THE ILLEGALITY DEFENCE AND SANCTION-SHIFTING: IN DEFENCE OF GRAY V THAMES TRAIN LTD

Abstract

This article considers the rule that a claimant who has been wronged will be denied recovery where the damage flowed from a sanction imposed as a result of their own illegal acts such that compensating the claimant would divert a sanction intended to be imposed on the claimant to the defendant. The article has two purposes. The first aim is to provide a counterweight to the overwhelming body of academic literature critical of Gray v Thames Trains Ltd. in which the House of Lords, in applying the illegality bar found it unnecessary to examine the purpose of the criminal sanction against the claimant, preferring to treat its existence as sufficient to lead to a denial of recovery. The article argues that academic support for adoption of an alternative test of “significant personal responsibility” rests on precarious grounds, depending, as it does, on the “unsatisfactory state of law” and “different policies” arguments. This article reconceptualises the rule in Gray and systematically examines the role played by the theme of consistency between the civil law and criminal law in judicial decision-making. The second aim is to evaluate Gray in light of Patel v Mirza. The article critiques the Supreme Court's inconsistent treatment of deterrence in Henderson v Dorset University NHS Foundation Trust and Stoffel v Grondona, and argues that the way the court in Henderson conceptualised the relationship between Gray and Patel discloses an approach which is more closely aligned with that adopted by the minority in Patel.

I. Introduction

“[A]s applied in tort, the ex turpi causa principle … has had a chequered history.” So uttered Justice McLachlin when delivering her judgment in Hall v Hebert.¹ As her Honour explained, the operation of the illegality defence in tort must be kept within narrow confines, and is only justified where allowing the claimant to recover would “introduce inconsistency into the fabric of the law”, thereby undermining “the integrity of the legal system”.² This concern is engaged in two situations, where allowing recovery would either permit the claimant to profit from an illegal or wrongful act; or enable the claimant to evade a penalty prescribed by the criminal law.³ As Ernest Weinrib observes, permitting recovery in these instances would create “an intolerable fissure in the law's conceptually seamless web”.⁴ This article is concerned with the illegality defence as applied in the latter situation which will be referred to as “sanction-shifting” scenario: that is, where the claimant is trying to shift the adverse consequences arising from a sanction imposed upon the claimant by the criminal law to the defendant through the instrumentality of tort law.

More than a decade has passed since Gray v Thames Train Ltd.,⁵ in which the House of Lords authoritatively laid down the applicability of the illegality defence to sanction-shifting scenarios. The result might on one analysis be seen as unremarkable and orthodox, since it was supported by “high authority in the Commonwealth”,⁶ including those at the highest appellate level.⁷ However, in the intervening period, a considerable body of academic commentary on Gray has accumulated,⁸ and while there is generally no disagreement as to the fundamental correctness of applying the illegality defence in the sanction-shifting context,⁹ most commentators are critical of applying the doctrine where the claimant had no “significant personal responsibility” for the criminal offence, or where the criminal sanction lacks a penal element.

During the same period, the law in relation to illegality has undergone “a period of considerable uncertainty”¹⁰ with different configurations of the Supreme Court¹¹ expressing divergent views as to whether the working of the illegality defence should be rule-based (reflecting the reliance rule championed in Tinsley v Milligan),¹² or policy-based, the latter of which involves balancing competing policy considerations which might weigh in favour of or against denial of the claim. An enlarged panel of nine Justices was assembled in Patel v Mirza to resolve that conflict.¹³ The majority of the court decided that the latter is the correct approach. Patel is, however, an unjust enrichment case. Four years later, in Stoffel v Grondona ¹⁴ and Henderson v Dorset Healthcare University NHS Foundation Trust,¹⁵ the Supreme Court was presented with the occasion to consider how the restated law of illegality in Patel would play out in the tort context. In Henderson the court revisited, and affirmed, the correctness of Gray.

This article has two purposes. The first aim is to mount a comprehensive defence of Gray. The academic response to Gray was overwhelmingly critical. Nevertheless, it was still endorsed – unanimously and in robust terms – by a seven-member panel of the Supreme Court in Henderson. This article argues that the Supreme Court is right to reject the alternative suggestion that “sanction-shifting” is only improper in cases where a claimant bore significant personal responsibility for the illegal act. However, in contrast to the brevity of the reasoning in Henderson, the article places its defence of Gray in the context of a wider exploration of judicial freedom to alter the law and the role played by the objective of maintaining consistency between the criminal law and civil law in the adjudication of private law disputes.

The second aim is to assess critically the relationship between Gray and Patel and how the illegality defence should be developed. A number of academics argue for a departure from Gray on the basis that its reasoning, and result, have been superseded by Patel in the sense that the application of the Patel framework to the sanction-shifting factual pattern (especially where the claimant is said to be lacking significant personal responsibility) would not support the illegality defence as formulated in Gray. The second aim of the article concerns a broader question, namely, whether the effect of Patel is limited to jettisoning the reliance approach or whether it fundamentally overhauls the entire corpus of case law on illegality in private law. The precedential scope of Patel, and how the previous case law should be re-evaluated in light of that decision, stand in urgent need of authoritative determination.

To develop the above arguments, this article will be structured as follows. Sections II to IV are concerned with the first aim of the article. Section II summarises the key reasoning in Gray and examines the relevant academic criticisms. Section III substantiates and supports the Henderson court's categorical repudiation of the notion of significant personal responsibility. Section IV first teases out the shortcomings inherent in the existing accounts of the narrower rule articulated in Gray and proffers a fresh perspective as to how the narrower rule should be conceptualised. It then constructs a conceptual framework within which to analyse the case law that has hitherto appealed, in its reasoning, to the value of consistency between the criminal and civil law. Against this conceptual framework, it confronts the argument that tort law should depart from the test of responsibility in criminal law and adopt a distinct criterion of “significant personal responsibility”, due to the differing policy aims of criminal law and tort law. Section V addresses the second aim of this article. It observes that there are points of inconsistency between how the issue of deterrence was analysed in Henderson and Stoffel, and argues that the way in which the court rationalised the relationship between Gray and Patel reveals an approach that is more closely aligned with the minority in Patel, which has important implications for the precedential reach of Patel in the development of the illegality jurisprudence. Section VI concludes.

II. What Did Gray Decide?

A. The House of Lords’ Reasoning

The facts of Gray may be shortly summarised. The claimant, Mr. Gray, was involved in a major railway accident caused by the negligence of the defendants, the train operator and the company responsible for the rail infrastructure. He developed PTSD and experienced personality change, as a result of which he stabbed a man to death. He was convicted of manslaughter on the basis of diminished responsibility and was detained in a mental hospital. The claimant sued the defendants to recover damages resulting from his criminal conduct and from his detention in the hospital. The House of Lords held that all heads of damages were irrecoverable by virtue of the illegality defence, in the form of a narrower rule and wider rule of public policy postulated by Lord Hoffmann.¹⁶ The narrower rule posits that one cannot recover for damage which flows from a criminal sanction consequent upon the commission of an offence, because to do so would result in inconsistency “between the criminal law, which authorizes the damage suffered by the plaintiff in the form of loss of liberty because of his own personal responsibility for the crimes he committed, and the claim that the civil law should require someone else to compensate him for that loss of liberty”.¹⁷ This has been termed “the consistency principle”.¹⁸ On the other hand, the wider rule dictates that one cannot be compensated for consequences which follow from one's criminal conduct because such an outcome “is offensive to public notions of the fair distribution of resources” and “risks bringing the law into disrepute and diminishing respect for it”. Moreover, this is an outcome “of which public opinion would likely to disapprove and would thereby undermine public confidence in the law” (“the public confidence principle”).¹⁹ Since this article focuses on the illegality defence as applied in the sanction-shifting context, nothing more will be said as regards the wider rule.²⁰

In relation to the narrower rule, it was suggested that the hospital order was imposed not for punishment of Mr. Gray, but for treatment of his PTSD. This argument was rejected by Lord Hoffmann in the following passage (which has subsequently caused a great deal of controversy):

But the sentence imposed by the court for a criminal offence is usually for a variety of purposes: punishment, treatment, reform, deterrence, protection of the public against the possibility of further offences. It would be impossible to make distinctions on the basis of what appeared to be its predominant purpose. In my view it must be assumed that the sentence (in this case, the restriction order) was what the criminal court regarded as appropriate to reflect the personal responsibility of the accused for the crime he had committed.²¹

Lord Rodger of Earlsferry likewise said that since the claimant did not challenge the criminal court's orders, the civil court must proceed on the basis that the claimant knew, and was responsible for, what he did.²² His Lordship, however, reserved his opinion as to the applicability of the illegality defence to a situation where the index offence of which the claimant was convicted was trivial but revealed that he was suffering from mental disorder, attributable to the defendant's fault, which necessitated his detention pursuant to a hospital order.²³

Although agreeing with the disposal of the appeal, Lord Phillips of Worth Matravers made two reservations, the second of which concerns the situation “where it is the criminal act of the defendant that demonstrates the need to detain the defendant both for his own treatment and for the protection of the public, but the judge makes it clear that he does not consider that the defendant should bear significant personal responsibility for his crime”.²⁴ Lord Phillips therefore disagreed with Lord Hoffmann as to the “general applicability” of the proposition that it is impossible to discern the predominant purpose underlying a particular criminal sanction so as to determine whether the claimant bore significant personal responsibility for the crime.

B. Academic Criticisms

The critical academic commentaries on Gray do not question the fundamental correctness of applying the illegality defence in the context of sanction-shifting, but they, consistent with Lord Phillips's second reservation, cast doubt on whether the defence should be applied where the claimant is found not to bear “significant personal responsibility” for the crime, or where the criminal sentence imposed is solely rehabilitative, rather than punitive, in nature. The thrust of the argument is that if the criminal court, in deciding what sentence the claimant should be subjected to, makes clear that the claimant is not “personally responsible” for the crime and that the sentence is not imposed for “punishment” or “deterrence”, but rather for “reform” or “treatment”, there is no inconsistency between the criminal and tort law if the latter enables the claimant to be compensated.

James Goudkamp argues that the narrower rule, if it is capable of being justified at all, can be explained by two (in)consistency rationales: “goal inconsistency” and “pronouncement inconsistency”.²⁵ The goal inconsistency rationale posits that if the claimant is compensated in tort law in respect of damage flowing from a criminal sanction, it would take some of the sting out of it and undermine the goals which the criminal law seeks to advance through the imposition of the sentence on the claimant. On the other hand, pronouncement inconsistency posits that allowing the claimant to “shift” the loss arising from the criminal sanction to the defendant would cast doubt on the criminal law's determination regarding the claimant's responsibility for the offence. Against this backdrop, Goudkamp argues that where the criminal sentences are purely rehabilitative, awarding compensation in respect of such sentences in tort law is conducive to the rehabilitation of the offender.²⁶

Along similar lines, Sarah Green contends that certain objectives of imprisonment, such as rehabilitation of the offender, can be furthered if the claimant is allowed to make recovery because vindicating the right of the claimant and respecting his personhood through an award of compensation is congruent with the object of rehabilitation of the criminal sentence.²⁷

The above arguments are further developed by Matthew Dyson²⁸ who argues that where the tortious claimant has little or no personal responsibility, the goals of prohibition and punishment usually underpinning a criminal sanction are not engaged. Hence, an award of compensation in respect of damage flowing from the criminal sanction would not contradict any goal of the criminal law. The fact that the actus reus and mens rea of an offence are satisfied and that the claimant is convicted only paints a partial picture of the prohibition. The civil court is entitled to look beyond the criminal conviction, and take into account other evidence, such as the sentencing remarks of the criminal court, to see if the criminal disposition carries a penal element.

III. Significant Personal Responsibility

A. A Test Case: Henderson

A decade after Gray was decided came the case of Henderson.²⁹ The claimant appellant, Ms. Henderson, a paranoid schizophrenic, stabbed her mother to death when she experienced a serious psychotic episode. She was initially charged with murder but later with manslaughter on grounds of diminished responsibility, and consequently was sentenced to a hospital order and unlimited restriction order. She brought an action in negligence against the relevant NHS trust, claiming damages for various losses caused by her mother's murder and damage consequent upon her compulsory detention. The defendant admitted liability for negligence in failing to return the claimant to hospital in light of her manifest psychotic state, but it successfully invoked the illegality defence by praying in aid Gray.

A notable feature of Henderson is the sentencing judge's explicit remarks that there was no suggestion in the claimant's case that “[she] should be seen as bearing a significant degree of responsibility for what [she] did”.³⁰ This brought the factual matrix of Henderson squarely within the territory of Lord Phillips's second reservation, giving the perfect occasion for the Supreme Court to undertake a fundamental re-examination of Gray, in light of all the critical academic commentaries it spawned, especially in relation to the notion of significant personal responsibility.

One of the major points of contention revolved around how Lord Hoffmann's remarks (quoted above) as to the impossibility of discerning the predominant purpose of a criminal sentence should be interpreted. Two interpretations appeared to be open to the Supreme Court:

Interpretation I: Lord Hoffmann was merely making an assumption as to the personal responsibility of Mr. Gray. This interpretation is supported by the fact that Mr. Gray had never argued that he bore no personal responsibility for the manslaughter and that the hospital order was not imposed for his manslaughter but for treatment of his mental illness;³¹ it was “on that premise”³² that Lord Phillips agreed with the majority's disposal of the appeal. On that analysis, there is in fact no disagreement between Lord Hoffmann and Lord Phillips because Lord Hoffmann did not exclude the possibility that the criminal sentencing judge may expressly find that the claimant did not bear significant personal responsibility or that the sentence imposed is for treatment rather than punishment. This is an interpretation that has the support of tort scholars such as Goudkamp and Dyson.

Interpretation II: Lord Hoffmann was saying that once the claimant was convicted for the criminal offence, the civil court must “assume” that he was personally responsible for the crime. As such, it is impermissible for the civil court to inquire into the degree or extent of the subject's personal responsibility, or second-guess what is the predominant purpose of the sentence. This was the interpretation adopted by the first instance judge in Henderson,³³ and left undisturbed by the Court of Appeal.³⁴

The Supreme Court opted for interpretation II. Implicit in interpretation I is the assumption that despite the claimant's unchallenged conviction and the criminal court's finding of criminal responsibility, the civil court could, on the basis of the sentencing judge's remarks, conclude that she did not bear significant personal responsibility. However, in the Supreme Court's view, it is impermissible for the civil court to go behind the criminal court's conviction and “move away from the M'Naghten approach to insanity”.³⁵ Three reasons were given by Lord Hamblen who delivered the sole judgment. The first is that if the civil and criminal courts adopt different tests of responsibility and take opposing stances on the issue of the claimant's responsibility, “inconsistencies” will be heightened.³⁶ A related concern is the “uncertainty”³⁷ accompanying the notion of significant personal responsibility: we are not told why it should be the appropriate threshold, what it means and how it is to be determined.³⁸ The third concern is that the claimant's approach may amount to “judicial legislation”, a point which was underdeveloped by the court and is given further analysis below.³⁹

B. Defending the Supreme Court's Rejection of Significant Personal Responsibility

In Henderson, the claimant proffered two reasons why it would be permissible for tort law to adopt a different standard of responsibility (i.e. significant personal responsibility).

The first reason concerned “the unsatisfactory state of the law governing criminal responsibility”⁴⁰ (the “unsatisfactory state of law” argument). The relevant test for negating criminal responsibility is the M'Naghten rule of insanity,⁴¹ which requires the accused to prove on the balance of probabilities that “at the time of the committing of the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing, or, if he did know it, that he did not know he was doing what was wrong”.⁴² The claimant alluded to the findings of the Law Commission that the existing law in the area of criminal responsibility has long been maligned as “unfair, out of date and failing to reflect advances made in medicine, psychology and psychiatry”,⁴³ and proceeded to endorse the Law Commission's principal conclusion that the lack of criminal responsibility should be extended to cover those who are incapable of controlling their actions, citing the Law Commission's suggested formulation that there is an absence of criminal responsibility if people “lack the capacity to conform their behaviour to meet the demands imposed by the criminal law regulating their conduct”.⁴⁴ This formulation, the claimant argued, should be adopted as the test governing “significant personal responsibility”, the satisfaction of which is to be decided by the trial judge in a civil claim as a question of fact.⁴⁵

The second reason concerned “the divergent functions of tort law and the criminal law”: the former is principally concerned with “the connection between the wrongdoing and the claimant's injury” while the latter focuses on the person's wrongdoing⁴⁶ (the “different policies” argument).

It will be argued that neither of the two reasons proposed for departing from the criminal test of responsibility in the realm of tort is persuasive. The Supreme Court in Henderson was therefore right to adopt interpretation II of Lord Hoffmann's reasoning and reject the proposed alternative that a claimant would only be barred by the illegality defence where they had “significant personal responsibility” for the illegal act. However, the court has not fully addressed the two reasons put forward by the claimant to justify the adoption of significant personal responsibility, and scholarly commentaries on Henderson to date have not given due attention to this issue.⁴⁷ The analysis below seeks to fill that lacuna.

1. The “unsatisfactory state of law” argument

With respect to the first reason, the fundamental complaint of the claimant is not about the illegality defence (in the form of the narrower rule) as such, but rather about the insanity defence, and more specifically, the high threshold for negating criminal responsibility. If the insanity defence were to be amended in the way proposed by the Law Commission, the claimant would likely have been acquitted of manslaughter (since she lacked the capacity to control her actions) and would not have borne any criminal responsibility. As a result, the issue of illegality would not have arisen in the first place.⁴⁸ The claimant sought to circumvent the failure to reform the law of insanity by modifying the illegality bar, deploying the novel concept of significant personal responsibility. In doing so the claimant put forward the test of whether the claimant lacked capacity to conform their behaviour to the demands imposed by the criminal law. Rejecting this proposal, Lord Hamblen said that “[w]hat the justification is for that proposed test was not really explained, nor was its meaning. Not only is it a recipe for uncertainty, but it risks being tantamount to judicial legislation”.⁴⁹

A possible retort against the “judicial legislation” objection is that since tort law is primarily governed by the common law rather than statutes (at least insofar as common law jurisdictions are concerned),⁵⁰ judges should not be unduly restrained in the development of the common law. This argument proceeds from the premise that the judiciary tends to give deference to the will of the legislature and refrain from developing the common law in a way that radically alters the operation of legislation. For example, in Attorney General for Jersey v Holley,⁵¹ Lord Nicholls of Birkenhead said that when Parliament altered the common law relating to provocation through the enactment of section 3 of the Homicide Act 1957, “it is not open to judges now to change (‘develop’) the common law and thereby depart from the law as declared by Parliament”,⁵² even if the present state of the law in relation to provocation, as codified by Parliament, is not satisfactory. Returning to the illegality defence, it might be said that arguments against judicial “legislation” are misdirected because the insanity defence (unlike, for example, the defence of provocation) is a judge-made rule, and that tort law is not primarily governed by statute. As a result, there is nothing to prevent the civil court from adopting a more nuanced and open-textured test of responsibility in the law of tort. Henderson involved an attempt to persuade the court that a novel development in one area of the common law (the significant personal responsibility test in tort law) is justified by reason of issues or problems arising from another area of common law (the insanity defence in criminal law).

There are, however, other concerns with significant judicial shifts even if they are within the common law. Some of these relate to the complex interconnections between different parts of the common law. As expressed by the minority of the Supreme Court's decision in Willers v Joyce,⁵³ when the majority sought to extend the tort of malicious prosecution to civil proceedings, there could be “unanticipated knock-on effects in other areas of law” (such as, there, the law of privilege).⁵⁴ A similar concern was also voiced in Holley by Lord Nicholls, who, although lamenting the unsatisfactory state of the law relating to provocation, opined that “the law on provocation cannot be reformulated in isolation from a review of the law of homicide as a whole”.⁵⁵ The counsels of Lord Nicholls in Holley and the minority in Willers are particularly fitting in the context of the M'Naghten rule of insanity, the operation of which is intimately linked with other areas of the law of homicide such as fitness to plead and the (partial) defence of diminished responsibility.

Of course, there have been many cases in which the highest courts have sought to reformulate common law rules that have become a source of injustice. For example, the Supreme Court has made considerable revisions in contentious areas of the law such as secondary liability in criminal law,⁵⁶ the approach to assessment of damages in tort,⁵⁷ and indeed the illegality defence itself.⁵⁸ In respect of reform of the law of insanity/criminal responsibility, as the Law Commission confessed in its own report, the lack of legislative action is not because the law in this area is free from error, but rather because there is “little evidence of a practical problem in relation to the operation of the insanity defence”, so that other aspects of the criminal law, such as unfitness to plead, should be given priority.⁵⁹ Given that Parliament might not have intervened through legislation as a matter of practicality and legislative priority, it might be argued that the court should not be bound by that political decision, and “accept this reality and shoulder their responsibility for the state of the common law”.⁶⁰

That said, the situation in Henderson was importantly different from those in Jogee, Knauer and Patel in which the Supreme Court took the initiative to reform the law. There, the issues or areas of law which were subjected to judicial development were directly relevant to the resolution of the appeals. In contrast, in Henderson, the criminal law (including the insanity defence) only forms part of the backdrop against which the issue of illegality defence fell to be decided. To understand why this is so, two distinct inquiries need to be clearly separated: (1) whether Ms. Henderson should have been convicted (the conviction issue); (2) whether the illegality defence should apply in the circumstances of this case (the illegality issue); the conviction issue is within the exclusive province of the criminal court, while the illegality issue belongs to the civil court. The insanity defence (and hence its merits or demerits) is only directly relevant to the conviction issue, but that is not the question which the Supreme Court was asked to answer in that case. Setting aside the issue as to whether the argument that tort law should adopt a test of significant personal responsibility because of the unsatisfactory state of the insanity defence would amount to re-opening the basis of the claimant's conviction in the criminal court⁶¹ (something which is beyond the competence of the civil court), Ms. Henderson's proposal on developing a separate threshold of significant personal responsibility in tort law should fail on a more fundamental basis that whether her claim in tort should be barred is ultimately dependent on whether she bore criminal responsibility for her act (as the illegality defence is premised on the consistency of criminal law and civil law), not on tort law's perception of her responsibility regarding her criminal act.

2. The “different policies” argument

Insofar as the claimant argued that tort should depart from the criminal law's test of responsibility because the policies underpinning criminal law and tort law are different, this argument appeared to be based on a misunderstanding of how argument premised on different policies in different areas of law should be properly applied. Understanding why this is so requires a more fundamental examination of the meaning of consistency in the context of the illegality defence and the role played by the theme of consistency in legal reasoning more generally, to which we now turn.

IV. Consistency Revisited

A. Renewed Rationalisation of the Narrower Rule

Although the narrower rule is said to be underpinned by the consistency principle, what exactly is “inconsistent” with allowing the claimant's recovery in the context of sanction-shifting has yet to be clearly articulated. This section examines three accounts of how the narrower rule avoids inconsistency, but finds each of them unconvincing. Instead, I suggest that the narrower rule is not just directed at achieving doctrinal tidiness between different branches of law, but also the institutional cohesion of the legal system.

The first account of the inconsistency that the narrower rule attempts to avoid, is inconsistency in findings of responsibility: to countenance the tort action would be to find the defendant responsible for a matter for which the criminal law has already determined that the claimant was responsible. In the words of Goudkamp, “sanction-shifting actions disturb the internal consistency of the law by reason of their incompatibility with the criminal law's reliance on the theory of free will”.⁶² The incompatibility arises from the civil court's conclusion that the claimant's conduct is not an expression of his choice but the result of the defendant's negligence, and on the other hand, the criminal court's finding that the claimant's conduct is a product of the exercise of his free will, the mens rea of the relevant offence having been established beyond reasonable doubt. A possible retort against the first rationalisation is that the law does not require proof that the defendant's act is the sole cause of the harm or injury: it is only a necessary condition. Therefore, a finding of criminal responsibility on the part of the claimant does not preclude a finding in a tortious claim that the defendant is responsible for causing the claimant to act in a particular way. However, a further counter-argument could be made against this retort: if, subsequent to the defendant's negligence, the claimant committed a voluntary, intentional act which also constitutes one of the causes of the harm, the claimant's act would operate as a novus actus interveniens, breaking the chain of causation between the defendant's negligent act and the claimant's injury and negating any responsibility on the part of the defendant. In this way, the result in Gray and the narrower rule can be defended. However, if the claimant's criminal act constitutes a novus actus interveniens, the tort claim would have failed at the causation stage and the issue of illegality does not arise.⁶³ Indeed, if we take this line of argument to its logical conclusion, we would find ourselves in a position where the illegality defence would be rendered wholly otiose, because “the only relevance of the illegal act is a question intrinsic to the establishment of the tort”, which is “whether the claimant's own act constituted a novus actus interveniens between the defendant's breach of duty and the ultimate loss for which he was claiming”.⁶⁴

The second account is that countenancing sanction-shifting claims would undermine, and therefore produce inconsistency with, the goals or objects which the criminal law is trying to promote. This is best exemplified in Goudkamp's “goal inconsistency” analysis. However, the difficulty with this account is that it goes against the clear guidance of Lord Hoffmann in Gray, as interpreted in Henderson, that the civil court is not allowed to speculate on the aim or policy which the criminal court is trying to advance through the criminal sentence. Moreover, this rationalisation is susceptible to the argument that tort law and criminal law have different policy aims. Further, as Goudkamp points out, there are exceptional situations in which the objects of the criminal law would be furthered by sanction-shifting claims, such as where the criminal offence is designed to discharge a moral function (and the defendant rather than the convicted claimant is “truly to blame”) or a market function.⁶⁵

The third account, unlike the previous ones, does not focus on the consistency between criminal law and tort law, but rather on the nature of loss for which damages are sought. As Weinrib points out, “[the claimant's] conviction forms the damage element of his tort”.⁶⁶ In the words of Allan Beever, “the claimant must assert a loss of something to which she is not entitled in law” and that “in law the violation of the right had no value”.⁶⁷ Beever's argument has subsequently been approved by Chief Justice McLachlin in her extrajudicial writing,⁶⁸ in which her Honour resiled from her previous view expressed in Hall v Hebert that the illegality defence involves the policy of coherence trumping the claimant's normal corrective justice entitlement.⁶⁹ According to Sharon Erbacher,⁷⁰ this line of analysis involves a reference to what Jane Stapleton terms “gist damage”, which is the “minimum actionable damage” necessary to trigger a claim in negligence.⁷¹ However, the third rationalisation suffers from essentially the same conceptual flaw as the first one insofar as they both involve reducing the illegality defence into one of the constituent elements of the tort of negligence. If the loss arising from the criminal sentence is not actionable in negligence, no tort would, ex hypothesi, have been committed.

How then should we satisfactorily rationalise the narrower rule? It is argued that in rationalising the narrower rule, there is an important gloss we must put on it: an award of compensation by the civil court in favour of the tortious claimant carries a normative message that the claimant ought not to suffer the injury. In the context of sanction-shifting, the injury is the direct consequence of a sanction or penalty prescribed by the criminal court due to the claimant's criminal responsibility and his consequent conviction. To borrow the words of one commentator, the injury sustained by the claimant – being inflicted by the courts themselves – is a lawful injury.⁷² Hence, if the civil court is to compensate the claimant for such injury, this would be tantamount to saying that the claimant ought not to have been penalised by the criminal court. Allowing sanction-shifting claim would therefore mean that the civil court is, through its award of compensation, casting doubt on the legitimacy and propriety of the sanction imposed by the criminal court. Thus conceptualised, the consistency principle is not just about doctrinal unity between different branches of law, but about the consistency between different organs or institutions within the same legal system. To put the point more bluntly, the consistency principle is not just about the doctrinal tidiness between tort law and criminal law, but also about the relationship between the civil court and the criminal court – the focus is not just doctrinal but also institutional.

This rationalisation of the narrower rule with an institutional dimension has a number of advantages. First, it explains why the Supreme Court would prefer interpretation II of the passage of Lord Hoffmann cited above: that the civil court must “assume” or “proceed on the basis” that the criminal court's disposition is an “appropriate” reflection of the claimant's personal responsibility for the crime committed underscores the respect and due deference which the civil court accords to the criminal court. An institutional rationalisation also helps us to make sense of the Supreme Court's remarks in Henderson that the narrower rule also engages the public confidence principle because “one of the reasons that the public would be likely to disapprove of the outcome is the inconsistency which it involves between the criminal law and the civil law”.⁷³ Apart from the outcome of the litigation and the consistency between criminal and civil law, it is submitted that the narrower rule more fundamentally promotes public confidence in the legal system at an institutional level: it is important that the civil court does not undermine the authority of, and the public confidence in, the criminal justice system through countenancing sanction-shifting actions.

B. Overview of the Role of Consistency Between Civil and Criminal Law in the Adjudication of Legal Disputes

The consistency between civil and criminal law is at the heart of the illegality jurisprudence, especially in the context of sanction-shifting. The illegality defence, however, is not the only area of law in which the question of consistency arises. In particular, the question has frequently arisen whether the legal tests or thresholds in the civil law should correspond to their criminal law counterparts. This section draws upon cases from different branches of private law, including tort, property and family law, to argue that there is no single (consistent) approach to the question of consistency between the criminal and civil law; rather the question of whether concepts deployed in each field must map on to those used in the other varies with the circumstances.

In some circumstances, the courts have rejected claims based on the desirability of consistency. In Ashley v Chief Constable of Sussex Police (Sherwood intervening),⁷⁴ for example, the House of Lords decided that the criteria for self-defence in tort law against a claim in battery are not the same as those in criminal law. Rejecting “the plea for consistency between the criminal law and civil law” as lacking “cogency”, Lord Scott of Foscote pointed out that “the ends to be served by the two systems are very different”.⁷⁵ A key function of the criminal law is “to identify, and provide punitive sanctions for, behaviour that is categorised as criminal because it is damaging to the good order of society”.⁷⁶ The function of tort law, on the other hand, is “to identify and protect the rights that every person is entitled to assert against, and require to be respected by, others”, and to “strike a balance between the conflicting rights”.⁷⁷ Hence, the test of self-defence in tort not only requires the defendant to establish that his mistaken belief that he is in danger of an imminent deadly attack from the assailant is honest, but also reasonable, because this can strike a balance between the right of a person “not to be subjected to physical harm by the intentional actions of another person” and the right of a person “to protect himself by using reasonable force to repel an attack or to prevent an imminent attack”.⁷⁸

Similarly, in R. v Hinks,⁷⁹ the House of Lords rejected an argument that the criminal law should be aligned with the civil law of property. The accused, the carer of the victim, was charged with and convicted of theft, in circumstances where the civil law would have recognised a valid gift from the victim, who was a man of limited intelligence, to the accused. The conviction was upheld on appeal. In the face of this apparent conflict between criminal law and civil law, Lord Steyn sought to defend the result by arguing that “the purposes of the civil law and the criminal law are somewhat different”, and that even though “in theory the two systems should be in perfect harmony”, “in a practical world there will sometimes be some disharmony”.⁸⁰ One reason for such disharmony in the context of theft is to eliminate the need for the judge to explain to the jury the civil law concepts (such as “ownership”) in respect of the element of appropriation in the offence of theft.⁸¹ Hinks has been subject to trenchant academic criticisms. It has been pithily pointed out that “consistency of criminal law and civil law concepts is essential in theft, which after all is concerned with the protection of property rights. It would be more than odd if D could be convicted of theft when she had not in fact interfered with V's property rights”.⁸² Beatson and Simester argue that theft, as a species of property offence, is essentially concerned with the protection of property rights;⁸³ hence, unlike crimes such as assault, the rights being protected by the offence of theft are necessarily rooted in the civil law.⁸⁴ In this sense, the policy aims of theft and property law (rather than that of criminal and civil law in the abstract) can be seen to be coterminous.

In contrast, in Local Authority X v MM (by her litigation friend, the Official Solicitor),⁸⁵ Munby J. held that the test of capacity to consent to sexual relations must be the same in the civil law and the criminal law.⁸⁶ There are sound reasons of policy, it was said, why the civil and criminal law should “speak with one voice” in this area.⁸⁷ Apart from the fact that such consistency “adds clarity”,⁸⁸ the more important reason is that “in this context both the criminal law and the civil law serve the same important function: to protect the vulnerable from abuse and exploitation”.⁸⁹ These judicial remarks were recently endorsed by the Supreme Court in A Local Authority v JB (Respond intervening),⁹⁰ although Lord Stephens delivering the judgment acknowledged the possibility for the civil law to impose a more demanding test of capacity than the criminal law.⁹¹

These examples suggest that while the consistency between the criminal and civil law (or consistency in the law more generally) has at times been described as a “principle”,⁹² it should be better seen as an aspiration or an overarching rationale which drives the fashioning of a specific legal rule. Therefore, the content of the concept of consistency between the criminal and civil law is incapable of exhaustive or precise definition, and can only be ascertained by reference to the specific context in which such concept is invoked. Therefore, considerable caution needs to be exercised when one transplants the consistency argument from one context to the other.

The survey of the case law also demonstrates the need to give vague concepts such as “consistency” or “coherence” substantive content. In the context of sanction-shifting, the notion of consistency is used in a very specific sense of ensuring that the civil court does not, by permitting sanction-shifting to take place, contradict the criminal court so as to maintain the institutional integrity of the justice system. In the second category, whether the consistency thesis would require the legal test in one branch of civil law to be aligned with its criminal law parallel would depend on whether the “functions” served by them are the same. This explains the courts’ diametrically opposite reception of the consistency argument in Ashley and Local Authority X: in Ashley, the functions of tort law and criminal law, in the context of self-defence, are not the same; in contrast, in Local Authority X, the civil law and criminal law are unified in their commitment to protecting vulnerable individuals in the context of determining whether a person has the requisite capacity to consent to sexual relations. The academic critiques on Hinks serve to accentuate the point that simply putting up the flag “consistency between criminal law and civil law” does not advance our understanding of what consistency actually means – one must be more specific (e.g. the consistency between theft (a specific criminal offence) and property law (a specific branch of civil law)).

C. Confronting the “Different Policies” Argument

When it comes to the “different policies” argument advanced in Henderson, we can find a similar argument being deployed in Ashley. However, it is inappropriate to transplant the reasoning in Ashley to the context of sanction-shifting. Recall that in Ashley, the subject matter against which the issue of self-defence is directed is battery, which constitutes both a tort and a criminal offence. Against this backdrop, the House of Lords held that if the claim is brought in tort and the defendant wishes to raise self-defence, the threshold for satisfying that defence is different from a case where the defendant is prosecuted criminally because the threshold should be formulated in a way which best gives effect to the policy aim which the claim in question is designed to vindicate, depending on whether the claim is criminal or tortious in nature. This is different from the claimant's argument in Henderson that tort law should have a different test from the criminal law to gauge the claimant's responsibility in respect of her criminal act.

The point is even more clear from Corr v IBC Vehicles Ltd.⁹³ There, as a result of the employer's negligence, the late employee suffered PTSD and developed depression, which caused him to commit suicide when suffering from an episode of severe depression. The issue was whether the suicide act of the deceased would amount to novus actus interveniens, breaking the chain of causation between the defendant's negligence and the death of the deceased. The answer depends on whether the suicidal act was “a voluntary, informed decision taken by [the deceased] as an adult of sound mind making and giving effect to personal decision about his future”.⁹⁴ Rejecting the M'Naghten rules of insanity as the appropriate test for determining whether that threshold was met, Lord Bingham of Cornhill said that:

[W]hatever the merits or demerits of the M'Naghten rules in the field of crime, and they are much debated, there is perceived in that field to be a need for a clear dividing line between conduct for which a defendant may be held criminally responsible and conduct for which he may not. In the civil field of tort there is no need for so blunt an instrument … it would be retrograde to bar recovery by the claimant because the deceased was not, in M'Naghten terms, insane.⁹⁵

The point that Lord Bingham intended to convey is that the test for determining a person's responsibility for a criminal offence, that is the M'Naghten rules of insanity, should not be transplanted to tort law to determine whether the claimant should bear responsibility for suicide, which is not a crime and on which the criminal court has nothing to say. Therefore, the fact that criminal law and tort law may be underpinned by different policy concerns does not lead to the conclusion that tort law should embrace significant personal responsibility as the threshold for determining the claimant's responsibility for his criminal conduct.

V. Where Should Gray Stand In The Patel Era

Having disposed of the first purpose of the article, we now turn to its second aim: the assessment of Gray in light of Patel. There, a majority of the Supreme Court, led by Lord Toulson, eschewed a reliance-based approach in favour of a policy-based approach. Under this approach, the operation of the illegality defence should be guided by a principled and transparent assessment of the public policy considerations which may impact on the coherence and integrity of the legal system. This is achieved through a “trio of necessary considerations”,⁹⁶ (a) “the underlying purpose of the prohibition which has been transgressed”, (b) “any other relevant public policies which may be rendered ineffective or less effective by denial of the claim” and (c) “the possibility of overkill unless the law is applied with a due sense of proportionality”.

A. Application of the Patel Trio in the “Sanction-shifting” Context

1. Application of Patel trio to the facts of Henderson

There is a view (of which Dyson is a prominent advocate) that Gray should be departed from, on the basis that the application of the Patel trio of factors to a situation where the claimant does not bear significant personal responsibility would lead to a different outcome.⁹⁷ Under Dyson's analysis, each of the three stages in the Patel trio should be resolved in favour of allowing the claimant to recover.

Dyson's argument that the application of stage (a) weighs in favour of the claimant rests on the following strands: (i) the criminal law has various ways to demonstrate an accused's lack of capacity or personal responsibility, and the much criticised M'Naghten approach to insanity is only one of them; (ii) the criminal court can use capacity questions (automatism and diminished responsibility) and, in particular, sentencing remarks to show that the claimant has no significant personal responsibility and that the criminal sanction is not imposed for punishment; (iii) it follows that any application of criminal prohibition is merely “minor and technical” and has no real force.

Moving on to stage (b), Dyson argued that denial of claimant's claim would “deny the importance of the duties owed by public authorities to those with mental disability and to the wider society with respect to those with mental disability”. This is a factor which the Supreme Court took into account but which it ultimately rejected as being trumped by more weighty considerations of the consistency and integrity of the legal system.

As to stage (c), Dyson argued that the balance of fault is tilted in favour of the claimant since there was no significant personal responsibility and the criminal courts did not opt to punish the claimant.

In light of the above academic criticism regarding Gray's alleged incompatibility with the Patel framework, Henderson presents the perfect fact pattern for testing the strength of this argument. This subsection will summarise the Supreme Court's application of the Patel framework to the facts of Henderson, and explain how the Supreme Court parts way with the reasoning of the critics who suggested that Patel implied a different outcome.

Coming first to stage (a) of the Patel trio, the policy rationales identified by the Supreme Court in Henderson are as follows: (i) deterrence of unlawful killing; (ii) protection of the public, given the fundamental right of the right to life; (iii) public condemnation of unlawful killing; and (iv) punishment of criminals who commit unlawful killing.⁹⁸ The court posited that stage (a) should not be confined to the specific purpose of the prohibition transgressed, but also include other policy considerations which may weigh in favour of the operation of the illegality defence,⁹⁹ citing Lord Wilson's two-step approach in Hounga.¹⁰⁰ In the present context, that would include the consistency principle and public confidence principle identified in Gray.¹⁰¹ The claimant submitted,¹⁰² in line with academic commentaries,¹⁰³ that it is absurd to suggest that a person, when experiencing a serious psychotic episode, may be deterred from killing by the inability to recover compensation from the tortfeasor. While the Supreme Court acknowledged the “force” of the claimant's submission, it ultimately took the view that “the question should not be considered solely at the granular level of diminished responsibility manslaughter cases”,¹⁰⁴ thereby rejecting the argument that applying the illegality defence would not meaningfully deter, hence prohibit, a person who lacks significant personal responsibility from committing crime. The court asserted that when the issue is “looked at broadly”, “there may well be some deterrent effect in a clear rule that unlawful killing never pays and any such effect is important given the fundamental importance of the right to life”.¹⁰⁵ The court also dismissed the argument that the criminal sentence lacks a penal element, since “the fact of a criminal conviction for manslaughter is itself punitive”.¹⁰⁶

When it comes to stage (b), the claimant put forward four countervailing policies which were said to militate against the application of the illegality defence.¹⁰⁷ These include the policies of (i) encouraging public authorities such as NHS to comply with their duty to care competently for the vulnerable, especially those with mental disability;¹⁰⁸ (ii) compensating the tortious victims who bore no significant responsibility for their criminal conduct; (iii) ensuring that the victim is duly compensated by public bodies which caused him injury; and (iv) ensuring that the accused receive sentences proportionate to their offending. Although the Supreme Court saw force in some of these policy considerations, it considered that they were outweighed by the policy of ensuring consistency and integrity of the legal system.¹⁰⁹

Turning last to stage (c), having taken into account the facts that (1) manslaughter is a serious offence;¹¹⁰ (2) the offence is central to all heads of loss;¹¹¹ (3) the offence involves culpable killing with murderous intent;¹¹² and (4) the lack of marked disparity in the parties’ respective wrongdoing,¹¹³ the court concluded that denial of the claim is not disproportionate.¹¹⁴

2. Critique of the Supreme Court's reasoning

Although the Supreme Court's robust endorsement of Gray and rejection of Dyson's critique of the same is commendable, the judgment nonetheless leaves a number of outstanding questions, which are principally concerned with (1) what sorts of considerations fall within which stage of the trio elaborated in Patel, and (2) how the issue of deterrence should be approached under the Patel trio.

The first question stems from the Supreme Court's assertion that stage (a) should encompass all sorts of policy considerations (including the consistency principle and public confidence principle) that support the application of the illegality defence. The court treated the Patel trio as originating from the two-step approach in Hounga, which entails asking two questions, namely, “What is the aspect of public policy which founds the defence?” and “Is there another aspect of public policy to which the application of the defence would run counter?”¹¹⁵ While both Hounga and Patel support a policy-based approach, this does not mean that it is correct to read across from the first question in Hounga's two-stage approach to tweak the first element of the Patel trio. This can be seen from an examination of how the first question was approached in Hounga. That case concerned a claim of unlawful discrimination brought against the defendant who had dismissed the claimant on the basis of her (Nigerian) nationality. However, Ms. Hounga was an illegal immigrant to the UK, and the question arose therefore whether Mrs. Allen, the defendant, could escape liability by pleading illegality. In answering the first question, Lord Wilson considered whether permitting the claim would allow the claimant to profit from her unlawful conduct viz. illegal entry into the UK and whether it would permit evasion of a penalty prescribed by the criminal law.¹¹⁶ This appears to lend support to the proposition that stage (a) should not be limited to the policy considerations underpinning the prohibition transgressed. However, when answering the first question, his Lordship also considered whether the illegality defence would encourage the defendant to enter into illegal employment contracts and discriminate against the employees;¹¹⁷ considerations which, if analysed under Patel, would fall squarely within stage (b). Hence, the two-stage approach in Hounga, while sharing some similarities with the Patel trio, should not be used as a springboard to broaden the range of policy considerations which Lord Toulson intended to include in stage (a) of the inquiry.

The second question concerns the inconsistent treatment of deterrence at stage (a) in Henderson and Stoffel. If the reasoning in Henderson is set along those in Stoffel, then Lord Hamblen's suggestion in Henderson that the issue of whether the deterrence rationale is workable should be “looked at broadly” instead of “at the granular level” can be seen as an application of the remarks made by Lord Lloyd-Jones in Stoffel, that “a court will be concerned to identity the relevant policy considerations at a relatively high level of generality” when applying stages (a) and (b).¹¹⁸ In contrast, stage (c) is concerned with a “close scrutiny to the detail of the case in hand”.¹¹⁹ Lord Lloyd-Jones posited that “the effectiveness of the criminal law in particular situations or the likely social consequences of permitting the claim in specified circumstances” need not be considered at stages (a) and (b).¹²⁰ However, the way in which stage (a) was approached in Stoffel reveals that the court does take into account “the effectiveness of the criminal law in particular situations” at stage (a).

In Stoffel, the claimant purchased a property with the assistance of a mortgage, but the purpose of putting the property in the claimant's name was fraudulent, in that she wished to use her good credit history to obtain mortgage finance from the lender in order to assist the vendor in raising capital which he would not otherwise have been able to obtain. The defendant, a firm of solicitors retained by the claimant, negligently failed to complete the registration of the transfer to her of the property or the new legal charge in favour of her lender. When the claimant defaulted, the lender sued and obtained summary judgment against the claimant, who in turn brought a claim against the defendant for their negligent failure to register the transaction. The defendant solicitors argued that the claimant's claim was barred by the illegality defence. The Supreme Court disagreed. It rejected the solicitors’ submission that applying the illegality defence would promote the deterrence of mortgage fraud (which is one of the purposes of its criminalisation), pointing out that “the risk that [someone in the position of the claimant] may be left without remedy if their solicitor should prove negligent in registering the transaction is most unlikely to feature in their thinking”.¹²¹

The important point for present purposes is that the reason given by the Supreme Court for repudiating the deterrence rationale in Stoffel falls squarely within the consideration regarding “the effectiveness of the criminal law in particular situations”, namely, whether the deprivation of an otherwise valid legal entitlement to recover compensation against a negligent conveyancing solicitor would be effective in deterring the commission of mortgage fraud.¹²² Thus, the approach taken in Henderson in relation to the deterrence rationale – to ignore the fact that the risk of being left without a remedy by virtue of the illegality defence is most unlikely, if not impossible, to “feature in the thinking” of a person who was struggling with a serious psychotic episode when the manslaughter was committed – does not sit well with the approach adopted in Stoffel.

B. Issue of “Patel Compliant” and Implications

In Henderson, the Supreme Court has laid to rest the academic debate as to whether Patel marks “year zero” for the law of illegality. Burrows, for example, had argued that Patel “wipes the slate clean” of the existing authorities.¹²³ On the other hand, Goudkamp had taken a more modest approach, arguing that the trio simply operates as a “crosscheck” on the pre-existing authorities.¹²⁴ The latter approach appears to have been vindicated in Henderson, since the decisions which predated Patel were said to “remain of precedential value unless it can be shown that they are not compatible with the approach set out in Patel in the sense that they cannot stand with the reasoning in Patel or were wrongly decided in the light of that reasoning”.¹²⁵ This can be interpreted as a presumption laid down by the Supreme Court that pre-Patel case law is compatible with Patel and hence retains precedential force.¹²⁶ The question is: how can one decide if a specific test is compatible (or otherwise) with Patel? Henderson supplies a good illustration of that exercise. The court ultimately concluded that Gray is “Patel compliant”, but what is the basis for such conclusion?

One possibility is that the application of the Patel trio to the facts of Henderson does not yield a different outcome from the one reached by applying Gray. However, in many cases, including Patel ¹²⁷ and Stoffel,¹²⁸ the same results could be reached by applying the reliance-based approach and policy-based approach, and it would be remarkable if a subsequent Supreme Court is to conclude that the two are consistent with each other.¹²⁹

Another difficulty with declaring Gray as “Patel compliant” is the different analytical structures they call for. While the Supreme Court is correct to conclude that Gray is “an example of a decision on illegality based on policy considerations rather than reliance”,¹³⁰ this should not obscure the fact that the relevant public policy considerations (i.e. the consistency principle and public confidence principle) are automatically and directly engaged “once it was ascertained that the loss claimed was a penalty imposed by the criminal court or the necessary consequence of the sentence”.¹³¹ However, in Patel, whether allowing recovery would be “contrary to public interest, because it would be harmful to the integrity of the legal system” can only be ascertained by undertaking a forensic examination of the trio of considerations. Moreover, in Gray, the consistency principle alone, as embodied in the narrower rule, would operate to defeat any sanction-shifting action. However, when analysed under the Patel framework, the consistency principle and public confidence principle are simply part of the policy mix at stage (a), which must be balanced against other countervailing policy considerations, if any, at stage (b). This is distinct from how the illegality defence operates in Gray, where the consistency principle is dispositive of the result and is not subject to any countervailing policy considerations.

Setting its face against these difficulties, the Supreme Court declared Gray as being “Patel compliant” because “it is how Patel ‘plays out in that particular type of case’”.¹³² In so declaring, Gray is rationalised as an exemplification and embodiment of the policy-based approach in Patel in the specific context of sanction-shifting. More remarkably, in adopting this rationalisation, the Supreme Court in fact comes close to embracing Lord Sumption's view (in Patel) that the range of factors are simply “policy factors which have gone into the development of the current rules”, rather than “factors influencing an essentially discretionary decision about whether those rules should be applied”.¹³³ As Lord Sumption pointed out, the most fundamental disagreement within the Patel bench lies not in whether the range of factors are relevant, but in how they are relevant: should they be regarded “(i) as part of the policy rationale of a legal rule and the various exceptions to that rule” (the minority), or “(ii) as matters to be taken into account by a judge deciding in each case whether to apply the legal rule at all” (the majority).¹³⁴ Proposition (i) (the minority's approach) can be illustrated by the exceptions in the fields of restitution of unjust enrichment and tort, in which certain considerations in the range of factors approach find their voices. In Patel, Lord Mance, one of the dissenting Justices, pointed out that the doctrine of locus poenitentiae, an exception to the reliance rule, “avoids windfall benefits and disproportionate losses, without involving the positive enforcement of or the recovery of profits based on illegal bargains”.¹³⁵ This exception therefore ensures that the claimant could make recovery (i.e. no “disproportionate” loss, a concept reminiscent of stage (c)) without undermining the policy underlying the prohibition against the enforcement of illegal contract (stage (a)). In Henderson, Lord Hamblen endorsed Lord Rodger's reservation in Gray,¹³⁶ holding that “trivial offences” and “strict liability offences where the claimant is not privy to the facts making his act unlawful” would not constitute turpitude and engage the illegality defence.¹³⁷ Treating these offences as exceptions to the narrower and wider rules is consistent with the view that denial of the claim would be a disproportionate response (stage (c)) where the illegal conduct is not “serious” (factor (i)) (trivial offences) or “intentional” (factor (iii)) (strictly liability offences).

Moreover, efforts have been made in Henderson to accommodate Lord Sumption's criticisms of the majority's approach (such as the criticism that “the ‘range of factors’ test discards any requirement for an analytical connection between the illegality and the claim, by making the nature of the connection simply one factor in a broader evaluation of individual cases”)¹³⁸ within the analytical framework of the Patel trio. Such accommodation is achieved, paradoxically, by elevating the importance of “reliance” in the working of stage (a) and (c) of the Patel trio. At stage (a), Lord Hamblen stressed that “the closeness of connection between the claim and the illegal act” and the “factor” of reliance is relevant to the issue of consistency and coherence in the law.¹³⁹ When it comes to stage (c), Lord Hamblen clarified that amongst the four non-exhaustive factors identified by Lord Toulson, centrality is a factor of particular importance, which entails asking “whether there is a causal link between the illegality and the claim, and the closeness of that causal connection”.¹⁴⁰ In Stoffel, Lord Lloyd-Jones also said that “the question of reliance may have a bearing on the issue of centrality”.¹⁴¹

It is also instructive to note that the Supreme Court in Patel, or at least some of the Justices,¹⁴² appear to have understood there to be a dichotomy between a reliance-based or policy-based approach before them for resolution. Construed in this way, Patel has arguably left other alternative formulations of the illegality defence in other areas of law, including the narrower and wider rules in Gray, untouched.

If one bears in mind (1) the presumptive compatibility of pre-Patel authorities (except for Tinsley) with Patel; (2) the vindication of Lord Sumption's conceptualisation of the role of a range of factors in the illegality defence (as evidenced from the treatment of the relationship of Gray and Patel in Henderson); (3) the elevated importance of reliance in the operation of Patel trio; and (4) the narrow task that the Patel court was pursuing (namely, whether the reliance approach should be eschewed in favour of the policy-based approach), then, despite Patel's purported far-reaching jurisprudential implication, its practical effect may be far more limited than has been envisaged by proponents of Patel.

VI. Conclusion

This article had two aims. First, it has sought to counter the considerable body of academic literature critical of Gray with a defence of Lord Hoffmann's approach. This article defends the Supreme Court's rejection, in Henderson, of the proposal that an illegality defence should be premised on the “significant personal responsibility” of the claimant. This article revisits the theme of consistency in the context of illegality, and systemically examines how consistency has been resorted to in judicial reasoning. Several points of general significance emerge, including (1) the danger of judicial legislation and of using litigation in one area of law as an engine for law reform in another area of law which is not directly in point; (2) a renewed rationalisation of the narrower rule; and (3) the need for specificity and sensitivity to contexts in invoking the consistency between criminal and civil law in judicial reasoning.

Second, this article has sought to assess whether Gray can survive the restatement of the illegality defence in Patel and the future trajectory of this area of law. It critically evaluates how Patel trio plays out in the factual pattern of Henderson, and highlights a number of unsatisfactory aspects of the court's reasoning, especially in relation to the treatment of deterrence. It is further contended that the court's perception of Gray as an application of Patel in the context of sanction-shifting echoes Lord Sumption's argument in Patel that the “range of factors” should merely be seen as policy considerations which underlay the various rules and exceptions, rather than as matters to be applied on a case-by-case basis. Supporters of Patel had hoped that it had initiated a process of fundamental change in the approach to the illegality defence in private law.¹⁴³ However, the new generation of the Supreme Court in Henderson and Stoffel seem anxious to reverse the radical course taken by some of their predecessors in Patel. The refusal to read Patel as representing a break with the past signals a preference to develop the common law based on established authorities and in an incremental manner. It should be warmly welcomed.