Beyond culpability

So we may quibble about the close – too close – identification that is sometimes made between mens rea and culpability. Admittedly, mens rea is indispensable to culpability, in that we need more than a bare actus reus to conclude that a defendant deserves the official sanctions of the criminal law. Yet the contribution made by mens rea is not sufficient either; and its role in establishing fault is not a simple one.

More importantly, however, we should question the conventional association between mens rea and culpability at a deeper level. Mens rea is relevant to culpability, yes. But it serves other purposes too, purposes that are quite distinct and just as important. The existence of these other functions is too often overlooked, and is liable to be concealed by the depiction of mens rea simply as a “fault” element.

Broadly speaking, mens rea serves two main classes of function, and within that grouping it serves at least four particular functions. First, it helps to establish the moral innocence or guilt of the defendant's conduct (and as such, affects sentencing as well as conviction). We have noted already that it does so partly by contributing to findings of culpability. Additionally, as we see in the next section, sometimes mens rea constitutes the morally wrongful character of D's behaviour in a more fundamental way, by identifying what kind of action D is performing. In these cases, the finding of mens rea is integral to the wrong for which D is being held responsible.

The other functions concern liberty. Mens rea has important roles to play in articulating, and notifying, the limits of citizens' freedom. More specifically, one purpose discharged by mens rea is to secure fair warning to defendants, ensuring they have sufficient advance notice that, by their conduct, they risk violating the criminal law. Secondly, mens rea plays a key mediating role in criminalisation, being part of the trade-off between the protection of potential victims and the preservation of liberties for potential defendants.

These further roles often demand more stringent mens rea standards for liability than is needed to establish culpability. They help to explain why negligence, and sometimes even recklessness, is excluded in certain contexts of the criminal law, notwithstanding that one so acting might be culpable. In what follows, we discuss each role in turn.

Establishing the wrong

The second function sometimes performed by mens rea, then, is to establish the morally wrongful nature of a defendant's behaviour. As a general proposition, criminal conduct must be wrongful. It must be the kind of behaviour that a person ought not to do,Footnote ¹⁹ for which blame and punishment are potentially apt. We don't blame people for doing good things. We blame them only for doing things that are morally wrong. Notice that this is not just a point about culpability. There is a general objection to punishing the blameless in criminal law, but conclusions of blamelessness can be attributable to different sources. One may be blameless because, although one's conduct was wrongful, one was not culpable in respect of that conduct – as when one lacked mens rea, or had an appropriate excuse. But one may also be blameless because one's conduct was not wrongful in the first place. Thus the primary step is to establish whether the conduct is suitable for blame at all. This primary step not only affects conclusions about the defendant's moral position; it also shapes the justificatory rights of others to assist or resist him.

But what makes conduct wrongful? Let's start by thinking about the actus reus in isolation. Often the answer is straightforward. What makes D's behaviour problematic, and a candidate for potential blaming judgments, is some observable harm that it causes. If D breaks V's arm, or vandalises his property, the conduct is undesirable – the actus is reus – just in virtue of its external elements, and independently of D's mens rea. In these kinds of cases, the conduct is the very source of the wrong. Correspondingly, the criminal law's interest in preventing such conduct arises out of the harmful nature of the actus itself. These are the classic types of cases likely to influence us toward thinking of mens rea as a fault element. The actus reus establishes the harm; and we can move directly to the second stage, of establishing culpability.

On other occasions, however, the actus is not inherently reus. The external elements of D's conduct may not by themselves disclose anything wrongful or undesirable. Suppose that D stalks toward V with fist drawn back. What is the moral character of D's conduct? It depends. Perhaps D intends to strike V. Or perhaps D is a criminal law professor, acting out a scenario in order to demonstrate the moral ambiguity of such cases to his students. Whether D's conduct is a wrong of attempted assault depends on D's intentions. It is only if he means to strike that D's otherwise lawful conduct becomes a wrong. The same is true of theft. Picking up a glass of beer from the bar counter is not the stuff of theft, although it is the actus reus of theft. It lacks the nature of theft, unless and until it is done dishonestly with intent to deprive. In cases like these, D's mental state is integral to the moral character of his action. Without it, there is no reason to think of his conduct as wrong, as the kind of morally problematic conduct that warrants preventive measures by the state.

Helping to constitute the wrong is a standard role for mens rea. Consider, for example, the former English law of indecent assault. In Court,Footnote ²⁰ D placed a 12 year old girl, fully clothed, across his knees and beat her on the buttocks. The House of Lords ruled that such conduct, if a disciplinary measure, was a mere assault; but if done for sexual gratification, the assault was indecent. The decision allows that some actions are not, by their external elements, inherently wrongful, but are capable of being made wrongful when done for certain kinds of reasons. Court's ulterior motivation was not a fault constituent. Rather, it made his conduct the kind of wrong it was.

There are many examples like Court to illustrate that the character of an action cannot always be determined by reference to its external, actus reus features alone. Glanville Williams put the point more generally: ‘the act constituting a crime may in some circumstances be objectively innocent, and take its criminal colouring entirely from the intent with which it is done.”Footnote ²¹ Where that is so, the law may have no regulatory interest in D's conduct unless it is done for certain reasons. For these varieties of crime, mens rea does not operate simply to establish fault (although it also does that): because, in itself, the actus reus does not specify a wrong. It is only when accompanied by mens rea that there is an action meriting the attention of the criminal law. As such, the mens rea requirement does far more than simply establish a connection between the eventual harm and D's culpability.

Our modern criminal law contains some very important classes of offences that involve mens-rea-dependent wrongs.Footnote ²² These include most economic wrongs, such as exploitation and predatory behaviour; and indeed blackmail, where the distinction between a threat and a warning depends on the defendant's motivation.Footnote ²³ To some extent the dependence of economic wrongs on mens rea is an inevitable feature of marketplace transactions, where the actions of an economic actor are always likely to affect the interests of others. D may know that, by pricing her goods lower, she will take business away from V, and perhaps even force V out of business. Yet D does no wrong unless she prices her goods accordingly in order to drive V out of business. Here, the distinction between a side-effect of economic activity and an intended effect is central to our assessment of the wrongfulness of D's conduct. V's economic interests are by their nature exposed to the adverse effects of others’ economic activity: there is no general duty to avoid injury to economic interests, and the mere fact that those interests are harmed cannot establish that D's act wrongs the victim. Hence it is only the intention with which D acts that makes her conduct properly described as a (wrongful) attack on V. In a broadly free market, persons are left alone to pursue their own interests. In so doing, they wrong no-one even if their self-advancing conduct has deleterious side-effects for others. Those side-effects are a natural feature of competitive interaction. However, where they are deliberately sought, D is no longer pursuing her own interests but, instead, attacking the interests of others. That is morally wrongful; and it is wrongful even if she attacks those interests merely in order to advance her own position (a characteristic of wrongs such as dumping).

Another significant class of intention-dependent wrongs comprises pre-emptive crimes. Pre-emptive crimes seek to prohibit conduct that is not in itself harmful, but which becomes problematic when perpetrated for nefarious ends. There is nothing wrong with buying fertiliser; but if done with intent to make a bomb, an offence is committed against section 5 of the Terrorism Act 2006.Footnote ²⁴ Pre-emptive offences often involve a trade-off between actus reus and mens rea. Sometimes, it will be better to criminalise a more specific actus reus, e.g. possession of an automatic firearm. The more precisely defined offence offers better, more detailed guidance to citizens about their freedoms and obligations, letting them know where the law stands by marking out what counts as wrongful conduct clearly in advance. But as the inchoate law of attempts itself demonstrates, sometimes this precision about the actus reus is not available. In such cases, we rely on the defendant's intention to help mark out those inchoate acts that are wrongs, and which constitute a criminal attempt, or a crime against the Terrorism Act 2006, or some other preparatory offence. Moreover such intention-rich offences, despite their intangible actus reus, will frequently supply greater accuracy when labelling D's criminality, because they more clearly identity the moral character of D's behaviour.Footnote ²⁵ A conviction of some (hypothetical) offence of buying controlled fertilizer may not adequately describe D's wrongdoing in a case where D's intention was to manufacture a bomb. In such cases, doctrines of mens rea enable the law to establish flexible prohibitions while controlling the risk of attaching liability to morally innocent conduct.

Fair warning

Let us turn now to the second class of mens rea functions, which concerns the protection of citizens from state intervention. One form of contribution is through what scholars often call the “fair warning” principle.Footnote ²⁶ This is not quite the same as the overlapping principle that the liability of citizens should be ascertained by determinate legal rules, and we need to begin by getting that distinction clear. Both involve rule of law values. The latter, which is sometimes known as the principle of legality, requires that no-one should be convicted of any offence unless a law exists that designates her conduct a crime: nullum crimen sine lege, no crime without law. It also requires that the law should be prospective, not retrospective. Nowadays, it is a constitutional matter: the common law contains a presumption against retrospective crimes, buttressed by Article 7 of the ECHR, which plainly interdicts retroactive criminal laws. The law, especially the criminal law, should be fixed and definite, not alterable. The potential for criminal liability of a would-be actor should exist before she acts, not originate in the subsequent discretion of others.

The legality principle requires that liability should be determined by the rule of law, and not the rule of officials. By contrast, concern for fair warning brings out a slightly different dimension of the rule of law: that its rules should offer advance guidance to the persons who are subject to them. An unpublished law may be constitutionally valid, for instance, but it is defective as part of a system of rules designed to guide society's behaviour – like the doomsday device in Dr Strangelove, it fails as a deterrent. The distinction is helpfully brought out by Lord Bingham in Rimmington and Goldstein:Footnote ²⁷

Bingham's second principle is the basic constitutional one: that if something is not illegal, no liability can follow even if it ought to be illegal, and indeed even if most people thought it was.Footnote ²⁸ His first principle, on the other hand, concerns predictability and fair warning. Individuals need to be able to predict the legal implications of their actions in advance, so that they have fair warning that their contemplated actions will risk incurring a criminal sanction; so that they know where they stand and can properly decide what to do in light of the law's guidance. With secure, reliable law, citizens are enabled to steer clear of criminal liability by choice, to navigate their way through life without unexpected legal consequences befalling them. Absent such law, the ability to plan one's life is undermined.

There is a growing tendency to constitutionalise this principle too. Fair warning ordinarily requires not only that criminal laws exist prospectively, but also that they be stated clearly and in a reasonably precise fashion. Extending the Article 7 jurisprudence, the European Court has emphasised the aim that “the individual can know from the wording of the relevant provision and, if need be, with the assistance of the courts' interpretation of it, what acts and omissions will make him liable.”Footnote ²⁹ That ambition seems hard to contest. None the less, caution is appropriate here. The goal of fair warning is not to ensure textual certainty for its own sake. What really counts is whether D can anticipate liability in the case at hand – whether criminal liability comes as a surprise, as something contrary to ex ante expectations.Footnote ³⁰ A focus on predictable liability rather than legislative clarity leaves space for what Andrew Ashworth calls a “Thin Ice” principle:Footnote ³¹

To be sure, obscure or discretionary (or retroactive) laws are likely to be insufficiently predictable, likely to offer inadequate ex ante guidance to citizens. But, if the objection is one of fair warning rather than the constitutional one of retroactive criminalisation, disappointing the ex ante expectations of a defendant may carry no weight when those expectations are unjustified. And expectations may be unjustified because of the kind of conduct the defendant contemplates, not just because of clearly-drafted prohibitions. One illustration of this is the well-known decision in R. v. R,Footnote ³³ where the House of Lords rejected spousal immunity in the common law crime of rape, thereby upholding the defendant's conviction of attempting to rape his wife. The decision is defensible on a thin-ice basis: when interpreting the current law of rape, R's anticipation of a legal immunity generated no reason for the courts to find one. It should have made no difference had R obtained an opinion from a lawyer, immediately prior to making his wife submit to intercourse, that in legal terms he would not be committing rape. His conduct was profoundly wrong, the very wrong of rape. This is not to say that courts may invent law retroactively in order to catch husbands who rape their wives. (That would violate the constraint of legality.Footnote ³⁴) But it is to say that, in considering what the current law is, the fact that the defendant thought it was different does not count for much in circumstances like these.Footnote ³⁵ The goal of fair warning is met here.

Understood in this wider sense, of liability's being anticipatable and not surprising, fair warning is also supported by mens rea requirements. H.L.A. Hart saw this clearly.Footnote ³⁶ Indeed, for Hart, warning rather than fault is the basic function of mens rea. By proving intention or recklessness, the prosecution establishes that D chose to perpetrate the actus reus; this, in turn, serves to ensure that D had (and rejected) the opportunity to avoid wrongdoing, and so to evade the attentions of the criminal law.Footnote ³⁷ Perhaps Hart exaggerated. Certainly, he underestimated the distinctive importance of fault in the criminal law.Footnote ³⁸ But he was right to observe that, by ensuring that the defendant has notice of his actions, mens rea requirements foster the rule of law.

The fair warning role of mens rea is especially important in the contexts of result crimes and omissions liability. Crimes of these sorts often involve no particular form of conduct, no conduct of a kind that might put the defendant on notice that – lo! – a conviction is at hand. Result crimes such as manslaughter, or criminal damage, require only that death, damage, etc., be caused by any act. If I walk into a new building and switch on the light, I may have no idea that, because of some fault in the wiring system, an explosion will result, perhaps causing fatalities. And there is nothing distinctive about switching on a light to alert me to the potential for danger.

The same holds for omission-based liability, which is by definition activity-unspecific. We do not complain of what the defendant does, but of what he does not. Yet there are so many ways in which one can fail to rescue another person, so many things one may be doing instead. Perhaps I am talking on the ‘phone, or walking to work, or eating my lunch, and romancing a new friend, or writing an article.… Any of these things may be engaging my attention while, at the same time, I am failing to rescue the victim. Easy to see, then, how liability might come as a shock.

Considerations of fair warning supply one ground why we tend to restrict omissions liability to specialist actors, such as parents and doctors. Part of the reason for our willingness to impose positive legal duties on such actors is that we can expect them, in virtue of their roles, to be on notice of the need to act; to look out for the interests of those to whom they owe such duties. Even then, positive obligations have the potential to surprise. There are so many things that a parent has to do each day, some involving her children and many not, that we cannot reasonably expect any parent constantly to be alert to the possibility that something may be amiss, constantly to be thinking about her charges to the exclusion of the rest of her life.

Mens rea requirements address this problem by helping to make sure that the result, or the omission, is not out of the blue. For those types of crime in particular, there is especial reason why stigmatic criminal liability should normally be confined to reckless or intentional wrongdoing. Otherwise, liability for inadvertent harms can make it too difficult for citizens, even citizens owing specific legal duties, to pursue their own lives without fear of unexpected criminal charges. If whatever activity they are engaged in does not itself give them notice of disaster, then we must rely on mens rea to ensure they have it.

John Gardner, however, would go even further. Advancing what he describes as “the mens rea principle”, he argues that “criminal wrongs should be such that one does not commit them unless one intends or is aware of at least one wrong-making feature of what one is about to do, such that (assuming one knows the law) one is also alerted to the fact that what one is about to do will be of interest to the criminal law.”Footnote ³⁹ Thus Gardner requires awareness rather than notice. The rationale for his principle is also rooted in fair warning arguments, along the lines advanced originally by Hart:Footnote ⁴⁰

There is much to admire in this analysis, including its rule-of-law commitment to protect subject-citizens from oppressive and unexpected use against them of the state's power. But Gardner's conclusion now strikes us as being too strong.Footnote ⁴¹ This is for two reasons. First, to the extent that the rule of law enjoins unpredictable interference by the State, liability for negligence is reasonably predictable. Moreover, unless negligence liability be no deterrent, the argument prima facie deprives victims of protection from similarly unlooked-for intrusions by other citizens. The demand that notice be “stark” seems unduly weighted in favour of defendants. At least where the interest at stake is sufficiently important, as in manslaughter, a mens rea standard of gross negligence may be defensible.Footnote ⁴²

Secondly, and perhaps more importantly, Gardner's requirement that a criminal wrongdoer must intend or be aware of at least one wrong-making feature of the action neglects the gateway role of some activity-specific prohibitions. Perhaps the best example of this is rape. There is nothing wrong, or wrong-making, about sexual intercourse. But one reason why it is acceptable to criminalise negligence with respect to non-consensual sex, as many jurisdictions have, is that matters of consent are so closely bound up with the very nature of sexual intercourse that those engaging in sex are, implicitly, on notice that questions of consent are in play. The context is so intimate that alertness to the will of one's partner can be expected of every responsible actor, and a failure to exhibit such alertness is itself ground for condemnation. It isn't like carelessly tossing a stone aside, only (surprise!) to hear the sound of breaking glass. To make criminal damage an offence of negligence would be unfair because there is not always reason to inquire about the circumstances of one's conduct. Sometimes the unexpected happens. But sex is a specific activity, and not like behaviour that happens to cause property damage. It is distinctive in a manner that generates an immediate and obvious duty to consider whether one's partner consents.

Gardner's principle is absolute, but in our view fair warning constraints admit of a trade-off; one in which mens rea plays a part. Notice of impending criminal liability must be adequate rather than stark. Some activities are more obviously dangerous than others, some interests more in need of protection. The same holds for omission-based duties, some of which are more specific than others. The anaesthetist may be surprised to find that a tube is detached,Footnote ⁴³ just as the railway gatekeeper may be startled to discover the gate left open.Footnote ⁴⁴ Neither can complain, however, that these events were unlooked-for. In other cases, by contrast, it may only be the presence of mens rea, and in particular of intention or advertence, that guarantees sufficient warning of imminent liability.

Negotiating the parameters of criminalisation

The fourth role of mens rea is the least familiar. It too concerns the relationship of citizens with the state. As well as helping to secure fair warning, the so-called “fault” requirements in offences can also play an important role in mediating the scope of the prohibitions themselves, especially for those offences where the mental element is not required to identify a wrong.

It is common, indeed too common, for criminal lawyers to think of the criminal law in ex post terms, where the defendant has already acted; and our task is to decide whether she satisfies the criteria of an offence, including all of its actus reus and mens rea components. At this ex post stage, the temptation is to focus on the particular defendant, the character and circumstances of her conduct, and on whether she deserves punishment for what she has done. Seen from that perspective, it is easy to conceive of the offence specification as setting the criteria of punishment. But, as our discussion of fair warning suggested, that conception is incomplete. The offence should exist even before X is done. This is because the primary role of an offence is, ex ante, to articulate a prohibition. Do not do X in future, the prohibition says, and it backs that up with a threat of punishment.Footnote ⁴⁵

This ex ante role is central to the operation of any offence as deterrent. It is the threat of punishment, not the punishment itself, that is meant to deter would-be wrongdoers. The question of punishment only arises, in a sense, once the prohibition has failed. So we should begin by considering offences from an ex ante, deterrent perspective: as measures of criminalisation. (And, when interpreting offences, we should be less tempted to respond to backward-looking considerations about the culpability of the particular defendant, and driven rather more by the forward-looking implications of a given interpretation for the scope of activity the offence prohibits.)

Seen from that perspective, our point here is that the content of the prohibition includes the mens rea specification, and that the nature of its mens rea requirement directly affects the deterrent scope of the proscription.

By way of example, let us start with a putative offence where the actus reus comprises sexual intercourse with a person under the age of 16 years. Consider, next, what this offence leaves us free to do. The answer depends on its mens rea requirement. If, to commit the offence, D must intend or know that her partner is aged under 16, then D is left with considerable freedom of action. She must refrain from sex only when she knows or purposes that her partner be underage.

Contrast a mens rea element of (subjective) recklessness. If recklessness suffices for a conviction, it follows that D must refrain from sex whenever she suspects that her partner may be underage. By contrast with the knowledge-based offence, the scope of the prohibition is become larger: the former version permits her to proceed despite realising V may be underage; the latter does not.

All the more so when, as currently under English law, the mens rea element is negligence. Now D is not at liberty to have sex unless she takes care to check V's age. We do not mean by this to suggest that a prohibition of negligent under-age sex is inappropriate. Our point is simply that it is broader, more intrusive, and that its wider scope needs to be justified. Broader still would be the extreme case of strict liability, in which case D could only be sure of avoiding liability by refraining from sex altogether.Footnote ⁴⁶

Similar analysis applies to an actus reus such as causing damage to another's property. Current English law leaves people with the latitude to be careless but not the freedom deliberately to take risks.Footnote ⁴⁷ The difference may look like one of culpability; but it is also, and perhaps even primarily, a difference of liberty. When contemplating whether to prohibit conduct, one should not only consider the magnitude and seriousness of the harms involved. One should also consider the value of the conduct to be prohibited and the impact of its proscription upon the freedom of citizens.Footnote ⁴⁸ Mens rea elements can help to negotiate these considerations. They help to balance the invasiveness of a prohibition against the interests of those whom the law seeks to protect. Setting a more stringent mens rea requirement means that the prospect of liability will have less impact on the day-to-day lives of law-abiding individuals. Simply put, persons will be required to alter their behaviour on fewer occasions – e.g., only when they know, or suspect, or ought to know (as the case may be) about the risk. By pitching the standard of mens rea higher, we make it easier for citizens to comply with the law, and we reduce the range of activities they have to sacrifice in order to avoid the risk of a criminal conviction.Footnote ⁴⁹

The criminal law is designed to safeguard us from threats of harm. But criminal prohibitions are themselves threatening, in so far as they coercively restrict our freedom. Ordinary citizens must not only be shielded from the predations of others; they need also to be shielded from the state, which should not intrude upon their everyday, innocent activities. Quite apart from its other roles, the requirement of a mental element has an important function to play in keeping citizens outside the scope of criminal prohibitions. Mens rea, that is to say, protects freedom.