Few doctrines in the law of torts have received as much scholarly attention as the duty of care in negligence. It is legitimate to ask what new contribution a treatise on duty can make. James Plunkett's The Duty of Care in Negligence provides a useful consolidation of the historical and modern evolution of duty and a thoughtful critique of some current scholarly debates. Those who work in the field may find this material to be largely familiar: the historical terrain has already been well-charted by David Ibbetson and others; and the journey from Donoghue v Stevenson [1932] A.C. 562 to Anns v Merton London Borough Council [1978] A.C. 728 and Caparo Industries v Dickman [1990] 1 A.C. 605 is all too well known. Plunkett also provides a catalogue of the various methodologies for assessing novel duty situations, and a discussion of the appropriateness of using so-called “policy” reasoning in duty cases. The latter debate has become, to some scholars, detached from reality.
The meat of Plunkett's book is contained in chs. 4, 5 and 6, which assess the respective concepts of “factual duty” and “notional duty”. Factual duty refers to the determination “whether harm to the plaintiff was a reasonably foreseeable consequence of the defendant's conduct”, while notional duty refers to the question “whether the broad circumstances in which the plaintiff suffered the injury ought to be subject to the laws of negligence” (emphasis original). With respect to the first inquiry, Plunkett agrees with critics who find the duty question superfluous in negligence analyses, as it overlaps with the more clearly factual questions of whether the defendant breached the standard of care and whether the loss was too remote. All of these questions boil down to an assessment of whether the defendant's actions posed a foreseeable risk of the injury suffered by the plaintiff. Plunkett thus argues that the factual inquiry should be removed from duty analyses, which should instead focus exclusively on the notional question.
After reviewing the various duty “tests” that have been employed over the last century (and their respective flaws), Plunkett proposes that the quest for a single notional duty test be abandoned and replaced with five broad duty categories, each with its own principles of recovery and non-recovery. For cases involving physical injury, property damage or psychiatric harm, the principles would explain a general rule of recovery, subject to narrow exclusionary exceptions. Conversely, for cases involving omissions and purely economic loss, the principles would explain a general exclusionary rule, subject to narrow inclusionary exceptions. By employing these more discrete duty categories, Plunkett argues, we would be able more clearly to identify which principles were relevant to any given duty scenario, and avoid having to use concepts like “proximity” in such a broad way as to be vague and unhelpful.
Plunkett's proposal seems sensible, and reflects, to a large measure, the way that appellate courts already approach these situations in practice. Indeed, the UK Supreme Court adopted a very similar approach in its 2018 decision of Robinson v Chief Constable of West Yorkshire [2018] UKSC 4, [2018] A.C. 736. The court indicated that the Caparo approach was unnecessary, since the claim involved a positive act by police that posed a foreseeable risk of physical injury (which is subject to a general rule of recovery). This distinguished the claim from prior authorities involving police omissions to prevent crimes by third parties (which are subject to a general rule of exclusion). Further, the majority stressed that the more discretionary “fair, just and reasonable” stage of the Caparo test is not available to every defendant who wishes to avoid liability; rather, it is restricted to novel claims, where “the existing principles do not provide an answer” and the court must reason by analogy to established cases.
Predictably, the analysis of notional duty entices Plunkett to engage with the now extensive literature about the appropriateness of using policy considerations in the assessment of novel duties of care. Plunkett rehearses the well-worn arguments of scholars like Stevens and Beever that judges are not technically competent to assess policy factors, and that it is not politically legitimate for the (appointed) courts to do so. He also describes the criticisms that policy factors are “incommensurable” with interpersonal justice considerations (see Weinrib, “Does Tort Law Have a Future?” (2000) 34 Val.U.L.Rev. 561), that their use violates the rule of law, and that they render the law incoherent (e.g. Neyers, “The Economic Torts as Corrective Justice” (2009) 17 Torts L.J. 162).
Happily, Plunkett presents a relatively balanced and critical assessment, referring to the positions of Bagshaw, Cane, Priel, Robertson and Stapleton, among others. He questions, in particular, whether judges actually are unequipped to deal with matters of policy, given that they are accustomed, if not mandated, to consider policy implications when deciding questions of public law or human rights. Further, he notes that the use of policy considerations is not unconstrained, and that judges are bound by professional responsibilities and do not decide cases capriciously. Plunkett also points out that so-called “principled” reasoning may still require courts to choose between competing rights, which may be no more predictable or commensurable than questions of policy.
The most novel section of Plunkett's book is his empirical comparison of Commonwealth duty analyses (ch. 7). Plunkett completed a study of all the duty decisions of the High Court of Australia (HCA), Supreme Court of Canada (SCC), and House of Lords/UK Supreme Court (HL/UKSC) delivered between 1985 and 2015, examining the methodologies used by the courts to determine the existence of a duty. This includes whether they used general tests, duty categories or more fact-specific determinations, the extent to which they relied on policy considerations, and the apparent influence of academic commentary.
The results of Plunkett's analysis reveal some striking, though not altogether unsurprising, differences among the jurisdictions. For instance, in Canada, which has maintained its outward loyalty to the Anns test, that test was employed in 80% of all duty determinations. (It was used in 95% of the duty cases decided since 1995.) In contrast, the HL/UKSC used the Caparo test in 24% of duty cases, but used no general test in 63% of cases. The HCA relied on proximity in 14% of cases, the salient features test in 10% of cases, and applied no general duty test in 69% of the cases.
Of the cases applying the Anns/Caparo test, the HL/UKSC was most likely to reject duty at the “fair, just and reasonable” stage, while the SCC's rejections of duty were more evenly split between the elements of proximity and policy. The HL/UKSC rejected duty based on proximity only 15% of the time, and never rejected it based on foreseeability. This suggests that the three-stage test most often comes down to a simple weighing of pro-duty and anti-duty factors at the fair, just and reasonable stage. Interestingly, while the HCA generally eschewed any general duty tests, it used a so-called “balancing approach” (i.e. where the existence of a duty “depends on the circumstances of the case”: Kuhl v Zurich Financial Services [2011] HCA 11; (2011) 243 C.L.R. 361) in 50% of its duty determinations. In short, it appears that a large proportion of duty cases in the highest appellate courts are based on an unstructured or semi-structured balancing of factors that the court deems relevant to the case.
Finally, with respect to the controversial reliance on policy considerations, all three high courts were much more likely to rely on policy considerations as anti-duty factors, but were also most likely to use policy factors to support the conclusions already reached on other grounds. In other words, policy considerations were determinative in only a small minority of cases (only 17% in the HCA, 28% in the SCC and 19% in the HL/UKSC). The most popular anti-duty policy factor was the risk of indeterminate liability. This portion of Plunkett's analysis suggests that the academic hand-wringing over the use of policy is greatly over-stated, and that the law of negligence may not therefore be the “mess” that some scholars make it out to be.
Of particular interest to scholars is Plunkett's analysis of the influence of academic commentary on duty decisions, as evidenced by citations. Obviously, “influence” is difficult to measure, and a lack of citations cannot be equated with lack of influence; as Plunkett notes, courts can be influenced by writing that they do not cite, as in Arthur JS Hall and Co. v Simons [2002] 1 A.C. 615, at 688. Nevertheless, it is interesting that the SCC was markedly more likely to cite academic literature in its duty of care decisions (58% of the time) than the HCA (37%) or HL/UKSC (33%). In all three courts, academic commentary was most likely to be cited as background or in support of the decision; commentary that was in conflict with the decision was cited in only 5–7% of cases.
Although it would be difficult to reliably measure academic influence, it would be interesting to engage in a more detailed qualitative analysis of the commentary cited by the courts to determine which authors or publications are most likely to be cited. In this vein, in as yet unpublished work Professor Katy Barnett of the University of Melbourne has recently analysed academic citations by the HCA, revealing that it frequently cited textbooks, often cited the same works multiple times, and was disproportionately likely to cite Oxbridge scholars. The court also cited men over women by a ratio of about 10:1.
It is disappointing that Plunkett's empirical analysis did not form a greater proportion of his book, as it was highly intriguing and presented the best opportunity for him to make a unique contribution to the field. Similar analyses would provide much-needed context for ongoing scholarly debates about duty methodology, and might reveal that academics are focusing their energies on matters that have little real-world impact.
