(p. 560) 19. Actions under the rule of Rylands v Fletcher
Read this problem question carefully, and keep it in mind while you are working through the chapter that follows. At the end of the chapter, you will be able to apply what you have learnt to the problem question and advise the relevant parties.
Grab-and-Buy supermarket owns land on which it has built a huge two-storey metal-framed customer car park. One day, after extremely stormy weather with strong winds and heavy rain, the top level of the car park buckles; some of the metal railing breaks free and falls onto the neighbouring petrol station owned by Low-Price-Pumps. The impact damages the pumps and injures one of Low-Price-Pumps’ customers. Furthermore, water that had collected on the upper level of the car park due to an inadequate drainage system pours on to Low-Price-Pumps, flooding the forecourt of the petrol station. The station has had to close for two days, causing £10,000 loss of profit.
Low-Price-Pumps spends £50,000 having the forecourt cleaned and making safe the pumps. Grab-and-Buy argues that damage to the pumps caused by high winds is something that Low-Price-Pumps could and should have insured against.
Advise the parties.
(p. 561) 19.1 Introduction
Consider the following examples:
➔ Townbury Council’s outdoor swimming pool has to be closed after an adjacent ornamental lake bursts its banks after heavy rainfall and lake water floods into the pool, making it unsafe.
➔ Townbury’s water supply is contaminated by chemicals which make it unsafe to drink. The contamination is traced back to historical spillage of chemicals onto the floor of a local factory. The chemicals concerned are no longer used in manufacturing, having been banned by a European Directive in 1995.
➔ A gas pipe owned by Racksco and supplying gas for domestic use is exposed after a tap is left on accidentally by an old lady living in a warden-assisted flat provided for her by Townbury Council. Over time, water seeps from her premises to an embankment supporting the pipe, causing it to weaken and eventually crumble.
➔ A large pile of wood and other materials is collected on land belonging to Townbury Fire Service, in preparation for its annual bonfire party on 5 November, to which all citizens of the town are invited. A group of youths is caught trying to set the pile of wood alight, and fire officers dampen the wood down until they are ready to light it themselves. However, wind transports some still smouldering pieces of wood to the neighbouring property, where they burn down a garden shed.
We have already considered the way in which trespass, private nuisance and, to a lesser extent, public nuisance protect people’s ability to exercise their rights to enjoy land without undue or unreasonable interference. In this chapter we consider a particular cause of action, a variation on the nuisance theme—the rule from Rylands v Fletcher . This protects an occupier against interference due to an isolated escape from (as opposed to an ongoing interference with) neighbouring land. This action has been shaped over time by three key cases: Rylands v Fletcher, Cambridge Water Co Ltd v Eastern Counties Leather plc  and Transco v Stockport Metropolitan Borough Council .
The Rylands rule is therefore another method for regulating the use of land. It emerged out of the Industrial Revolution’s widespread impact on society, beginning in the eighteenth century, changing the face of the country by the mid-nineteenth century—to far less agricultural and increasingly industrial. With industry came increased risk of harm from things such as water, noxious chemicals and so on ‘escaping’ from factories, reservoirs and the like. Rylands was a direct legal response to some of these risks. It created so-called ‘strict liability’ for such escapes—that is, liability without fault.1 Owners of factories, reservoirs and so on would be liable even if they (p. 562) had taken all reasonable care to ensure that escapes did not occur. As in the previous chapter, we will also look here at the contribution the Rylands rule makes, or could make, to the protection of the environment—given its very nature it perhaps has a clearer role to play in this respect than nuisance. However, we will see not only that the rule has limited use in the modern day, but also that the strictness of liability under it appears to have become somewhat diluted.
19.2 The rule in Rylands v Fletcher
The rule in Rylands holds that where there has been an escape of a dangerous thing in the course of a non-natural use of land, the occupier of that land is liable for the damage to another caused as a result of the escape, irrespective of fault. A strict liability rule seems somewhat incongruous against the increasing incidence of fault-based considerations in modern nuisance cases.2 That said, liability under the rule, while strict, is not absolute. There are defences to the action and latterly the courts have decided that defendants are to be liable only for the foreseeable consequences of escapes. In fact, until relatively recently, the rule was largely considered to be of little real significance—a relic of interesting legal history. As we shall see, the action has been rarely used. Though the decision over 100 years later in Cambridge Water reawakened interest (particularly academic), any enthusiastic revival of the rule might seem overly ambitious given that Cambridge Water actually appears to further limit the scope of Rylands actions and arguably, therefore, its usefulness. In Transco, Lord Hoffmann remarked that ‘it is perhaps not surprising that counsel could not find a reported case since the Second World War in which anyone had succeeded in a claim under the rule’ (at ). This seems unlikely to change following further limitations from Transco.
The defendant mill owner employed independent contractors to build a reservoir on his land to provide water for his mill. During construction, the contractors discovered disused shafts from an old coal mine on the defendant’s land, which they assumed had been blocked up. However, when the reservoir was filled, the water in it burst through the old shafts, flooding the claimant’s operating mine. The claimant sought compensation from the mill owner for the damage caused.
The independent contractors had clearly been negligent in failing to ensure that the mineshafts were properly blocked up, but the claimant’s action was against his neighbour and he had not (and could not reasonably have) known of the existence of the shafts on his land and so could not be shown to be negligent (nor was he vicariously liable for the contractors’ negligence). The court also doubted whether an isolated escape of something from one area of (p. 563) land to another (as opposed to an ongoing emanation) could found an action in nuisance. Nevertheless, the claim succeeded.
Although the case went to the House of Lords, it was Blackburn J in the Court of Exchequer Chamber (Fletcher v Rylands and another ) who has long been considered to have given the classic statement of principle, finding that:
the true rule of law is that the person who for his own purposes brings on his land and collects and keeps there anything likely to do mischief if it escapes, must keep it in at his peril, and, if he does not do so, is prima facie answerable for all the damage which is the natural consequence of its escape.
The House of Lords approved this principle (at 339–40), Lord Cairns adding the stipulation that the defendant must also be engaged in a ‘non-natural’ use of land (at 338).
In Rylands, the defendant was strictly liable for all the damage caused by the escape of the water from the reservoir, even though he was in no way to blame for it.3 While this may not, at first glance, seem very ‘fair’, like many cases it is a product of its time. It was decided in a climate of rising social and political concern about the dangers of reservoirs and other processes of industrialisation at a time when another serious reservoir incident had recently occurred.4 This may have persuaded the courts that strict liability in such circumstances was necessary in order to ensure ‘social justice’ in the sense of protecting people against the ravages of industrialisation.5
There are four requirements that claimants must establish to succeed under the rule. The first three derive from Rylands itself; the first two from Blackburn J’s often-quoted statement in the Court of Exchequer and the third added by Lord Cairns in the House of Lords. The fourth requirement was not added until much later and comes from the leading opinion of Lord Goff in Cambridge Water.
(1) The defendant brings on his land for his own purposes something likely to do mischief …
(2) … if it escapes …
(3) … which represents a non-natural use of land …
(4) and which causes foreseeable damage of the relevant type.
We will look at each of the requirements in turn.
(p. 564) 19.2.1 ‘The defendant brings on his land for his own purposes something likely to do mischief …’
This requires a voluntary act of bringing something on the land, and is perhaps the natural result of concern about industrial practices. Blackburn J justified his rule by explaining that it could deal with the effect of industry on those who had no control over it, saying:
the person whose grass or corn is eaten down by the escaping cattle of his neighbour, or whose mine is flooded by the water from his neighbour’s reservoir, or whose cellar is invaded by the filth of his neighbour’s privy, or whose habitation is made unhealthy by the fumes and noisome vapours of his neighbour’s alkali works, is damnified without any fault of his own. (at 280)
The interesting question, however, is what constitutes things that can be brought onto land that are ‘likely to do mischief’—though ‘brought onto’ in this sense probably means ‘kept on’. Notably, most of the things the cases have been concerned with would be perfectly harmless in and of themselves, if they were not allowed to escape. In Rylands, water (collected in large volume in one area) was obviously within the ‘likely to do mischief’ category and, since then, other cases have determined that electricity (National Telephone Co v Baker ), gas (Batchellor v Tunbridge Wells Gas Co ), noxious fumes (West v Bristol Tramways Co ) and even a flagpole (Shiffman v Order of St John ) or a part of a fairground ride (Hale v Jennings ) also fit the description. Put simply, the thing or substance brought onto land need not be ultra-hazardous, but must be capable of causing damage—and therefore be dangerous—if it escapes.
This requirement seems to have been reinforced and made harder to satisfy by the House of Lords in Transco.
The defendant owned a block of flats next to a disused railway embankment, which it also owned. A pipe carrying water to the flats from the water main leaked and a large quantity of water travelled through a crack in the ground to the embankment, over time saturating it and causing it to collapse. There was no evidence that the leak was due to any carelessness on the part of the defendant. When the embankment collapsed it exposed and left unsupported a high-pressure gas pipeline owned by Transco (formerly part of British Gas), who had to act promptly in order to prevent the pipe fracturing. Their preventative actions cost about £94,000.
The first question for the court was whether the defendant had ‘brought something on to his land likely to do mischief’. In answering, Lord Bingham said:
I do not think that the mischief or danger test should be at all easily satisfied. It must be shown that the defendant has done something which he recognised, or judged by the standards appropriate at the relevant place and time, he ought reasonably to have (p. 565) recognised, as giving rise to an exceptionally high risk of danger or mischief if there should be an escape however unlikely an escape may have been thought of. (at )
This set a very high threshold, which the law lords found was not satisfied in Transco. In addition, Lord Scott stated that Transco’s claim should fail because there had been no ‘escape’ from the defendant’s land (see later)—the water from the leaking pipe had simply moved from one part of the defendant’s land to another; it had not exited the land or gone over any boundary.
So, the test as to what is ‘likely to do mischief if it escapes’ is in fact quite strict. Only when it can be shown that the defendant recognised (or should have recognised) that an exceptionally high risk of danger arose if the thing or substance accumulated on their land was able to escape will there be any possibility of liability arising. This would not change even if the risk of an escape taking place was low.6
Essentially, the ‘likely to do mischief’ test has been transformed by the Transco decision into a foreseeability test. A defendant will not now incur liability if they could not have foreseen an exceptionally high risk of danger should the thing brought onto their land escape. As we indicated in our discussion of private nuisance (section 220.127.116.11), foreseeability, as a requirement in land torts (traditionally understood to be distinct from the tort of negligence and stemming from the trespass torts), seems rather misplaced—it hints at there being a requirement of fault. This is particularly troubling in relation to Rylands liability, which is supposed to be strict. So, it should not matter whether the defendant could or should have ‘recognised’ (foreseen) that a thing would give rise to an exceptionally high risk of danger if it escaped.
19.2.2 ‘… if it escapes …’
The Rylands rule was established to deal with isolated escapes from land and therefore proof of an actual escape of a ‘thing likely to do mischief’ is vital to a claim. In Read v Lyons , a munitions inspector was injured while visiting a munitions factory when an artillery shell exploded during the manufacturing process. As there was no indication that the manufacturers had been negligent, the claimant brought an action in Rylands. The House of Lords rejected her claim. There had been no ‘escape’—the shell had not left the defendant’s premises. Crucially, the claimant was still on the property when she was injured: had she stepped over the threshold, she would have succeeded. (p. 566) The law lords held that an escape occurs only when the substance or item causing damage actually moves from the defendant’s premises to a place outside the defendant’s occupation or control. This was most recently confirmed by the Court of Appeal in Stannard (t/a Wyvern Tyres) v Gore . Tyres stored on the defendant’s land caught fire, subsequently destroying both the claimant’s and the defendant’s premises. While agreeing that it would be possible for Rylands liability to be based on fire, the court found that this would be rare, as it would need to be the fire itself that was brought onto land and then escaped. As the tyres themselves had not escaped, there could be no liability.7
Lord Macmillan held (obiter) in Read v Lyons that he was not prepared to allow personal injury claims under Rylands. He said that ‘as the law now stands an allegation of negligence is in general essential to the relevancy of an action of reparation for personal injuries’ (at 16–17).8 Do you agree with this? Should liability be confined to property damage or do you think that when people keep ‘dangerous’ things or substances on their land which injure someone if they escape, they should pay compensation?
There has been some debate as to whether intentionally releasing something from one’s land is capable of being viewed as an ‘escape’ and therefore falling within the Rylands sphere of liability. In Rigby v Chief Constable of Northamptonshire , police officers, in an attempt to catch a criminal hiding in a shop, fired CS gas into the premises to ‘flush’ the man out. This resulted in a fire which damaged the shop. The court held that where direct harm had been caused by an intentional act, trespass would be the correct action. However, in Crown River Cruises Ltd v Kimbolton Fireworks Ltd , Potter J suggested that intentional releases (in this case fireworks) could be deemed to be ‘escapes’ if not deliberately aimed in the direction of the claimant or their property.
19.2.3 ‘… which represents a non-natural use of land … ’
As we have seen, the third requirement that there be a ‘non-natural’ use of land was added by Lord Cairns in the House of Lords in Rylands. This was his interpretation of the original formulation from Blackburn J in the lower court, who referred to the defendant bringing onto or accumulating on the property something ‘which was not naturally there’. This means that the rule does not apply to things that are naturally to be found on a particular area of land and subsequently escape, causing damage. However, what is naturally on land—or what is ‘non-natural use’—has been subject to various interpretations over time.
(p. 567) It does appear clear, for example, that no liability can arise in respect of trees, shrubs or other plants that are naturally found on the defendant’s land, even if some part of these ‘escapes’ onto the claimant’s property (Giles v Walker ). It might be different, however, if the trees or plants in question were introduced by the defendant. The classic definition of ‘non-natural use’ is found in Rickards v Lothian :
It is not every use to which land is put that brings into play [the] principle. It must be some special use bringing with it increased danger to others, and must not merely be the ordinary use of land or such use as is proper for the general benefit of the community. (Lord Moulton at 280)
In Rickards, a tap was left running by an unknown party in the part of a building leased by the defendant. This caused a flood, damaging the claimant’s goods stored on the floor below. Despite there being an escape, no liability was found as the defendant was found to have been using the premises in an ordinary way.
‘Non-natural’ use appears to have become ‘non-ordinary’ use—a much narrower definition, as many things that are not ‘natural’ on land will in fact be quite ordinary. Further, what is ‘ordinary’ will depend on the time, place and context of the use of the land in question. This definition begins to make the test look a little like negligence (or like the reasonable user test in nuisance) and has allowed the courts to decide that various industrial activities are in fact ‘natural’ uses of land. This not only goes against the rule’s strict liability origins but would seemingly water down any impact that the rule from Rylands could have on environmental protection.
In British Celanese v Hunt , strips of metal escaped from the factory where the defendants manufactured electrical components. These hit overhead power lines, causing a power failure to the claimant’s factory. However, because the area in which both factories were sited was given over to an industrial estate, the defendants escaped liability under Rylands (although they were on other grounds found liable in both negligence and nuisance). The substance in question was certainly not ‘naturally’ on the land: it had been brought there by the defendants and had escaped, causing harm—but the use of the land was ‘ordinary’ in the context of industry, as well as being beneficial to the community (in the sense of providing a necessary commodity, creating employment, etc). Similarly, in Read v Lyons, the manufacture of artillery shells in wartime was found to be a natural use of land given the political/social context (although it should be noted that the Read decision was later doubted in Transco). The most recent interpretations of ‘non-natural’ use come from the Court of Appeal. In Stannard (t/a Wyvern Tyres) v Gore the storage of around 3,000 vehicle tyres on land was deemed natural use, despite them igniting and the fire escaping, destroying the neighbouring premises, and even though the capacity of a typical storage facility was exceeded and the stacking of the tyres was done ‘haphazardly and untidily’.9 In Northumbrian Water Ltd v McAlpine Ltd , in the context of large quantities of wet concrete escaping into a water (p. 568) drain (later setting, causing extensive damage), Moore-Bick LJ stated that ‘redevelopment of land in an urban setting cannot in my view be regarded as other than normal and reasonable, unless it involves the use of unusual methods of working’ (at ).
The definition of ‘ordinary’ use of land will inevitably change with time (for example, keeping a car was ‘non-natural’ in 1919, but would not be regarded as so today) and context. But having such a broadly interpretable rule allows the courts considerable flexibility when deciding whether to apply the rule in a new case. Overall, in our view, this had a negative effect on the use of the rule in Rylands as it means that a claimant is far less likely to succeed in any claim taken against general industrialised activity, which, of course, are the very activities likely to do the most damage (and the very activities from which the strict liability principle arose). There is no real explanation why the courts felt that this change of direction was necessary. Perhaps it was to limit the number of claims in a similar way to the duty of care in negligence, which may in turn be a consequence of changed emphasis in the law of tort and whose interests it is used to protect.
19.3 Cambridge Water v Eastern Counties Leather plc
A major case that substantially changed the meaning of ‘non-natural use’ (or ‘non-ordinary use’) of land was Cambridge Water. In the background to this case sit a number of common law decisions (such as Ballard v Tomlinson ) which recognised that downstream landowners hold a ‘natural right’ to receive water in its ‘natural quality and quantity’.10
The defendants owned and operated a tannery and, in the process of degreasing leather pelts, used a particular chemical solvent—perchloroethene (PCE)—for many years, ceasing its use in 1976. In the tanning process, large quantities of PCE had been spilled onto the concrete floor of the tannery. Over time, this seeped through the concrete and into the soil below, from where it ‘travelled’, eventually entering the watercourse and polluting a well situated 1.3 miles away and owned by the claimants for extracting drinking water for the city of Cambridge. The PCE pollution in the water was only discovered in 1983 when the water company became obliged by an EC Directive to test for certain chemicals, including PCE. Cambridge Water was forced to move its well, at a cost of more than £1 million, and sued the defendants for the costs associated with the relocation in negligence, nuisance and under the rule in Rylands.
(p. 569) By the time the case reached the House of Lords, the claims in negligence and nuisance had failed, largely because the harm that had occurred had been unforeseeable to the defendants. This left only the Rylands action. Giving the leading opinion in the House of Lords, Lord Goff felt bound to say that ‘the storage of substantial quantities of chemicals on industrial premises should be regarded as an almost classic case of non-natural use’ (at 309), notwithstanding any benefit to the community that is served by the industry or the fact that this might be seen (especially in a tannery) to be an ‘ordinary use’ of the land in question. He also found that the PCE entering the watercourse could be viewed as an isolated escape, despite the fact it had happened over a period of time and was caused by many separate spillages. However, despite these liberal interpretations, the claimants could not recover because the damage was too remote.
19.3.1 ‘… and which causes foreseeable damage of the relevant type’
Lord Goff’s view on the ‘escape’ and the issue of ‘non-natural’ use of land appears to contradict the direction taken by the courts in the mid-twentieth century. While this may be viewed as a good thing, in that courts would subsequently be more able and therefore more likely to find ‘non-natural’ uses of land (a triumph for environmentalists),11 the reason for the changed approach shows that this is not indeed the case. As we mentioned earlier, a fourth requirement to the Rylands test was added in Cambridge Water—what prevented the claimant’s recovery of damages was the fact that the harm suffered was regarded as too remote. Lord Goff added the requirement that any damage caused by the escape must be foreseeable and, in considering this, held that it was not foreseeable even to the skilled person that quantities of chemical spilt on concrete would cause damage to the claimant’s water. Although some damage from the spillage of the PCE might be imagined, the damage to the claimant’s well could not be, particularly given that the EC Directive had not been in force.
Lord Goff justified his findings by analogy to nuisance (which, as discussed in section 18.104.22.168, also has a foreseeability requirement) and by allusion to Blackburn J’s original formulation in Rylands, in which he referred to ‘anything likely to do mischief if it escapes’. In fact, it is said that as a student at Oxford Lord Goff had read and been very impressed by an article by F H Newark which argued against the separation of Rylands liability and private nuisance.12 The Cambridge Water situation provided him with an opportunity to put this into practice. He closed the gap between Rylands and nuisance in two ways. First, by equating non-natural use with reasonable use he essentially established the rule as an application of nuisance to one-off escapes (a position later confirmed by the House of Lords in Transco). Secondly, he introduced the remoteness principle from The Wagon Mound (No 1) .13
(p. 570) As we have already indicated, ‘non-natural use’ had come to be interpreted so as to limit liability under Rylands where the courts felt it necessary or appropriate to do so. However, Lord Goff recognised that the addition of the foreseeability requirement would sufficiently limit the scope of liability, so there remained no need to do so with the ‘ordinary use’ test.
Why do you think the courts felt it necessary to try and rein in liability under the rule in Rylands? If the objective of the rule was originally to protect claimants (or society as a whole) from the negative environmental and other effects of industry, do such limitations seem appropriate? And for whose benefit were these limitations being put in place?
There has been some discussion as to how far the foreseeability test goes and about what exactly must be foreseeable—the harm alone or also the escape in the first place? Lord Goff’s opinion is not entirely clear on these points (perhaps a reflection of his priorities at the time), but the commonly held view is that it is not the escape that must be foreseeable, merely the damage resulting from the escape. As Conaghan & Mansell point out:
authority for this position can be derived from the Rylands decision itself: Blackburn J’s allusion, for example, to ‘anything likely to do mischief’ suggests that the defendant must reasonably recognise the dangerous nature of the substance under his control but, having done so, he ‘must keep it at his peril’, implying liability regardless of whether or not he could reasonably have foreseen and/or guarded against an escape. (p 144)
Therefore, the correct question is: once there has been an escape, is the damage suffered by the claimant of a type or kind that was a reasonably foreseeable consequence of such an escape?14
Lord Goff also considered the ongoing position. There was still clearly PCE pollution affecting the particular well—once the testing of the water had included tests for PCE post-EC Directive, the problem was known. On this point, Cambridge Water argued that, even though the damage (pollution by PCE) may have been unforeseeable when it was spilt, it had clearly become foreseeable by the time the case reached the court, and the PCE was still escaping. Could the defendant therefore be liable for continuing escapes? Lord Goff dismissed this argument, holding that because the defendants could not possibly have foreseen the fact that the well would become polluted at the date of the spillages, strict liability for the damage (which became foreseeable at a later date, but by which time nothing could be done to prevent the escapes continuing) should not be imposed upon them. The PCE had escaped beyond the defendants’ control (even though it would have been possible—but expensive—for them to clean up the water).15 (p. 571)
Before leaving Cambridge Water, consider the following passage from Conaghan & Mansell:
the interesting question is not whether Cambridge Water was correctly decided but rather why the Law Lords chose to decide the case as they did. To some extent, the result can be attributed to good housekeeping instincts—Lord Goff considered that the application of a foreseeability criterion to the determination of liability in Rylands would lead to ‘a more coherent body of common law principles’ (at 76) … More generally, however, it is instructive to consider the broader social and political context within which Cambridge Water was decided. [Eastern Counties Leather] is judicially perceived as a model company which excites the admiration of right-thinking people. Mann LJ notes that it is of ‘high repute’ locally, taking a proper ‘pride in its history’ as a business which was first incorporated in 1879 (at 56). Lord Goff commends the company’s ‘good standard of housekeeping’ and its ‘modern and spacious’ accommodation (at 63). Into this idyllic scene of industrial industriousness comes a European regulation which, for no demonstrable health reason, renders water from the Sawston borehole no longer ‘wholesome’ and thereby threatens the whole future of this well-established business with a damages bill of almost £1,000,000. Anyone could be forgiven for thinking that this offends justice and common sense, particularly in circumstances where the alleged polluters couldn’t possibly have known that their activities would have such expensive consequences … However the effect of placing the cost of pollution on Cambridge Water is equally disquieting—they will inevitably recoup their loss through raised water rates, but in what sense is it more just to require the residents of Cambridge to foot the bill? (pp 146–7)
In economic and environmental terms, then, the Cambridge Water decision favours industry over society (and distributive rather than corrective justice); the losses or costs are spread among the citizens who pay for their water. The question is whether this is a better solution than one requiring the polluter to pay. An environmental perspective may reflect an entirely different outcome—making the polluter pay is surely preferable to making society pay for both the environmental and social cost of industry, and may operate as a deterrent against future polluting practices. Lord Goff considered the environmental question—and whether this should be a policy consideration that weighed heavily against Eastern Counties Leather—but thought that the creation of mechanisms for protecting society from pollution was more properly the role of Parliament, not the judiciary through the development of one small rule in tort. Put simply, he believed that pollution should be regulated by public, not private, law.
It is also questionable just how much ‘development’ of the tort the imposition of liability on Eastern Counties Leather would involve. In Rylands, the defendant landowner could not possibly have foreseen that one potential consequence of his erection of a reservoir would be the flooding of his neighbour’s mine. The point was that despite the unforeseeable nature of the harm, he should be held strictly liable for it happening, as by building the reservoir he had assumed the risk of all harms that might be caused by the water escaping, whether foreseeable to him or not. Why should this be any different when we are faced with a modern ‘model company’?
Do you think that the Cambridge Water decision was correct? Could it have been decided differently? Remember that the House of Lords found major disagreements on some points thought to be determinative by the Court of Appeal (which had found in favour of Cambridge Water). Does this mean the House of Lords was necessarily right and the Court of Appeal was necessarily wrong? You may like to take into consideration the fact that although PCE was not known to have any deleterious health consequences to humans, it had, by the time Cambridge Water was decided, been found to cause liver cancer in mice.
Also consider what the damages claim in Cambridge Water was for. The costs incurred related to the finding of a new site for a water borehole and the sinking of a new well. What this means is that the old well was never cleaned up. Should there have been an obligation on one of the parties to clean up the PCE from the old well? If so, which party, and why?
For more information go to the Online Resource Centre.
19.4 Transco v Stockport MBC
The final requirement of Rylands comes from the House of Lords in Transco, which approved and expanded upon Lord Goff’s opinion in Cambridge Water. Dealing with what could be considered to be ‘non-natural’ land use, the law lords found that the provision of a piped water supply from the mains to a block of flats was a totally natural use of land. Transco now makes it even harder for claimants to establish a defendant’s liability. Not only will the ‘mischief test’ be harder to satisfy, the ‘non-natural use’ test will also be. The House of Lords noted the link between these two concepts and all five law lords thought that a correct redefinition or re-interpretation of ‘non-natural use’ should require there to be some use that is extraordinary and unusual according to the standards of the day. Lord Bingham suggested the question should be whether the use of land was ‘extraordinary and unusual’ (at ); Lord Walker agreed, also adding the word ‘special’ to the requirement (at ). Piping water to a block of flats in an urban area clearly does not fit this definition, therefore there would have been no liability in Transco on this point either.16 Typically, Lord Hoffmann reverted to economic analysis, suggesting that the role of insurance in cases of this type was also relevant, saying that ‘a useful guide in deciding whether the risk has been created by a “non-natural” user of land is to ask whether the damage which eventuated was something against which the occupier could reasonably be expected to have insured himself’ (at ). (p. 573)
Should it matter whether the party was—or is considered by the court to ought to have been—insured? The question, when it comes down to economic analysis, is about which party is better placed to bear the costs. If Transco had to bear the cost, it could then redistribute this cost by increasing the prices paid by all its customers. Similarly, if Stockport MBC had to pay, an increase in council tax for local residents would cover the cost. Similar points arise if we consider payment ‘in advance’, via the payment of insurance premiums. Do you think it should be councils who insure themselves against eventualities of this type, or the companies who provide the services (for example, gas), usually in the interests of making profit? Or, more accurately, should citizens in a particular area bear the costs, or the customers of a profit-making enterprise?
Lord Hoffmann seems to suggest that if an occupier of land could have foreseen such an eventuality, they should (a) have insurance and (b) the use of land would be classified as ‘ordinary’ so the Rylands test would not be satisfied. This puts the insurance burden firmly upon occupiers (potential claimants). Put simply, on this view a claimant cannot say that they thought that there might be a risk of harm occurring, but they chose not to insure, and then go to court seeking damages under the rule in Rylands. Lord Hoffmann’s approach, however, was not favoured by the other law lords and was in fact explicitly rejected by Lord Hobhouse (at ).
19.5 Standing and defences
19.5.1 Who can sue?
Cambridge Water seemed to move the Rylands rule closer to nuisance, with the addition of the foreseeability requirement. Indeed, this was Lord Goff’s intention. Only a few years later, the House of Lords’ decision in Hunter v Canary Wharf Ltd  made having a proprietary interest in land a prerequisite to a claim in nuisance. This meant that the question arose as to whether this would extend to those taking claims in Rylands. It is clear that others have successfully claimed under the rule in the past (Perry v Kendrick’s Transport Ltd ), though this was criticised in Read v Lyons. However, if the focus of both actions is the same, namely the protection of rights to and over land, as Lord Goff suggested, it would seem logical that the right to be able to take a claim would be the same for each. This position was confirmed by the House of Lords in Transco;17 therefore only parties with ownership or exclusive possession of the land concerned have standing to claim under Rylands. (p. 574)
As we indicated earlier, following Transco it is also no longer possible to claim for personal injury under the rule in Rylands (in the same way that it is not technically possible in nuisance).18 This limitation has, however, been criticised by some scholars as being at odds with the House of Lords’ belief that there may be ‘a category of case, however small … in which it seems just to impose liability even in the absence of fault’ (Lord Bingham, Transco at ).
Roderick Bagshaw argues that it is strange to limit the rule to damage to real property especially given that one of the reasons for the rule in the first place was as a response to a loss of life in a number of disasters. For example, Lord Walker in his speech in Transco refers to the Aberfan tragedy in which 144 people (including 116 school children) were killed when on 21 October 1966 thousands of tons of colliery waste from the Merthyr Vale Colliery slid down a mountain side above the village of Aberfan in south Wales, destroying 20 houses and the village school. To suggest (in the absence of negligence) there would only have been good reason to compensate the property owners and not for the deaths and personal injuries of those involved seems extraordinary. This, however, is the result of subsuming Rylands within nuisance.
For more information go to the Online Resource Centre.
It also appears that other losses are excluded from the rule, now it has been subsumed into private nuisance and (see further later) more closely aligned with negligence. In D Pride & Partners (A Firm) v Institute for Animal Health ,19 for example, Tugendhat J excluded claims for pure economic loss from both nuisance and Rylands actions.
The initial defences available under the rule in Rylands come from Blackburn J’s original judgment. He indicated that there should be only three defences to the (strict) liability. These are outlined in Table 19.1.
Table 19.1 Defences to a claim under the rule in Rylands v Fletcher
Fault of the claimant or express or implied consent
Escape caused by the unforeseeable act of a stranger
Escape caused by an ‘act of God’
The defendant will not be liable where an escape happens solely because of natural causes, in circumstances where no human foresight or prudence could reasonably recognise the possibility of such an occurrence and take preventative measures. Arguably, however, modern technological and scientific advances render this defence largely defunct. Do you think that exceptionally heavy rains or violent winds would be unforeseeable today? What kind of exceptional situations would now be considered (after Greenock) an ‘act of God’?
Blackburn J’s defences are not the only ones applicable to Rylands claims, others more familiar from nuisance can also arise. Statutory authority is the primary example and the approach to this is the same as is taken in nuisance.20 Contributory negligence and volenti non fit injuria are also applicable.21 The fact that defences are available at all is somewhat controversial—they essentially suggest that an element of fault resides in an area of tort that is supposed to have created strict liability. This, in our opinion, weakens the action’s potential use in environmental protection. The subsuming of the action into nuisance has also created doubt about its usefulness in the modern day and, as we shall see in the following section, it appears that to some extent any residual application of Rylands is edging closer and closer in some circumstances to a fault-based land tort.
19.6 The nuisance/Rylands v Fletcher/negligence overlap
Lord Walker, in Transco, suggested that liability in nuisance (and presumably, therefore, under Rylands) ‘overlaps with (indeed, is a sort of condominium with) that of negligence’ (at ). A series of cases supports this—that is, cases in which there is clearly fault on the part of the defendant, but where the harm is to the claimant’s interest in land and liability has been found in nuisance or Rylands.22 These cases started within a category known as ‘continuing or adopting a nuisance’, and refer to those situations where a pre-existing nuisance (created usually by a third party) was allowed to continue by the defendant, sometimes resulting in the ‘escape’ of something from the defendant’s land. Furthermore, in Burnie Port Authority v General Jones Pty Ltd , the Australian High Court expressly incorporated the Rylands rule into negligence. In Cambridge Water a foreseeability requirement—more familiar from negligence and considerations of fault—was added to both claims in private nuisance (p. 577) and under Rylands. Lord Walker, in Transco, thought that the scope of the action had been restricted by developments in negligence (at ). Thus, it seems that questions must now be asked about the role of fault within these types of claim, particularly in those cases that seem, on their facts, to straddle private nuisance and Rylands.
A drainage pipe on the defendants’ land had been constructed by the local authority without the defendants’ knowledge. When it was built, its end should have been covered with a grille or grate to stop debris blocking it and causing flooding. A grate had been provided but was put in the wrong place, rendering it useless. The claimant’s land was flooded and damaged on more than one occasion when the pipe became blocked following heavy rainfall.
The House of Lords held the defendants (an order of monks) liable on the ground that they had ‘continued or adopted’ an existing nuisance. While it was clear that the nuisance itself—that is, the problem—had been created by the local authority in its negligent construction of the pipe, the defendants had done nothing to rectify the problem. Because at least one of the monks was found to have known about the existence of the pipe, and the fact that the problem would have been easy to rectify, it was held that the defendants had not taken reasonable steps to stop the nuisance continuing.
This clearly illustrates what we mean by a claim ‘straddling’ the tort of private nuisance and an action under Rylands. The applicability of a particular action turns on what is categorised as the ‘nuisance’ or problem in the first place. If it is the pipe, this is an ongoing situation (at least, the risk created would be) and would fall in the private nuisance category. However, if it is the flooding of the claimant’s land then it could be viewed that the water, once accumulated behind the blockage in the pipe, had ‘escaped’, making this a Rylands situation. However, it also seems eminently possible that this is, in fact, negligence—the defendants did not act reasonably or failed to take reasonable steps to do something about the grate and the potential problem that the pipe could become blocked. This is clearly a fault-based consideration and shows how the three different tort actions can overlap. Is what is actually being said that there is a duty to do something about a known and existing state of affairs, if it could pose a potential problem?23 If so, this is clearly breached by not doing so, and is what causes the damage complained of, subject to any foreseeability considerations. Conor Gearty argues that ‘[t]his was negligence pure and simple, confused by an ill-fitting and woolly disguise of nuisance’ (1989, at 237).
The harm in Sedleigh-Denfield originated from a third party’s actions, but similar effects arise in situations caused by acts of nature. In Goldman v Hargrave , a gum tree on the defendant’s property was struck by lightning. The defendant took what turned out to be inadequate steps to alleviate the risk that the tree would catch fire. (p. 578) The tree continued to smoulder, eventually igniting, the fire travelling to the neighbouring property and causing damage. The Privy Council followed Sedleigh-Denfield, finding the defendant liable for the damage, based on his knowledge of the foreseeable danger.24 Again, this seems both like an escape (of fire) and a fault-based consideration, aligning it more closely with negligence, and indeed the Privy Council treated it as such, though they refused to state whether the case could also be decided under nuisance.
19.6.1 The ‘measured duty of care’
The cases in this area seem to involve an escape of one sort or another and without the added limitations to the Rylands rule would appear to be almost classic examples of the type of harm that Blackburn J imagined would attract strict liability. In relation to them a new term of art has been coined: a ‘measured duty of care’ in nuisance, in which positive duties are owed by landowners to act to prevent damage to others.25
In Leakey v National Trust , a portion of a large mound of earth on the defendant’s ground collapsed (as exposure to weather conditions had weakened it over time), damaging two houses owned by the claimant which were situated at the bottom of the mound. As the defendants had long been aware of the effect of the weathering on the earth, they were found liable. Here, again, the Court of Appeal explicitly considered the relationship between nuisance and negligence in claims of this type, essentially coming to the point that the two actions were one and the same in these situations. We would add that given the ‘type’ of nuisance being considered, Rylands is subsumed in these claims as well, by virtue of the ‘escapes’ that cause the damage in all of them. The result of Leakey is that a claim can now be taken in nuisance or negligence, but cases since have tended to rely on nuisance, possibly because it is easier to do so than to go back to ‘duty, breach, cause’ formulations. It is also perhaps arguable that where the only harm is to amenity interests, the claim must by its very nature be framed in nuisance and, therefore, all claims of this type might as well be.26 However, at least where there is actual damage to property, the claims seem as though they would equally as easily succeed in negligence if framed in that way (see, for example, Holbeck Hall Hotel Ltd v Scarborough Borough Council ; Lippiatt v South Gloucestershire Council ; Bybrook Barn Centre Ltd v Kent County Council ; Delaware Mansions Ltd v Westminster City Council ; Vernon Knight Associates v Cornwall Council ), yet all are considered nuisance actions. Indeed, in Delaware Mansions, Lord Cooke went as far as to say that the nuisance/negligence distinction in these cases ‘is treated as of no real significance’ (at ).
(p. 579) That said, a more recent case indicates that the scope of the duty may be used to limit claims even where a measured duty of care seems appropriate. In Lambert and others v Barratt Homes Ltd and another , a local authority succeeded in its appeal against a finding that it should have prevented the flooding of the claimant’s land from surface water that accumulated on its own land by constructing a catch pit and drainage system for the water at considerable cost. As the accumulation of the surface water had in fact been caused by Barratt (by blocking a drainage ditch and culvert on land adjacent to the local authority’s)—and because Lambert could recover the entire cost of drainage work from Barratt in nuisance—the scope of the authority’s ‘measured duty’ extended only to cooperation and to allowing others to facilitate the drainage of water from its land.27
19.7 Where does Rylands v Fletcher fit today?
In Cambridge Water, Lord Goff questioned whether the rule from Rylands was best seen as analytically distinct from nuisance or if they should be viewed as two parts of the same thing. Even though the Rylands rule had developed out of a desire to protect landowners from the risks of isolated escapes from industry and other risky activities, he held that ‘it would … lead to a more coherent body of common law principles if the rule were to be regarded as essentially an extension of the law of nuisance to isolated escapes from land’ (at 306). In Transco, the House of Lords confirmed that Rylands actions should be viewed as a sub-species of private nuisance. This, to a large extent, brought to an end a long period of uncertainty and academic speculation on the matter. Their lordships also reviewed the scope, relevance and application of the rule in the modern day and provided guidance as to its future application.
In particular the House of Lords in Transco:
• rejected the suggestion that the rule should be absorbed into negligence as in Australia and Scotland;
• rejected the suggestion that the rule should be more generously applied, confining it to ‘exceptional’ circumstances where the occupier has brought something onto his land which poses an ‘exceptionally high risk’ to neighbouring property should it escape and which amounts to an ‘extraordinary and unusual use of land’ given the place and time in which it occurs; and
(p. 580) • clarified that only those with rights to land could sue, bringing the action into line with private nuisance, following Hunter, and confirming that no claim for personal injury could be brought under it.
Transco limits the rule in Rylands to a role in the protection of interests in land (as in nuisance). In so doing it prevents it from taking on a broader role in ensuring that those who wish to indulge in ultra-hazardous (but socially beneficial) activities make appropriate provision for absorbing the inevitable costs of occasional catastrophes and environmental harms. In this respect, the function or purpose of this area of tort law as a regulatory system—if it ever had one—is lost.
It is clear that the Rylands rule has always been difficult to rationalise, and particularly so in the modern day, where it seems largely unnecessary to hold on to it. It was judicially created in an industrialisation context to deal with perceived dangers caused by large-scale building of reservoirs for industry, making it seem like an exception to nuisance through its dealing with one-off events via the imposition of strict liability. However, by the mid-1990s, as Cambridge Water shows, its impact and potential as a tool for environmental protection was constrained—again, a product of the time, post-Thatcher, but still while the country was under Conservative government and business concerns were prioritised. Furthermore, many of the practices on which it may have had an impact had, by this time, already been regulated in other ways—reservoirs, for example, by then coming under the Reservoirs Act 1975. Sadly, things have not really improved following the reconsideration of the action by the House of Lords in Transco, as to all intents and purposes Rylands is now merely a sub-species of nuisance and, as we argued in the previous section, potentially more accurately viewed as a sub-species of negligence. As a result, its deliberate differences from nuisance, including the ability to recover for personal injury and ‘wrongful’ conduct—as well as most of its environmental potential—have either been lost or become (unintentional) inconsistencies. Bagshaw contends that:
[i]t is often asserted that ‘hard cases make bad law’, but Transco confirms that bad law is also easily made when a case is too straightforward. Perhaps it was because it was so clear to their Lordships that the claim should not succeed (the defendant’s activity being too commonplace, insufficiently risky, there being no escape and the claimant’s loss being easily insurable) that they allowed inconsistencies and controversial propositions to stray into their speeches. But given that very few claims rely on the rule in Rylands v Fletcher and Transco did nothing to encourage greater use of the rule, the main effect of these imperfections will probably be to challenge law students and textbook writers. (2004, at 392)
In this chapter we considered the rule from Rylands, a judicially created action allowing claims against those who allow substances to escape from their land and cause damage to others’ property. This rule was completely a product of the time—there were societal concerns about the effects of mass industrialisation, including, as in the case itself, the building of artificial reservoirs.
(p. 581) Due to its historical context, the rule has been (and can be) used only in limited circumstances and the primary bulk of case law comes from the late nineteenth and early twentieth century. But it should be remembered that the tort of nuisance was already in existence before the Rylands rule was created, and this would allow people to sue for uses of land that interfered with their own interests on a more frequent or ongoing basis. It should also be noted that the tort of negligence emerged as a separate entity in the early twentieth century, and many of the cases under the rule in Rylands could, in some ways, be construed as negligence actions. These days, it is only those situations where a completely accidental escape occurs that could be said to fall outside the bounds of negligence.
The rule today is best understood through a trilogy of cases: Rylands itself, then Cambridge Water in the mid-1990s, followed by Transco early this century. These decisions, all from the House of Lords, have in their own way shaped the way the rule can be used, either by adding new requirements (Cambridge Water), or redefining the way the existing requirements should be interpreted (Transco). That said, there is potential—subject to its now stringent requirements—to use the rule to help to protect the environment from one-off catastrophes, not in a preventative sense but perhaps with more regard to clean-up measures when things go wrong. Whether this was what was intended when the rule was created is debatable—it was more likely intended as a vehicle to compensate (wealthy) landowners than to protect the land itself.
The usefulness of the rule is certainly questionable in the modern day—it appears from Cambridge Water and, also, the nuisance case of Hunter, that its scope (particularly as a tool for environmentalists) has been severely restricted. The rule has moved closer to nuisance (and negligence), with all that that entails, including a foreseeability requirement and a proprietary interest in land in order to be able to claim. In other jurisdictions, the rule has been swallowed by negligence. Without doubt, this is a rule now very limited in scope and application—potentially having more historical interest than practical application in the twenty-first century.
After reading the chapter carefully, try answering the questions which follow. If you would like to know what we think, visit the Online Resource Centre (www.oxfordtextbooks.co.uk/orc/horsey_tort/).
1. How could the action under Rylands be used to help to solve environmental problems?
2. Does the action under Rylands serve any useful purpose in the modern day?
3. Should we follow the Australian High Court and formally view the action under Rylands as a species of negligence?
4. Consider the problem question at the start of this chapter—now having read about the topic what would be your advice to Low-Price-Pumps? If you need some pointers in thinking about how to answer this question, turn to the Appendix where each problem (p. 582) is annotated with issues and cases to consider, or log on to our Online Resource Centre (www.oxfordtextbooks.co.uk/orc/horsey_tort/) and check your ideas against our suggested outline answer.
A good place to start your further reading is the chapter by A W Brian Simpson, which helps to contextualise the Rylands litigation. More up-to-date and insightful interpretations are provided in numerous other articles, particularly by Bagshaw and Nolan.
Amirthalingam, Kumaralingam ‘Rylands Lives’  CLJ 273Find this resource:
Bagshaw, Roderick ‘Rylands Confined’ (2004) 120 LQR 388Find this resource:
Gearty, Conor ‘The Place of Private Nuisance in a Modern Law of Torts’  CLJ 214Find this resource:
Lee, Maria ‘Private Nuisance in the House of Lords: Back to Basics’ (2004) 15 King’s LJ 417Find this resource:
Nolan, Donal ‘The Distinctiveness of Rylands v Fletcher’ (2005) 121 LQR 421Find this resource:
Simpson, A W Brian ‘Bursting Reservoirs and Victorian Tort Law: Rylands and Horrocks v Fletcher (1868)’ in Leading Cases in the Common Law (OUP, 1995)Find this resource:
Wightman, John ‘Liability for Landslips’ (2000) 2 Env L Rev 285Find this resource:
3. So, would the owner of the ornamental lake in the first scenario at the beginning of this chapter be liable to Townbury Council? The question will always be whether the requirements of a claim under Rylands have been met—have they?
4. With severe loss of life. See A W Brian Simpson ‘Legal Liability for Bursting Reservoirs: The Historical Context of Rylands v Fletcher’ (1984) 13 LS 209, 219.
5. Though it is interesting to note that the reservoir bursting in Rylands—essentially an accident, albeit large-scale, causing only property damage—convinced the courts to act to create a strict liability rule. This stands in stark comparison to other ‘large-scale’ accidents and personal tragedies we have seen in tort (e.g. the Hillsborough Stadium disaster, see further, section 5.6), following which the courts have sought to restrict liability.
6. Have you now changed your mind about whether Townbury Council would be able to recover damages from the owner of the ornamental lake? Similarly, would Racksco have any claim since the Transco decision?
7. Note also that the storage of tyres in this case was held to be a ‘natural use’ of land, meaning no liability could have arisen even if there had been an escape. See, however, LMS International Ltd & Ors v Styrene Packaging and Insulation Ltd & Ors , especially –.
9. cf Harooni & Anor v Rustins Ltd .
10. Though note that the decision in Ballard was not about the reasonableness or otherwise of upstream use but was more a policy decision reflecting the fact that people downstream tend to suffer more from pollution.
11. Though the trend may now be back in the other direction: see Stannard (t/a Wyvern Tyres) v Gore and Northumbrian Water Ltd v McAlpine Ltd.
12. F H Newark ‘The Boundaries of Nuisance’ (1949) 65 LQR 480.
14. Might it be argued, therefore, that the factory owners, in the scenarios outlined at the beginning of the chapter would, unlike Eastern Counties Leather, be liable, as such pollution is foreseeable since Cambridge Water itself?
15. Note, then, that the decision goes against the environmental principle that the ‘polluter should pay’.
16. See also Ward LJ in Stannard (t/a Wyvern Tyres) v Gore who found that tyres are not ‘exceptionally dangerous or mischievous’ things and use of land as a tyre-fitting business neither ‘extraordinary’ nor ‘unusual’ (at ), and Moore-Bick LJ in Northumbrian Water Ltd v McAlpine Ltd at .
17. See e.g. Lord Hoffmann (at ), in the context of who is best placed to insure against property damage, Lord Hobhouse (at ) who indicates that ‘other parts of the law of torts’ cover personal injuries and, perhaps most conclusively, Lord Bingham (at ).
18. In the past personal injury claims had been included without question: Hale v Jennings ; Perry v Kendrick’s Transport Ltd .
21. As was indicated in Colour Quest v Total Downstream UK Plc . Steele argues that ‘consent’ in this form must mean something closer to ‘shared benefit’ if it is to be consistent with the purpose of the rule (p 689). See further Chapter 10.
22. Steele states that Rylands ‘provides a direct alternative to negligence in some circumstances’, adding: ‘In terms of long-term survival, that has been its problem’ (p 679).
23. In negligence this would be dealt with as a question of duty regarding the actions of a third party which then cause damage to the claimant—see, for an example of a case with no real discernible difference, Smith v Littlewoods , section 4.5.4.
24. Thus it seems that Townbury Fire Service, in the scenarios outlined earlier, could also be liable.
25. Steele points out that the notion of ‘duty’ is alien to nuisance, but also that the analysis of these cases is not typical of negligence, either, due to their subjective rather than objective standard (p 620).
26. Though as we have seen, to claim for personal injury, a claim would have to be formulated in negligence.
27. cf Vernon Knight Associates v Cornwall Council .