Arscott & Ors v The Coal Authority & Anor

This is an appeal against the decision of Royce J, sitting at Merthyr Tydfil, when on 16 July 2003 he dismissed the appellants’ claims in nuisance brought against the first and second respondents. Permission to appeal was granted by Mantell LJ on 13 October 2003.

The case concerns a serious flood which took place at Aberfan, Merthyr Tydfil, on 22/23 October 1998. That night there was an exceptionally heavy rainfall. The river Taff, which lies to the east of Aberfan and flows from north to south, overtopped its banks. The homes of the 32 appellants were flooded to a depth of approximately one metre. They all suffered substantial damage. The appellants blame the respondents for what happened. That is because colliery waste or spoil had been deposited, at the instance of the second respondent, on a piece of land called Grove Fields belonging to the them. Grove Fields borders the western bank of the river Taff and is roughly to the east/south east of the appellants’ houses. The waste deposit raised the level of Grove Fields by ten to twelve feet. The result was that on 22/23 October 1998 flood water which would otherwise have flowed on to Grove Fields had to find another place to run. It flooded the appellants’ homes.

I should say that the terrible catastrophe which befell Aberfan on 21 October 1966, when part of a coal waste tip above the village engulfed the school killing 116 children and 28 adults, plays its part in the background to the case for reasons which I will explain.

Amongst the issues canvassed between the parties before trial was whether the infilling of Grove Fields in fact caused the appellants’ properties to be flooded. However on the first day of the hearing it was accepted by the respondents that the infilling was indeed a material cause. So the judge did not have to decide that question. He dismissed the claims primarily because he held, relying on a long line of authority going back to the 18^th century, that a doctrine known as the “common enemy” rule applied to the case. The judge articulated this rule as follows: (paragraph 27):

“…an owner or occupier of land is entitled to use or develop his land so as to prevent flood waters coming on to his land. If in times of flood waters which would have entered his land in consequence damage another’s land – that does not provide a cause of action in nuisance.”

A major issue in the appeal is whether the judge was right to apply this rule at all. This issue requires us to consider the nature and scope of the rule as well, of course, as the facts of the case. In the alternative the appellants say that the rule, if it is potentially applicable, ought now to be modified in light of the court’s obligations under the Human Rights Act 1998 (“HRA”) in particular having regard to Article 8 of the European Convention on Human Rights (“ECHR”) and Article 1 of the First Protocol.

The judge dismissed the claim also on an alternative basis, which proceeded on the assumption (contrary to his finding) that the respondents could not take advantage of the common enemy rule. It was and is common ground that reasonable foreseeability of harm is a necessary condition for the recovery of damages in private nuisance. The judge held that it was not reasonably foreseeable (at the time the work was done) that the infilling of Grove Fields might cause flood damage to the appellants’ properties. It is said that on a proper appreciation of the nature of the requirement of reasonable foreseeability this finding was not open to him on the evidence.

Although it was agreed on all hands that there was no issue of “reasonable user” of Grove Fields by the respondents relevant to the claim against them in nuisance, the judge entered into the question whether the infilling amounted to a reasonable user, and held that it did. The appellants say that this finding was irrelevant or, if not irrelevant, wrong.

Lastly, the judge found free-standing reasons for absolving the first respondent, the Coal Authority (“the Authority” – successor for all relevant purposes to the National Coal Board (“the NCB”)) of any liability. However in the course of his submissions Mr Darling QC for the Authority accepted that (leaving the common enemy rule aside) if the judge’s finding on foreseeability could not be sustained, he could not support this further conclusion in his clients’ favour.

Now I must rehearse the facts of the case in greater detail. It is convenient to start with the geography.

As I have said the river Taff lies to the east of Aberfan and flows from north to south. On the other side of the village to the west, the land rises very steeply. That is where the colliery tips had been which collapsed in October 1966. A massive slide of tipped material, mud and water cascaded down on to the village inflicting the terrible disaster which happened then. Afterwards these tips were moved up the mountain to a safe location out of sight and landscaped. The tips whose spoil or waste is involved in this case were in a different place. There were two of them. They were known as the 19^th century tips because they were in the places or areas first used for waste from the Merthyr Vale Colliery, between the years 1869 and 1918. They were situated on flat ground in the centre of Aberfan. One of them, the one lying more to the east, lay immediately to the south of Grove Fields, and (like Grove Fields) abutted the west bank of the Taff. The other lay due west of the first. Aberfan Road ran north – south between the two of them. To the north of Grove Fields, and again abutting the west bank of the Taff, lay an area called the Plantation. In the 1970s it was a sewage farm. Roughly to the west and southwest of the Plantation (and west and northwest of Grove Fields) lay the area where the appellants’ homes were situated. Just as Aberfan Road runs between the two 19^th century coal tips, so also it runs between Grove Fields/The Plantation and the appellants’ properties. All save one of these lie to the immediate west of Aberfan Road.

As I have said there are 32 appellants. Most of them lived in an estate called Pantglas Fawr. The estate had been constructed by the second respondent, the Merthyr Tydfil County Borough Council (“the council”), not long after the 1966 disaster. It was completed in 1968. The appellants living in the Pantglas Fawr houses were either tenants of the council or freeholders after purchase of their houses from the council pursuant to the “right to buy” provisions contained in the Housing Act 1985.

A minority of the appellants lived at other addresses. Two lived at Police Houses which is immediately to the south of Pantglas Fawr. Another – rather a married couple – lived at Cartref Bungalow to the north of Pantglas Fawr. Another couple lived at Ynysygored Cottages. That is the northernmost property with which we are concerned, and unlike the others lies to the east of Aberfan Road.

Next I will describe the material history of the area before the October 1998 flood. That date by no means marked the first flood from the Taff at Aberfan. The right (that is, the west) bank of the Taff upstream of Aberfan was an area of historic flood risk. It seems that the riverbanks had been artificially raised since at least 1903. Sometime before 1960 a concrete flood wall in the right bank of the Taff was constructed around Grove Fields. Still there was a major flood over Grove Fields on 4 December 1960 causing substantial damage. A contemporary newspaper report recorded that the Taff rose to its highest level for 30 years. Between 60 and 70 houses in two streets in Merthyr Vale were flooded. There is some evidence that this flood was more severe than the one of October 1998. The estate at Pantglas Fawr, of course, had not yet been built.

In Part I Section 5 paragraph 35 of the Report of the Tribunal which conducted the inquiry into the 1966 disaster it was stated:

“From 1949 onwards certainly, and possibly from even earlier there were complaints of damage to property due to flooding in the Pantglas area of Aberfan. The main areas affected were the backs of houses in Pantglas Road in the vicinity of Pantglas Farm, especially near Aberfan Road and in the Cottrell Street area. The floodwater was frequently between 2 feet and 2 feet 6 inches in depth and under Black Bridge over the Aberfan Road it reached a depth of 7 feet on one occasion.”

The Pantglas area and Pantglas Road there referred to lay about 300 metres downstream from the Pantglas Fawr estate, which as I have said was constructed after the disaster and completed in 1968. Section 5 of the Report proceeds to give details of some of the many complaints about flooding made by local people. Then there follows this passage, to which Lord Thomas QC for the appellants drew our attention:

“We have set out these examples of complaints made by the inhabitants of Aberfan – and there were very many more which we might have quoted – in order to illustrate the state of mind in the village and to explain why there was so much concentration on the menace of flooding, perhaps to the disregard of the even greater menace which threatened them. We do not claim to understand even today why the Borough Council and the National Coal Board were unable to reach an amicable settlement of what appears to be a simple drainage problem and so avoid a great deal of misery in the village over many years. The former can no more be acquitted of lack of vigilance than the latter can be absolved from grave failure to face up to their substantial responsibility in relation to what was for decades undoubtedly a scandalous state of affairs.”

The importance of this passage is not to rehearse again the Tribunal’s excoriation of these defaults, but because it is a graphic illustration of the existence of long-standing local anxiety about flooding, a problem which had become notorious long before the infilling of Grove Fields with spoil or waste which, in circumstances I shall shortly explain, was done in the 1970s. This dimension of the case is material to the debate about foreseeability. Grove Fields was a frequent and familiar flood location.

After the 1966 disaster there was a political impetus towards the reclamation of derelict land in Wales, which consisted very largely of disused spoil heaps. In November 1966 the Secretary of State for Wales made a speech in the House of Commons in which he announced the establishment of a special unit within the Welsh Office to co-operate with other bodies in carrying forward the work. One of its members was introduced in a meeting of the appropriate committee of the council (or its predecessor) on 22 February 1967.

Later in 1967 there was another flood. Petitions from affected residents were received by the council. Among these, it seems, were some of the early residents of Pantglas Fawr, though there is no evidence that any water entered their homes at this time. (Mr Palmer QC for the council readily conceded that the estate was liable to flooding; his case was that it was not reasonably foreseeable that the infilling of Grove Fields would divert water so as to flood the houses on the estate.) By early 1968 the Glamorgan River Authority (“GRA”), which owed legal responsibilities relating to flood protection, had in mind to raise the embankment where Grove Fields abutted the Taff. That was duly done, the work being completed in around 1970 (I think we lack evidence of the exact date), but in any event before the infilling of Grove Fields was carried out.

The council minutes of meetings after the 1966 disaster show much consideration given to the removal of the tip complex to the west of Aberfan which had been the source of the catastrophe. As I have said these tips were at length moved up the mountain to a safe location out of sight and landscaped. The infilling of Grove Fields with the spoil from the 19^th century tips was of course a different matter. The immediate events leading up to it begin, so far as one can trace the chronology from the documents we have, with a suggestion at a council meeting of 6 January 1968 that Grove Fields was the most suitable site to fulfil an initiative that had been put forward for the provision of playing fields at Aberfan. Plainly this was seen as a kind of regeneration following the disaster; and that is an important dimension in the history. Reference was made at the meeting (I think by the appropriate official reporting to the councillors) to “… the close proximity of tip spoil… preventing flooding etc…”. There followed correspondence with the Welsh Office. At length, in or after August 1968, the Town Clerk reported the Secretary of State’s view which favoured infilling: “[t]he treated land could be used for a variety of purposes and material from the tips could be used to improve the low-lying land nearby, at present subject to flooding”. Further discussions ensued, some involving the NCB. Gradually the scheme was firmed up. Cost estimates were prepared by or in July 1969. There was discussion how far the tips, after “washing”, might yield marketable coal. Contemporary documents show a combined aspiration to prevent flooding and dedicate the land for the community’s use. These ambitions were, I think very obviously, two sides of the same coin.

By August 1971 the scheme was effectively finalised. Certain drainage works on Grove Fields were carried out: previous drainage had become ineffective with time; in 1968 the land had been described as “soft and miry”. Areas were prepared to receive the discard from the tips. A washery was to be constructed. By October 1971 there was a good deal of public pressure to get ahead with the scheme. A planning consent was issued by the council to the NCB to which conditions were attached. I should set out condition 4, in light of one of Lord Thomas’ submissions for the appellants:

“Where serious flooding, erosion or other damage to land within or outside the area so defined in the planning application as a direct result of the removal of the Grove Tips and the spreading of the material [occurs] the necessary remedial measures are to be taken.”

On 14 September 1972 a formal agreement was entered into between the council and the NCB, and the work went ahead. At length it became apparent that there was no economic prospect of reclaiming coal from the tips, so that by 18 June 1973 the operation had become a “muck shifting” exercise. It was completed in 1975. As I have said the level of Grove Fields was raised by ten to twelve feet. New playing fields were provided for, as had been proposed. Changing rooms were built.

In December 1979 there was widespread flooding in South Wales, and the Taff overflowed again. The houses in the Pantglas Fawr estate were flooded. Families had to be rescued with dinghies and boats. We were told that at trial the experts were broadly in agreement that the severity of the flood was much the same as that of 22/23 October 1998; though it was the recollection of at least one resident that in 1979 there had been a greater depth of water, perhaps by as much as six inches. By this time the Welsh Water Authority (“WWA”) had succeeded to the responsibilities of the GRA. In consequence of the flood they erected new flood defences downstream of Grove Fields. It seems clear that there was no perception or suspicion, certainly none recorded, that the infilling of Grove Fields a few years earlier had been a cause of the flooding in 1979 of the houses in the Pantglas Fawr estate. The residents of the estate appear to have attributed the flood of that year to run off from the mountainside. In fact the experts in due course instructed for the purpose of this litigation, to whom I shall refer shortly, were to agree that the 1979 flood was caused in the same way as that of 1998.

I have already given a thumbnail sketch of the flood of October 1998. The Taff overtopped its banks to the north of Grove Fields. The scope and area of the flood were much as they had been in 1979. The flood defences erected following the 1979 flood proved ineffective. There was some evidence that the point where the overflow occurred was a “low spot” in the embankment, occasioned by some failure of survey or design; but nothing, I think, turns on that. At all events the Plantation, Pantglas Fawr, the Police Houses and other areas to the north (including Ynysygored Cottages) and west of Grove Fields were inundated. Grove Fields, with its heightened surface, was not. As I have said, the homes of the appellants were flooded to a depth of approximately one metre. Some were in poor health. There were rescue operations in the middle of the night. The clean up operation took a long time. Not all of the appellants were insured. It takes no imagination to picture the hardships suffered.

By now, the Environment Agency (“EA”) had succeeded the WWA which as I have said had earlier succeeded the GRA. They instructed consulting engineers and the flood was investigated. At length it was decided to raise the embankment upstream of Grove Fields. There were temporary works in late 1998. In late 2000 the embankment was raised permanently, and linked with the raised level of Grove Fields. This would protect the Plantation and Pantglas Fawr from flooding, so it was and is envisaged, to a standard measured in terms of time of 1:100: that is, a risk of one flood in 100 years.

In due course experts were instructed by the parties to the litigation: Dr Samuels for the appellants, Mr Philpott for the Authority, and Mr Frith for the council. They prepared and exchanged reports as to the cause of the 1998 flood. They deployed data from an earlier report by SRK Consulting, which had been commissioned by the EA. Each had additional data also. There was a considerable measure of agreement: for example as to the historic flood risk, and the fact of overflow from the Taff as the principal source of flooding. As I have said, on the first day of the trial it was accepted that the infilling of Grove Fields was indeed a material cause of the flooding of the appellants’ properties. That had by no means been clear from the start. Before the concession, causation looked like being a major – perhaps the major – issue in the case. It is I think plain, not least from the experts’ Agreed Review prepared after meetings on 9 April 2003 and 22 May 2003, that this conclusion as to causation was very significantly promoted by the use of a computer programme, the ISIS river modelling software, which enabled the construction of a model of the Taff simulating the flows and depths of water in the course of the 1998 flood. It is also to be noted that, as Dr Samuels accepted, if Grove Fields had been raised to the minimum height necessary to facilitate recreational use (which would have required 50,000 – 60,000 cubic metres of infill material), the appellants’ homes would still have been flooded, just as in fact they were. The additional infill, raising Grove Fields to ten or twelve feet, was not a cause of the flood.

Now, we are not of course directly concerned with the cause of the flood, given the concession in the appellants’ favour. The means of its ascertainment is relevant only to the different question, whether such a consequence of the infilling of Grove Fields was reasonably foreseeable 25 years or so before, when the infilling was done; and when no ISIS software was available. This issue of foreseeability will require me to travel over some other aspects of the evidence; but for the present I have done with the facts, and will turn to the law.

As I have foreshadowed, there were at trial two principal live issues, upon both of which the judge decided in the respondents’ favour: the application of the common enemy rule, and the question of foreseeability of flood damage to the appellants’ properties being caused by the infilling of Grove Fields. Lord Thomas challenges both conclusions. Though the second depends largely on the state of the evidence, it is in my judgment not possible to do justice to either issue without first setting out, at the risk of unoriginal repetition, certain general characteristics of the law of nuisance.

To this end I find it most convenient to start with the speech of Lord Wright in the well known decision of their Lordships’ House in Sedleigh-Denfield v O’Callaghan (Footnote: ¹)at 903:

“The ground of responsibility is the possession and control of the land from which the nuisance proceeds. The principle has been expressed in the maxim ‘Sic utere tuo ut alienum non laedas’. This, like most maxims, is not only lacking in definiteness but is also inaccurate. An occupier may make in many ways a use of his land which causes damage to the neighbouring landowners and yet be free from liability… Even where he is liable for a nuisance, the redress may fall short of the damage… A balance has to be maintained between the right of the occupier to do what he likes with his own, and the right of his neighbour not to be interfered with. It is impossible to give any precise or universal formula, but it may broadly be said that a useful test is perhaps what is reasonable according to the ordinary usages of mankind living in society, or more correctly in a particular society. The forms which nuisance may take are protean.”

As I shall show the balance here referred to is of the first importance both in considering the scope of the common enemy principle and in seeing whether it should be modified in light of the obligations arising under ECHR imposed on public authorities by HRA. But first there is more to say about the general law of nuisance. In Transco plc v Stockport MBC (Footnote: ²) Lord Walker of Gestingthorpe said this at paragraphs 95-97:

“95 Nuisance has been described as ‘protean’ (Lord Hope of Craighead in Hunter v Canary Wharf [1997] AC 655, 723, echoing Lord Wright in Sedleigh-Denfield v O'Callaghan [1940] AC 880, 903). Lord Wilberforce said in Goldman v Hargrave [1967] 1 AC 645, 657: ‘the tort of nuisance, uncertain in its boundary, may comprise a wide variety of situations, in some of which negligence plays no part, in others of which it is decisive.’ Lord Lloyd of Berwick (in Hunter, at p 695) provided a simple classification:

‘Private nuisances are of three kinds. They are (1) nuisance by encroachment on a neighbour's land; (2) nuisance by direct physical injury to a neighbour's land; and (3) nuisance by interference with a neighbour's quiet enjoyment of his land.’

Encroachment by the branches or (as in Delaware Mansions [2002] 1 AC 321) roots of trees is an example of the first category. The second category can be seen as including (but is certainly not limited to) the emission or escape of dangerous substances. Within this category at least, nuisance and Rylands v Fletcher are ‘congeners’ (as it was put by Lord Macmillan in Read v J Lyons & Co Ltd [1947] AC 156, 173). Nuisance by noise or smell is an example of the third category and it is in this category that the principle of ‘give and take, live and let live’ has most part to play (see Bramwell B in Bamford v Turnley (1862) 3 B & S 62, 84). The unifying factor in all three categories is that there is some sort of invasion of the claimant’s land, or his enjoyment of it.

96 Although the boundaries of nuisance are uncertain (and perhaps shifting) it is possible to sketch in some salient features of particular relevance to this appeal. One part of the territory overlaps with (indeed, is a sort of condominium with) that of negligence. That is particularly the case where a failure to take reasonable care may result in the owner or occupier of land ‘adopting’ or ‘continuing’ a nuisance for which he was not initially responsible. Here the line of authority includes Job Edwards Ltd v Birmingham Navigations Co Proprietors [1924] 1 KB 341; Sedleigh-Denfield v O’Callaghan [1940] AC 880; Goldman v Hargrave [1967] 1 AC 645; Leakey v National Trust for Places of Historic Interest or Natural Beauty [1980] QB 485; Holbeck Hall Hotel Ltd v Scarborough Borough Council [2000] QB 836 and Delaware Mansions Ltd v Westminster City Council [2002] 1 AC 321. In the factual situations in those cases (an accumulation of burning refuse, a blocked drain, a redgum tree struck by lightning, and so on) the defendant's actual neighbour was also his neighbour for the purposes of the principle in Donoghue v Stevenson [1932] AC 562, 580.

97 Elsewhere in nuisance’s extensive territory, however, negligence (in the sense of a demonstrable failure to take reasonable care) has traditionally been regarded as irrelevant. If the noise and smell from stabling for two hundred horses (used to pull trams) is intolerable in a densely-populated residential neighbourhood, it is no defence that the defendant has used all reasonable care to minimise the annoyance: Rapier v London Tramways Co [1893] 2 Ch 588, 600. That was reaffirmed by your Lordships’ House in Cambridge Water, at p 300, where Lord Goff referred to the relevant control mechanism being that of reasonable user. And when the focus moves to the species of nuisance known as the principle in Rylands v Fletcher, strict liability is its essential characteristic: liability arises (apart from particular defences) without the need for proof of negligence.”

It is, I think, no surprise that the law of nuisance is protean: the modes in which a person may make use of his land (or his rights associated with land use) are themselves protean. So are the consequences of his doing so. And it is no surprise that the boundaries of the law of nuisance are uncertain and perhaps shifting: the balance to be struck between the landowner’s enjoyment of his property and the avoidance of harm to his neighbour will never be hard-edged, unless the law is degraded to a comprehensive rule-book, and that would be coarse and unjust. But in this as in other fields the common law’s benign empiricism comes at a price. The price is a relative loss of legal certainty. The antidote is to see how far the disparate instances of actionable nuisance (and of what has been held not to be actionable nuisance) may be assembled, not within a rule-book, but within a principle or at least a coherent set of principles or themes.

Now I would certainly not presume to offer a general synthesis of the law of nuisance; but I think it possible to discern in the cases certain overlapping themes, whose relationship one with another goes some distance in setting a framework to illuminate a reasoned basis for the limits or edges of the common enemy principle, and to decide (to the extent that we are required to do so) whether it needs any modification in light of our domestic law obligations arising under ECHR. I have two such themes in mind. The first is the relation between the notion of a natural use of land and the notion of a reasonable use of land. The second is the requirement of reasonable foreseeability of damage as a condition of liability in nuisance.

As for the first of these: I think it may broadly be said that a landowner will not be liable in nuisance for the consequences of what would be recognised as a natural use of his land by him, unless the quality or extent of that use by him was unreasonable. This is, as it seems to me, the effect of the dictum of Bramwell B in Bamford v Turnley (Footnote: ³) at p. 83 (cited by Lord Goff of Chieveley in Cambridge Water Co v Eastern Counties (Footnote: ⁴) at 299D-E):

“… those acts necessary for the common and ordinary use and occupation of land and houses may be done, if conveniently done, without subjecting those who do them to an action.”

“If conveniently done”: thus, I take it, if the use is excessive and plainly inconsiderate, it will not be reasonable. One might compare the observation of Brett J in Benjamin v Storr (Footnote: ⁵) that not every “fleeting and inconvenient” interference will be an actionable nuisance. Now, what is a natural use of land is by no means self-evident. In Cambridge Water (Footnote: ⁶) Lord Goff expressed the view that the most authoritative statement of the principle of natural use of land was to be found in the advice of the Privy Council delivered by Lord Moulton in Rickards v Lothian (Footnote: ⁷), when he said of the rule in Rylands v Fletcher (Footnote: ⁸):

“It is not every use to which land is put that brings into play that principle. It must be some special use bringing with it increased danger to others, and must not merely be the ordinary use of the land or such a use as is proper for the general benefit of the community. ”

But if, albeit broadly, a landowner will not be liable in nuisance for the consequences of what would be recognised as a natural use of his land by him, unless the quality or extent of that use was unreasonable, still if he pursues a non-natural use, then generally no appeal to reasonableness will help him. This is most plainly exemplified by the well known rule in Rylands v Fletcher: anyone who in the course of a non-natural use of his land brings onto it for his own purposes anything likely to cause harm if it escapes is answerable for all direct damage thereby caused. Although in much of the learning, judicial and academic, this rule has been treated as a separate tort from nuisance, it seems to me that is a needless multiplication of categories, and that Rylands v Fletcher should be regarded as a species of nuisance following Lord Goff’s discussion in Cambridge Water at 297D-306D: indeed Lord Walker specifically so describes it in the passage from his opinion in Transco plc v Stockport MBC which I have set out.

I should add that, while no doubt the rule in Rylands v Fletcher is alive and well, and what are plainly natural – perhaps “ordinary” is a better word – uses of land will continue to be respected and allowed unless pursued excessively, still the importance of the distinction in our law between natural and non-natural uses is receding. So much is suggested by such cases as Leakey v National Trust (Footnote: ⁹) (decided in this court), referred to by Lord Walker in Transco. It is stated in the headnote:

“… an occupier of land owed a general duty of care to a neighbouring occupier in relation to a hazard occurring on his land, whether such a hazard was natural or man-made; that the duty was to take such steps as were reasonable in all the circumstances to prevent or minimise the risk of injury or damage to the neighbour or his property of which the occupier knew or ought to have known; that the circumstances included his knowledge of the hazard, the extent of the risk, the practicability of preventing or minimising the foreseeable injury or damage, the time available for doing so, the probable cost of the work involved and the relative financial and other resources, taken on a broad basis, of the parties; and that, in the present case, it being accepted by the defendants that the quantity and cost of the work required had not gone beyond their financial or other capacities or been greater than had been necessary to deal with the actual damage to the plaintiffs, the plaintiffs had been entitled to judgment…”

Since Leakey is not directly in point, I will not take further time with it, save to say that it was by no means a revolutionary departure; it is plain from the leading judgment of Megaw LJ that the decision built on earlier learning and in particular the reasoning of their Lordships in Sedleigh-Denfield v O’Callaghan, which I have cited.

The second theme to which I referred earlier was the requirement of reasonable foreseeability of damage as a condition of liability in nuisance. This is a perfectly general requirement to which claimed instances of liability under the Rylands v Fletcher rule are subject no less than other classes of nuisance (Footnote: ¹⁰). Unlike some aspects of my first theme – the relation between the notion of a natural use of land and the notion of a reasonable use of land – this requirement of foreseeability is hard-edged. Its application in practice, it goes without saying, depends on the evidence in the case: which, of course, may very well not be hard-edged.

We have, then, a constellation of themes within which the common enemy rule has to be understood. They are: a bias in favour of natural user, subject to its being no more than reasonably enjoyed; a bias (effectively a conclusive rule) against non-natural user where that involves the escape of something noxious onto a neighbour’s land; a bias against the harbouring of a danger, a hazard, on one’s own land whether the hazard is natural or man-made. And in no case will there be liability without reasonable foreseeability of damage.

So far as can be discerned from the books the rule was first recognised in the Court of Session in 1741 in Farquharson v Farquharson (Footnote: ¹¹). The report – it is only a note – states:

“It was found lawful for one to build a fence upon his own ground, by the side of a river, to prevent damage to his ground by the overflow of the river, though thereby a damage should happen to his neighbour by throwing the whole overflow in time of flood upon his ground. But it was found not lawful to use any operation in the alveus.”

I shall come back to the alveus. It means an established watercourse (even if dry for part of the year or from time to time) as opposed to a flood plain, which is an area of land liable to flooding not contained in a specific watercourse or alveus. In the course of argument Lord Thomas made it clear, notwithstanding assertions in his skeleton argument to the contrary, that he accepted that Grove Fields was a flood plain, and so was not (nor did it comprise) an alveus.

The common enemy rule has consistently been accepted in the English cases. I will not cite all the learning. The first case in the books is R v The Commissioners of Sewers for the Levels of Pagham (Footnote: ¹²). There, the common enemy was not a river’s overflow, but the inroads of the sea. The Commissioners erected groynes and other works to defend the stretch of coast for which they were responsible against the sea’s encroachment. But the consequence was that the sea flowed with greater force upon adjoining land, whose owner brought proceedings. Lord Tenterden CJ said at 361:

“I am… of opinion that the only safe rule to lay down is this, that each land-owner for himself, or the commissioners acting for several land-owners, may erect such defences for the land under their care as the necessity of the case requires, leaving it to others, in like manner, to protect themselves against the common enemy.”

R v Trafford (Footnote: ¹³), decided quite shortly after the Pagham case, provides a striking instance of the rule’s limitations. The river Mersey and an associated brook overflowed their banks in wet weather at the place in question, and the waters went north and west over adjoining lands, at length flowing back into the Mersey. The affected landowners raised banks (referred to as “fenders”) to confine the flood waters to the river and the brook. A canal, carried over the Mersey on an aqueduct, was constructed by authority of Parliament. The aqueduct had three arches taking the canal across the land where the flood waters naturally went. Thereafter the flow of water in the Mersey was increased by improved drainage higher up the river. So the landowners raised the height of their fenders. This pushed up the height of the flood water so as to endanger the canal and its aqueduct. The canal owners prosecuted the landowners on indictment for a nuisance. The jury delivered a very complicated special verdict which, under the procedure of the time, was then referred to the Court of King’s Bench. Lord Tenterden, who as we have seen was party to the decision in Pagham, delivered the judgment of the court in favour of the prosecutor (887-888):

“Now, it has long been established, that the ordinary course of water cannot be lawfully changed or obstructed for the benefit of one class of persons, to the injury of another. Unless, therefore, a sound distinction can be made between the ordinary course of water flowing in a bounded channel at all usual seasons, and the extraordinary course which its superabundant quantity has been accustomed to take at particular seasons, the creation and continuance of these fenders cannot be justified.”

However the case did not end there. The decision of the King’s Bench was appealed – removed on error – to the Court of Exchequer Chamber. The court’s judgment was delivered by Tindal CJ who made it clear that a verdict of guilty would only be sustainable if (among other things) it appeared that (210):

“… the course which the flood water is stated in the special verdict to have taken, and by which it was carried again into the river at a lower point, was the ancient and rightful course for it to take.”

The Exchequer Chamber held that this was not shown with sufficient clarity, and they ordered a venire de novo.

Trafford was thus a case in which the landowners’ potential liability depended on whether or not their raised fenders interfered with the alveus – the established watercourse – of the river; the judgment of the King’s Bench was specifically overturned because it was not shown that that was so. The law’s prohibition of such an interference can be seen in other cases to which we were referred: Vinnicombe (Footnote: ¹⁴), Gartner (Footnote: ¹⁵), Woolner (Footnote: ¹⁶) Gerrard (Footnote: ¹⁷), and Marriage (Footnote: ¹⁸). Nield (Footnote: ¹⁹) is interesting as showing (Footnote: ²⁰) that “there is no duty on the owners of a canal analogous to that on the owners of a natural watercourse, not to impede the flow of water down it”.

I have already referred to Lord Thomas’ acceptance that on the facts Grove Fields was not, and did not comprise, an alveus. I have drawn specific attention to the learning which establishes that the law prohibits interference with the alveus because I think it important to be clear about the limits of the common enemy rule, of which this prohibition is one, and an important one.

The cases disclose another limit to the common enemy rule. While the rule essentially permits the landowner to erect defences whose effect will be that water which would otherwise have flowed onto his land will be diverted onto his neighbour’s, it distinctly does not permit the landowner to take measures so as to cause water which has already or will in any event come onto his land, to flow from it onto that of his neighbour. A leading instance is to be found in Hurdman v NE Ry Co (Footnote: ²¹). The defendants raised the surface of their land, with the consequence that rainwater falling onto it penetrated a wall and ran into the plaintiff’s land, causing substantial damage. Delivering the judgment of the court Cotton LJ indicated that the heap or mound erected on the defendants’ land had to be considered as “an artificial work”. He continued (173):

“Every occupier of land is entitled to the reasonable enjoyment thereof. This is a natural right of property, and it is well established that an occupier of land may protect himself by action against any one who allows any filth or any other noxious thing produced by him on his own land to interfere with this enjoyment. We are further of opinion that… if any one by artificial erection on his own land causes water, even though arising from natural rain-fall only, to pass into his neighbour’s land, and thus substantially to interfere with his enjoyment, he will be liable to an action at the suit of him who is so injured…”

The judgment proceeds to discuss and reject an argument to the effect that the court’s conclusion could not stand with a mine-owner’s immunity from suit where his works cause water to gravitate into the mine of another lower down. I need not take time with that.

The other case I will cite for the proposition that a landowner may not remove floodwater which is on his own land to his neighbour’s land, so causing damage, is Whalley v Lancs and Yorks Ry Co (Footnote: ²²). The headnote is admirably crisp:

“By reason of an unprecedented rainfall a quantity of water was accumulated against one of the sides of the defendants’ railway embankment, to such an extent as to endanger the embankment, when, in order to protect their embankment, the defendants cut trenches in it by which the water flowed through, and went ultimately on to the land of the plaintiff, which was on the opposite side of the embankment and at a lower level, and flooded and injured it to a greater extent than it would have done had the trenches not been cut. In an action for damages for such injury the jury found that the cutting of the trenches was reasonably necessary for the protection of the defendants’ property, and that it was not done negligently:-

Held, that though the defendants had not brought the water on their land, they had no right to protect their property by transferring the mischief from their own land to that of the plaintiff, and that they were therefore liable.”

With great respect to the judges in the case, I think the real basis of the decision was put most forcefully by Gully QC in the course of argument for the plaintiff (134):

“There is a great difference between a right to a landowner to protect his property against a common enemy, as was the case in Nield… and Pagham Commissioners, and a right to pass such enemy on to the land of a neighbour.”

And so the court held. The Master of the Rolls stated (137):

“… if [the water] had been left alone and allowed simply to percolate through the embankment, even though all of it would have gone on to the plaintiff’s land, it would have gone without doing the injury which was done by reason of its passing through the cuttings which the defendants made. The defendants did something for the preservation of their own property which transferred the misfortune from their land to that of the plaintiff, and therefore it seems to me that they are liable.”

Where on the facts it is the effects of flood waters that give rise to legal dispute, Hurdman and Whalley may be seen (as indeed I have described them) as marking limits to the common enemy rule. You are entitled to protect yourself against the common enemy’s incursions; but if the incursion upon your land has already happened or is about to happen, you may not export it to your neighbour. This is a pragmatic drawing of the line. Mr Palmer submitted, I think rightly, that Whalley illustrates a point of departure in the policy of the law: the point between those situations where the landowner must accept a burden in favour of his neighbour, and those where he is entitled to protect his own land, even at his neighbour’s cost.

It is a point of departure which to some eyes will look fragile, as perhaps will the distinction between the alveus and the flood plain. But both of these, in the context of the common enemy rule, are mechanisms to achieve the balance which, as Lord Wright said in Sedleigh-Denfield, “… has to be maintained between the right of the occupier to do what he likes with his own, and the right of his neighbour not to be interfered with”. So it is at this point, as it seems to me, that the common enemy rule can be seen to conform with the general law of nuisance. It represents a resolution of the balance between self interest and duty to neighbour which, in the broad context of land use of which it is a particular instance, makes up the body of law called nuisance. I have said (paragraph 6) that it was agreed on all hands that there was no issue of “reasonable user” of Grove Fields by the respondents relevant to the claim against them in nuisance. That is right; but it is only so, in my judgment, because the limitations of the common enemy rule – no interference with an alveus, no discharge from one’s own land – are themselves partial guarantors of reasonable user. Where works done to protect land from the common enemy are within these limits, they may be considered a natural use of the land which is prima facie also a reasonable use. They might still be actionable if they were in some way excessive: as, for instance, might have arisen if Grove Fields had been raised to a height in excess of what was needed to prevent its being flooded, and the extra height was an independent cause of damage to the Pantglas Fawr estate.

On the facts of this case, it is clear that the two limitations are met; clear also that any additional height of the infill did not at all contribute to the damage to the appellants’ properties.

I must turn to the applicability or otherwise of the common enemy rule to the present facts. However I find it convenient first, while still dealing with the law, to confront Lord Thomas’ particular argument that the rule needs to be modified in light of the obligation owed by the court, as a public authority within the meaning of the HRA, to vouchsafe ECHR rights.

Lord Thomas refers to ECHR Article 8 (respect for one’s home) and Article 1 of the First Protocol (right to the peaceful enjoyment of one’s possessions). His submission is essentially to the effect that the common enemy rule ought to be more narrowly confined than it has been in the principal cases which I have cited, because its unbridled application on the present facts, leaving the appellants wholly uncompensated for the damage they have suffered, subjects them to a “disproportionate burden”. It is not entirely clear how Lord Thomas would have the rule reformulated. In his skeleton argument (paragraph 24) it is suggested that the law be developed so that the rule will not “provide a developer of land with a complete immunity to what might otherwise be an actionable wrong”. But if that means that these claims should simply have succeeded, and full compensation awarded, the common enemy rule would not be reformulated but abolished. Such an outcome was not, however, distinctly urged by Lord Thomas. If what is intended is the adoption of some middle way, conditioned perhaps by the economic resources of the injured claimants, nothing was offered for our consideration to show how the common law might sensibly articulate such a modified rule.

As it happens this court, and their Lordships’ House, have already been required to give some consideration to the impact of ECHR upon a state of affairs in which interference with an occupier’s enjoyment of his land has been said to give rise to a claim in nuisance: Marcic v TWUL (Footnote: ²³). In that case the plaintiff’s house and garden had been repeatedly affected or flooded by sewage discharged from sewers operated by the defendant, a statutory sewerage undertaker. A section of the sewer had become overloaded because of the increased user which followed the connection, as of right, of new parties to the system. The defendant had a points-based system for the allocation of resources to deal with flooding incidents. But its application meant that there was no prospect of anything of substance being done in the foreseeable future to remedy the flooding of the plaintiff’s property, although it was practicable to carry out the necessary work. This court (overturning the judge below and applying Leakey) held that the defendant was liable to the plaintiff in nuisance at common law. But it also considered and upheld the claim, which the judge had accepted, that the defendant’s failure or omission constituted a violation of the plaintiff’s rights under Article 8 and Article 1 of the First Protocol. However their Lordships’ House disagreed. At paragraph 34 Lord Nicholls of Birkenhead stated that “the cause of action in nuisance asserted by Mr Marcic is inconsistent with the statutory scheme”. As regards the ECHR this court, after citing the Commission’s decision in S v France (Footnote: ²⁴), had stated (paragraph 117):

“This suggests that where an authority carries on an undertaking in the interest of the community as a whole it may have to pay compensation to individuals whose rights are infringed by that undertaking in order to achieve a fair balance between the interests of the individual and the community.”

But the House of Lords held there was no ECHR claim. At paragraph 71 Lord Hoffmann stated:

“That leaves only the question of whether the remedies provided under the 1991 Act do not adequately safeguard Mr Marcic’s Convention rights to the privacy of his home and the protection of his property. The judge, who found for Mr Marcic on this ground, did not have the benefit of the decision of the Grand Chamber of the European Court of Human Rights in Hatton v United Kingdom Application No 36022/97, (unreported) 8 July 2003. That decision makes it clear that the Convention does not accord absolute protection to property or even to residential premises. It requires a fair balance to be struck between the interests of persons whose homes and property are affected and the interests of other people, such as customers and the general public. National institutions, and particularly the national legislature, are accorded a broad discretion in choosing the solution appropriate to their own society or creating the machinery for doing so.”

In the present case the respondents have, first and foremost, a short point to the effect that in October 1998 the HRA was not in force and the ECHR rights are, accordingly, simply irrelevant to the dispute between the parties. I would accept this. With respect I will not take time with authority: Lord Thomas has not established to my satisfaction any arguable basis upon which the court should give retrospective effect to the ECHR rights on the facts of the case.

However I would not quite leave the matter there. I hope it is useful to state my view that the common enemy rule, subject perhaps to exceptional instances, is in principle inoffensive to Article 8 and Article 1 of the First Protocol. In its context it strikes a proper balance “between the right of the occupier to do what he likes with his own, and the right of his neighbour not to be interfered with”: Lord Wright in Sedley-Denfield; and “between the interests of persons whose homes and property are affected and the interests of other people, such as customers and the general public”: Lord Hoffmann in Marcic. This latter formulation in terms engages the balance between private right and public interest. Every landowner has his own claims and interests. But the public interest requires a just and ordered resolution of the competition between one landowner’s claim and the next. The limits, which I have discussed, of the common enemy rule provide as much. This statement of principle made in the European Court of Human Rights in Sporrong (Footnote: ²⁵) has often been repeated:

“… the Court must determine whether a fair balance was struck between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights. The search for this balance is inherent in the whole of the Convention…”

In my judgment the common enemy rule, subject as I have said to the possibility of exceptional instances not exemplified by the present case, meets this balance. This is a department of the law which (like so many: perhaps most) has always involved a painstaking search for a fair result between self and other. The quest of the ECHR is just the same. I see no reason to suppose that in this field the common law does not answer both.

Aside from his case on ECHR, Lord Thomas advanced these submissions. (1) The common enemy rule has no application on the facts, because his clients’ properties were inland of the Taff and did not, like Grove Fields, abut its banks. Inland occupiers such as the appellants could not be expected to take protective measures against flooding as if they were riparian owners. (2) The infilling of Grove Fields was a public project in which flood prevention was at most an incidental goal; the damage to the appellants’ properties was out of proportion to such public benefit as the infilling obtained; the risk of harm should accordingly be underwritten by the public developer. (3) The council were, or had been, the appellants’ landlord. “It cannot be acceptable for a local authority to be able to deflect floodwater from one area of land in its ownership on to another if the consequence is flood damage to the properties of the tenants from whom it collects rent.” (Footnote: ²⁶)

The first of these submissions is more elusive than it appears. The contention must be that as between the appellants and the council (as occupiers of Grove Fields) the Taff was not to be treated as a “common enemy” at all. However that would only be true if either (a) the appellants’ properties, by virtue of their inland location, were at no risk whatever of flooding from the Taff, or perhaps (b) any such risk to their properties was not reasonably foreseeable. But if (a) were right the flood of October 1998 would not have happened, and if (b) were right the respondents would not in any event be liable in nuisance and would need no recourse to the common enemy rule. It is a necessary premise of the appellants’ case that the flooding of their houses was reasonably foreseeable; but if it was, the waters of the Taff were a common enemy: one which confronted Pantglas Fawr as surely as Grove Fields. That being so the fact that the appellants’ properties were some distance back from the riverbank possesses no legal significance.

Submission (2) conflates two distinct points which need to be separated. First it is said that flood prevention was only a minor or incidental aim or motive of the infilling of Grove Fields. Secondly it is said that (not least given the modest resources of the appellants) the council as a public developer, expecting to reap objective public amenity benefits from the works, ought in justice to make good the harm inflicted by the works on the appellants and their properties. In my judgment the first of these propositions carries the appellants nowhere. A rule which required the court to measure, in a case like the present, the level of importance subjectively attached by the decision-maker to the goal of flood prevention in comparison with other possible benefits would be obviously unworkable and therefore disreputable. It is enough that flood prevention be a perceived and actual benefit. I do not find it necessary to discuss the possible but eccentric case where flood prevention was neither perceived nor intended as a consequence of the proposed works, but in fact eventuated from their execution.

The second proposition under submission (2) – that the council as a public developer ought in justice to bear the costs (here, the appellants’ damages) associated with the public amenity benefits which the works will generate – is a political position which cannot, I think, steer the common law of nuisance. There is certainly no trace of it in the books.

Submission (3) – the council’s position as landlord – is in my judgment rhetorical and without substance. The tenant has all the rights which his lease gives him. Barring any public law claim in judicial review where as here the landlord is a public body, and in this case nothing of the kind is advanced, it is impossible to discern any more ample rights in the tenant’s hands than his tenancy gives him. But it is not suggested here that the appellants have any claim against the council arising under their leases (or the terms of transfers under the right to buy legislation).

In all these circumstances I would hold, assuming for the purposes of the point that the flooding of the appellants’ homes in consequence of the infilling of Grove Fields was reasonably foreseeable, that the learned judge below was right to apply the common enemy rule.

There remains then the issue of reasonable foreseeability itself. I am conscious that this might be said to be the logically prior question. If such a chain of causation was not reasonably foreseeable, there could in any event be no liability on the shoulders of the respondents and the common enemy rule would not need to be considered. It has however seemed to me to be appropriate to deal with the common enemy rule first, given the importance of the point and out of deference to counsel’s careful arguments.

Though we were referred to a good deal of evidence by Lord Thomas on this issue, I can deal with it relatively shortly. It is convenient to start with the approach taken by the learned judge below. He stated (paragraph 42) the undoubted truth that it was reasonably foreseeable that the Taff might flood. Then he cited the planning condition which I have set out above at paragraph 17, and observed (paragraph 43) that “[t]here is a difference between what is theoretically capable of being foreseen and what is reasonably foreseeable.” He proceeded to state that at the time when the infilling scheme was carried out, no one – no relevant party – in fact foresaw that the works might cause flooding at Pantglas Fawr (and the reference must be to the interior of the houses). He listed the parties in question: the council, the NCB, the Welsh Office, the inhabitants of Aberfan and the GRA. At paragraph 44 he stated that had the GRA, as the experts with responsibilities relating to flood risk, foreseen any such risk it is inevitable that they would have objected to the scheme; but they did not do so. Then (paragraph 45) he observed that the WWA had not concluded that the infilling was a cause of the 1979 flooding at Pantglas Fawr, and finally (paragraph 46) he noticed the scale of investigation, assisted by up-to-date computer technology, which had to be undertaken to yield the experts’ conclusion that the infilling was “probably” causative of the 1998 flooding in the appellants’ homes. In light of all these factors the judge concluded (paragraph 47) that it was not reasonably foreseeable at the time of the scheme’s execution that the infilling would or might cause flood damage to the appellants’ properties on the Pantglas Fawr estate.

This issue of reasonable foreseeability was a question of fact: a matter of inference and judgment, no doubt, rather than one of primary fact, but fact nonetheless. The judge had all the evidence unfold before him over the course of a substantive trial, and undertook a site visit of the relevant area. Although the issue was not distinctly one of this or that witness’s credibility, I do not think that this court should interfere in the judge’s decision on this question unless at least it is shown that there was a mistake in his reasoning or he misapprehended the material evidence. This approach is I apprehend in line with the reasoning in Assicurazione (Footnote: ²⁷) on which Mr Darling for the Authority relied. I need not, with respect, cite the text.

The judge derived his conclusion that the causal link between the infilling and the damage to the appellants’ properties was not reasonably foreseeable essentially from three premises: (1) none of the parties which might have been expected to foresee such a risk, if in truth it could reasonably be foreseen, did so at the time; (2) there was no perception at the time that the infilling had caused the 1979 flooding of the Pantglas Fawr houses (which, as I have indicated, was no less severe than that of 1998, and which the experts agreed had in fact been caused in the same way); and (3) it took a good deal of expert work and up to the minute technology to establish that the infilling had actually been a cause of the damage at all. In my judgment there is nothing wrong with this logic. At first Lord Thomas sought to assault its impact by an argument put forward as one of legal principle. Referring to authority such as Hughes v Lord Advocate (Footnote: ²⁸) he submitted that in order to establish reasonable foreseeability his clients were not obliged to show that the precise damage which eventuated, and the precise mechanics by which it happened, had been reasonably foreseeable: only that damage of the same kind as the actual damage had been so foreseeable. That being so, since it was plainly foreseeable that the Taff would burst its banks and cause flooding, and there was really no contest but that the Pantglas Fawr estate might be affected, reasonable foreseeability was made out.

Mr Palmer had no quarrel with those two factual premises. He accepted that it was indeed reasonably foreseeable that the Taff would flood and the Pantglas Fawr estate would be affected, at any rate as far as the doorsteps. Historic flood levels, to which the experts referred, showed as much. Mr Palmer’s distinct case, however, was that it was not reasonably foreseeable that the infilling of Grove Fields would worsen the scenario by diverting water so as to cause floods inside the houses on the estate.

Now, I readily accept that an event may be reasonably foreseeable even though the precise mechanics of its causation are not. I need not, with respect, cite authority. The contrary proposition, at least if taken literally, would impose a wholly unreal and therefore unjust burden on claimants in many classes of case. But reasonable foreseeability must imply some understanding of the chain of events which is putatively foreseen; otherwise we are looking not at foresight, but divination. In fairness Lord Thomas, pressed in the course of argument, effectively conceded as much. He accepted that he had to show that it was foreseeable that the infilling of Grove Fields caused changes in the flow patterns of the water which, in the event of the Taff again overtopping its banks, would expose the appellants’ properties (and I understood him to mean their interior) to flood damage. That was a correct concession.

Mr Palmer’s argument is that given the matters of fact upon which the judge relied (and upon which Mr Palmer expanded in the course of his submissions), there was solid ground on which to conclude that the causal link between the infilling of Grove Fields and the flooding inside the appellants’ houses was not reasonably foreseeable when the works to Grove Fields were done. In my judgment this argument is correct. It involves no hidden premise to the effect that the appellants had to demonstrate what was the precise mechanism constituting the causal link. In these circumstances the foreseeability issue is in my judgment concluded in the respondents’ favour, unless the factual premises underlying the judge’s conclusion can be undermined.

In an endeavour to assault the judge’s approach to the facts Lord Thomas was in particular at pains to establish that the GRA had given no positive approval to the infilling of Grove Fields and had not in truth addressed their minds to its possible effects. The GRA’s contemporary records were not available. Lord Thomas cited many passages from the transcript of Dr Samuels’ evidence. I will not set them all out. Dr Samuels stated that the decision to proceed with the infilling rested with the local planning authority, though it should have been taken in consultation with the GRA; and indeed it appears, as Dr Samuels acknowledged after being shown a council minute of February 1971, that the GRA had been content that the scheme should go ahead. They had responsibilities for flood prevention matters. They had been concerned to raise the Taff’s embankments in 1969 to afford protection to a standard in terms of time of 1:40: that is, a risk of one flood in 40 years. Dr Samuels also accepted, as I read the transcript (Footnote: ²⁹), that the GRA may not have expressly considered what would happen if the Taff overtopped its banks after the level of Grove Fields had been raised. As a free-standing point, Lord Thomas was able to demonstrate that plans of the area which were used for the purposes of the infilling operations were out of date and did not show the Pantglas Fawr houses.

Having looked over all the passages in the oral testimony on which Lord Thomas relied, I have concluded that they afford his case, at best, marginal assistance. It is true that the GRA were not the actual decision-makers for the infilling scheme, and may not have expressly addressed their minds to its potential consequences in terms of heightened flood risk. But in my judgment such points as these do not begin to touch the general force of the factual premises which led the judge to conclude as he did on the foreseeability issue. As for the planning condition on which Lord Thomas relied to demonstrate foreseeability, Mr Palmer established to my satisfaction that this was a standard term whose imposition in the permission for the Grove Fields scheme carried no implication that floods of the kind in question were foreseen or foreseeable.

For what it is worth I would for my part lay particular emphasis on two facts: (1) no one thought the infilling was the culprit of the very similar 1979 flood; (2) the determination that the infilling had in fact caused the flooding in the appellants’ houses, originally vigorously disputed, was the fruit of a painstaking and sophisticated exercise involving the use of technology by no means available in the 1970s. But wherever one puts the emphasis I am in no doubt that the judge was entitled to conclude as he did.

I should record that in the course of his reply Lord Thomas sought to rely, on the foreseeability issue, on some observations contained within a report prepared by SRK for the EA. SRK reports were contained in the trial bundles because, as I have indicated, the experts deployed data contained in them. But there had been no interlocutory permission to adduce this material as free-standing expert evidence; no party sought so to adduce it or refer to it, and no SRK witness was called; it was not relied on before the judge; and in my judgment it cannot fairly be relied on now.

For all these reasons I would dismiss the appeal.