, that the lack of complaining witnesses leads to the use of intrusive police practices, bribery, and discriminatory enforcement. However, many of these problems arise in the enforcement of criminal laws that have not been labeled as victimless or proposed for repeal (for example, carrying an unregistered weapon). The victimless crime concept may draw some of its rhetorical appeal from largely unarticulated philosophical premises: if a crime is truly victimless, efforts to enforce it may not only be difficult but illegitimate (Morris and Hawkins, 1970). But whether the victimless crime criterion is best viewed as an argument of principle or one of practicality, the concept is of very limited utility in deciding the more difficult issues of criminalization. The criterion lacks a clear definition, fails to cover some of the offenses to which it has been applied, and applies equally well to other offenses that have not been proposed for repeal. The relative victimlessness of an offense is closely related to several important practical issues in the criminalization decision (discussed below). However, labeling a crime as victimless only begins what is, in most cases, a very difficult process of assessing complex empirical facts and fundamental value choices.

The cost-benefit approach. A more sophisticated (but less rhetorically effective) practical approach to the criminalization question seeks to identify the specific advantages and disadvantages of invoking the criminal law, in an effort to determine whether the total public and private "costs" of criminalization outweigh the benefits (Kadish, 1967). Strictly speaking, cost-benefit analysis involves the weighing of variables that are measurable in dollars or other quantitative units. The proponents of a cost-benefit approach to criminalization generally concede that there is little quantitative data in this area, but they argue that the approach is still a useful way of thinking about criminalization problems (Kadish, 1972).

The costs of criminalization include public and private burdens, both tangible and intangible. Beginning with tangible public costs, it is necessary (but often difficult) to separate out the police, court, attorney, and correctional expenditures properly attributable only to the enforcement of a specific criminal law. (Occasionally this "marginal" cost is fairly clear—the salaries and expenses of the police narcotics division, for example.) In some cases, apparent costs may not entirely disappear with decriminalization: many of the previous enforcement efforts (such as collecting and jailing public drunks) may have simply taken up the slack in police and jail resources that must still be maintained to handle peak loads. Moreover, decriminalization may require the police and other public officials to respond to the underlying behavioral problem in alternate ways that have their own costs: in the example above, removal of penalties for public drunkenness may lead to more arrests for disorderly conduct, or may increase the use of costly public and private medical facilities.

Private costs of criminalization include not only concrete items such as attorneys' fees and other litigation expenses, but also several factors that are harder to measure. These include the individual's loss of the freedom, pleasure, or other value derived from engaging in the forbidden conduct (for example, having sex with a prostitute; owning firearms); the anxiety and social ostracism imposed upon offenders, whether or not they are detected and prosecuted; the reduction or elimination of the offender's economic earning power during the period of pretrial and trial proceedings, while sentence is being served, and perhaps for the rest of his life; detention and other losses of physical freedom, before and after conviction; and the uncompensated costs and inconveniences imposed on witnesses and jurors.

Another set of costs, which may affect both public and private interests, is the tendency of criminalization to produce more, rather than less, socially undesirable behavior. Examples of illegal behavior that may result from criminalization are bribery of the police or other enforcement officials; extortion by officials of money or other favors in return for nonenforcement; private blackmail by threats to expose the offender; discriminatory enforcement of the law against unpopular groups or individuals, or in favor of defendants with more political or social influence; and the use of illegal methods of obtaining evidence, such as unauthorized searches and electronic surveillance, coercive interrogations, and entrapment. Indeed, some categories of crimes (for example, vagrancy and disorderly conduct) seem to have been specifically designed to undercut constitutional limitations on arrest, search, and interrogation (Kadish, 1967).

Even where the methods of law enforcement are not clearly illegal, they may be so contrary to widely held feelings of privacy or fairness that they cause a lowering of public respect for the law, particularly among social groups already alienated from society, including ethnic minorities and the poor. Examples of such questionable tactics include selective enforcement in order to conserve resources; the use of undercover agents, decoy officers, and informers from the criminal milieu who are paid in money or leniency; arrests for purposes of harassment or to "clean the streets," with no effort to prosecute; and "legal" searches, electronic surveillance, and intrusive physical surveillance (for example, peering through holes in the ceilings of public washrooms in order to observe possible homosexual or narcotic offenses). As may be noted, the problems listed above arise primarily when the behavior involves consenting parties and few, if any, witnesses. It is partly this relative invisibility of victimless crimes that makes their enforcement so costly.

Criminal laws may violate principles of equal justice even if not intentionally enforced in a discriminatory or selective manner. For example, poor or uneducated women are less likely to obtain a safe, although illegal, abortion and thus must either bear unwanted children or suffer the risk of death or severe medical complications. Efforts to enforce prostitution laws against "call girl" operations are costly and difficult, and therefore most arrests of prostitutes involve street solicitations by lower-class or minority-group women (Morris and Hawkins, 1970). Drug-law enforcement is likewise heavily biased against poor, nonwhite street dealers and users (Tonry). Unequal justice is inherently wrong, but it also has important practical consequences. Perceptions of unfairness, either in the law's procedures or its impact, undercut the legitimacy of legal prohibitions, making citizens less willing to obey the law (Tyler).

Other costs of criminalization arise when the prohibited conduct involves goods, activities, or services that are in great demand, such as gambling, drugs, liquor, illegal weapons, abortion, commercial sex, and pornography. When there is high demand, prohibition tends to limit supply more than demand, thus driving up the black-market price and creating monopoly profits for those criminals who remain in business. Organized criminals tend to have advantages over less organized ones in exploiting illegal markets and coping with law enforcement pressures, and consequently, criminalization tends to foster the growth of sophisticated, well-organized, and powerful criminal groups. Once in existence, organized crime tends to diversify into other areas of crime. Its high profits provide ample funds for bribery of public officials, as well as capital for diversification. Finally, whether or not the participants in black markets are highly organized, they tend to use violence to resolve their disputes; thus, the enactment and increased enforcement of alcohol and drug prohibitions in the twentieth century led to increased rates of lethal violence (Miron).

Although higher prices tend to discourage some would-be participants in prohibited activities, the underlying high demand, combined with restricted supply, maintains both high prices and high participation rates. In extreme cases of high and inflexible demand (for example, heroin or cocaine addiction), exorbitant prices force many participants to commit other crimes to pay for the illegal goods or services they want. These are generally nonviolent property crimes, such as shoplifting, or other forms of vice such as prostitution or sale of drugs, but violent property crimes may also be encouraged. Finally, because the illegal goods or services are in great demand, a large number of otherwise lawabiding citizens are driven into association with the criminal elements who supply these goods and services. There is a danger that these citizens will come to view themselves as criminals, since society labels them as such. As members of the criminal subculture, they may lose respect for the law and are more likely to be drawn into other forms of crime.

Another concrete cost of criminalization is the barrier that the law erects between the criminal and important social services and protections. For example, when the law forbids abortion or drug use, consumers are forced to make use of unsanitary instruments or medically unsound procedures, increasing the risk of death, injury, or infection (to themselves and, in the case of HIV, to all of their sexual or syringe-sharing partners). If harm occurs they are unwilling to seek the medical attention they need, for fear of exposing their criminal behavior; pregnant drug users may avoid all prenatal care. Laws against prostitution and homosexuality may have a similarly adverse effect: participants who contract a venereal disease are less likely to seek timely medical treatment. Moreover, although prostitution laws seek to prevent the exploitation and physical abuse of female prostitutes by their pimps, the existence of criminal penalties and enforcement efforts probably makes women more likely to seek the support and protection of the pimp, while discouraging them from seeking legal protection from exploitation and abuse.

Further costs of criminalization include overloading the criminal justice system with a mass of petty cases; creating a law enforcement bureaucracy with a "vested interest in the status quo" (Packer, p. 333), thus thwarting efforts at reform or even research; and fostering the illusion that a social problem has been taken care of, thereby discouraging the development of more effective alternative measures (Kadish, 1967).

Finally, it can be argued that extending the criminal law to behavior that is widely believed to be morally neutral or that is engaged in by the vast majority of citizens dilutes the stigmatizing quality of criminal sanctions generally, thus robbing them of their peculiar effectiveness in dealing with more serious conduct (Packer; Kadish, 1967). Of course, lack of widespread moral condemnation may also make the law difficult or impossible to enforce, thus limiting the benefit of prohibition.

Benefits of criminalization. On the other side of the ledger, the benefits achieved by criminalization fall into several categories. To the extent that the criminal law is enforced with a view toward preventing specific social or individual harms, the likelihood of achieving such preventive benefits depends on the following factors:

1. The probability that the behavior defined as criminal will be observed or detected by anyone other than the immediate participants;
2. The probability that various parties will invoke formal criminal processes—that witnesses or participants will report the crime to the police and support prosecution efforts, that the police will be able and willing to make an arrest, that the prosecutor will approve the filing of formal charges, and that judges and juries will be able and willing to find the defendant guilty and impose significant sanctions;
3. The likelihood that conviction and sentence will reduce the future incidence of the behavior defined as criminal, either through general deterrence of other potential offenders, the "educative" effect of punishment, special deterrence or rehabilitation of the punished offender, or incapacitation of the offender; and
4. The likelihood that reducing the incidence of the behavior defined as criminal will reduce any more remote harms sought to be prevented (for example: the likelihood that reducing acts of drunken driving at low alcohol-concentration levels will reduce accident frequency or severity).

Of course, to the extent that punishment of criminal behavior is considered proper for its own sake (to impose "deserved" punishment, for example), the last two factors are irrelevant.

Even if criminal prosecution is not successful or is not even attempted, the mere existence of criminal prohibitions might have some indirect effect on the incidence of the behavior defined as criminal and thus, even more indirectly, on any more remote harm to be prevented. The labeling of behavior as criminal represents a social judgment that such behavior is morally wrong or at least undesirable, and this judgment may serve to reinforce similar feelings among members of the public. As with actual enforcement efforts, the extent of this symbolic effect depends on the strength of the perceived relationship between the prohibited behavior and the more remote social harm, if any, sought to be prevented (for example, the relationship between euthanasia and the devaluation of human life). Another important set of considerations here involves attitudes about unenforced law; to the extent that most people believe that the law should either be enforced or repealed, unenforced law promotes cynicism and disrespect for the law, particularly the criminal law (Kadish, 1967). Thus, legislative attempts to denounce certain behavior symbolically, with no intention or ability to enforce the law, may do more harm than good.

Perhaps the most critical determinant of the total "benefit" of criminalization, whether by means of actual enforcement efforts or symbolic denunciation, is the importance of the social harm involved. In the case of heroin possession or sales, for example, it must be decided how seriously society views the use of heroin itself—as well as undesirable behaviors (e.g., driving) of persons under the influence of heroin, and the risk of heroin addiction—in order to decide whether the costs of criminalizing heroin use and sale are worth bearing. A host of tangible and intangible factors must then be considered: the values of human rationality and full consciousness; the losses of life and health that are not attributable to prohibition itself; the potential loss of the economic productivity of users; and perhaps even the anguish that heavy use or addiction may impose on relatives and close friends of the user.