Enforcing Morality


A. P. Simester


Suppose that we are considering whether an action is immoral. Its harmfulness is sometimes, but not always, our starting point. Some actions, such as murder and attempted murder, are wrongs because of the harm to which they conduce. Others are not. Actions such as rape, perjury and blackmail are wrongs prior to any consequences they cause. Their wrongfulness originates elsewhere. It may spring from the means by which the agent does something: not in where you go, as it were, but in how you get there—notably, how you treat people along the way. (Did you deceive her?) In special cases, it may be motive-based, resting on why a person does something. (Was it a warning or threat? An offer or blackmail?) These kinds of action involve what might be termed nonderivative wrongs, in as much as their basic wrongfulness is not dependent upon an outcome. If asked, “Which comes first, the wrong or the harm?” We can only say: it depends.

When deciding whether to prohibit such actions, however, a parallel question may not receive the same reply. Famously, for Mill (1859 ch. 1), “the only purpose for which power can rightfully be exercised over any member of a civilised community against his will is to prevent harm to others.” Perhaps, like Feinberg (1984: 34–36), we might require that the harm also be wrongful, but part of the point of the Harm Principle is to focus attention on the harm—and to reject immorality, or wrongfulnessper se, as a ground of prohibition. On the other hand, Patrick Devlin (1965: 12–13) controversially asserted that “it is not possible to set theoretical limits to the power of the State to legislate against immorality ⋯ or to define inflexibly areas of morality into which the law is in no circumstances allowed to enter.” Indeed, according to a school of thought known as Legal Moralism, an action can warrant proscription simply on the ground of its moral wrongfulness.

Ultimately, these approaches are incompatible. But they share common ground, and Devlin’s challenge helpfully focuses our attention on the requirements of the Harm Principle itself. There is nothing special, or objectionable, about confining prohibitions to morally wrongful actions, i.e., actions that one ought not to do. Those are exactly the sorts of action that prohibitions should address. As we shall see, it is certainly arguable that there are harm-based constraints on state intervention to regulate wrongs. If so, however, they complement rather than displace the wrongfulness requirement. Indeed, a concentration on harm may divert attention from the more general inquiry whether, and if ever when, we should prohibit wrongful conduct.

What Counts as Immoral?

Depending on one’s views about the scope of morality, it is possible to narrow the gap. Michael Moore (1997: 662), for example, rejects the thought that morality has anything to say about consensual sexual practices. Hence, even on his retributivist view, the state has no basis to criminalize such activities. More generally, if one thinks of morality as an antidote to selfishness (cf. Mackie 1977, ch. 5), reasons may be “moral” ones when they address how we should treat each other. On that view, the legal moralist’s claim to enforce morality will tend to converge with the Harm Principle’s focus on conduct that, directly or indirectly, affects other people’s lives.

But that would artificially truncate our inquiry. Devlin’s skepticism about limiting state enforcement of morality finds its strongest expression if we take it that an action is “immoral” whenever it is morally wrongful; and that it is morally wrongful whenever, all things considered, one ought not to do it. In turn, one ought not to do an action whenever the reasons favoring its performance are, all things considered, defeated by the reasons against. For an action to be immoral, therefore, does not require that it is seriously or profoundly wrong, that it be evil or wicked; only that it should not be done. Most wrongful conduct is venial.

On this view, practical morality is concerned not merely with how one should treat others but with the question, what should one do? Thus no categorical distinction is drawn here between “moral” and other kinds of reasons, such as prudential ones. There are, of course, many distinctions that we can draw, in particular between guiding reasons (which in fact apply to an action—Raz 1990: 16ff) and explanatory reasons (by which the agent is, subjectively, motivated). The present essay is concerned with guiding reasons—that is, with the reasons why we ought, or ought not, to do something. Yet within that realm, reasons, whether prudential, altruistic or of some other character, are either good—i.e., valid—reasons for doing something, or they are not. Those narrower labels are meaningful, but they are not foundational to the moral question, what ought one to do? Obviously, different kinds of reasons may have differing weights and priorities; what we tend to call “moral” reasons, especially concerning the interests of others, often have relatively greater importance. Yet a reason that is, say, prudential in character can still be a good reason for doing something. Imprudence can be a vice too. Imagine the case of a successful, patriotic businessman, who decides to kill himself should the UK do badly in the next Eurovision Song Contest. He may justly be criticized morally by his friends for resolving to “throw away” his life on such a foolish basis.

We do need, however, to distinguish an action’s wrongfulness, or immorality, from its beinga wrong. In the usage I adopt here, an action is a wrong when it breaches a duty or violates a right. On occasion, it may be permissible—not wrongful—to perpetrate a wrong, as when D breaks into V’s house in order to call an ambulance to the accident on the road outside. Our main concern in this essay is with the converse issue: whether, as Jeremy Waldron (1981) would put it, one has a right to act wrongfully.

Three Theses Concerning Immorality

Taking moral wrongfulness as a starting point may seem counterintuitive. It is often said that the law should not be in the business of prohibiting immoral behavior, and at least one version of that claim is surely right. But we need to be careful about what is meant by the claim, and about how convincing it really is. Clarification is required. Here are three possible interpretations:

1.That φing is immoral is insufficient to justify its criminalization (Insufficiency Thesis).

2.That φing is immoral is necessary to justify its criminalization (Necessity Thesis).

3.That φing is immoral is insufficient to establish even a pro tanto ground for its criminalization (Non-qualifying Thesis).

(For convenience, I will focus primarily on enforcement through criminal law. As we see below, however, similar principles apply to coercion through the civil law.) The first thesis, that moral wrongfulness is insufficient to justify criminalization, seems uncontroversial. Even Devlin could embrace it. Suppose that φing ought not to be done. Accept too, for the moment, that this generates a reason to prohibit it. It does not follow that, all things considered, we should prohibit φing, because the reason favoring prohibition may be defeated by other considerations.

One set of counter-considerations is operational. Even if a prima facie case can be made for prohibition, and φing lies within the range of conduct that there is reason to criminalize, that case must still overcome various negative constraints that militate against criminalization generally. At the very least, to make an all things-considered case for criminalization, we need to show that the criminal law offers an appropriate method of controlling φing, and is preferable to other methods of legal regulation available to the state. Recall the disastrous attempt by many western governments in the twentieth century to regulate alcohol using criminal prohibitions, which created a black market ripe for extortion and racketeering. Rightly, alcohol licensing and taxation laws are now preferred. Other constraints include the practical challenges of crafting an offense definition in terms that are effective, enforceable and which meet rule of law and other concerns (Simester and Sullivan 2007, § 16.5–7). It may be, if these demands cannot be met, that the state ought not to prohibit φing despite the prima facie case for doing so.

The in-principle case must also be weighed up against the burdens of prohibition itself, most notably in terms of freedom and lost opportunities (Feinberg 1984: 216). No doubt extramarital affairs constitute wrongful betrayals. But perhaps they should not be criminalized because of the extensive intrusions that their prohibition would involve. For all of these reasons, we should concede the Insufficiency Thesis. But we can do so without concern. For it is a long step from that thesis to concluding (i) that immorality is unnecessary to justify criminalization, or (ii) that even an in-principle case for criminalizing φing requires more than that φing is immoral.

The Need for Moral Wrongfulness

A very long step. Indeed, conclusion (i) would be a misstep, because the truth is the other way around. Preventing immorality is an indispensable condition of criminalization. It is the Necessity Thesis, not an “un-necessity” thesis, to which we should subscribe: any prohibition of φing can be justified only when φing is morally wrongful action.

Within the criminal sphere, the Necessity Thesis is most easily defended by reference to the distinctive nature of criminal law, which punishes, and censures, the offender for having done wrong. The criminal law is a blaming institution, and one cannot blame a person unless that person does something morally wrong; that is, unless she does something that, all things considered, she ought not to do. One can, of course, also judge people morally for their good deeds: yet such judgments are not blaming judgments. Blame lies only for conduct that, all things considered, one should not do.

As it happens, while that argument is adequate for the criminal law, a version of the Necessity Thesis holds also for the civil law; and, indeed, for all of us. No moral agent should act wrongfully. And the state is, like the rest of us, subject to the requirements of morality. It too should act in accordance with undefeated, all things-considered reasons. Where it fails to do so, it acts wrongly, just like the rest of us.

This matters because if, all things considered, D has undefeated reason to φ, it generally follows that no one, including the state, should stop D from φing. Notwithstanding that D’s reasons to φ may be personal to D, the existence of those reasons is itself a general matter (cf. Gardner 2007: 131). That D has reason to φ, therefore, is something that commands our allegiance too. This is not so much because D herself is entitled to respect—since respect for another human being does not imply that we must always respect the reasons for which that person acts. Rather, it is because the reasons themselves are entitled to respect—that is, because they are (good) reasons.

(In passing, we should allow some provisos to this claim. On occasion, the existence of reasons may depend on the status of the agent; thus it is possible that some reasons for individuals to φ may be excluded in the hands of certain other agents, such as the state, and vice versa. More on this possibility later. Neither do I suggest that we should all care just as much about reasons that are personal to others as we should about those personal to us.)

Legally Created Wrongs?

But what about regulatory laws? It is a commonplace that the state frequently prohibits conduct that is not pre-legally wrong. Indeed such offenses, which Anglo-American lawyers callmala prohibita, form the major part of the criminal canon. They vastly outnumber mala in se proscriptions of conduct that is pre-legally wrong. Yet, if φing is not morally wrongful, how can we justify its prohibition and subsequent punishment? Is the Necessity Thesis incompatible with mala prohibita offenses? This worry has concerned many writers. As Douglas Husak (2007: 112) complains, “I fail to understand why persons behave wrongfully when their conduct is malum prohibitum but notmalum in se.” While accepting that some mala prohibita prohibitions may give substantive content to underlying, preexisting wrongs, Husak doubts that many modern offenses do specify such pre-legal wrongs.

This line of thought is partly right and partly misleading. It is right in so far as it reflects the truth—and it is a truth—that φing does not become morally wrong just because the state declares it so. But it is misleading in that it doesn’t sufficiently distinguish prohibition, which is forward looking, from punishment, which is retrospective andex post. The justification of an act of criminalization is not the same as the justification of an act of punishment. They are different acts. The former can play a role in justifying the latter.

How so? Because the state sometimes creates specific moral reasons. Moreover, it can do so in a variety of ways (e.g., Honoré1993; Finnis 1980: 284ff). Indeed, that power is not restricted to the state. When the soccer referee shows a player a red card, the player thereupon has a reason, indeed a duty, to leave the pitch. Any moral agent in a position of authority can create reasons: the power to do so is part of the very idea of authority (cf. Raz 1979, ch. 1). Prior to law, there is no reason to drive on any particular side of the road, but one arises as soon as the state stipulates on which side the citizens should drive. If the state rules that we must drive on the right, it thereby creates a post-legal reason so to do.

Admittedly, this involves a contingency: that the authority is not merely legitimate but effective, so that there is a reasonable expectation of conformity with the rules it creates. But this does not seem too much to require, since effectiveness is a general condition of the instrumental reasons underpinning a state’s authority. In one sense, the law’s role here is to generate authoritative conventions. At the same time, the rule is not mere convention. Imagine that (as sometimes occurs in certain countries) one arrives at a road being unsure what is the local driving practice. One knows the legal rule—drive on the left, say—but not whether it is observed. Other things being equal, one should drive on the left. Slowly.

Many standard examples of successful norm creation involve conventions, typically as coordinating rules or as content-determinations of some more abstract, pre-conventional norm. How many players to field on a soccer team? How should we return the ball into play? What side to drive on? Th answers to these questions may be to some extent arbitrary, even suboptimal; yet the very existence of an authoritative answer is itself valuable. Now we can have an organized game. Now we can drive with more safety. And so on. The precise content of these coordinating rules may be less important than the purpose they serve. Even a rule about the age of consent in underage sexual intercourse, which most people would regard as a mala in se offense, is partially conventional. It varies widely across jurisdictions and history. But the rule is valuable—morally valuable—in virtue of helping to articulate one boundary of permissible sexual activity, benefiting potential offenders as well as potential victims by its clarity.

In all these kinds of cases, the state has good ex ante reason for passing the relevant law. In effect, the state acts as a conduit, crystallizing those ex ante reasons into a more particular, practicable form—the moral force of which derives not from the enactment itself, but from its function. Generally speaking, the wrongfulness of a rule violation depends on the moral force of the rule. For a malum prohibitum rule, the moral force comes from its instrumental value, which depends, in turn, on the reasons the rule serves and how well it serves them.