2. CRIMINAL V. CIVIL SANCTIONS

Because I propose to justify acts of punishment for the violation of the criminal law hence, as criminal sanctions, in this section I explicate a distinction between criminal and civil sanctions. The validity of the conventional distinction between criminal and civil law is, for the sake of argument at least, granted throughout (see Altman Reference Altman2001: 111–19). In addition, although I do not propose to justify court-imposed sanctions in civil cases, I am not casting doubt on the justifiability of civil sanctions. It is simply, I suppose, that the justification of criminal sanctions deserves its own treatment. It is worth noting therefore that there are good reasons to consider criminal sanctions (or punishments) as distinct from civil sanctions. Some of these reasons highlight the fact that the distinction between criminal and civil sanctions is, in the US at least, widely accepted. Influenced by centuries of Anglo-American legal history and jurisprudence, lawyers, jurists and judges in the US are generally undisturbed, even comfortable holding that the purpose of a criminal sanction is to punish, while the purpose of a civil sanction is to compensate (Keeton Reference Keeton1984: 7). In addition, drawing at times upon the authority of the US congress, the US Supreme Court ruled and affirmed that a civil remedy is ‘different in nature’ in comparison with punishment.Footnote ⁵ And in the philosophical literature, ‘punishment’ is often reserved for discussions of legal sanctions on crime, as the justification of civil sanctions is glossed over (but cf. Coleman Reference Coleman1992). Moreover, it is instructive to think of a civil case in terms of ‘an arbitration.’ Given a dispute between two or more parties, the judge is an arbiter who is authorized to determine the resolution of the dispute as she is guided by principles of justice or law. The judge determines what justice or principles of law would have directed in the particular case before her and orders the parties to follow these directives. A civil case may therefore call upon the judge to make the law rather than merely ipso facto rule on whether the parties complied with the law. Obeying the judge's verdict, the parties are then complying with the law rather than being punished. Finally, under a civil judgement, the parties may be said to be complying with the law rather than being punished even when they incur costs, since to bear costs by obeying the law is not (as I stress in the next section) the same as to be punished by the law.Footnote ⁶

3. THE LEGAL ORDER AS A PUBLIC GOOD

I discussed a distinction between civil and criminal sanctions because this paper focuses on the justification of criminal sanctions, i.e. punishments, alone. I shall now develop and defend the claim that punishment is justifiable as a sort of tax (which I label ‘a quasi-tax’), maintaining along the way that raising taxes to fund the production of public goods is justifiable as long as the amount paid as tax by each individual is fair (appropriate) and arguing that the legal order is a pure public good.Footnote ⁷

To explain the uniqueness of the provision of certain goods and services, economists have distinguished between private goods (or services) and public goods (or services). People may think that the government is the provider of public goods while privately owned firms competing in the marketplace provide us with private goods. It is not, however, because the government does the provision that a good is considered public, for the government may choose to provide private goods or services, while public goods or services may be provided privately. Grain, for instance, is a pure private good even as the governments of ancient Rome provided it, sometimes freely, to hundreds of thousands of people (Rostovtzeff Reference Rostovtzeff1960: 174). Supplied by private organizations, the ringing of a church bell is a pure public service.

A pure public good is distinguishable from a pure private good by means of two criteria (Browning and Browning Reference Browning and Browning1989: 584–85). First, the benefit of a public good is said to be non-excludable, namely, it is not feasible to exclude non-payers from receiving the benefit of the good once the good is available to payers. Second, a pure public good is non-rival in consumption; that is to say, the consumption of the good by one person need not reduce the quantity available for consumption by other persons. National defence, street lighting and environmental protection are often-given instances of pure public goods (Pearce Reference Pearce1986: 348). I shall now demonstrate that the legal order, the peace and security stemming from the credibility of legal threats, fits the definition of a pure public good.

A person may be excluded from entering a state or removed from it once he has entered, but if, as a tourist, a person is spending his time visiting a state, he may enjoy the protection of the law on his visit without having to pay for it. He may spend money on many things but he does not have to pay a special, additional fee for the protection of the law. It is not feasible, then, to exclude tourists as well as children or wards of the state, all of whom are non-payers, from enjoying the legal order. In addition, as one person enjoys the protection afforded him by the legal order, there need not be a smaller quantity available for other people. If I am going to sleep at night comfortably knowing the safety of my possessions, my neighbour may be enjoying the same, comfortably knowing the safety of his possessions, without having to split with me any fixed quantity. To avail oneself of the benefits of the legal order, the peace of mind and the ability to plan and execute one's plans, is not to slice a piece of a pie which then leaves a smaller piece for other people to consume, hence the legal order is non-rival in consumption. The legal order, therefore, is both non-excludable and non-rival in consumption, so it is a pure public good.

As a pure public good, the legal order may be supported by several kinds of contributions, and I will discuss non-punitive contributions to the legal order before I discuss legal punishment as my main concern. One kind of contribution to the legal order has to do with money paid as taxes or fees to a local or national authority, which money may be used to hire police, judges and law-makers or build prisons and courts. Taxes and fees may be calculated with a view to the future funding of operation of the justice system. They are then forward-looking contributions to the maintenance of legal order.

The second kind of contribution to the legal order derives from compliance with the law that is costly to the complier. Whether compliance is costly to the complier depends on the preferences of the complier. If, for example, a person complies with the legal speed limit simply because he prefers driving at a speed below the legal limit (say, 60 mph), the person's compliance need not be costly to him. And it matters not, it seems to me, whether the preference for driving within the legal speed limit is motivated by convenience or by moral reasons. All things considered, a person whose preferred driving speed happens to be within the legal limit does not incur costs due to his compliance. A person's compliance with the law is costly to her, however, if she would have exceeded the speed limit but for the fear of a fine. There is room, in addition, to distinguish between pecuniary and non-pecuniary costs. If I can increase the value of my property by adding a room to my house though in fact I am complying with the law's prohibition on the addition of the room, I make a pecuniary contribution to the legal order because, to obey the law, I forsake an increase in the value of my property. But if I am merely frustrated because the law forbids me from taking my dog along with me to the park, the cost of my compliance is non-pecuniary, as is the cost to one who is driving within the speed limit for fear of punishment though he may not expect to receive a pecuniary benefit from exceeding the speed limit. Needless to say, I am neither claiming nor suggesting that it is in principle wrong for the law to impose costly compliance. In fact, as advocates of the fairness theory of punishment have urged, it is difficult to imagine a legal system not based in large part on costly compliance (Dagger Reference Dagger1993, Reference Dagger2008).

4. A QUASI-TAX FAIRLY ASSESSED

A third kind of contribution to the legal order derives from acts of punishment. The pain, deprivation or suffering that is imposed on convicted fellows constitute, it seems, the most significant sacrifices for the sake of maintaining the legal order. The relative security and peace attained by civil society, as well as the amenities of civil society and the creation of its culture, indeed depend upon the suffering of the convicted. As an individual contribution to the maintenance of the legal order, an act of punishment may be analysed in terms of ‘quasi-taxation’, a quasi-tax being a tax that need not be paid with money, the fairness of which depends on backward-looking considerations. But let us elaborate.

When a person is inoculated to prevent the spread of an epidemic, her pain or discomfort is an individual sacrifice toward the production of a public good, if the sum total of numerous, similar sort of sacrifices by many individuals would prevent the epidemic from spreading, thus securing the health of the public. When a young man is conscripted, drafted to serve in the military forces of the state without compensation for his labour, his service is his own contribution toward the maintenance of the national defence, which is a public good. Now, if the law had not credibly threatened punishment, lawful conduct would have been minimized at the expense of society. Since an act of punishment makes a legal threat credible, it functions to deter, for the sake of the legal order, violations of the law. As he is being punished, a person is made to contribute to the maintenance of the legal order, which is a public good. The pain, deprivation or harm imposed on a person guilty of breaking the law is a sacrifice on the part of the person toward the legal order. Like an act of inoculation or an act of conscription, an act of punishment, as it forces a personal sacrifice toward the provision of a public good, may properly be viewed as the imposition of a quasi-tax.Footnote ⁸ Thus, another way to identify a quasi-tax is to point out that, unlike the assessment of a tax, where the money assessed is itself the resource of funding a public project, in quasi-taxation the sacrifice of the contributor, like the pain of inoculation, is not in itself the only resource used to fund the public good.

It may now be asked, given that there are various kinds of sacrifices to sustain the legal order, why should criminal violations be sometimes subject to imprisonment or the punishment of death? Why, in other words, should a guilty criminal be imprisoned or even pay with his life for his crime rather than merely forced to make a monetary contribution to sustain the legal order? We reply to this important question by drawing on the work of Cooter and Ulen (Reference Cooter and Ulen2012). First and foremost, there is a problem of collecting information about the social cost of crime. We may agree that a burglar, for instance, in addition to compensating the direct victims of his actions, should reimburse the state for the costs of his arrest, trial and detention. But what about the costs of his burglary to the neighbours of the burglarized home? In an attempt to prevent similar losses to their own properties, some of the neighbours may pay for security systems; some may install bars on their windows, some may purchase insurance and still some others may incur costs due to other kinds of attempts to prevent burglaries to their homes. Needless to say, it tends to be exceedingly difficult for the state regularly to get an accurate estimate of the total costs of burglaries and other crimes to the entire community. ‘The information requirements for pricing crimes are prohibitive’, writes Cooter, ‘while the information requirements for sanctioning crimes are relatively low’ (Reference Cooter1984: 1523). Therefore, certain kinds of offences, like burglaries, where calculating the social cost of the offence is, as a matter of policy, exceedingly difficult if at all possible, are better handled with punishment such as detention rather than with having the offender reimburse the state.

Second, keeping in mind the goal of uniformity in the law, consider a case of rape. While a rich man guilty of the crime may, if he is permitted to do so, discharge of his penalty by reimbursing the state, a poor man in the same situation may not do the same. Thus a poor man guilty of rape may spend twenty years in prison, while a rich man may hardly feel the adverse effect of his penalty for the same kind of crime, when the goal of uniformity in the law requires that, as much as possible, both may suffer the same degree of pain, loss or deprivation for the same kind of crime. Again, punishment rather than a monetary contribution seems to be the more suitable reaction to rape because it tends to be closer than a monetary contribution to the purpose of equalizing the harshness of the penalty.Footnote ⁹

Not to deny that taxes may be unjust or even confiscatory, it is justifiable to tax the people in order to provide them with a public good, provided that the amount of the tax is fairly imposed on every taxed individual. Since an act of punishment is an act of taxation, the harshness or severity of punishment, analogous to the burden of a tax, has to be fair in order for the act of punishment to be justifiable. Fairness in quasi-taxation, however, differs from fairness in taxation.

Back in the 19th century, interest in taxation theory gave rise to a concept of fair, just or equitable tax, later developed by Lindahl, Bowen and Samuelson into the so-called ‘benefit approach’ to fair taxation (Musgrave Reference Musgrave1959: 73–8, Allan Reference Allan1971: 99–108). According to the benefit approach, a person's share in the total tax needed to fund a public good ought to be equivalent to the person's share in the benefit of the good. If, for example, the total benefit of a public good is at $60, and a person's share of the total benefit is worth $15, it is fair to require the person's tax to constitute 15/60 of the total cost of producing the public good. If the cost of producing the public good is $40, it would be fair for the person's tax to reach $10. The amount of the tax is related to the future benefit of the taxed individual.

Fair contribution is differently viewed in cases of quasi-taxation. We no longer seek to equate the magnitude of the contribution with some future benefit to the individual. The facts taken into account in determining the burden of the sacrifice must be related to the past of the person. We do not, for example, attempt to estimate the future benefit of public health to a person who carries in his body some pathogenic microorganism. Justifiably, the person may be inoculated irrespective of his share in the benefits of the public's health because he poses a risk to the public if he is not so treated. What happened determines the fairness of his sacrifice, the fact that the person has contracted a contagious disease. Similar considerations apply to the quasi-tax of conscription. If a person is able to contribute by his efforts to the success of a military campaign, it may be fair to conscript the person irrespective of how much he may expect to benefit from participating in the campaign. Fairness in punishment, given that punishment is justifiable as a quasi-tax, is to be determined by backward-looking considerations rather than by forward-looking ones.

Since punishment is a quasi-tax and a tax has to be fair in order to be justifiable, the burden imposed on the offender – the severity or harshness of the punishment – has to be fair in order for the punishment to be justifiable, and this fairness is explained in terms of the retributivist outlook. In thinking of the justification of punishment in terms of quasi-taxation, therefore, we propose a shift in the traditional understanding of the role of retribution in the theory of punishment. We now think of retribution not as a complete justification of the act of punishing offenders, but rather as providing reasons for taking the harshness of the punishment to be fair or unfair. Importantly, the retributivist standards on the basis of which the fairness of punishment is to be explained may consist of the requirement that the severity of the punishment ought to match the wrongfulness of the offence, when the wrongfulness of the offence consists, in addition to the harm caused by the offence in terms of property and liberty, of an estimate of the moral culpability of the offender. Thus, the same sort of violation may be subject to a more severe punishment when the harm done by the offence is the same, but the offender has acted deliberately or maliciously or if the offence has been repeated.

Several authors contend that, as an exclusive justification of punishment, retributivism is question begging (Shafer-Landau Reference Shafer-Landau, Feinberg and Coleman2000: 771) or, as Stanley Benn has written, ‘to say “it is fitting” or “justice demands” that the guilty should suffer is only to reaffirm that punishment is right, not to give grounds for thinking so’ (Reference Benn and Murphy1985: 10). We thus note that given our proposed reassignment – using retribution to explain the fairness of the punishment rather than justify the very act of punishing the offender – the role of retribution is commensurate with our sentencing practices which, as punishment is meted out, take into account both the harm caused by the offence and the culpability of the offender. In addition, it is no longer plausible to level the charge of question-begging against retributivism for, although it need not be self-evident, it is not question begging to suppose that the severity of punishment has to track the moral culpability of the offender. Finally, should it be objected that, if my argument requires a justifiable punishment to be one that is fairly imposed according to retributivist principles, all the philosophical work needed to justify punishment is done by the retributivist theory of punishment, my reply is the following. Retributivism may indeed require a fair punishment, but it does not require the punishment to have a social benefit. On my account, punishment is not justifiable unless it has a deterrent effect which promotes the good of society by maintaining the legal order.

We conclude this section by noting that, so conceived, fairness in punishment preserves the deterrent effect of punishment. For to have a deterrent effect, the punishment has to make the offence unprofitable, and if, as retributivism requires, the severity of the punishment matches the wrongfulness of the offence, an offender, from the perspective of rational choice, would not have a sufficient reason to commit the offence.Footnote ¹⁰

To sum up the argument of this section, an act of punishment amounts to the assessment of a quasi-tax, that is to say, a tax that (a) need not be paid with money and (b) the fairness of which depends on backward-looking considerations. The retributivist approach provides the justification of the fairness of punishment in terms of its harshness or severity, allowing both the harm done by the offence and the moral culpability of the offender to be taken into account.

5. REPLIES TO OBJECTIONS

Kant considers a hypothetical case in which executing a justly convicted murderer, given that the state in which the murder has taken place is about to be disbanded, may not be expected to have a deterrent effect. Kant maintains that, in such a case, the murderer ought nevertheless to be executed so as to uphold just retribution and to discharge of the state's duty of punishing the criminal. ‘The murderer must be executed’, Kant claims, ‘so that everyone will duly receive what his actions are worth and so that the bloodguilt thereof will not be fixed on the people because they failed to insist on carrying out the punishment’ (Reference Kant and Murphy1985: 22). Given the case described by Kant, it may be objected that, on the theory here proposed, it is justifiable to punish for a criminal violation only if the punishment amounts to an individual contribution to the production of the legal order as a public good. If the legal order as a public good does not have to be funded, which seems to be the case in Kant's hypothetical case, punishment is unjustifiable. Hence the murderer in Kant's case may not justifiably be executed for his crime. Moreover, given the extraordinary situation alluded to by Kant, there seems to be no need for policing the streets, for operating courts to punish offenders or for building prisons as places to confine offenders. As a public good, the legal order need not be funded at all, either by direct taxes or by quasi-taxes. If punishment in the criminal law is a quasi-tax, it would seem unjustifiable then, not only to execute the murderer; to the extent that the legal order does not need to be funded, it would be unjustifiable to make the murderer pay the state a fine of $5! I say in reply that, even if my theory rules out punishing the murderer for violating the criminal law, there is nothing in my theory to prohibit civil suits from being brought against the murderer. Those unjustly harmed by the murder are entitled to bring suits for damages, hence the murderer may be severely punished.

A perennial objection to the utilitarian justification of punishment is that, for the sake of the common good, it allows for excessive punishments, even for the punishment of the innocent. And a protracted, fierce defence of the utilitarian standpoint against the objection notwithstanding, we suppose that the objection has not been deflected (see Boonin Reference Boonin2008: 62–75). In light of our approach, however, an innocent person may not justifiably be punished, for in the first place, on our view, punishment is justifiable only if it is fairly imposed in accordance with retributive principles. It is unjustifiable, therefore, to compel, by means of punishment, a person to pay for a wrong he is not guilty of causing. Secondly, quasi-taxes may rightly be levied selectively, so only the members of a certain group may justifiably pay the tax even if the entire community benefits from the tax. Only comparatively young persons justifiably may be conscripted when all the people of the state benefit from their service (see Dimock Reference Dimock, Feinberg and Coleman2000: 28). To prevent a contagious disease from spreading, only potentially sick people justifiably may be inoculated. In the same manner, only a portion of the people, those convicted of an offence, may justifiably be punished, for the sake of the common good. A person innocent of breaking the law is not a member of the group whose members only may justifiably be required to help fund the legal order by means of being punished, the group of those guilty of breaking the law.

A common objection to the utilitarian theory of punishment has to do with the charge that the theory allows for the treatment of offenders merely as a means to an end. To punish exclusively in order to deter violations of the law, irrespective of the moral culpability of the offender for the actual violation, is to force a person to suffer not because he is guilty of causing others to suffer but, even if his punishment is neither excessively harsh nor too mild, only because his suffering is expected to deter others from violating the law. And this, the objection goes, amounts to treating the offender merely as a means to an end. It surely is reasonable to wonder whether our reformulation of the mixed theory, given that it justifies punishment partly in reference to its deterrent effect, may also be open to this sort of objection.

I say in reply, with some repetition of claims already made, that for punishment to be justifiable, the punishment, first, must have a deterrent effect and second, the punishment must be fair in the terms of its severity. The fairness of the punishment, given my proposal, has to take into account both the harm done by the offence and the moral culpability of the offender. The moral culpability of the offender has to be correctly assessed in determining a fair punishment. An act of punishment may be unfair if the moral culpability of the offender is incorrectly assessed. It is not the case, therefore, that on my account, justifiable punishment allows for the treatment of offenders merely as a means to an end.

If the utilitarian theory of punishment is true, it has been objected, it follows that unsuccessful attempts to commit crime ought to be punished just as severely as successful attempts, which is contrary to common practice and to common sense. Our reformulation of the mixed theory, however, is immune to this sort of objection, for we require fairness in punishment, implying that the severity of punishment ought to fit the wrongfulness of the offence. Thus, if an unsuccessful attempt is not as wrongful as a successful attempt to commit crime, a fair punishment for it need not be as severe as the punishment for a successful one.

Finally, it has been argued that no deterrent effect may be expected from punishing some persons convicted of crimes against humanity, such as the leaders of Nazi Germany put on trial at Nuremberg. Hence, the argument proceeds, supposing that the punishments imposed at Nuremberg were justifiable, deterrence is not essential for the justification of punishment (Ten Reference Ten1987: 47–51). In reply, I say that ‘punishment’ is used in a number of different contexts and senses. We talk about punishing criminals, but we also talk about punishing dogs or children, hence the term ‘punishment’ in one context need not mean exactly the same as in others. This paper deals with legal punishment, when a person is punished by the government whose law he has violated. At the time of the violation, the person is under the jurisdiction of the government that punishes him. This, however, is not the case with many of the so-called ‘crimes against humanity’. When he committed the actions for which he has been convicted and executed, Goering, for instance, was not under the jurisdiction of any of the allied states. When he committed Hungarian Jews to Auschwitz, Adolph Eichmann was not subject to the laws of the state of Israel. These executions, therefore, though justifiable, were not cases of legal punishment in the strict sense that the term has in this paper.

To make the point in a different way, if capital punishment by the state is justifiable, intentional killing is justifiable. However, not all cases of intentional killing are cases of capital punishment. Killing in self-defence, for example, or killing in war may be cases of intentional killing that are not capital punishments. The executions of Goering and Eichmann were justifiable intentional killings, as the governments that executed them wanted to present their cases against the accused before the international community, to make it clear that they had not arbitrarily singled out certain people for harsh retaliation. These executions were not capital punishments in the strict sense of the term, as a legal action of the state against a person subject to its laws.

6. A CONCLUDING REMARK

To justify punishment as a criminal sanction, we make sure in this essay to uphold throughout the traditional distinction between civil and criminal sanctions (and, ipso facto, the traditional distinction between civil and criminal law). Surprisingly, perhaps, given that we justify criminal sanctions as quasi-taxes, we now face the prospect of a comprehensive analysis of legal sanctions, both civil and criminal, in economic terms. Let us initially turn our focus to the work of Richard A. Posner, in particular the summary of his analysis of the major fields of the civil law: property, contract and torts. He says:

Posner's view is that, in large part, the civil law may be justified as an instrument of promoting or protecting economic efficiency. If so, given that a purpose of a legal sanction is to make the law and its threat of sanction credible, the promotion of economic efficiency may also be the basis of justifying civil sanctions. Building on Posner's view, we propose then that, if the justification of civil sanctions is grounded in economic efficiency, the justification of criminal sanctions is grounded in the fairness of taxes (or quasi-taxes). We consequently contemplate a comprehensive analysis of legal sanctions in economic terms.