ROBINSON v Chief Constable of West Yorkshire Police [2018] UKSC 4, [2018] 2 W.L.R. 595, is a landmark in the tort of negligence. Unlike many of its predecessors, its status as such is not based on breaking new ground, but on charting the existing one with sufficient clarity. While Robinson does not bring about a significant change in the law, it provides an authoritative guide to the landscape of determining the foundational question within the tort of negligence: when does a duty of care arise?
The facts in Robinson were simple. Two police officers attempted to arrest a suspected drug dealer in a busy street in the centre of Huddersfield. The suspect resisted and a struggle broke out, during which the three men knocked into Robinson, a relatively frail 76-year-old woman who was passing by. Having suffered injuries, she sued the police in negligence. Her claim failed at first instance and in the Court of Appeal, but it succeeded in the Supreme Court. Although the Supreme Court's decision was unanimous, the judges’ reasoning was not. The majority approach is found in Lord Reed's judgment, with which Lady Hale and Lord Hodge agreed. The judgment clarifies a number of important issues regarding the correct approach to determining when a duty of care exists.
The first concerns the establishment of a duty of care in general. Lord Reed held, at [21], that “the proposition that there is a Caparo test which applies to all claims in the modern law of negligence … is mistaken”. In Lord Reed's view, the whole point of Caparo Industries plc v Dickman [1990] 2 A.C. 605 was to retreat from the idea in Anns v Merton London Borough Council [1978] A.C. 728 that there is a single universally applicable test. Properly understood, Caparo advocated an incremental approach, which is based on the use of established categories as guides into the existence and scope of a duty of care in a novel situation (at [25]). In Lord Reed's analysis, the correct approach today is as follows: (1) where there is a line of authority establishing that a duty of care is owed or not owed, the courts must follow that authority. In such a case, it is “unnecessary and inappropriate” to consider whether the existence or non-existence of a duty of care is “fair, just and reasonable”, because the decision to recognise or not a duty is already based on considerations of justice and reasonableness (at [26]); (2) where the Supreme Court is invited to depart from an established line of authority, it can appropriately examine what is “fair, just and reasonable” (at [26]); (3) where there is a novel case in which the existing authorities do not provide an answer on whether a duty of care exists, the courts must develop the law “incrementally and by analogy with established authority” (at [27]). This includes considering what is “fair, just and reasonable”. Overall, the relevance of the Caparo factors (reasonable foreseeability, proximity and fairness, justice and reasonableness) lies in providing a framework of inquiry into whether the law ought to take an incremental step by analogy with established authority.
Lord Reed's analysis was not original, nor was it meant to be. Yet the fact that it proved necessary to undertake shows the level of confusion which persisted after Caparo and the importance of removing it. It is nonetheless still not entirely clear when a type of case is novel, as illustrated by the disagreement between Lord Reed, at [28], and Lord Mance, at [86]–[87], over whether or why Smith v Ministry of Defence [2013] UKSC 41, [2014] A.C. 52, was such a case. Robinson suggests that a novel situation is one which falls outside an established category of liability, or which cannot be resolved by reference to established principles in the existing case law. In this respect, Lord Reed treated physical loss resulting foreseeably from positive conduct as constituting axiomatically an established category, irrespective of the precise factual circumstances.
Lord Reed's analysis also affirmed the legitimacy of examining policy arguments when deciding whether a duty of care exists, albeit only in the situations outlined above (at [42] and [69(2)]). Similarly, at [84], Lord Mance regarded the view that the courts should not be influenced by policy considerations in this context as “unrealistic” and contrary to the way that the common law has traditionally developed. Few judges, if any, would nowadays subscribe to such a view (cf. Lord Scarman in McLoughlin v O'Brian [1983] A.C. 410, 430–31), but it finds some support in academic literature (e.g. R. Stevens, Torts and Rights (Oxford 2007)). Robinson is a reminder that judges occupy a world in which theoretical concerns co-exist with pragmatic ones.
The second issue relates to the establishment of a duty of care in cases involving public authorities. Building upon Lord Toulson's judgment in Michael v Chief Constable of South Wales Police [2015] UKSC 2, [2015] A.C. 1732, Lord Reed held, at [32], that “at common law, public authorities are generally subject to the same liability in tort as private individuals and bodies”. This can be traced to A.V. Dicey, Lectures Introductory to the Study of the Constitution (London, 1885), 177–78, and is sometimes known as the “equality principle”. Public authorities are generally treated like private individuals and ordinary negligence principles apply to them. Several consequences follow. First, a public authority owes a duty of care for positive acts which, if done by a private individual, would give rise to a duty of care. Secondly, the omissions principle applies to a public authority; therefore, as a general rule, it does not owe a duty of care to protect a person from harm by another. Thirdly, a public authority does not come under a positive duty to act just because it has a statutory duty or power which, if exercised, might have prevented the harm. Fourthly, recourse to policy considerations in cases involving public authorities takes place only in a novel situation or where established authority is challenged.
For all the clarity and simplicity of the analytical framework put forward, questions remain about its foundations. According to Lord Reed, at [31], it represents “a return to orthodoxy”, echoing Lord Hoffmann's reasoning in Stovin v Wise [1996] A.C. 923 and Gorringe v Calderdale Metropolitan Borough Council [2004] UKHL 15, [2004] 1 W.L.R. 1057, and restoring the position in East Suffolk Rivers Catchment Board v Kent [1941] A.C. 74. However, the claim to orthodoxy was challenged by Lord Mance and Lord Hughes on the basis of a long line of cases which justified the non-liability of public authorities in various situations by reference to policy arguments, most notably in Hill v Chief Constable of West Yorkshire [1989] A.C. 53. Approving those cases, Lord Hughes held that although the omissions principle may provide another reason why the police do not owe a duty of care to victims when investigating crime, it cannot be the only or a sufficient reason (at [117]); instead, the ultimate reason lies in policy (at [118]). Lord Reed's answer, at [69(3)], was to explain that the relevant cases, such as Hill, involve the application of the omissions principle, with policy discussed because they also raised novel legal issues or a challenge to established authority. In any case, Lord Reed acknowledged, at [31] and [40], that the reasoning in some of those cases has now been superseded. In truth, the relevant cases were not decided as omissions cases, and, to that extent, Lord Reed's judgment involves an element of re-interpretation of their juridical basis. All in all, Lord Reed's analysis amounts to choosing one line of conflicting authority over another, which is entirely legitimate. In so choosing, he avoided the use of the policy arguments in Hill which are problematic.
More fundamentally, the application of the omissions principle to public authorities has been questioned, first, on the basis that its justifications do not apply with equal force to public authorities, and, secondly, on the basis that there are normatively important differences between public authorities and private individuals which validate treating them, in this context, differently to a degree (S. Tofaris and S. Steel, “Negligence Liability for Omissions and the Police” [2016] C.L.J. 128). Even so, a majority of the Supreme Court has recently applied the omissions principle to public authorities twice in Michael and Robinson. In the interests of certainty, the law should now be regarded as settled.
The third issue concerns the negligence liability of the police. Having rejected a special treatment of the police when investigating or suppressing crime on the basis of the policy arguments endorsed in Hill (at [69(1)]), Lord Reed applied the general framework of negligence liability outlined above. This made the resolution of the case easy. As this was a case of a positive act, under the application of ordinary principles the police owed a duty of care to the claimant not to cause her foreseeable physical injury. The only remaining issue was whether the police had breached their duty. It was held, at [77]–[78], that they had done so, because, intending to arrest the suspect in a way that would minimise the risk of harm to passers-by and thus to cancel the arrest if someone was in harm's way, they failed to notice Robinson who had just walked by.
Lord Reed's judgment in Robinson has sought to revert to a position that he regarded as orthodox. Notwithstanding questions about the prevalence of that position in the pre-existing case law and its general desirability, the judgment has resolved several ambiguities in the law of negligence and has provided a blueprint for its future development.
