I. Introduction
This article argues for the uniform definition of mens rea terms, such as recklessness and intention, in English criminal law.Footnote 1 It will be demonstrated in section II that the English courts have defined these terms in various ways in different circumstances. Section III will then consider a defence of this approach offered by Victor Tadros. This will be attacked from two perspectives in section IV. First, the role of mens rea terms in providing ex ante guidance to citizens hoping to avoid the threat of punishment will be considered. It will be contended that the criminal law's communicative function is compromised when the same words are used in different ways depending on the context. Secondly, it will be suggested that clear, uniform definitions of mens rea terms are required to limit adequately the discretion accorded to criminal justice actors, including trial judges and juries deciding questions of blame. Then, in section V, some practical difficulties emerging from this account will be discussed.
II. Two Examples
This section demonstrates that the English courts have shied away from defining mens rea terms uniformly. The jurisprudence on intention and recklessness will be focussed upon here, as these are the most widely used and discussed fault terms.
A. Intention
In R. v Woollin,Footnote 2 it was accepted that intention means either acting to bring about a particular result (“direct intention”) or acting in the face of the acknowledged virtual certainty that a particular result will come about (“indirect intention” or “oblique intention”).Footnote 3 This definition of intention was not provided in general terms. Lord Steyn was clear that “it does not follow that ‘intent’ necessarily has precisely the same meaning in every context in the criminal law. The focus of the present appeal is the crime of murder.”Footnote 4
This leaves open the possibility that intention might be taken to mean something other than purpose/foresight of virtual certainty in other contexts. Textbook writers thus have to write equivocal sentences such as “English criminal law consistently, though not invariably, permits inferences of intent to be drawn with respect to side effects which are the virtually certain outcome of [the defendant's] conduct”.Footnote 5 Trial judges are similarly cautious: they might have to give the jury directions on intention in the law of murder, whilst leaving the word undefined in relation to counts on the indictment alleging other offences.Footnote 6 The conservatism of the House of Lords in Woollin leaves room, in short, for inconsistency and uncertainty in English criminal law's understanding of intention.Footnote 7 The appellate courts will have to confirm whether foresight of a virtual certainty will lead to a permissible inference or finding of intention in relation to each offence,Footnote 8 or each area of the law (for instance, attempts).Footnote 9
The legislature has managed to complicate matters further, with the Serious Crime Act 2007 demonstrating how legislation might contain uncertainty over the meaning of intention, leading to guess work on the part of lawyers (and, it must be assumed, bemusement on the part of any laypersons who try to establish what the law is). Section 44 of the Act makes it an offence intentionally to encourage or assist an offence. By section 44(2), a defendant is not “to be taken to have intended to encourage or assist the commission of an offence merely because such encouragement or assistance was a foreseeable consequence of his act”. This is consistent with the definition in Woollin as the core question in the House of Lords' approach to intention is whether it is virtually certain that a consequence will be brought about and the defendant realised this. Mere foreseeability is not enough.Footnote 10 Foresight of a virtual certainty that somebody will be encouraged to commit, or assisted in committing, an offence will, nevertheless, not satisfy the mens rea requirement of section 44.Footnote 11 This is because it is a separate offence, under section 45, to encourage or assist an offence believing that it will (not might) be committed.
Foresight of a virtual certainty that the defendant's conduct will (not might) encourage or assist the commission of an offence is thus presumably not enough to lead to the finding of intention under section 44.Footnote 12 In this context, foresight of a virtual certainty that somebody will commit an offence constitutes “belief”, and will lead to liability under section 45 (if the defendant believes he will encourage or assist the commission of a certain offence) or section 46 (if he believes a number of offences might be committed, and is sure one will be, but is not sure which one).Footnote 13 Lawyers might well struggle to grasp these niceties in the definition of intention (law students do!). It is thus doubtful that laypersons, who must (at least in an ideal world) plan their conduct around what the law declares to be wrong, if they wish to avoid the threat of punishment, can understand it.
B. Recklessness
This same analysis can be applied to the case law and legislation on recklessness. Like intention, recklessness continues to mean different things depending on the situation. It is well known that, in R. v G and Another,Footnote 14 the House of Lords defined recklessness as follows: the defendant must foresee an unjustified risk attendant upon his conduct and go on to take it regardless.Footnote 15 In common with Lord Steyn's approach in Woollin, Lord Bingham started off his opinion in G and Another by noting that he wanted to “make it as plain as I can that I am not addressing the meaning of ‘reckless’ in any other statutory or common law context”.Footnote 16 If the context changes, so might the definition of recklessness. Once again, authors are forced into being equivocal: “English law has progressed to the point where there is, for almost all offences, now only one test of recklessness”.Footnote 17
The courts' approaches to recklessness support this line of thinking. First, where the defendant is acutely, voluntarily intoxicated, he can be “reckless” with regard to a circumstance or consequence, even if he was too drunk to notice the relevant risk at the time of acting.Footnote 18 There is no need to inquire into his foresight of risk at the time of becoming intoxicated.Footnote 19 All that need be established is that the risk was one that, if the defendant had been sober at the time of acting, he would have noticed.Footnote 20
Secondly, juries trying cases of “gross negligence” manslaughter can be told that they may convict if the defendant was “reckless” – so long as the “ordinary” sense of that word is being employed by the trial judge.Footnote 21 “Gross negligence” does not require advertence of risk,Footnote 22 and thus it can be assumed that “ordinary” recklessness does not either. Indeed, the Court of Appeal has made clear that G and Another as no relevance in “gross negligence” manslaughter cases.Footnote 23
Thirdly, where the defendant “closes his mind” to a risk, and claims that he was not consciously aware of it at the time of acting, he will still be treated as reckless.Footnote 24 The courts here will engage in a fiction that the defendant had decided to ignore the relevant risk, even in circumstances where the defendant claims he never foresaw it. Such a fiction is unconvincing because, surely, a person can only choose to ignore that which he believes exists.Footnote 25 What this rule seems to allow, then, is the conviction of inadvertent risk-takers who missed the relevant risk for unacceptable reasons.Footnote 26 This does further damage to the idea of recklessness as advertent risk-taking.
Finally, the legislature might create offences that define recklessness in a more expansive way than the House of Lords did in G and Another. For instance, the Uranium Enrichment Technology (Prohibition on Disclosure) Regulations 2004Footnote 27 provide that:
(2) For the purposes of these Regulations, a person is reckless as to whether a disclosure made by him might assist or enable any person who is undertaking or proposes to undertake a specified activity to undertake that activity if–
(a) at the time he makes the disclosure–
(i) he has recognised that the disclosure would create a risk that any person who is undertaking or proposes to undertake a specified activity might be assisted or enabled to undertake that activity; or
(ii) he is indifferent as to whether the disclosure would create such a risk or not; or
(b) the disclosure creates an obvious risk that any person who is undertaking or proposes to undertake a specified activity might be assisted or enabled to undertake that activity, but at the time he makes the disclosure he has failed to give any thought to the possibility that the disclosure would create such a risk.Footnote 28
Regulation 2(a)(i) coheres with the understanding of recklessness in G and Another. The other two conceptions (indifference and failure to think of an “obvious” risk) do not, and resemble understandings of recklessness developed following R. v Caldwell Footnote 29 (a decision which was departed from by the House of Lords in G and Another).
Foresight of risk is therefore central to recklessness in English criminal law, but it is not essential to it in all contexts. For all the talk of “general principles [being] laid down” in G and Another,Footnote 30 recklessness still means various things depending on the context.Footnote 31 Again, lawyers and laypersons might both develop a sense of real confusion when looking at recklessness as it applies throughout the criminal law. This makes decisions about what is, and is not, prohibited seem unnecessarily complicated. Despite this point, the English approach to defining mens rea terms has received theoretical defence from Victor Tadros. The next section outlines his argument.
III. Tadros on “Localised Concerns”
Tadros's defence of the approach to definition adopted in cases like Woollin and G and Another arises in the context of a wider discussion of the “general part” of the criminal law – those rules and principles that apply across all, or at least two categories of, offences.Footnote 32 Mens rea terms are, clearly, aspects of the “general part”Footnote 33 because they are used across different categories of offences. It is murder, an offence against the person, to cause the death of another human being whilst intending to kill or cause grievous bodily harm.Footnote 34 It is theft, an offence against property, to appropriate dishonestly the property of another, intending to deprive the owner permanently of that property.Footnote 35 Intention thus cuts across these different classes of offences. Although there are notable exceptions (for instance, dishonesty is only really an element of one category of offences),Footnote 36 most other mens rea terms operate similarly.
Tadros does not, however, think that this means that mens rea terms such as intention and recklessness must be overly “general”, i.e. defined uniformly throughout the criminal law. This is because, Tadros explains, different categories of offence, and even different offences within each category, are sensitive to what might be called “localised concerns”. The principles, policies and other types of moral and political argument that bear upon a polity's understanding of culpable wrongdoing will vary from context to context.Footnote 37
For instance, Tadros thinks Woollin's definition of intention works well in some offences (his examples include murder, theft and causing grievous bodily harm with intent),Footnote 38 but is less appropriate where the harm threatened is usually less serious (his examples include criminal damage, and attempts at less serious offences such as common assault).Footnote 39 In those latter circumstances, finding intention in foresight of a virtual certainty is illegitimate as it is likely to miss the core wrong that the criminal law is aiming to combat. In consequence it would be “crude” to define such a crucial mens rea term as intention, which is used throughout the criminal law, uniformly.Footnote 40
Tadros was writing before the decision in G and Another, but his thinking could be applied to the law on recklessness following that decision. The localised concerns which impact on how wrongs should be conceived of in the law of property offences mean, the argument would go, that a foresight-based understanding of recklessness should be applied there. Where these concerns change, perhaps because the defendant was at fault for not recognising a risk, recklessness might be understood in a wider sense. In the context of manslaughter, where the law is concerned with the taking of risks with people's lives, it might be appropriate to have an even wider conception of recklessness. Given these points, the adoption of one understanding of the word “reckless” would be over-simplistic and potentially damaging to the criminal law's enterprise of defining wrongs accurately.
Before challenging Tadros's defence of the current English approach towards definition in the following sections, it is important to be clear about what aspects of his account are not disputed here. There is no problem with Tadros's endorsement of a conservative approach to defining culpable wrongdoing, which is opposed to attempts to create “grand theories”, which explain in general terms large swathes of the criminal law and are likely, as a result, to miss important distinctions.Footnote 41 Tadros's fear, for instance, is that if the understanding of intention in Woollin were applied across the criminal law, some offences of intention would not be defined in a way that reflects fairly the relevant moral and political concerns relating to the defendant's culpability or the nature of his wrongdoing. Grand theories will tend to do this, and there is an obvious evil to be avoided here. The criminal law might well fail to maintain its necessary moral legitimacy if its rules are clumsy and indifferent to what are no doubt important concerns in the eyes of the citizens bound to live according to the criminal law's declarations, lest they be punished.
Furthermore, Tadros is correct that, to avoid this evil, different offences need different fault elements – a view shared by other writers.Footnote 42 What is taken issue with here is the idea that this makes it proper for courts and legislatures to use the same word to mean different things in different contexts. Tadros and the English courts (and particularly the House of Lords, in the mens rea cases mentioned in section II, above) appear to think that the argument from competing localised concerns of principle, policy, etc. validates this approach. They seem to view important mens rea concepts such as intention and recklessness as mere words – empty vessels to be filled with whatever meaning courts think the context calls for – rather than key tools in a communicative enterprise.
It is the increased popularity of the view that criminal law and criminal justice ought to be communicative exercisesFootnote 43 that first gives rise to the suspicion that this apathy towards the uniform definition of core mens rea terms is misguided. The next section explains how a focus on communication and discretion can be used to build a case for the clear and uniform definition of general mens rea terms such as intention and recklessness, whilst meeting Tadros's worry about localised concerns.
IV. An Alternative Approach
The principle of “fair warning” – that citizens should not be surprised to find out that they have committed a criminal offence – is of core importance to any defensible system of criminal law.Footnote 44 It helps to secure the rule of law, insofar as it ensures that the criminal law is capable of guiding conduct.Footnote 45
A crucial aspect of fair warning is that citizens should have access to clear and stable guidance on what is proscribed as culpable wrongdoing, so that they can avoid criminal punishment.Footnote 46 A key aspect of the communicative enterprise of criminal law is to provide such guidance. Effective communication, it will be contended here, requires a language that is comprehended by all potential and existing parties in a criminal case. It is not fostered by malleable mens rea definitions that vary with context. This argument will be developed from Antony Duff's discussion of punishment, language and communication.Footnote 47
A. Duff on Communication
Duff's core point is that, if trials are communicative endeavours leading to punishments which communicate authoritative censure, it is important that those involved in the criminal justice process can communicate with each other throughout, and this requires a shared language. It is not simply lawyers and judges who must be able to communicate effectively with the language of the law and be able to understand its nuances and inconsistencies. The language of the criminal law should be accessible to those lay actors who must argue about it to avoid the pain of punishment. It must also be accessible to those laypersons called to give judgment on their peers, be they jurors or magistrates. And it must be accessible to those lay members of the public who hear of the case. Otherwise, the wider message of criminal liability could be lost in translation.Footnote 48
A crucial aspect of this accessible language, for Duff, is that it allows citizens to argue about, and (if they disagree with the courts' view) ultimately to understand the nature of their culpable wrongdoing. He is concerned primarily that those being tried are able to understand the normative nature of their acts (why they are wrongful) and their mental states and attitudes (why they are culpable), but it will be contended here that this kind of argument goes beyond Duff's ex post perspective, and applies also in the area of ex ante guidance. It is from the anticipatory perspective that the need for clear and uniform definitions of mens rea terms begins to become clear.
In planning their lives, citizens must be able to understand properly the nature of what the law declares to be a criminal offence, and it is obvious that this will usually be conducted without the benefit of legal advice. Duff's points about the ability to translate the law for laypersons (developed in a discussion on the trial) are thus less crucial here.Footnote 49 In the trial, such translation is readily available; in fact the state provides it through arrangements such as legal aid. But by the trial stage, a vital avenue for communication has already been missed. It is surely vital that citizens can avoid being put on trial in the first place by understanding clearly the law's normative notions of wrongdoing and culpability and acknowledging the steps needed to avoid committing offences.Footnote 50 The law should thus be clear in its message, and applied consistently, so that non-specialists can utilise it effectively at the stage of planning their conduct, as well as in later stages of the criminal process. This need for clarity is particularly vital in the criminal law context, as misunderstandings can have severe consequences.
It is necessary to clarify exactly what effective communication is thought to involve.
B. The Boundaries of Definition
The paragraphs above might be read to constitute an argument for a return to the search for “ordinary” understandings of core criminal law concepts,Footnote 51 which could be grasped easily by laypersons by (moral) intuition. For those familiar with the history of mens rea terms in England and Wales, this will bring up (bad) memories of the search for “ordinary” recklessness in Caldwell,Footnote 52 and the search for a suitable definition of intention, which culminated in Woollin.Footnote 53 These exercises were not entirely useless, because it is important that, if laypersons are to understand the language in which it expresses itself, the criminal law is related sufficiently to terms that non-lawyers employ in their everyday lives.Footnote 54
To see why, consider if recklessness were to be defined as desiring to bring about a certain result. This would surely be an understanding of recklessness which had strayed too far from “ordinary” understandings, as it seems more akin to “everyday” understandings of intention. The language of the law, if it bore absolutely no relation to the language of laypersons, would be beyond citizens not because it was applied inconsistently (perhaps it would be applied consistently by faithful judges and juries), but rather because it had drifted too far from the normal language of blame utilised by citizens.Footnote 55 The law must be anxious to avoid this result. This is because a convincing account of the criminal law's legitimacy, on communication grounds, is that citizens should be able to identify with the criminal law and view it as theirs, rather than some “alien” imposition.Footnote 56 Inaccessible language is likely to appear “alien”.
This is, however, as far as the argument for “ordinary” understandings of criminal law concepts goes. It sets boundaries within which the law should operate when defining its concepts in semantic terms.Footnote 57 Recklessness can sensibly be defined, as it is in a range of contexts in English law, as the conscious taking of an unjustified risk. This is close enough to the “everyday language” and normative understanding of recklessness to be comprehended by a layperson, and this ensures that – although the law has adopted a definition of recklessness which might not quite match perfectly a layperson's intuitive understandingFootnote 58 – communication is possible because it is close enough and semantically graspable. Citizens could understand easily that they ought not to take certain risks that they have foreseen as being attendant on their conduct, or else they might be penalised.
The difficulty complained of here arises where that definition of recklessness is not applied consistently. As the citizen grasps recklessness in one context, he is surprised or embarrassed to learn that it might mean something else in another (see above). Although both of these definitions might end up being close to an intuitive understanding of recklessness, matters are complicated by the difference in semantic understanding propagated by the courts and the legislature. This is not conducive to citizens feeling confident in their ability to plan and communicate via the language of their law; a law that respects them as planning agents.Footnote 59 This has implications for the certainty with which citizens can view their prospective conduct or argue about their alleged misdeeds. Again, given the special context of the criminal law, such uncertainty can have severe consequences, which ought to be avoided.
If the language of the law and the language of extra-legal discourse are to be bridged successfully, so that a communication about culpable wrongdoing can be engaged in at the planning stage of conduct, it is crucial that citizens can be on firm ground in their understanding of the normative and semantic contours of mens rea concepts, rather than on shifting sands. They should not feel, when trying to engage with the criminal law's content on such a core matter, that they are engaged in a discussion with Humpty Dumpty, where the meanings and content of certain terms can change unexpectedly, frustrating real dialogue.Footnote 60 If a statute or common law offence can be committed intentionally, or recklessly, that should mean one thing, rather than various things.
This account raises immediate objections, which will be dealt with in the following sections.
C. Robinson on Conduct and Adjudication
The first challenge to the communicative account sketched above arises from the work of Paul Robinson, who has done most to divide criminal prohibitions and defences into “rules of conduct” and “principles of adjudication”.Footnote 61 Rules of conduct are addressed to citizens, and ought to be clear and concise in order to give meaningful ex ante guidance and communicate to the public the criminal law's prohibitions.Footnote 62 Principles of adjudication, by contrast, do not guide the actions of citizens. They are useful instead from an ex post perspective; when a court is assessing the blameworthiness of the defendant's alleged wrongdoing.
Given that courts are accorded time for thought and reflection about the defendant's actions (time which he might often be denied), Robinson contends that principles of adjudication can be drafted more imprecisely than rule of conduct, lest their ability to assess blameworthiness accurately be compromised.Footnote 63 A simple answer to the above confusion in the English jurisprudence, then, is that mens rea terms like intention and recklessness are always principles of adjudication: they play no part in giving ex ante guidance to citizens, and thus it does not matter if they are confusing or not.Footnote 64 If lawyers can understand these terms, and judges and juries can – with suitable direction – apply them, then that is sufficient. The argument for clear and uniform definitions of mens rea terms presented above thus fails.
There are three responses to this argument. First, Robinson provides uniform definitions of mens rea terms in his draft “Code of Adjudication” appended to the most worked-through model of his theory.Footnote 65
Secondly, Robinson himself notes that mens rea terms are not always principles of adjudication, because sometimes they will play a part in defining a wrong.Footnote 66 His work has concentrated on inchoate offencesFootnote 67 – where mens rea is an essential part of establishing when a person has “crossed the line” into criminal liability – but Andrew Simester and Winnie Chan also give the example of theft.Footnote 68 It is not wrong, in itself, to pick up an item of property such as a bag. It is wrong, however, to pick up a bag knowing that it belongs to someone else and intending to deprive the true owner permanently of her property in dishonest circumstances.Footnote 69 If this is the case, then surely those mens rea terms which make conduct wrongful need to be defined clearly to ensure that citizens are capable of planning their affairs so as to avoid potential criminal liability. Those mens rea terms are, after all, elements of conduct rules. They ought, then, to be expressed in clear and consistent language upon which confident plans regarding liability can be made.Footnote 70
This point might be taken further: if the criminal law prohibits the intentional causing or risking of a certain result, then a person who takes the time to think about whether or not he is intending a result (or simply taking the risk of its occurrence) is entitled, surely, to guidance on the possible penal consequences. If this is the case, then surely it does make sense to direct citizens not to do certain actions when they are doing them with certain mental states.Footnote 71 In fact, some writers seem to note that this is the role of mens rea terms: to ensure that fair warning and ex ante guidance were available to the defendant prior to the alleged offence.Footnote 72 This suggests that Robinson is being too restrictive in his concession that mens rea terms are not always aspects of principles of adjudication.
As argued above, since the criminal law is a communicative enterprise, such ex ante guidance as can be provided through the law's definitions of its mens rea terms should be presented clearly. Malleable definitions of mens rea terms hinder this process unnecessarily, and reduce the polity's confidence that the criminal law belongs to it. Uniform definitions would help to avoid this difficulty.
Thirdly, even if English law were reformed in order to implement Robinson's dichotomy between conduct and adjudication, it would presumably still use the same mens rea words (intention, recklessness, etc.) to give both ex ante and ex post guidance about wrongdoing and culpability. It would seem then that the definition of terms like intention would have to be consistent and clear in some contexts (where intention forms part of a rule of conduct for citizens), but not in others (where intention becomes part of a principle of adjudication for courts). This would produce a perplexing account of the criminal law, and it must be remembered that not all of those who Robinson would give principles of adjudication to will be trained legal specialists. Jurors would soon get confused if they understood intention in one context, but not in another. Those planning their conduct might be puzzled when they hear of court judgments that can only be explained by reference to the defendant's mens rea (or lack thereof), and fail to accord with their “conduct rule” understandings. Again, this would shake their confidence in the criminal law's declarations of culpable wrongdoing. It is therefore submitted that uniform definitions and a wide range of fault terms used across the criminal law would remain necessary from a communicative perspective, even if Robinson's conduct/adjudication distinction were applied faithfully.
Robinson's conduct/adjudication dichotomy does not, then, render redundant the communicative argument outlined above. It is now necessary to consider a second objection.
D. “Thin Ice” and Discretion
The second attack on the communicative account offered above concerns its transition from normative to semantic clarity. It might be wondered whether such a move is necessary, or whether it would defeat the purpose of the criminal law in punishing culpable wrongdoing.
This objection arises from a qualification to the requirement of “fair warning” of potential criminal liability: the “thin ice principle”.Footnote 73 The idea is that, as long as citizens can appreciate that their conduct is of dubious legality, they are sufficiently warned about the prospect of punishment.Footnote 74 This would mean, in the context of mens rea words, that variation in definition would be permissible so long as all definitions remained sufficiently close to the morally intuitive understanding mentioned above. In such cases, citizens presumably could sense their conduct sailed close to the wind, and would not be surprised unfairly if they were held liable for an offence. If so, the argument for uniformity in definition falls away: all that can be argued for is that mens rea words should be defined within the broad parameters of everyday (moral) understanding.Footnote 75
Again, there are two responses available to this claim. The first is that the “thin ice principle” is itself not conducive to a clear criminal law that citizens can feel confident in utilising when planning their conduct. In short, it undermines the requirement of “fair warning” to a very large extent. It is hence a troubling principle. This will no doubt strike some as being unproblematic because of the very function of the principle – to catch those who, although not caught by a core understanding of fault, are still, given their awareness of the dubiety of their conduct, bad enough to warrant conviction.Footnote 76
But acceptance of that viewpoint would lead to something of a Pyrrhic victory, because of a second, more troubling, concern over discretion.Footnote 77 The “thin ice” principle might allow a wide margin of appreciation for criminal justice actors, including those in the courts, in their definition or application of mens rea concepts. The risk is that this allows arbitrary distinctions (just when is this definition to be applied?) and bias to creep into the criminal justice system, as well as usurping the legislature's role in criminalisation decisions.Footnote 78 An additional benefit of adopting clear and consistent definitions of mens rea terms is that it would ensure that the discretion accorded to bodies like the police and courts to “do justice” in individual cases, would be reduced to a degree that these dangers could be minimised.Footnote 79
There is still the point, however, of what to do if there are perceived gaps, which those who should be convicted of an offence are falling through. Furthermore, there remains a need to respond to the localised principles that Tadros identified as being so important to the present English approach. These points are discussed in the next section.
V. Practical Concerns
It has been argued that, if the criminal law is to live up to its communicative ideal, citizens must be able to understand (normatively and semantically) and use the language of wrongdoing and liability adopted by the courts and the legislature in planning their conduct. Similarly, if the discretion accorded to criminal justice actors is to be controlled adequately, the criminal law's mens rea terms should be defined clearly and consistently.
But what are the courts and the legislature to do if they accept the argument from communication made here, yet want to ensure that the law can react appropriately to the competing principles and policies which affect different areas of the criminal law? As noted above, Tadros suggests that the courts may, in interpreting terms such as intention, be conservative and not seek to make generalisations which might damage some parts of the law. This approach results, however, in the courts using a small number of mens rea terms in different ways and leads to the potential confusion pointed to above. A better solution, it is submitted, would be to expand the range of mens rea terms presently utilised in the criminal law, and ensure that, when they were used, these words are defined consistently across offences.Footnote 80
A. A Way Forward
Consider again the definition of intention. Sometimes there does not seem to be a significant moral difference between meaning to bring about a certain result (for instance, death) and failing to be motivated sufficiently by the realisation that it is virtually certain that, if the defendant proceeds with his plan, that result will be brought about. It might be argued that the law would be missing something morally significant if it defined murder as only desired and purposeful killings. Therefore, foresight of a virtual certainty might be treated, for the law of murder, as being sufficient mens rea to secure culpability. This is the position at present in English law (with the added complication that nobody is sure if foresight of a virtual certainty is intention, or just something morally equivalent to it).Footnote 81
Rather than doing what the English courts have up until now done, however, and simply expanding the concept of intention in this area to include foresight of a virtual certainty, it would be more conducive to clarity if the mens rea of murder were expressed in terms of an intention to kill (or cause grievous bodily harm)Footnote 82or foresight of virtual certainty that death (or grievous bodily harm) would result from the defendant's action. “Foresight of virtual certainty” could be conceived of as a separate category of mens rea from intention. This is a solution which TadrosFootnote 83 and othersFootnote 84 have alluded to,Footnote 85 but it is submitted that the argument should be made far more strongly, and rooted in the communicative enterprise of the criminal law and the need for control over the discretion accorded to its institutions. It is unacceptable, because it makes inaccessible the message (and undermines the communicative enterprise) of the criminal law, to be conservative in the way that the English courts have been and, as a result, use the same words in different ways in different contexts.Footnote 86 The courts should thus refuse to do so, and should define mens rea terms in a uniform fashion wherever they are encountered.
The benefit of using different terms, rather than different definitions, is obvious. It would avoid the need for the courts to decide when intention ought to include foresight and when it ought not to. This decision would be left to Parliament in drafting an offence.Footnote 87 It would avoid problems such as in the Serious Crime Act 2007, where lawyers have to work out how the legislature meant “intention” to be defined in a particular context. Intention would mean, if foresight of virtual certainty were separated from it, meaning to bring about a certain result – a definition that coheres sufficiently with the “everyday” understanding of that term that it could be utilised effectively by citizens in planning or arguing about their conduct. It would also be clear to citizens what “foresight of virtual certainty” means, and it would not be difficult to explain to them that, where foresight of virtual certainty is viewed as being sufficiently culpable to constitute mens rea then the definition of the crime is satisfied. Plans could be made fairly, on this basis, to avoid liability.
By now, an obvious practical objection to this argument will have arisen in the mind of the reader: the courts cannot just invent new mens rea terms in the law of murder. The main difficulty focussed on by this suggestion is obvious when one considers the House of Lords' decision in Woollin. The Lords were tasked with setting out the law on murder, a crime with two forms of mens rea: an intention to kill or an intention to cause grievous bodily harm. In endorsing, with minor modification, the direction on intention developed by the Court of Appeal in R. v Nedrick,Footnote 88 the Lords accepted that foresight of a virtual certainty could lead to a finding Footnote 89 of intention in murder cases. They reached the conclusion that, in such cases, there is no significant moral difference between these two types of killing. In Tadros's terms, localised concerns pointed towards an expanded concept of intention, as that was the word with which the courts could work.
The argument above was that this was wrong as it complicated the law's understanding of intention and a preferable course would have been for the criminal law to declare that foresight of a virtual certainty was sufficient mens rea for murder. The House of Lords could not, of course, achieve this result. It would have been conspicuous, to say the least, to hold that, although the modern mens rea of murder had been settled for almost twenty years,Footnote 90 there was now a new form of guilty mind (foresight of virtual certainty) sufficient for conviction where the defendant had caused the death of another person. Concerns about precedent, and of judicial law making, prevented such a course.Footnote 91 These are the same concerns that ought to make the variable use of mens rea terms by courts troubling.
A similar difficulty is visible in the context of the definition of recklessness developed in the law of rape in the 1980s. By the late 1970s, rape required an intention to have sexual intercourse and either knowledge that the woman was not consenting, or recklessness as to her consent.Footnote 92 If recklessness was understood to mean advertent risk-taking in this context, then that would mean that a man, who failed to notice blatantly obvious signs of non-consent from his partner, would have to be acquitted of rape (in theory, at least). At least some such actors are going to be more morally reprehensible than some advertently reckless actors (i.e. the man who notes a minuscule risk of non-consent, but concludes it is not worth investigating). The wholesale exclusion of those who were simply indifferent – and, as a result, inadvertent – “rapists” from punishment was unacceptable; again, localised concerns pointed towards the expansion of the law's understanding of recklessness in this context. It is therefore not incredibly surprising that the courts moved to consider an indifferent actor as being reckless.Footnote 93
This no doubt did violence to the idea of advertent recklessness (which was already in tatters following Caldwell), but it is quite clear that the judges might have felt that they had no other choice. Once again, they could not simply declare that indifference was a new form of mens rea, distinct from recklessness (as, following the argument above, would have been ideal). So the integrity of recklessness as advertent risk-taking was further surrendered, beyond the extent to which the majority's decision in Caldwell had already done so, to ensure that those who surely deserved punishment for their culpable wrongdoing got it.
These decisions probably “did justice”, but only at a price: the clarity of the law for laypersons. Furthermore, the courts exercised discretion in the definition of mens rea terms that was not subject to clear boundaries. It is submitted that it would have been preferable for the courts to stick to a uniform definition of mens rea terms and require Parliamentary intervention. For instance, the legislature has, through the Sexual Offences Act 2003, amended the definition of the mens rea of rape to more clearly locate culpable wrongdoing, without endangering the clarity of the definition of recklessness in other contexts. If the defendant lacks a reasonable belief in consent, he is to be convicted if his partner was, in fact, not consenting.Footnote 94 The type of case mentioned above is thus captured well, and recklessness can go back to meaning (usually)Footnote 95 advertent risk-taking. This makes it easier to argue about the application of the law of rape to particular circumstances, and makes it clear to citizens (in normative and semantic terms) what the polity takes to constitute rape.
Judges would do well to spur on such legislative clarifications. If a statute defines a crime in terms of recklessness, but the defendant clearly did not notice the relevant risk, he might still appear culpable (and perhaps, though not always, was on “thin ice”). The trial judge might reach the conclusion that, had Parliament thought about it, they would have included inadvertent risk-taking within the ambit of the offence (by outlining a distinct definition of recklessness, or including a mens rea term that includes inadvertent risk-taking, such as negligence). The judge may feel inclined, in light of this conclusion, to fudge the issue of foresight for the jury.
The appellate courts might well support this, claiming that this is one instance where – because of concerns local to that offence – recklessness includes inadvertence (it has happened before and, in G and Another, Lord Rodger very clearly left the door open for it to happen again).Footnote 96 But the point is that this decision is based on supposition and no real democratic discussion of the localised concerns that the criminal law is meant to reflect. It would be preferable to permit the occasional “injustice” to occur, in order to allow a proper discussion to take place in the legislature.
An analogy can usefully be drawn here from the law on conspiracy. Trial (and appeal) judges were presented, in section 1 of the Criminal Law Act 1977, with an offence which very clearly required knowledge or intention as to every element of the contemplated offence before a conspiracy would be made out. They tried, nevertheless, to interpret that section in terms of mere suspicion to catch seemingly culpable actors.Footnote 97 This did violence to the notion of knowledge as it is applied more generally in the criminal law, and the decision in R. v Saik Footnote 98 – which re-emphasised the clear meaning of the statutory text – ought therefore to be celebrated. It makes clear that, if Parliament wants to capture such suspicious actors, it must say so specifically. The discretion to think up mens rea terms to capture distinct types of culpability is with them, in an ideal system, not the courts.
B. Two Concerns
Things being as they are, the obvious danger with the argument above is that Parliament would, hoping to avoid unmerited acquittals, make every offence overbroad, with the lowest form of fault element possible. They could, for example, make all offences capable of commission through negligence or (if they can justify it) offences of strict liability, to ensure that no “culpable” wrongdoers go free. This is no doubt a concern, but it would make it very clear to the public what Parliament was doing. However tough on crime “the public” is thought to be,Footnote 99 if every offence had a flimsy fault element (or no fault element at all) attached to it, convictions would be avoided through “jury equity”, and citizens would soon become attuned to the fact that the criminal law no longer reflected anything like their ideas about wrongdoing and culpability. As contended above, this would compromise the moral legitimacy of the criminal law, and jeopardise the sense of obligation that citizens have towards it. It must be asked whether Parliament would take this sort of risk. If not, then the concern about overcriminalisation is not unreal – and would have to be guarded against – but it is less significant than it might otherwise be.Footnote 100 What is required is keen attention to the role of different mens rea terms in criminalisation decisions, a topic which cannot be done justice here.Footnote 101
A second concern is to do with the potential for achieving rationality and coherence in a system as messy as English criminal law. The argument in this article presupposes a clear range of distinct terms applied consistently, and this would be difficult to impose on English law. One (perhaps easy) answer is that a wholesale codification exercise would have the best potential to realise this ambition. If a code could employ consistent language throughout, and develop a rich vocabulary to ensure its messages about wrongdoing and culpability were clear, then this would be a large step forward in endorsing the point defended in this article.
One objection to this is that the law might end up with a very rich vocabulary. There might be “Caldwell recklessness” (for the unthinking), “Majewski recklessness” (for the acutely voluntarily intoxicated) and so on, and this would result in a clunky and confusing text. This is not necessarily the case. Many of these judicial devices overlap with each other and they could be rationalised down to a manageable number. What would need to happen is a debate, and a long reflection, upon what it means to be culpable. The issues of morality and politics would have to be engaged with head-on, with an eye on the general landscape of the criminal law, rather than smuggled in through scattered decisions that scarcely look beyond the offence or circumstances at issue.Footnote 102 Parliament, not the courts, is the place to have such a debate, after preparatory work by institutions such as the Law Commission.
The real difficulty is, of course, that such a comprehensive clarification exercise is extremely unlikely to happen.Footnote 103 The Law Commission has – despite its raison d'etre being to reform the law, including in particular the codification of that lawFootnote 104 – abandoned the idea of a criminal code,Footnote 105 and it is easy to see why. Law Commission proposals for reform of the criminal law, if they are acted upon at all, often undergo an unfortunate metamorphosis during the Parliamentary process. The result is sometimes scarcely ever what the Commission envisioned, and their views might be utterly irrelevant when it comes to divining sense from a statute.Footnote 106 The reality is that a code which went through Westminster would almost necessarily be confused. The most plausible way out of the doctrinal “make-do-and-mend” attitude towards general definitions, outlined in Part II above, is probably closed.
This is no doubt a disappointing conclusion, but that is not because the theoretical point defended in this article is impractical in itself. It is the politicisation of issues of criminal law and justice that makes the quest for clarity and coherence – for mens rea terms that citizens can identify with readily and utilise effectively in planning their affairs or defending themselves in questioning or in court – beyond simple reach. This should be a cause for much concern, given the problems with the present approach to the definition of mens rea terms identified in this article. If the courts can make these difficulties more immediate for Parliament, so that a more sensible debate over criminal fault can commence, then they should.
VI. Conclusion
Tadros is correct to point out that different localised concerns may apply to the ascertainment of culpable wrongdoing in different contexts, and that – if it is to fulfil its function properly – the criminal law should be sensitive to this. This has, in English law, traditionally meant that a limited number of mens rea concepts are defined in various ways, but it was argued in this article that this approach is unhelpful. It impedes the communicative endeavour of the criminal law and its institutions, and the answer is not simply to point out that lawyers might translate this language for citizens charged with offences. Citizens should feel confident in their assertions about the law applicable in certain situations, and this means utilising a rich vocabulary which is applied consistently to identify, in normative and semantic terms, wrongdoing and culpability. They should also feel confident that criminal justice officials cannot frustrate such plans through exercising undue discretion with regard to the definition of mens rea terms in distinct contexts.
Once this is accepted, any mask of theoretical legitimacy should be removed from English law's present approach to defining key concepts used throughout the criminal law. The full realisation of law as a communicative endeavour is prevented at present by the politicisation of criminal justice and fears about what codification (the easiest manner in which to attempt coherence and consistency in the law) might bring. This should be a source of much regret.
