No one would deny that there is some important relationship between moral wrongness and the appropriate content of the criminal law. Many criminal offenses, including the most familiar offenses such as rape and murder, are also moral wrongs. And it is difficult to believe that their wrong-making features do not also sometimes provide reasons to criminalize them. This idea commits us only to a very weak view of the relationship between moral wrongness and the content of the criminal law. For example, it is consistent with the view that harm is central to the justification of criminalization. Some conduct may be wrong in virtue of its being harmful. And that characteristic of the conduct provides a reason to criminalize it. But the view leaves open the possibility that it is permissible to criminalize conduct that is not wrong if that conduct is harmful, and also the possibility that there is some wrongful conduct, namely non-harmful wrongful conduct, that there is no reason to criminalize.
For example, assault may be wrong in virtue of the harm that it causes. Those who believe that the criminal law is justified only because it prevents harm might justify criminalization of assault in virtue of the harm that it causes, and not because it is wrong. If this is true, harm explains the wrongness of assault and also justifies criminalizing it. In contrast, writing racist books in private may be wrong but not harmful. It may be wrong in that it expresses, through action, contempt for others. But if it is not harmful, writing racist books in private may not be criminalized. The mere fact that the conduct is wrong, defenders of this view might suggest, provides us with no reason at all to criminalize it.
Many people believe in a closer relationship between wrongness and criminalization than this. Here are two ideas. One is that moral wrongness is a constraint on criminalization decisions. On this view, only morally wrongful conduct can be criminalized. A stronger version of this idea claims that conduct may be criminalized only if that conduct is morally wrong independently of the creation of the criminal law (Duff 2010a). This view rules out what we might call “pure mala prohibita offenses.” Pure mala prohibita offenses are offenses which criminalize conduct that is wrong only in virtue of its having been criminalized. It is consistent with this view that it is permissible to criminalize conduct which becomes wrong in virtue of having been regulated in some other way.
The latter possibility is essential to the plausibility of the strong view. For example, it is implausible to rule out criminalizing driving on the wrong side of the road. But driving cannot be on the wrong side without some formal or informal regulation governing driving. Defenders of the strong view can explain this as follows. We are permitted to create regulations governing driving. That renders some kinds of driving wrongful. In virtue of that fact the conduct may be criminalized. The strong view only excludes the criminalization of conduct that is rendered wrong only by being criminalized.
A weaker version of the view that wrongness is a constraint on criminalization holds that it is permissible to criminalize some conduct only if the conduct, once criminalized, is wrong (Husak 2008: 72–77). On a strict reading, all tokens of the conduct prohibited must be wrong. This view admits the possibility of pure mala prohibita offenses. It may nevertheless be a significant constraint on criminalization. For we may be tempted to criminalize some conduct in order to pursue a social goal even if some tokens of the conduct are not wrong even after it has been criminalized.
For example, prohibiting sexual intercourse with a person below the age of 16 may be the most effective way of protecting vulnerable young people from sexual exploitation. But perhaps not all of the conduct prohibited by a criminal law of that kind is wrong even after the conduct has been criminalized, for example if the defendant is a young person himself or herself and his or her partner was a particularly mature 15-year-old. It depends on whether the defendant has a moral obligation to obey a law prohibiting this conduct once it is created. If he does, defenders of the weaker version of the view, unlike defenders of the stronger version, have no objection to its being criminalized. If he does not, defenders of the weaker version of the view will regard the criminalization of the conduct as troubling.
A second idea is that moral wrongness always creates a reason in favor of criminalization. Michael Moore endorses this “legal moralist” view (Moore 1997, ch. 16). It extends the very weak view that I outlined in the introduction to all cases of wrongness. It holds that any characteristic in virtue of which something is wrong provides a reason to criminalize that thing. This view is consistent with a range of constraints on criminalization. It may be that it is permissible to criminalize conduct only if that conduct is harmful. We might justify this on the grounds that the enforcement of the criminal law inevitably causes harm, and the harm that it causes must be outweighed by prevention of harm. But harm, on this view, may only play the role of defeating reasons against criminalization. Wrongness is a reason in favor of criminalizing all conduct. But a further constraint imposed by the harm principle must not be breached to justify criminalizing the conduct all things considered. This bears some similarity to Joseph Raz’s account of the harm principle within his perfectionist liberal political philosophy (Raz 1986; see also Moore 1997, ch. 18).
Here are some principles that reflect these views.
1) The Wrongness Justification. It is always a reason in favor of criminalization of some that the conduct is wrong.
2) The Strong Wrongness Constraint. It is permissible to criminalize some conduct only if that conduct is wrong independently of its being criminalized.
3) The Weak Wrongness Constraint. It is permissible to criminalize some conduct only if that conduct is wrong either independently of its being criminalized or as a result of its being criminalized.
My aim in this essay is to evaluate the first two principles. I leave consideration of the third principle for another occasion.
Before proceeding further, it is worth clarifying the ambiguous idea of wrongness. Wrongness, in this discussion, might be thought to refer either to prima facie wrongness, or wrongness all things considered. We might evaluate the principles as applied to either kind of wrongness. I will assume that the principles refer to wrongness all things considered. Providing defendants with a justification defense provides a way of decriminalizing conduct that is prima facie wrong but justified all things considered. So, for example, if killing in self-defense is prima facie wrong, but justified all things considered in morality, killing in self-defense can be decriminalized by making self-defense available as a defense to murder.
This invites reflection on a further question—what is the role of excuses? Typically, excused conduct is understood to be wrong all things considered, but the defendant is not blameworthy of perpetrating it. We might object to the Wrongness Justification on these grounds—there is no reason to criminalize conduct that is fully excused. We can respond to this trivial objection by amendment:
The Wrongness Justification (as amended). It is always a reason in favor of criminalization of some conduct that the conduct is both wrong and blameworthy.
This is the version of the Wrongness Justification that I will evaluate.
The Wrongness Justification
The Wrongness Justification is false if there are some things that are wrong that there is no reason to criminalize. Antony Duff and Sandra Marshall have developed an increasingly influential view that holds that there are such things (Marshall and Duff 1998; Duff 2007; Duff and Marshall 2010; see also Husak 2008: 135–37). On their view, there is good reason to criminalize all and only those things that are publicly wrong. Public wrongdoing is a restricted class of wrongdoing. It includes only those wrongs that the public ought to be concerned with in their role as members of a political community.
Here is how they develop this view. One component of understanding what it means for a person to be responsible for committing some wrong involves considering to whom one is responsible for committing that wrong. A public wrong is a wrong which an agent is responsible to the public for committing. The public is constituted by shared membership of a political community. The idea that an agent is responsible to the public for wrongdoing is to be understood as the agent having a duty to answer to the public for that wrongdoing. It follows that if an agent does not have a duty to answer for a certain kind of wrongdoing to the public, there is no reason for the state to criminalize that wrongdoing.
One further specification of this view is important in helping us to evaluate it. In Duff and Marshall’s most recent account of it, the idea of public wrongdoing is not intended to provide an account of criminal wrongs. It is intended as an account of wrongs that the state can regulate, where the criminal law is one among a number of potential state responses to that wrongdoing (Duff and Marshall 2010). This evades a concern that I have previously expressed about their account (Tadros 2010): that their view seems to justify incorporating civil wrongs within the criminal law. In their recent development of the idea of public wrongdoing, they suggest that whether some public wrong ought to be criminalized depends on alternative methods for dealing with the conduct. On this account, though, the idea of public wrongs is not really an account of the criminal law at all, but an account of the kind of wrongdoing that the state in general is concerned with.
On Duff and Marshall’s view, a person may lack a duty to answer for some public wrongdoing all things considered. For example, suppose that the police entrap me to commit theft. I may have a prima facie duty to answer for theft. Theft is a public wrong. But in this case I don’t have a duty to answer to the state for my theft, for the state lacks standing to demand that I answer to it (for more discussion of standing see G. A. Cohen 2006; Duff et al. 2007; Tadros 2009; Duff 2010b). Similarly, I may lack an all things considered duty to answer for theft because requiring me to answer would seriously interfere with my health, or the health of another, or because doing so would be too costly. These considerations do not undermine the idea that theft is a public wrong. It is a public wrong as a type of conduct even though it is not true that each token of theft results in an all things considered duty to answer for the conduct.
What if there is some conduct which cannot be criminalized by type for the reason that it will typically be the case that requiring a person to answer for that conduct will be too burdensome or costly? Consider dropping litter. It may be that in every case of litter-dropping there is no duty to answer to the state for dropping litter because requiring a person to answer to the state for dropping litter is always too costly. Let’s work on the supposition that this is true. My question is—on the Duff/Marshall view, what are we to say about this kind of case? Does the fact that there is never a duty to answer to the state for dropping litter imply that dropping litter is not publicly wrongful and hence that there is no reason to criminalize it?
This question could be answered in the negative, though as I will suggest in a moment, this is not what I take Duff and Marshall’s view to be. If it is, the following argument might be given:
1) Dropping litter is wrong in virtue of the nuisance that it causes to members of the public.
2) That provides a reason to consider it a public wrong.
3) The fact that demanding citizens to answer for it would always be too costly defeats this reason to consider it a public wrong.
4) Hence it is not a public wrong. And hence there is no reason to criminalize it.
The difficulty with this argument is that it adds unnecessary complications to an argument available to friends of the Wrongness Justification. Friends of the Wrongness Justification argue as follows:
1) Dropping litter is wrong in virtue of the nuisance that it causes to members of the public.
2) That provides a reason to criminalize it.
3) The fact that demanding citizens to answer for it would always be too costly defeats this reason to criminalize it.
It is not obvious what the extra complications in the first argument add to the second. If we object to the Wrongness Justification only because of the extra steps added by the first argument, the objection is trivial. In contrasting their view with Michael Moore’s legal moralism, Duff and Marshall think that we should distinguish between wrongs that we should leave the parties merely to sort out between themselves and wrongs which the state should intervene in (Duff and Marshall 2010). But Moore could respond that it is consistent with his view that we should leave it to people to sort out litter-dropping between themselves and 3) is the reason why.
The better interpretation of Duff and Marshall’s view, the one that I suppose they intend for this kind of case, suggests that dropping litter is a public wrong even if it cannot be criminalized. They may argue as follows:
1) Dropping litter is wrong in virtue of the nuisance that it causes to members of the public.
2) This renders it publicly wrongful.
3) That provides a reason to criminalize it.
4) The fact that demanding citizens to answer for it would always be too costly defeats this reason to criminalize it.
This argument implies that the costs of criminalizing conduct are independent of whether the wrong is public. It also implies that there is a difference in kind between the considerations that determine whether some wrongful conduct is publicly wrongful and the considerations that determine whether a public wrong ought to be criminalized. To demonstrate that the Wrongness Justification is false, the quality of these special considerations needs to be clarified.
The natural way to do this is to show that there are practical considerations, such as cost, that may render the criminalization of some public wrongs unwarranted that are distinct from the principled considerations that render some wrongs nonpublic. We require a robust way of distinguishing between these “in principle” and “in practice” considerations. This way of understanding Duff and Marshall’s view is clearly implied in Duff and Marshall (2010). In that piece they explore alternative methods of dealing with some conduct that they believe is publicly wrongful. The principles that distinguish public wrongs from the more general class of wrongs remain elusive, however. If all of the considerations that determine whether a wrong is public are of the same kind as the consideration of cost in the example of litter-dropping, there is no important difference between Duff and Marshall’s view and legal moralism.
It is by no means easy to make a case for principled restrictions on criminalization. Take Duff’s favorite example of adultery (Duff 2007: 144–45). Adultery may reasonably be thought at least sometimes to be wrong—it is a breach of a legitimate expectation of fidelity created by marriage. But is it publicly wrong? Duff tends to think not (though he is not completely unequivocal about it). But we might think that the reason why we should not criminalize adultery is because doing so would tend to result in unhelpful intrusions into relationships, or because it is very difficult to determine whether adultery is wrong in any particular case. These do not seem different in kind from considerations of cost. So why not conclude that adultery is publicly wrongful?
In Answering for Crime, Duff does provide an account—though one that I think is not fully developed—of what might do the work here. When he discusses public wrongs, he indicates that there are some wrongs which also violate the values of the political community, and it is this fact that renders those wrongs public. He writes:
A justification of criminalization will need to begin by specifying some value(s) that can be claimed to be public, as part of the polity’s self-definition; show how the conduct in question violates that value or threatens the goods that it protects; and argue that that violation or threat is such as to require or demand public condemnation.
(Duff 2007: 143)
He then indicates that examples must be discussed on a case-by-case basis. This view seems to depend on the idea that there are some values that render conduct wrong, but as they are not values of the political community, the conduct cannot be criminalized in virtue of those values.
One difficulty with this idea is in identifying the values of the political community. Obviously, the account must be normative—we are concerned not with the values that the political community actually endorses, but rather with the values that it ought to endorse. The fact that the political community values things that it ought not to value cannot provide it with reasons to criminalize things that appear wrong to it in the light of those values. And the fact that the political community fails to value things that it ought to value cannot provide it with a reason against criminalizing things that are wrong in the light of those values.
Perhaps there is a more modest relativism in Duff’s account whereby the values that a political community ought to endorse depend to some extent on its particular features. That leaves open the possibility that there are values that some people rightly endorse and conduct can be wrong in the light of those values. But as the political community lacks a reason to endorse that value, it also lacks a reason to criminalize that wrongful conduct. I am not sure that this idea can be made compelling.
The core of my concern relates to the idea that we should begin with the polity’s “self-definition.” I am not sure what a polity’s “self-definition” is, or what it means to claim that a value is public as a part of it. Take property rights. The state ought to protect property rights. As theft violates those rights there is good reason to criminalize theft. But I am not sure how the value of property rights or the goods that it protects form part of the state’s self-definition. What advance does the idea of self-definition make over specifying the legitimate functions of the state?
Furthermore, I am not clear that the polity’s self-definition, insofar as it is distinct from the proper functions of the state, can be established prior to the criminalization of conduct. There is a whiff of circularity about Duff’s suggestion. What a state criminalizes seems to determine its self-definition, but its self-definition is supposed to govern what it criminalizes. Perhaps we need some clear examples of things that are wrong in virtue of some value that is not a value of the political community. I doubt that adultery is a good example, as it is difficult to see why the state has no interest in the value of people abiding by the legitimate expectation of fidelity that they create through marriage. There are many reasons not to criminalize adultery, but I doubt this is one of them.