By Per Laleng
In Part I, I noted that the threshold question (“Does the conduct engage the public interest?”), and the Patel question, (“Would allowing the claim damage the public interest?”), are different inquiries. The first is conduct-based, and the second is systemic. The court’s assertion that “The risk of producing inconsistency and disharmony in the law, and thereby damaging public confidence in the integrity of the legal system … is of central importance in determining the threshold in this case” (Judgment para [127]) blurs these questions. By importing systemic concerns into the threshold, the court conflates engaging the public interest with damaging it. That conflation deepens during the Patel assessment. Applying Patel, the Supreme Court arrives at what it calls the “inevitable conclusion” that the illegality defence applies (Judgment para [170]). The conclusion was indeed inevitable, but not because the case was clear. It was inevitable because the Patel framework has been rebuilt to produce only one answer.
As originally conceived by Lord Toulson, the Patel test is a structured framework designed to answer an overriding question: would allowing recovery for something tainted by illegality produce inconsistency and disharmony in the law, thereby damaging the integrity of the legal system (Patel v Mirza [2016] UKSC 42 at [100]). (Note the absence of “public confidence”). To answer this question, courts consider whether denying the claim would: (a) enhance the policies underlying the rule broken by the claimant; and/or (b) weaken any other relevant public policies; and if the balance favours denying the claim, whether (c) that denial would be a proportionate response. The first two stages balance competing policy considerations from the perspective of the legal system. The third is a case-specific fairness check, oriented to principle rather than policy.
The purpose of the prohibition
When turning to stage (a) of Patel, the Supreme Court said the purpose of the rule against taking human life is the preservation of life and the promotion of respect for the sanctity of life. The court identifies four public interests served by the rule: public protection, deterring unlawful killing, public condemnation of unlawful killing and punishment, and an acknowledgement of the wrong done to the victims, their family and friends (Judgment para [141]).
The court recognized that the deterrent effect of the rule was questionable in relation to an insane claimant but echoed Lord Hamblen in Henderson: “there may be some deterrent effect in a clear rule that unlawful killing never pays” (Lewis-Ranwell Judgment para [143]). Or maybe not.
Condemnation without punishment
In Henderson, Lord Hamblen paired condemnation with punishment (Henderson para [129]). Henderson was a diminished responsibility case, so punishment was still in play. The Supreme Court in Lewis-Ranwell unbundles the pair. Instead, condemnation is paired with public protection: “Although he is spared criminal responsibility for his conduct, and the law focuses on the protection of the public as opposed to punishment, his conduct is neither justified nor excused. His conduct is unlawful and deserves to be condemned” (para [142]).
This passage is confusing. When it considered the threshold question, the Supreme Court said: “The claimant’s acts were not justified by his insanity. He was simply excused criminal liability” (para [130]). If condemnation rests on the premise that the conduct is “neither justified nor excused”, then that premise appears to have been undermined by the court’s own earlier statement. The equivocation on “excused” is compounded by the conflation of criminal responsibility and criminal liability throughout the judgment, a problem replicated by a similar conflation of civil responsibility and civil liability when the court turns to inconsistency.
If condemnation rests on the premise that the conduct was unlawful, it is difficult to see the functional distinction between the threshold question and Patel stage (a). The unlawful conduct is sufficient to engage both the state’s interest at the threshold and public condemnation at Patel (a). But why should this claimant’s conduct be condemned? He is not at fault, not blameworthy, not being censured. The only sense in which he is being “condemned” is via his hospital order – imposed for public protection, not punishment. That hospital order is doing double, if not treble duty: at the threshold, for condemnation, and later for inconsistency.
To these weak justifications grounded in deterrence and condemnation, the court adds the public interest in acknowledging that a grievous wrong has been suffered (para [141]). This consideration is new in Lewis-Ranwell, unsupported by cited authority, and never developed. It may be doing the heaviest lifting in the first part of the stage (a) analysis and may also explain why the court spent considerably more time focusing on inconsistency and public confidence.
Consistency and circularity
As noted above, the overarching question the Patel test is designed to answer is whether allowing the claim would produce inconsistency, disharmony and thereby damage the integrity of the legal system. Stage (a) was focused narrowly on the policies underlying the specific prohibition transgressed. In Henderson, Lord Hamblen expanded stage (a) to include “other general policy considerations that impact on the consistency of the law and the integrity of the legal system … In the present case, for example, that would encompass the public policy considerations identified in Gray … namely the consistency principle and the public confidence principle” (Henderson, para [119]). Quite apart from the uneasy conflation of policy and principle, what Lord Hamblen appears to be doing is placing the concept the test is supposed to evaluate – consistency – inside the test itself. Henderson closed the distance between the test and the question it is supposed to answer.
Lewis-Ranwell completes this collapse. The court writes: “The court is entitled and required to take a broad view of the underlying policies, including those relating to the internal consistency of the law and the integrity of the legal system” (para [144]). The court then identifies five specific inconsistencies across nine paragraphs (paras [144-150]). This is all done within stage (a). The court appears to be identifying inconsistencies in order to demonstrate that allowing the claim would lead to inconsistencies. It is difficult to see why this isn’t circular.
In addition to inconsistency, the court considers public confidence, with which inconsistency is “closely bound up” (para [151]). At [127], the court said inconsistency risks damaging public confidence. Inconsistency is a risk factor for public confidence. But in its conclusion on stage (a), the court breaks this connection: “For these reasons, in particular for reasons of consistency and public confidence, we conclude that there are very weighty considerations supporting the view that the underlying purpose of the prohibition would be enhanced by denying the claim” (para [154]). Consistency and public confidence are here identified as distinct reasons supporting a conclusion about the purpose of the prohibition – itself a sub-question within the very framework designed to test for inconsistency. More circularity.
The inevitable collapse
If stage (a) already asks “would permitting the claim lead to inconsistencies?”, the overarching question is substantially answered before reaching (b) or (c). At para [151], the court says the identified inconsistencies are “necessarily detrimental”. The outcome was effectively decided. Stage (b) gets three short paragraphs and stage (c) gets two: it is difficult to see what independent work remains for counter-policies or proportionality. The tripartite structure of the Patel test becomes illusory.
This matters beyond Lewis-Ranwell because the Patel framework must work across future illegality cases. The test was designed as a structured framework that requires judicial judgment: courts examine the situation through three lenses and then assess consistency. Henderson began collapsing the framework by putting the conclusion concept inside stage (a). Lewis-Ranwell completes the collapse. The result is a test that predetermines its own conclusion using undefined or loosely defined policy labels – consistency, public confidence, condemnation – that are open-ended enough to support almost any conclusion. The Unruly Horse.
The Supreme Court may not have needed to take the route it adopted. In Gray, Lord Hoffmann argued that causation analysis might offer an alternative understanding of why claims like these fail: the claimant’s own acts may present insurmountable causation hurdles (Gray, paras [51-4]). The claim under the Human Rights Act 1998 proceeds despite the denial of the negligence claim. Given the tiny number of claims of this nature, one is left to wonder whether the financial and analytical cost of ventilating this issue in the Supreme Court was worth it.
In Patel, Lord Toulson stated “The public interest is best served by a principled and transparent assessment of the considerations identified” (Patel, para [120]). It is questionable whether the Supreme Court achieved either despite its “inevitable” conclusion.