IT is a truth universally acknowledged that the English law on negligence liability for mental injury to secondary victims is unsatisfactory. What is less readily acknowledged is what to do about it. In White v Chief Constable of South Yorkshire [1999] 2 A.C. 455, the House of Lords stated that “the search for principle was called off in Alcock v Chief Constable of South Yorkshire [1992] 1 A.C. 310” (p. 511 (Lord Hoffmann)) and “the only sensible general strategy for the courts is to say thus far and no further” and “by and large to leave any expansion or development… of the law to Parliament” (p. 491 (Lord Steyn)). In the quarter of a century since, Parliament has done nothing, despite detailed recommendations by the Law Commission. Lower courts have responded variously: some by relaxing the Alcock straitjacket at the margins, others by keeping it firm. Fine distinctions have become finer, and the law more complex. In Paul v The Royal Wolverhampton NHS Trust [2022] EWCA Civ 12, the Supreme Court was invited by the Court of Appeal to consider the law on secondary victims in the context of medical negligence. This fresh opportunity for an authoritative examination of the law brings us back to the central question: what to do about negligence liability for secondary victims?

Paul concerns three joined cases with similar facts. The defendant allegedly failed to diagnose a person's life-threatening condition causing their traumatic death sometime later. The claimants were close family members of the deceased who suffered mental injury by witnessing their relative's death or coming upon them immediately after death. Whether they were owed a duty of care as secondary victims turned on the application of the Alcock framework. This applies in addition to the need for a recognised mental injury and reasonable foreseeability of mental injury to a person of normal fortitude. It requires: (1) a close tie of love and affection between the claimant and the person imperilled; (2) physical and temporal proximity to the (horrifying) event of imperilment or its immediate aftermath; (3) direct perception through unaided senses of the event or its immediate aftermath; (4) sudden shock in the sense of a sudden appreciation of a horrifying event which violently agitates the mind. In Paul, all claimants witnessed the horrifying death of their relative or its immediate aftermath, but not the defendant's misdiagnosis months earlier. The issue was whether element (2) was satisfied, namely, whether the relative's death could constitute the relevant event for the purposes of physical and temporal proximity. The Court of Appeal hesitantly held that it could not, consequently the claims failed.

The issue was not strictly governed by Alcock or its precursor McLoughlin v O'Brian [1983] 1 A.C. 410 (see Lord Wilberforce's speech), both cases which involved a typical accident scenario where the defendant's careless act/omission and the physical consequences to the imperilled person giving rise to the “horrifying event” occur at, or approximately at, the same time. Paul is a typical medical negligence scenario involving a delay between the careless act/omission and the physical consequences to the imperilled person.

Nevertheless, the Court of Appeal held that it was bound by its own decision in Taylor v Novo (UK) Ltd. [2013] EWCA Civ 194. In that case, the claimant's mother was injured at work when a stack of racking boards was carelessly tipped over her (the accident). She collapsed and died three weeks later at home in front of the claimant. Lord Dyson M.R. held that there was a “single event”, the accident, which had two consequences: injury suffered by the mother in the accident and her subsequent death. It was not enough for the claimant to witness the latter as that did not qualify as a shocking event under Alcock; to recover, she had to witness the accident itself. The court in Paul treated Novo as binding authority for the proposition that there can be no delay between the defendant's careless act/omission and the physical consequences that the claimant witnessed (at [12], [96] (Vos M.R.); at [104] (Underhill L.J.)). This covers two situations. The first is where the careless act/omission produces physical consequences immediately and sometime later, at which point they are witnessed by the claimant; the second occurrence cannot be the qualifying shocking event. These are the facts of Novo. The second is where the careless act/omission does not produce physical consequences immediately but only does so later, at which point they are witnessed by the claimant; that occurrence cannot be the qualifying shocking event. This follows from Lord Dyson M.R.'s approval of Auld J.'s judgment in Taylor v Somerset Health Authority [1993] P.I.Q.R. 262 but is undermined by the incompatibility with North Glamorgan NHS Trust v Walters [2002] EWCA Civ 1792, where a baby's epileptic seizure and death, caused by careless misdiagnosis months earlier, constituted a qualifying shocking event. The claims in Paul fell within the second situation and so failed. However, the court expressed reservations as to whether Novo correctly interpreted the relevant House of Lords authorities and gave permission to the claimants to appeal to the Supreme Court.

It is respectfully submitted that the Court of Appeal's decision should be reversed. There are three ways of doing this. The first is to distinguish Novo on the basis that the defendant's careless act/omission in that case had two physical consequences – the injury in the accident and the death, whereas in Paul it had one – the death. There is only one point in time when physical consequences in Paul materialise to give rise to a shocking event. In other words, Novo covers only the first of the two situations described above. While this reasoning is doctrinally defensible, it has the unwelcome effect of introducing even finer distinctions into a body of law that is riddled with hair-splitting and arbitrary distinctions.

The second way is to overrule Novo. Novo is based on the notion that the policy considerations justifying the Alcock framework also dictate that the framework must be applied to new factual situations restrictively so as not to generate liability. Such a reading of “thus far and no further” is questionable. It results in the Alcock framework being applied too rigidly and contrary to “the need for flexibility in dealing with new situations” that Lord Slynn affirmed in W v Essex CC [2001] 2 A.C. 592, 600. It makes it almost impossible for secondary victims to recover in cases involving careless medical treatment of someone else. In short, it is possible to affirm Alcock and overrule Novo. To do so, however, would be a missed opportunity. This is because the policy justifications underpinning Alcock are themselves unconvincing and ought to be confronted head on.

The third way is to reform the Alcock framework. This is the boldest yet most convincing approach. It is in line with principle, policy and historical development. First, it has long been accepted that Alcock is not justified as a matter of principle. Neither “corrective justice”, which provides the structure of tort law, nor “distributive justice”, which operates within that structure, justify treating pure mental injury and physical injury differently. This conclusion is also likely to follow if one were to ask what the public think, as Lord Hoffmann did in White under the false banner of “distributive justice” (p. 510).

Second, policy does not justify the Alcock framework. Of the considerations articulated by Lord Steyn in White (pp. 493–94), the most potent is “floodgates” – the notion that a single event will lead to a flood of claims. The reason is that there is no physical immediacy in the causation of pure mental injury to restrict the class of claimants, unlike with physical injury. However, experiences in other jurisdictions show that the concern is overstated. While it may support some restrictions, it does not support Alcock.

Third, reforming Alcock will be in keeping with the historical development of the law, admirably explained in numerous studies (see e.g. D. Mendelson, The Interfaces of Medicine and Law (Aldershot 1998)). Attempts to limit recovery have traditionally been based on judicial understandings of mental injury as informed by contemporary medical understandings. For example, developments in the early law are explained by reference to medical perceptions of the relationship between body and mind. It is true that law has always lagged behind medical science as older decisions, fortified by precedent, are not easily dislodged. But it is also true that law has for most of its history been informed by medical science. Yet, since Alcock, the two have been seriously out of step, with medical advancement ceasing to animate the law. It is time for law and science to march again in the same direction, but with law limping a little behind (cf. Mount Isa Mines v Pusey (1970) 125 C.L.R. 383, 395 (Windeyer J.))

If the Alcock framework is to go, what should replace it? One option is to hold that the test is whether it is reasonably foreseeable that the claimant would suffer mental injury as a result of the defendant's careless act/omission, and explain that in determining whether that is satisfied, courts can take into account several factors, including the relationship of the claimant with the person imperilled, the physical and temporal proximity to the event, and how her mental injury was caused (see e.g. Review of the Law of Negligence (Ipp Review) (2002); Tame v New South Wales / Annetts v Australian Stations Pty Ltd. [2002] HCA 35; Dillon v Legg 68 Cal. 2d 728 (1968) (cf. Thing v La Chusa 48 Cal. 3d 644 (1989))). Another (less expansive) option is to affirm the test of reasonable foreseeability of mental injury, but discard physical and temporal proximity, direct perception of the event and sudden shock as additional requirements. This would restrict claims to those with close ties of love and affection to the person imperilled (see Law Commission, Rep 249, 1998). In either case, the test would be additional to the requirement for a recognised mental injury.

One concern is that reform might result in a flood of successful claims. The requirement of a recognised mental injury, however, makes this unlikely. Nor should it be assumed that the test of reasonable foreseeability of mental injury can always be readily satisfied. Evidence from comparable jurisdictions does not suggest so. Another concern is that it might create uncertainty. Reform, however, does not mean starting from scratch. Even the adoption of the first (more expansive) option would not land English law in uncharted territory; instead, it would return it to the position espoused by Lord Bridge in McLoughlin. What is more, previous decisions can be rationalised within the adopted framework (or discarded as incompatible with it) to offer guidance on how it applies in particular situations. And while certainty is important, it is not the only desideratum; the aim in developing the law of negligence is to balance the need to maintain certainty with the need to achieve justice (Robinson v Chief Constable of West Yorkshire [2018] UKSC 4, at [27] (Lord Reed)). In this connection, were the first approach to be adopted, it is paramount that it does not result in a mechanical application of the factors guiding reasonable foreseeability, otherwise they would simply transform from overt to covert control mechanisms, robbing the approach of its flexibility.

Finally, should such reform not be left to Parliament? There are good reasons why courts might generally take that course. Judges have no relevant expertise or easy access to the sort of material that informs legislative policy making. Nonetheless, the case for deferring to Parliament in this area is less strong. The law with its policy restrictions has been exclusively developed by the courts. Much preparatory work has been done by the Law Commission. Parliament has not only repeatedly failed to act, but the Government has taken the view that the development of the law is best left to the courts (Department for Constitutional Affairs, The Law of Damages (CP 9/07), 2007). Pleas from judges, the Law Commission and individual Members of Parliament have fallen on deaf legislative ears. In these circumstances, expecting Parliament to act is a pious hope. The ball is now firmly in the Supreme Court's court. It is hoped that it will pick it up and bring the law into line with contemporary societal and medical perceptions. It will no doubt be tempting to conclude again that legislation is the best, perhaps the only, option for reform. Tempting, but wrong.