A Middle Ground Approach to Rylands v Fletcher Liability for Ultra-Hazardous Activities, Lecture notes of Law

Download A Middle Ground Approach to Rylands v Fletcher Liability for Ultra-Hazardous Activities and more Lecture notes Law in PDF only on Docsity! OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 80 A Thing Less Likely to Do Mischief: An Alternative Application of the Rule in Rylands v Fletcher to ‘Ultra-Hazardous’ Activities Ryan Chan-Wei* Introduction hrough the past century, the rule in Rylands v Fletcher1 has been the subject of much debate, to the point where Francis Bohlen commented that yet another analysis of Rylands v Fletcher would merely be ‘a thrashing out of old straw’.2 Considering how crowded the field is, it might be difficult to see what value yet another article can add, let alone one by an undergraduate. However, I humbly suggest that there is still more to be said, especially with respect to how Rylands v Fletcher has been applied in the specific context of ‘ultra-hazardous’ activities. In particular, the English courts have not adopted a distinct approach3 in applying the rule in Rylands v Fletcher to ‘ultra-hazardous’ activities despite valid concerns raised to the contrary, 4 instead preferring to subject both ‘ultra- hazardous’ and non-ultra-hazardous’ activities to the same Rylands v Fletcher standard. This article will demonstrate that such a position can, and should, be rethought. 1. Overview The existing debate concerning Rylands v Fletcher liability for ‘ultra-hazardous’ activities has largely been framed as a stark choice between two polarities, to the point where it was once described as ‘a storm centre’.5 On the one hand, some see Rylands v Fletcher as an opportunity to impose a blanket rule of strict liability for ‘ultra-hazardous’ activities, by embracing the enterprise liability approach that has significantly shaped American tort law, and leaving the non-‘ultra-hazardous’ Rylands v Fletcher cases to negligence.6 On the other hand, some see the principle as obsolete, arguing instead that the rule in Rylands v Fletcher is an anachronism in light of the development of the fault principle and that it should be subsumed into the broader tort of negligence, even in the context of ‘ultra-hazardous’ activities, as the Australian courts have done.7 In contrast, this article will suggest that a middle ground can be found. An over-extension of the rule in Rylands v Fletcher (i.e. by imposing a general regime of strict liability for ‘ultra-hazardous’ activities) is just as undesirable as subsuming it entirely into the law of negligence, as both are sweeping responses to a category of cases that often have unique fact situations and instead require a * St. Anne’s College, Oxford. I am deeply grateful to Dr Imogen Goold and Benjamin Mak for their comments on an earlier draft. I would also like to express my heartfelt thanks to the editorial team of the Oxford University Undergraduate Law Journal for reviewing this article. Any and all errors remain my own. 1 [1868] LR 3 HL 330 (HL). 2 Francis Bohlen, ‘The Rule in Rylands v Fletcher’ (1911) 59 U Pa L Rev 298, 373 and 423. 3 This reluctance is primarily underpinned by concerns over the ‘uncertainties and practical difficulties’ of pursuing a separate approach for ‘ultra-hazardous’ activities, as highlighted by Lord Goff in Cambridge Water Co v Eastern Counties Leather Plc [1994] 2 AC 264 (HL) 305. These concerns will be addressed in Section 3. 4 For instance, see John Fleming, The Law of Torts (9th edn, Law Book Company 1998) 341. Arguing in favour of strict liability, he criticizes Read v Lyons [1947] AC 156 (HL) for prematurely stunting the ‘development of a general theory’ of Rylands v Fletcher liability for ‘ultra-hazardous’ activities. 5 Cecil Wright, ‘The Adequacy of the Law of Torts’ (1961) CLJ 44, 52. 6 Cambridge Water Co (n 3) 304. 7 Burnie Port Authority v General Jones Pte Ltd [1994] HCA 13. T OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 81 more calibrated approach. For instance, it will be shown that the Australian approach has the right normative instinct (i.e. to focus on fault),8 but the wrong practical implementation (i.e. the evidential requirement of proving duty of care and breach of that duty to make out the tort of negligence may be too onerous a burden for the defendant to bear). Instead, Rylands v Fletcher liability should be kept as it stands in English law, with the exception that the test for ‘ultra-hazardous’ activities should be separate, because such activities carry a uniquely high risk of harm. Unlike ordinary activities, when engaging in an ‘ultra-hazardous’ activity, even a small degree of fault can lead to the manifestation of an inordinate amount of harm; there is thus value in treating them as distinct, in order to better account for the disproportionate risks they pose. This article will therefore propose an alternative framework for assessing Rylands v Fletcher liability in the context of ‘ultra-hazardous’ activities, which consists of 1) a working definition of an ‘ultra-hazardous’ activity and 2) a ‘reduced fault’ liability test. It may not be a flawless approach in practice, but it is arguably better for the courts to adopt an untested framework that attempts to offer a more nuanced and context-specific analysis of individual cases, than to apply a rigid, broad-brush approach which swings to doctrinal polarities and ignores the factual differences across the broad spectrum of Rylands v Fletcher cases. After all, in the former, we are approximately right; in the latter, we are precisely wrong.9 2. Understanding the Rule in Rylands v Fletcher In Rylands v Fletcher, the factual scenario was fairly straightforward. The defendants constructed a reservoir in order to provide water for their mill, with the permission of the owner of the adjacent land. They employed independent contractors to do so and they themselves took no part in the construction. While the defendants were not guilty of any negligence in their selection of the site for the reservoir, it eventually transpired that there were some long abandoned mine workings underneath the site. In particular, there were several old shafts, which appeared to have been filled in. When excavating the bed of the reservoir, the contractors came upon these shafts, but did not make their existence known to the defendants. It was found that the contractors were negligent in proceeding with the construction of the reservoir despite their knowledge of the shafts. The plaintiffs were lessees of a mine that lay under land close to the reservoir. In working their mine, they had broken into some of the aforementioned ‘long abandoned mine workings’. After the contractors completed work on the reservoir, the defendants filled it with water. The shafts (which were filled in by soil) gave way under the pressure of the water, allowing the water to flow into the abandoned mine workings. The water then flowed into an opening the plaintiff had made when breaking into the abandoned mine workings, and from there the water finally flowed into the plaintiff’s mine. At trial, the main point of contention was whether the defendants could be held liable irrespective of proof of negligence on their own part, or on the part of the independent contractors. Although the court had some doubts about whether an isolated escape, as opposed to a continuous state of affairs, could found an action in nuisance, the defendants were ultimately held liable. In delivering the opinion of the Court of Exchequer Chamber, Blackburn J laid down the 8 The complex jurisprudential debate surrounding the role of corrective justice and strict liability in tort law is beyond the ambit of this article. For the sake of clarity, it will suffice to note that the former should be preferred in context of Rylands v Fletcher liability. For more details as to why the latter should be approached with caution, see Lord Goff’s judgment in Cambridge Water Co (n 3) 305-306. Also see John Fleming, The Law of Torts (9th edn, Law Book Company 1998) 369. 9 To paraphrase a quote from famed investor Warren Buffett, who once said, ‘It is better to be approximately right than precisely wrong.’ OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 84 However, a greater cause for concern is the incoherence of the notion that an ‘ultra-hazardous’ activity by definition has to be one ‘not of common usage’. From a policy standpoint, it is hard to see why this should be, because there are clearly examples of activities in ‘common usage’ that one might instinctively regard as ‘ultra-hazardous’. For instance, nuclear power plants have been recognized as ‘ultra-hazardous’ by the Nuclear Installations Act 1965, but it is plausible that, at one point, those living near Three Mile Island or Fukushima (or any city that depends on nuclear power to support its economy and energy needs) would have regarded nuclear power plants as being of ‘common usage’, in the sense that it was ‘customarily carried on by many people in the community’.21 Consider also the use of explosives to blast open a mine, which is likely to be regarded as an ‘ultra-hazardous’ activity. In an urban centre, this would clearly not be of ‘common usage’; however, by contrast, in a mining town, it is likely that this would be considered ‘common usage’ and not an unacceptable risk, because many residents of that town are likely to be involved in the mining activities too (and probably accustomed to the occasional debris flying into their property from the explosions).22 Nevertheless, Section 520 of the First Edition of the Restatement of Torts can still serve as a useful definitional guide, even though it is somewhat unclear, because its emphasis on ‘common usage’ rightfully draws attention to the notion that the level of risk considered to be ‘unacceptable’ for the purposes of an ‘ultra-hazardous’ activity should depend on the context. Instead of imputing a consistent objective standard of ‘common usage’ to all fact situations, which is a rather broad-brush approach; a more context-specific approach would be beneficial, because the extent to which an activity is of ‘common usage’ can depend significantly on the norms of a particular locality. B. Proposed Definition The proposed definition builds on the foundations set out in the First and Second Editions of the Restatement of Torts, in particular by modifying the test from the Second Restatement. It should be noted that although the Second Restatement uses the term ‘abnormally dangerous’ activity instead of ‘ultra-hazardous’ activity, this is largely a matter of semantics. ‘Abnormally dangerous’ is meant to be a slightly more accurate descriptor of how such activities tend not to be of ‘common usage’, but for practical purposes it can be used inter-changeably with ‘ultra-hazardous’, as was done in the cases of Langan v Valicopters23 and Bennett v Larsen.24 It is also important to note that this stage of determining whether an activity is ‘ultra- hazardous’ in the first place is a necessary but insufficient prerequisite for imposing liability; liability only attaches in the second stage, after applying the ‘reduced fault’ liability test (which will be discussed in the next section). If an activity does not fulfil the ‘ultra-hazardous’ test in the first place, then the liability test subsequently applied will be that of the normal fault test under the tort of negligence or nuisance, rather than the ‘reduced fault’ test. The test proposed in this article is modified to account for more context-specific criteria and to offer more clarification on issues such as the risk of harm. My submission is that the defendant would be engaging in an ‘ultra-hazardous’ activity if a reasonable man in the defendant’s shoes would conclude that: i. The activity is inherently dangerous, in that it involves a high risk of severe harm that cannot 21 Section 520(e) of the First Edition of the Restatement of Torts and Section 520(i) of the Second Edition of the Restatement of Torts. 22 It is important to note that on occasion ‘there will be a regime of statutory liability in force, which will displace the common law rule, and frequently exceed it’, as observed by Jenny Steele in Tort Law (3rd edn, Oxford University Press 2014) 697. For instance, consider Section 209 of the Water Industry Act 1991. However, barring the presence of legislation, it is still necessary to craft a working definition of an ‘ultra- hazardous’ activity that is broad enough to cover the examples above, lest they fall through the cracks. 23 Langan v Valicopters Inc 567 P 2d 218 (Wash 1977). 24 Bennett v Larsen Co 348 N W 2d 540 (Wis 1984). OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 85 be eliminated despite the exercise of reasonable care; ii. And the activity is not of ‘common usage’ (in the sense that it is neither customarily nor contextually appropriate to the locality in which it is carried out), such that the risks that accrue from it are deemed ‘unacceptable’. The first limb accounts for the higher risk of harm threshold inherent in ‘ultra-hazardous’ activities. Additionally, it highlights a key distinguishing factor between liability for ‘ultra-hazardous’ activities and ordinary negligence liability, which is that liability under the former can still be incurred notwithstanding the exercise of reasonable care. In contrast, it is unlikely that ordinary negligence liability would be found if the harm could not be eliminated despite the exercise of reasonable care, as it would not be ‘fair, just and reasonable’ (under the third limb of the Caparo25 test) for the courts to implement a duty of care in that context. The second limb addresses variations in context that are not accommodated by the Second Restatement’s original definition, as discussed earlier, because the extent and type of risk that one might deem abnormal or ‘unacceptable’ ultimately depends on the context. Furthermore, it also addresses the possibility that the same activity can carry a different risk of harm, depending on the circumstances. For example, storing nuclear waste in an underground bunker in the middle of the desert carries a lower risk of harm than storing it in an underground bunker next to a densely populated city. The test might be subject to criticism on the basis that its objectivity is merely a veil for the judiciary to apply a policy decision, as an objective standard tends to be inherently vague and how it is applied depends largely on who decides the case. Nevertheless, this has not prevented a workable objective test from being applied in numerous other circumstances (e.g. negligence and product liability), and thus should not pose a significant obstacle in this instance either. 4. An Alternative Application of the Rylands v Fletcher Rule to ‘Ultra-Hazardous’ Activities A. Lord Bingham’s Obiter Dictum in Transco26 There is value in treating ‘ultra-hazardous’ activities as distinct, because they carry a uniquely high risk of harm. Unlike ordinary activities, when engaging in an ‘ultra-hazardous’ activity, even a small degree of fault can lead to the manifestation of an inordinate amount of harm; and such activities should therefore be afforded a separate test in order to better account for the disproportionate risks they pose. Although the English courts have been hesitant to treat ‘ultra-hazardous’ activities as distinct, due to difficulties in defining ‘ultra-hazardous’, the preceding section has shown that it is possible to come up with a working legal definition. Having addressed that, this article will now explore how Lord Bingham’s obiter dictum in Transco could hypothetically be applied, by extrapolating from his passing remarks in that case to create a legal framework for thinking about Rylands v Fletcher liability in the specific context of ‘ultra-hazardous’ activities. In obiter,27 Lord Bingham observed that there was a ‘category of case, however small it may be’ in which it ‘seems just’ to extend the rule in Rylands v Fletcher and impose liability ‘even in the absence of fault’. He did not offer any concrete explanations, but instead merely offered a few cases as examples of what would constitute this category, such as Cory Brothers, 28 Rainham Chemical Works29 and Cambridge Water Co, in which the defendants were all arguably engaged in some form of ‘ultra-hazardous’ activity. However, although Lord Bingham referred specifically to an ‘absence of fault’, a common thread running through each of these cases is in fact the presence of a fault element 25 Caparo Industries plc v Dickman [1990] UKHL 2, [1990] 2 AC 605. 26 Transco Plc v Stockport MBC [2003] UKHL 61, [2004] 1 AC 1 [6]. 27 ibid. 28 Attorney General v Cory Brothers & Co Ltd [1921] 1 AC 521 (HL). 29 Rainham Chemical Works v Belvedere Fish Guano [1921] 2 AC 465 (HL). OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 86 of unreasonable risk-taking. However, the degree of fault in these cases is too insignificant to be captured by traditional, currently legally recognized standards of fault (such as negligence). For instance, Lord Bingham suggests that liability should have been imposed in Cory Brothers, even if the claimants had failed to prove negligence, since ‘memories of the tragedy at Aberfan are still green’.30 The facts of Cory Brothers are similar to that of the Aberfan tragedy, where a waste tip slid down a mountainside into a village and killed many people, as the defendants placed up to half a million tons of mineral waste on a steep hillside, some of which eventually slipped off and damaged nearby properties. Lord Bingham appears to be suggesting that liability in the absence of negligent fault might be appropriate because the defendant was aware of the potential for severe harm to occur, and is therefore at fault for proceeding with the activity without making any effort to take precautions to mitigate the risk of harm materializing. This echoes the view of Viscount Haldane in Cory Brothers, who argued that instead of proving negligence (as the claimants did), it would suffice merely to ‘show that the hillside was steep, and that to pile rubbish on it in a large heap was to put a dangerous structure there’.31 Similarly, in Rainham Chemical Works, the defendants were involved in manufacturing picric acid (an explosive chemical) for the British government to make munitions during the First World War. However, they were not experts in chemistry and did not realize that that the picric acid was in fact highly explosive. Although there was no proof of negligence in their production process, they were nonetheless held liable for damage to neighbouring property when a fire occurred in their factory and the picric acid led to an explosion of ‘terrific violence’.32 A fault element is less obvious than in Cory Brothers, but was arguably still present: the chemical clearly had the potential to cause serious harm (the defendants knew that they were being manufactured as part of a war effort to make explosive munitions), and the defendants could be faulted for not attempting to determine the exact extent to which their chemicals were explosive. A reasonable man would arguably have made an effort to identify the specific explosive potential of the chemical, before proceeding with their production, in order to be better placed to take precautions to prevent the risk of harm from materializing. Finally, Lord Bingham held that an extension of the rule in Rylands v Fletcher would be justified in Cambridge Water Co if the damage had been foreseeable by a reasonable man, and it is clear a fault element would have been present on the part of the defendant tannery in such a case, as a reasonable man would have taken more steps to prevent their industrial solvents from percolating down to the water table – after all, it was established in the case that the solvents were ‘bound to cause mischief’ if they escaped.33 It is important to note that although Lord Bingham might appear to be proposing the imposition of a strict liability regime (albeit limited to the abovementioned fact situations), since he mentions the imposition of liability ‘even in the absence of fault’,34 this is unlikely to be the case. It would be inordinately difficult to justify the imposition of a limited strict liability regime,35 as any arguments used to establish this (e.g. economic efficiency) are likely to apply with equal force to the 30 Transco Plc (n 26) [6]. 31 Attorney General (n 28) 536. 32 Rainham Chemical Works (n 29) 471. 33 Cambridge Water Co (n 3) 288-289. 34 Transco Plc (n 26) [6]. 35 Even if a case could successfully be made for the imposition of a limited strict liability regime (e.g. ‘only in the absence of fault’), it would then be very difficult to justify circumscribing it only to that particular context, because of the all-encompassing nature of strict liability, which imposes liability regardless of whether the defendant was at fault or not. Therefore, it is likely that any arguments for a limited strict liability regime would apply with equal force to a general strict liability regime, because the former is not easily distinguishable from the latter. OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 89 Australia was right to conclude in Burnie Port Authority that a duty of care analysis is appropriate since Rylands v Fletcher cases often reflect a relationship where the claimant is in a special position of dependence or vulnerability, it is precisely because of this vulnerability that it might not be appropriate for a Rylands v Fletcher case to be brought under a negligence action. After all, it is eminently conceivable that the disparity between the plaintiff and defendant could be so large (e.g. the defendant might be a large industrial enterprise) as to be ‘highly improbable that a (relatively impecunious) claimant would be able to establish the necessary lack of reasonable care for the purposes of a negligence action’. 48 Essentially, the Australian approach has the right normative instinct (to focus on fault), but the wrong practical implementation (the evidential requirement of proving duty of care and breach of that duty is too onerous a burden for the defendant to bear). In contrast, this alternative approach of applying a ‘reduced fault’ test attempts to tread the middle ground and offers a more calibrated approach. It lowers the bar from the Australian approach, such that claims can still be brought as long as some fault element can be established. At the same time, this does not impose a regime of strict liability as broad as the American approach, which may prevent cases from slipping through the cracks, but also brings with it a host of undesirable attendant consequences. By way of illustration, consider two defendants – one storing hazardous chemicals, and another transporting nuclear waste. The former is aware that his chemicals are somewhat hazardous, but does not make an effort to determine exactly how hazardous they are, beyond placing them in ordinary storage barrels conventionally used for storing hazardous chemicals. However, the chemicals are more hazardous than he expects, eventually corroding the barrels and leaking out, causing severe property damage. In contrast, the latter takes precautions that reduce the level of risk to miniscule proportions, such as transporting the nuclear waste in an armoured train, using several layers of durable protective casts.49 However, the waste leaks out when the train is derailed due to a fault in the tracks, and causes equally severe property damage. According to the American approach, which imposes a general regime of strict liability for all ‘ultra-hazardous’ activities, it is likely that both defendants will be found liable. On the other end of the spectrum, according to the Australian approach, it is plausible that neither defendant will be found liable. In the latter case, there was no negligence on the part of the defendant; in the former, it may not be possible to prove negligence, as the defendant’s fault may not have been sufficient to establish a breach of duty (as in Rainham Chemical Works). However, under the proposed test, a more calibrated response can be found, such that the defendants’ fault is taken into account and they are not treated with the same broad-brush approach despite their different factual situations. Applying this test to the first defendant, he would have been found liable, as it is 1) it is plausible that he should have reasonably foreseen a risk of harm materializing from storing hazardous chemicals and 2) knowing of the risk of harm, a reasonable man would have first determined the extent to which the chemicals could pose a hazard, before proceeding with storage, in order to ensure that adequate protective measures were taken to prevent their escape. In contrast, applying this test to the second defendant, he would not have been found liable. Even though 1) it is reasonably foreseeable that transporting nuclear waste might lead to a high risk of harm materializing, 2) a reasonable man in the defendant’s shoes would have taken such a risk anyway, because of the extensive precautions that had been taken and the fact that he is unlikely to have had any way of knowing about a fault in the tracks. Finally, a corollary of the ‘reasonable man’ element in the second limb of the test is also that a defendant will not liable for any damage that it would have been unreasonable to expect him to know 48 John Murphy (n 46), 660. 49 An example borrowed from Ken Kress, ‘The Seriousness of Harm Thesis for Abnormally Dangerous Activities’ in David G. Owen (ed), Philosophical Foundations of Tort Law (Oxford University Press 1995) 281. OXFORD UNIVERSITY UNDERGRADUATE LAW JOURNAL 90 about (e.g. damage arising due to unforeseen occurrences). By way of yet another example, consider a defendant engaging in an ‘ultra-hazardous’ activity such as using dynamite to blast a tunnel through a mountain (this carries with it the risk of landslides). As long as he has taken adequate precautions such as putting up warning signs well in advance and scanning the immediate vicinity to make sure there are no civilians present, he would not be liable if some hikers disregarded his warnings, set up a camp near the mountain at the last minute and consequently were injured or killed by the landslide resulting from the explosion. After all, it is plausible that a reasonable man in the defendant’s shoes would have gone ahead with the activity anyway, as all necessary precautions were taken and there was no way to know about the presence of the hikers. 5. Conclusion This article has proposed an alternative framework that analyzes Rylands v Fletcher liability for ‘ultra- hazardous’ activities in a way that seeks to tread the middle ground between the doctrinal extremes of the American approach and the Australian approach. Although untested and probably far from perfect, it is hoped that this framework is nevertheless a useful contribution to the existing debate on Rylands v Fletcher, because it endeavours to apply an ‘approximately right’ calibrated approach to a problem that has thus far been vulnerable to the ‘precisely wrong’ broad-brush approaches which swing to the doctrinal extremes of either imposing a general regime of strict liability, or subsuming the rule in Rylands v Fletcher entirely into the law of negligence.