BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales Court of Appeal (Civil Division) Decisions

You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Leakey & Ors v National Trust [1979] EWCA Civ 5 (31 July 1979)
Cite as: [1979] EWCA Civ 5, [1980] QB 485

[New search] [Context] [Printable version] [Buy ICLR report: [1980] QB 485] [Help]


Neutral Citation Number: [1979] EWCA Civ 5

(On appeal from Order of Mr. Justice O'Connor)

Royal Courts of Justice
31st July 1979

B e f o r e :




DORIS IRENE LEAKEY (his wife) and





(Transcript of the Shorthand Notes of The Association of Official
Shorthandwriters, Ltd., Room 392, Royal Courts of Justice,
and 2 New Square, Lincoln's Inn, London, W.C.2)


Mr. IAIN GLIDEWELL, Q.C. and Mr. JOSEPH HARPER (instructed by Messrs. Dawson & Co., London, W.C.2) appeared on behalf of the Appellants (Defendants).
Mr. HUBERT DUNN (instructed by Messrs. Gamlens, London, W.C.2, Agents for Messrs. Pardoe, David & Shaw, Bridgwater) appeared on behalf of the Respondents (Plaintiffs).



Crown Copyright ©

    LORD JUSTICE MEGAW: This appeal from the judgment of Mr. Justice O'Connor raises questions which are of importance in the development of English law. The learned judge held that the defendants are liable to the plaintiffs in damages, on a claim framed in nuisance, based on the fact that soil and other detritus had fallen from property owned and occupied by the defendants on to the plaintiffs' properties. It was accepted by the parties that the instability of the defendants' land which made it labile, and which had caused, and was likely to continue to cause, falls of detritus on the plaintiffs' land, was not caused by, nor was it aggravated by, any human activities on the defendants' land. It was caused by nature: the geological structure, content and contours of the land, and the effect thereon of sun, rain, wind and frost and such-like natural phenomena. It was held by the learned judge, and is not now in dispute, that, at least since 1968, the defendants knew that the instability of their land was a threat to the plaintiffs' property because of the possibility of falls of soil and other material. Although requested by the plaintiffs to take steps to prevent such falls, the defendants had not taken any action, because they held the view, no doubt on legal advice, that in law they were under no liability in respect of any damage which might be caused to neighbouring property in consequence of the natural condition of their own property and the operation of natural forces thereon.

    Mr. Justice O'Connor has held that that view of the law is wrong. He bases his decision on the judgment of the Judicial Committee of the Privy Council in Goldman v. Hargrave [1967] ChC 645. The main issue in this appeal is whether Goldman v. Hargrave accurately states the law of England. If it does, the appeal fails, and the defendants are liable.

    The opinion of the Judicial Committee in Goldman v. Hargrave, an appeal from the High Court of Australia, emanating originally from the Supreme Court of Western Australia, was delivered by Lord Wilberforce, the Board consisting of Lord Reid, Lord Morris of Borth-y-Gest, Lord Pearce, Lord Wilberforce and Lord Pearson. We in this Court are not bound by that decision, as we should have been bound if their Lordships had been sitting as an Appellate Committee of the House of Lords. But there can be no suggestion that the Board regarded the law of Western Australia, on the issues which they decided, as differing in any way from the law of England. I need scarcely say that in those circumstances it would be only after the most mature consideration that this Court could regard it as right to conclude that the unanimous view of the Board did not accurately represent the law of England. Nevertheless, as the defendants in this appeal, perfectly properly and with powerful and carefully marshalled arguments, ask us to say that Goldman v. Hargrave is not the law of England, we must, equally, give these submissions the mature consideration to which they are entitled.

    I shall, of course, have to return to these submissions in more detail later. But it may be helpful if at this stage I indicate briefly that the criticism offered of the decision in Goldman v. Hargrave is, in essence, that that decision involves the disregard, perhaps by inadvertence, of a principle of law laid down by the -House of Lords in Rylands v. Fletcher [1868] LR 3 House of Lords 330, which has, it is said, ever since been accepted and followed. If the Judicial Committee's decision assumes, or involves the proposition, that the House of Lords in Sedleigh-Denfield v. O'Callaghan [1940] AC 880 had departed from that principle, then that was a decision which the House of Lords themselves had not been entitled to reach in 1940, in the light of the then-prevailing doctrine of precedent in the House of Lords: and it should not be assumed that the House of Lords had so decided, contrary to that doctrine. Alternatively, the defendants submit that, at most, the ratio decidendi of Sedleigh-Denfield v. O'Callaghan leaves it open for, but not obligatory upon, lower courts thereafter to decide that a landowner could be liable for damage to his neighbour's property although there was no unnatural use of the land and the nuisance resulted solely from the operation of the forces of nature. If that be the position as regards authority, then, the defendants submit, despite the great weight which ought to be given to the view of the Privy Council, this Court ought, with the greatest respect to that view, to reject it and to decide the question in the opposite way. It is free to do so. It should, it was submitted, hold that there is no such liability.

    Having thus stated, as I thought to be desirable at an early stage of this judgment, the broad nature of the important issue which arises on this appeal, I shall now return to a more detailed statement of the facts and events which led up to this appeal. But I do not propose to go into great detail, for the issues to be decided do not depend on detail. I shall try to say no more than is desirable as giving the background to the appeal.

    Near Bridgwater, in Somerset, there is a hill, rejoicing in the name of Burrow Mump. It has since 1946 been owned by the defendants, the National Trust. It is a conical hill which rises steeply, on all its sides, from the Somerset plain. Geologically, it is composed of Keuper Marl, which, as I understand it, makes it peculiarly liable to cracking and slipping as a result of weathering. At the base of its western side lie two adjacent houses, which belong to the plaintiffs. "Elm Glen" belongs to Mr. and Mrs. Leakey, the first and second plaintiffs. "Hillside", which is immediately to the north of "Elm Glen", belongs to Mr. Storey, the third plaintiff. The latter house had been constructed by the amalgamation of 2 cottages which had stood on the site for, perhaps, 200 or 300 years. "Elm Glen" was built in 1915, but there had been an earlier house or houses on the same site, going back to a time, to use the picturesque phrase, "whereof the memory of man runneth not to the contrary".

    The lower part of the western hillside of the Hump which looms over these two houses, is particularly steep. At that place it takes the form of a bank. At some unknown time in the past, it is thought, there was some cutting, by human hands and implements, for some unknown reason, of the hillside at this point. The consequence may have been to make what I have called "the bank" steeper than it would have been if nature had been left to its own workings. But, though issues were raised about this in the pleadings in the action, nothing now turns on the fact, or the possibility, that there had been this remotely distant human intervention. It is accepted by all parties that the legal rights and liabilities are to be ascertained on the assumption that the contours of the defendants' property are as nature made or developed them.

    It is also to be assumed that, as admitted by the defendants in their pleading, they were not only the owners, but also the occupiers, of Burrow Mump at all relevant times. We were told by defendants' counsel that in fact the Mump, or part of it, had been occupied on lease or licence by someone to whom the defendants had given grazing rights. But, for the purpose of this appeal, the defendants, fairly and properly, do not seek to go back on their admission that they are the occupiers. It is not, therefore, a question for consideration in this appeal whether it would have affected the issue of the defendants' liability if someone else had been the occupier.

    Another fact which I have already mentioned as not being in dispute, but which it is right should be emphasised, is that there has been no human activity, whether the possible long-distant cutting of the hillside already mentioned, or any bringing of anything on to the defendants' land, or the carrying out of any work or operations of any sort on the land, which has in any way affected or increased the labile condition of the land. Its instability, its propensity to slip, which admittedly exists, is caused by nature: the contours, the geological structure and material, and the effect thereon of sun, rain, wind, frost and other natural agencies.

    As a result of the operation of natural agencies on the steep contours of Burrow Mump, where its western slope rises in a bank at the back of the plaintiffs' houses, there have from time to time over the years, for many years past, been slides, often quite minor, sometimes more substantial, of soil, rocks, tree-roots and such-like detritus, from the bank, which belongs to the defendants, on to land belonging to each of the plaintiffs: on to a strip of land lying at the backs of the houses, between them and the bank, and also, on occasion, on to a piece of the Leakeys' land to the south of their house, where there was an out-house belonging to them. There were earlier complaints and discussions, into which I need not go. But, as I have already said, there is now no dispute but that, since 1968 at least, the defendants appreciated that the bank was a part of their property and that it was a threat to the houses below because of the real possibility of falls of material from it.

    The present action has its origin in developments following the exceptionally hot, dry summer of 1976 and the unusually heavy rainfall of the ensuing autumn. Mrs. Leakey, in September of that year, noticed that a big crack had opened up in the bank above her property. She drew the attention of the defendants to it, pointing out that there was a grave danger of a major collapse of the bank on to her house. The defendants told her that this was a natural movement and that they were not obliged to do anything about it. She offered to pay half the cost of reducing the bank. At that time she thought that the cost would be about 1,000. That offer was rejected.

    A few weeks later there was a large fall of the bank. Part of the intrusion into the plaintiffs' properties involved the piling up of detritus against the outside wall of the Leakeys' kitchen. A letter was sent by solicitors acting for the Leakeys to the defendants saying that the house was at risk of further falls and that there were large tree-stumps in the material which had already fallen. They asked the defendants to remove the fallen debris. The defendants refused, denying responsibility; but they said that they would be glad to give the Leakeys a licence to go on to the defendants' land to abate the nuisance to their property. The Leakeys and Mr. Storey then joined forces in launching the action out of which this appeal arises. I mention at this stage, simply to show the seriousness of the case, that the cost of clearance and of protective works hitherto undertaken is about 6,000.

    The writ was issued on 31st January, 1977, claiming, first, a mandatory injunction, restraining the defendants from permitting soil, earth and tree-stumps to move on to the plaintiffs' properties; secondly, an order to remove the soil, etc., which had already fallen; and, thirdly, damages for nuisance. The defendants stress that there was no express pleading of negligence, nor any formulation of the claim other than in nuisance. The defendants by their defence denied any liability. I need not consider the pleadings more deeply.

    The plaintiffs sought an interlocutory injunction. The defendants sought, courageously in the light of Goldman v. Hargrave, to strike out the statement of claim as disclosing no reasonable cause of action. The two interlocutory applications were heard by Mr. Justice Bristow on 1st March, 1977. He dismissed the defendants' application. He granted the plaintiffs an interlocutory injunction, on the plaintiffs' undertaking as to damages. The judge ordered the defendants to take steps to prevent soil etc. moving from Burrow Mump on to the plaintiffs' properties. The defendants carried out work on the land. That work was completed in April, 1977, at a cost to the defendants of 2,175. No further falls occurred before the hearing of the action.

    The action was thereafter tried before Mr. Justice O'Connor. On 20th January, 1978, he delivered judgment. He held that the defendants were liable in nuisance. He awarded damages of 50 to Mr. and Mrs. Leakey and nominal damages of 2 to Mr. Storey. The damages were relatively small because, first, the defendants had undertaken and paid for the protective work under the interlocutory injunction, which included the removal of the fallen soil from the Leakeys' property; and, secondly, because, while refusing to grant an injunction, the judge made it clear that if there should, unfortunately, be further encroachments, the plaintiffs, or either of them, would be free to present a fresh claim for damages. In fact, since Mr. Justice O'Connor's judgment, the defendants, no doubt wisely and properly, have caused further protective works to be carried out, which have cost, we have been told, approximately a further sum of 4,000. Hence the 6,000 to which I have referred earlier as having been the cost of clearing and protective works hitherto.

    At the trial, an undertaking was given on behalf of the defendants that, if the defence were to succeed, the defendants would not seek to enforce the undertaking given by the plaintiffs on the granting of the interlocutory injunction. That undertaking would have involved a payment to the defendants of a sum of 2,175, as the amount then was. Before us, counsel for the defendants renewed that undertaking and further undertook that, if the appeal were to succeed, the defendants would not seek to recover from the plaintiffs any part of the further 4,000 or so which has since been expended by the defendants on protective work. While it is right to record this generous offer by the defendants, it cannot, of course - and was not intended to - influence the decision in any way. It was made clear, further, that the undertaking did not relate to the cost of any further protective works.

    From Mr. Justice O'Connor's judgment, the defendants appeal. They submit that the question is not only one of great moment as affecting themselves, the National Trust, by reason of potential liabilities, if the decision is upheld, in respect of many properties which they own for the benefit of the public in many parts of the United Kingdom; but also, they submit, it is a question of general importance in the law. If Goldman v. Hargrave were to be treated as a part of the law of England, that, the defendants say, would be a development of the law which would not only be inconsistent with long-standing authority binding on this court, but also would be undesirable from the point of view of the general public interest and of fairness to persons, wealthy or poor, who own or occupy land, large or small in extent.

    There was a cross-notice by the plaintiffs (strictly, it should have been a cross-appeal) asking that injunctions should be granted. However, it was abandoned.

    For the defendants in this appeal, the fundamental proposition was formulated by counsel as follows:-

    "In English law, neither the owner nor the occupier of land from which, solely as the result of natural causes, natural mineral material encroaches on to, or threatens to encroach on to, adjoining land, causing damage, is under any liability to the adjoining land owner".

    Certain points should be made at once as to that proposition.

    First, the opening words, "In English law", are properly and deliberately included so as to emphasise that, even if the proposition has to be treated as being inconsistent with the ratio decidendi of Goldman v. Hargrave, that case, however persuasive, does not have the status of a binding authority as to English law.

    Secondly, the phrase "solely as the result of natural causes" is intended to connote, if the negative form of words were to be used, "not being the result of any human action, activity or use of land".

    Thirdly, the proposition applies irrespective of whether or not the owner or occupier of the potentially encroaching land has knowledge of the risk that the encroachment will or may take place.

    Fourthly, the proposition applies both to natural material making up the solid ground and to water naturally on or under land. In relation to this "fourthly", I believe that the adjective which was included in the proposition as formulated by counsel, "mineral" in "natural mineral material", was included as providing a possible basis of distinction according as the substance of the encroaching material is, on the one hand, mineral, such as rocks or soil (if soil is mineral), and, on the other hand, vegetable, such as a tree trunk or the roots or the branch of a tree. This, at first sight, unattractive distinction of "animal, mineral or vegetable?" is, I believe, introduced into the proposition in order to prevent the proposition from coming into obvious conflict with authorities, such as Davey v. Harrow Corporation (1958) 1 Queen's Bench 60, binding on this Court, as to damage caused by trees.

    The defendants' second proposition, which I propose to consider in the course of considering the first proposition, is that, if the first proposition be wrong, so that Goldman v. Hargrave does represent the law of England, nevertheless the liability which is imposed under the Goldman v. Hargrave principle is a liability in negligence not in nuisance. The present claim was pleaded, and pleaded only, in nuisance. Hence, it is said, it must fail.

    There is a third proposition of the defendants as to the scope of the duty, if duty there be, affecting the defendants. I shall leave the consideration of that proposition until I have dealt with the earlier two propositions affecting the existence of the duty.

    Counsel for the plaintiffs formulated, and at our request put before us in a document, a series of twelve propositions. It involves no disrespect to that clear and very helpful formulation of the arguments for the plaintiffs that I do not find it necessary to set them out and consider them specifically as they were framed. So far as I find it necessary to reach conclusions on those propositions, I shall do so in the course of my discussion of the defendants' propositions.

    Because of the conclusion which I have reached on this main issue, and my reasons for that conclusion which I do not think differ for the most part in any substantial respect from the reasons expressed in the judgment of Mr. Justice O'Connor I think that the simplest and most satisfactory method of dealing with the main issue is to start by a brief summary of the facts of, and the decision in, Goldman v. Hargrave.

    The relevant facts of Goldman v. Hargrave were simple. A red gum tree, 100 feet high, on the defendant's land was struck by lightning and caught fire. The defendant caused the land around the burning tree to be cleared and the tree was then cut down and sawed into sections. So far there could be no complaint that the defendant had done anything which he ought not to have done or left undone anything which he ought to have done, so as in any way to increase the risk which had been caused by this act of natural forces setting fire to the tree. Thereafter the defendant (this was the state of the facts on which the Judicial Committee based their decision) did not do anything which he ought not to have done. He took no positive action which increased the risk of the fire spreading. But he failed to do something which he could have done without any substantial trouble or expense: which would, if done, have eliminated or rendered unlikely the spreading of the fire: that is, to have doused with water the burning or smouldering sections of the tree as they lay on the ground. Instead, the defendant chose to allow or encourage the fire to burn itself out. Foreseeably (again it was the forces of nature and not human action), the weather became even hotter and a strong wind sprang up. The flames from the tree spread rapidly through the defendant's land to the land of neighbours where it did extensive damage to their properties.

    The judgment of the Board was delivered by Lord Wilberforce. It was held that the risk of the consequence which in fact happened was foreseeable. This,' it is said, "was not really disputed". The legal issue was then defined (page 656/F): "....the case is not one where a person has brought a source of danger onto his land, nor one where an occupier has so used his property as to cause a danger to his neighbour. It is one where an occupier, faced with a hazard accidentally arising on his land, fails to act with reasonable prudence so as to remove the hazard. The issue is therefore whether in such a case the occupier is guilty of legal negligence, which involves the issue whether he is under a duty of care, and, if so, what is the scope of that duty".

    It is to my mind clear, from this passage and other passages in the judgment, that the duty which is being considered, and which later in the judgment is held to exist, does not involve any distinction of principle between what, in another sphere of the law, used to be known as misfeasance and non-feasance. A failure to act may involve a breach of the duty, though, since the duty which emerges is a duty of reasonable care, the question of misfeasance or non-feasance may have a bearing on the question whether the duty has been broken. It is to my mind clear, also, that no distinction is suggested in, or can properly be inferred from, the judgment as between a hazard accidentally arising on the defendant's land which, on the one hand, gives rise to a risk of damage to a neighbour's property by the encroachment of fire and, on the other hand, gives rise to such a risk by the encroachment of the soil itself, falling from the bank on to the neighbour's land. There is no valid distinction, to my mind, between an encroachment which consists, on the one hand, of the spread of fire from a tree on fire on the land, and, on the other hand, of a slip of soil or rock resulting from the instability of the land itself: in each case, the danger of encroachment, and the actual encroachment, being brought about by the forces of nature.

    If any such distinctions as I have referred to in the previous paragraph were sought to be made, I should have thought that their acceptance as being material - as leading to different conclusions of principle in law - would make the law on this topic, incoherent, artificial, uncertain and unpredictable. In other words, they would lead to bad law.

    At the point in the Board's judgment immediately following the passage which I have quoted above, the judgment goes on to deal briefly with the question of the appropriate description of the cause of action. Their Lordships in that case found it unnecessary to decide (page 656/G) "whether if responsibility is established it should be brought under the heading of nuisance or placed in a separate category.... The present case is one where liability, if it exists, rests upon negligence and nothing else; whether it falls within or overlaps the boundaries of nuisance is a question of classification which need not here be resolved".

    It is convenient at this stage to deal with the second proposition put forward by the defendants in the present appeal. The plaintiffs' claim is expressed in the pleadings to be founded in nuisance. There is no express reference to negligence in the statement of claim. But there is an allegation of a breach of duty, and the duty asserted is, in effect, a duty to take reasonable care to prevent part of the defendants' land from falling on to the plaintiffs' property. I should, for myself, regard that as being properly described as a claim in nuisance. But even if that were, technically, wrong, I do not think that the point could or should avail the defendants in this case. If it were to do so, it would be a regrettable modern instance of the forms of action successfully clanking their spectral chains; for there would be no conceivable prejudice to the defendants in this case that the word "negligence" had not been expressly set out in the statement of claim. The suggestion that if it had been so pleaded the defendants could have raised a defence of volenti non fit injuria, which they could not raise as against a claim pleaded in nuisance, is, in my judgment, misconceived. As counsel for the plaintiffs submitted, while it is no defence to a claim in nuisance that the plaintiff has "come to the nuisance", it would have been a properly pleadable defence to this statement of claim that the plaintiffs, knowing of the danger to their property, by word or deed, had showed their willingness to accept that danger. Moreover, I find it hard to imagine circumstances in which the facts which would provide a defence of volenti non fit injuria would not also provide a defence in a case such as the present in the light of the scope of the duty which falls to be considered hereafter.

    If the defendants' first and main proposition is wrong, I do not see that they can succeed on their second proposition.

    I return to the judgment in Goldman v. Hargrave. The law of England as it used to be is set out in the following passage (page 657/C): " is only in comparatively recent times that the law has recognised an occupier's duty as one of a more positive character than merely to abstain from creating, or adding to, a source of danger or annoyance. It was for long satisfied with the conception of separate or autonomous proprietors, each of which was entitled to exploit his territory in a 'natural' manner and none of whom was obliged to restrain or direct the operations of nature in the interest of avoiding harm to his neighbours".

    The judgment of the Board then goes on to review the development of the law which, as the Board held, had changed the law so that there now exists (page 661/G) "a general duty upon occupiers in relation to hazards occurring on their land, whether natural or man-made".

    That change in the law, in its essence and in its timing, corresponds with, and may be viewed as being a part of, the change in the law of tort which achieved its decisive victory in Donoghue v. Stevenson [1932] AC 562: though it was not until eight years later, in the House of Lords decision in Sedleigh-Denfield v. O'Callaghan [1940] AC 880, that the change as affecting the area with which we are concerned was expressed or recognised in a decision binding on all English courts: and, even then, the full, logical effect of the decision in altering what had hitherto been thought to be the law was not immediately recognised. But Goldman v. Hargrave has now demonstrated what that effect was in English law.

    The Sedleigh-Denfield case, as I shall call it, approved the dissenting judgment of Lord Justice Scrutton in Job Edwards v. Birmingham Navigation Proprietors (1924) 1 King's Bench 341 (the Job Edwards case). Thus, as the Judicial Committee recognised in Goldman v. Hargrave, it was that great judge whose judgment in the Job Edwards case "marked a turning-point in the law", eight years before Donoghue v. Stevenson. In the Job Edwards case, Lord Justice Scrutton took the view that the case required to be sent back for a new trial because insufficient facts had been found. But the principle which, differing from Lord Justice Bankes and Mr. Justice Astbury, Lord Justice Scrutton stated was (page 36C of the report): "....the landowner in possession is liable for a nuisance created by a trespasser, which causes damage to others, if he could, after he knows or ought to have known of it, prevent by reasonable care its spreading". At page 357 the learned Lord Justice had used words which are quoted in the judgment in Goldman v. Hargrave [1967] 1 AC 645 , which I need not here repeat. I should merely mention that, for the defendants in this case, much stress is laid on the fact that Lord Justice Scrutton in that passage used the phrase: "....if a man finds a dangerous and artificial thing on his land". The defendants submit that the inclusion of the words "and artificial" makes all the difference in the subsequent development of the law. For reasons which I shall give, I think not.

    Again, in his judgment in the Job Edwards case, Lord Justice Scrutton referred with approval to what was said in the 5th edition of Salmond's Law of Torts, 1920, page 260, in a passage which was later given express approval by Viscount Maugham, at page 893, and by Lord Wright, at page 910, in the Sedleigh-Denfield case. Sir John Salmond wrote: "When a nuisance has been created by the act of a trespasser or otherwise without the act, authority or permission of the occupier, the occupier is not responsible for that nuisance unless, with knowledge or means of knowledge of its existence, he suffers it to continue without taking reasonably prompt and efficient means for its abatement".

    After the Job Edwards case and before the Sedleigh-Denfield case came the judgment of Mr. Justice Rowlatt in the Queen's Bench Divisional Court in Noble v. Harrison (1926) 2 KB 332, in which he said, at page 338:

    "....a person is liable for a nuisance constituted by the state of his property: (1) if he causes it; (2) if by neglect of some duty he allowed it to arise; and (3) if, when it has arisen without his own act or default, he omits to remedy it within a reasonable time after he did or ought to have become aware of it".

    It is head (3) which is relevant for present purposes. If this is the law of England, then Goldman v. Hargrave correctly represents the law of England in its assertion of "a general duty upon occupiers in relation to hazards occurring on their land, whether natural or man-made". I shall refer later to a passage in the judgment of the other member of the court in Noble v. Harrison, Mr. Justice Wright as he then was.

    In the Sedleigh-Denfield case, a local authority had trespassed on the defendant's land, without the defendant's knowledge or consent, and had placed a culvert in a ditch on that land. By the improper placing of a grid at the mouth of the culvert, instead of further back, those who did the work created a danger of flooding which would be likely to spread to the plaintiff's land. The defendant, through his servants, came to know what had been done. He should have realised that it created a real risk of flooding of his neighbour's land. He did nothing. A heavy rainstorm caused the ditch to flood, because of the trespassers' work. The plaintiff's land was damaged. The House of Lords held that the defendant was liable. The defendant himself had not done anything which was an "unnatural user" of his land. He had not himself brought anything "unnatural" on to his land. But when he knew or ought to have known of the risk of flood water from his land encroaching on his neighbour's land, he had done nothing towards preventing it.

    Prevention could have been achieved without any great trouble or expense.

    The approval by the House of Lords in the Sedleigh-Denfield case of Lord Justice Scrutton's judgment in the Job Edwards case meant, at any rate unless it could properly be said that it was a decision inconsistent with an earlier decision of the House of Lords, that it was thereafter the law of England that a duty existed under which the occupier of land might be liable to his neighbour for damage to his neighbour's property as a result of a nuisance spreading from his land to his neighbour's land, even though the existence and the operative effect of the nuisance were not caused by any "non-natural" use by the defendant of his own land. But the liability was not a strict liability such as that which was postulated by the House of Lords in Rylands v. Fletcher [1868] ChR 3 House of Lords 330 as arising where damage was caused to another by an "unnatural" user of land. The obligation postulated in the Sedleigh-Denfield case, in conformity with the development of the law in Donoghue v. Stevenson, was an obligation to use reasonable care. A defendant was not to be liable as a result of a risk of which he neither was aware nor ought, as a reasonably careful landowner, to have been aware.

    That decision was in a case where, on the facts, something which might be described as "not natural" had been introduced on to the defendant's land in the building of the culvert: but not by the defendant. It had been done by a trespasser without the defendant's knowledge or consent. It was not a case in which the potential damage to the neighbour's land had been brought about by natural causes. Therefore it may be said that the Sedleigh-Denfield case did not decide, so as to bind lower courts in England, that an owner or occupier of land was under a duty to exercise reasonable care where natural causes, as distinct from the act of a trespasser, brought about the dangerous condition of the land, of which he, the owner or occupier, knew or which he should have realised. If I had taken the view that the Sedleigh-Denfield case does not bear upon the question raised by the present appeal (and therefore alsoought not to have influenced the decision in Goldman v. Hargrave), I should have reached a different conclusion on this appeal. I do not, however, accept the suggested distinction.

    My first comment is that the whole tenor of the speeches in the Sedleigh-Denfield case suggests that the view of their Lordships, if not their decision, was that the same duty arose. The approval of the passage in Salmon's law of Torts, to which I have previously referred, so suggests. A passage in Lord Wright's speech [1940] AC 914 gives, I believe, a fair reflection of the attitude of their Lordships:

    "The liability for a nuisance is not, at least in modern law, a strict or absolute liability. If the defendant by himself or those for whom he is responsible has created what constitutes a nuisance and if it causes damage, the difficulty now being considered does not arise. But he may have taken over the nuisance ready made as it were, when he acquired the property, or the nuisance may be due to a latent defect or to the act of a trespasser or stranger. Then he is not liable unless he continued or adopted the nuisance, or, more accurately, did not without undue delay remedy it when he became aware of it, or with ordinary and reasonable care should have become aware of it. This rule seems to be in accord with good sense and convenience".

    I am confident that Lord Wright's words "latent defect" were intended to include a defect in the land itself. Lord Wright was making the same point - the lack of any valid distinction, in this context, between a trespasser's act and an act of nature - as he had made when he was Mr. Justice Wright, in Noble v. Harrison.

    There, after referring to Barker v. Herbert (1911) 2 King's Bench 633, he said ([1926] 2 KB 341 ):

    "The nuisance in that case was caused by the act of a trespasser, but I think the same principle applied to a nuisance (in this case the latent crack in the branch with the resulting risk that some day it would fall) caused by a secret and unobservable operation of nature".

    So long as the defect remains "latent" there is no duty on the occupier, whether the defect has been caused by a trespasser or by nature. Equally, once the latent becomes patent, a duty will arise, whether the causative agent of the defect is man or nature. But the mere fact that there is a duty, does not necessarily mean that inaction constitutes a breach of the duty.

    My second comment on the suggested distinction is that it involves a fallacy. I cite a passage from the judgment in Goldman v. Hargrave[1967] 1 AC 645 at page.66l/E, which, I respectfully suggest, makes this clear beyond dispute:

    "It was suggested as a logical basis for the distinction that in the case of a hazard originating in an act of man, an occupier who fails to deal with it can be said to be using his land in a manner detrimental to his neighbour and so to be within the classical field of responsibility in nuisance, whereas this cannot be said when the hazard originates without human action so long at least as the occupier merely abstains. The fallacy of this argument is that, as already explained, the basis of the occupier's liability lies not in the use of his land: in the absence of 'adoption' there is no such use; but in the neglect of action in the face of something which may damage his neighbour. To this, the suggested distinction is irrelevant".

    The late Professor Goodhart, in an article published in 1932 in 4 Cambridge Law Journal, page 13, at page 26 made trenchant criticism of the suggested distinction. He said:

    "In brief, an occupier must use reasonable care to see that his land does not cause harm to others outside the land. Is there an exception to this rule when the harm is caused by a natural condition? An artificial pond in my land becomes stagnant and infects the neighbourhood. I am liable. A natural pond becomes stagnant. Kay I sit by and do nothing? Refuse is thrown on my land by strangers. It is my duty to use reasonable care to abate the nuisance. Refuse is blown on my land by the wind. Kay I permit it to remain? A trespasser injures one of my planted trees so that it is about to fall into the highway. It is my duty to cut down the tree. A storm half uproots one of my self-grown trees. Kay I allow it to crash into the highway? Such an astonishing exception to the general rule that an occupier must take reasonable care of his property ought to be established by clear authority before it is recognised".

    Is there, then, anything in the ratio decidendi of Rylands v. Fletcher [1868] LR 3 House of Lords 330, or in any subsequent authority binding on this Court, which requires or entitles us to disregard the decision in the Sedleigh-Denfield case or to prevent us from accepting the logical extension of it (so far as it is an extension) which was regarded as proper in Goldman v. Hargrave?

    The application of the decision and of the dicta in Rylands v. Fletcher has given rise to continual trouble thereafter in the law of England. But, as I see it, the true ratio decidendi of Rylands v. Fletcher is not relevant to the issue with which we are concerned. In Rylands v. Fletcher the defendant was held to be liable because he had erected or brought upon his land something of an unusual nature, which was essentially dangerous in itself. That, said Mr. Justice Wright in Noble v. Harrison (1926) 2 KB 242 "expresses the true principle of Rylands v. Fletcher". The decision was that, on such facts, there was strict liability.

    It would be no answer for the defendant to say "I did not know of the danger and had no reason to know of it". It was no part of the decision, as distinct from dicta, in Rylands v. Fletcher that one who has not himself brought something of an unusual nature on his land, nor has used his land in an unnatural way (whatever that may mean or include), is in no circumstances liable, if something from his land encroaches on his neighbour's land. That was why Viscount Maugham in the Sedleigh-Denfield case [1940] AC 880, at page 888, said:

    "My Lords, I will begin by saying that in my opinion the principle laid down in Rylands v. Fletcher does not apply to the present case. That principle applies only to cases where there has been some special use of property bringing with it increased danger to others, and does not extend to damages caused to adjoining owners as the result of the ordinary use of the land: see Rickards v. Lothian [1913] AC 263, 280".

    If Rylands v. Fletcher was thus irrelevant in the Sedleigh-Denfield case, it is not relevant in this case.

    Rylands v. Fletcher does not impose strict liability except where there has been some non-natural use of the land. But it does not hold, by way of binding authority, that there can be no duty where there has not been a "non-natural" use of the land. The well known passage, cited with approval by Lord Cairns, Lord Chancellor, at page 339, from the judgment of Mr. Justice Blackburn when Rylands v. Fletcher was decided in the Court of Exchequer Chamber, which does provide the ratio decidendi of Rylands v. Fletcher, is concerned only with strict liability where a man "for his own purpose brings on his land and collects and keeps there something likely to do mischief if it escapes". He must, it is said, "keep it in at his peril". But it is obiter dictum when Lord Cairns says, at page 338: "The defendants....might lawfully have used that close for any purpose for which it might in the ordinary course of the enjoyment of land be used; and if, in what I may term the natural user of that land, there had been any accumulation of water either on the surface or underground, and if, by the operation of the laws of nature, the accumulation of water had passed off into the close occupied by the plaintiff, the plaintiff could not have complained that that result had taken place. If he had desired to guard against it, it would have lain upon him to have done so, by leaving, or by interposing some barrier between his close and the close of the defendants in order to have prevented the operation of the laws of nature".

    I do not believe that in that dictum, Lord Cairns had in mind, or was purporting to express any view upon, the question with which we are here concerned. I believe he was merely intending to stress that the strict duty, in no way dependent on the existence of negligence, which he was holding to exist in respect of "unnatural user of land", did not apply where the user was not "unnatural". Certainly, the authority which he cited for absence of liability for damage where the defendant's user was "natural", Smith v.. Kenrick (1849) 7 Common Bench 575, was a very special and peculiar case on its facts. A trespasser, for whose acts the defendant was in no way responsible, had bored holes through the coal-face which belonged to the plaintiff, on the plaintiff's land, with the result that water from the defendant's coal-workings could and did run through those holes. Presumably the plaintiff knew that these holes existed in his own coal-face and that his own natural defences on his own land had been rendered ineffective - and yet he took no steps to repair his own defences. I find nothing in the elaborate pleadings for the plaintiff in that case to suggest the contrary. This explains the reference by Lord Cairns to "interposing some barrier between his close and the close of the defendants". It is to be noted, also, that in the judgment of the Court of Common Pleas in Smith v. Kenrick, Mr. Justice Cresswell, delivering the judgment of the court, said, at page 564: " would seem to be the natural right of each of the owners of two adjoining coal-mines.... to work his own in the manner most convenient and beneficial to himself, although the natural consequence may be that some prejudice will accrue to the owner of the adjoining mine, so long as that does not arise from the neglect or malicious conduct of the party".

    It is, I think, that passage from the judgment in Smith v. Kenrick, and not anything which was said in the speeches of Lord Cairns or Lord Cranworth in Rylands v. Fletcher, which led the reporter of the House of Lords decision in Rylands v. Fletcher to begin the headnote with the words "Where the owner of land, without wilfulness or negligence, uses his land in the ordinary manner of its use...."

    So I find nothing in Rylands v. Fletcher, or at least in its ratio decidendi, which could properly be used to justify the suggestion that the House of Lords in 1940 in the Sedleigh-Denfield case departed, consciously or unconsciously, from the law as laid down in Rylands v. Fletcher, or which was inconsistent with the extension, if it be an extension, of the Sedleigh-Denfield decision to defects naturally arising on land which constitute nuisances and give rise to damage to the land of neighbours. The House of Lords was not in 1940 precluded by earlier decisions of the House from following the Donoghue v. Stevenson approach or from holding that the neighbour in Lord Atkin's speech in Donoghue v. Stevenson included one who was a neighbour in the literal sense as being the owner of adjoining land.

    Is there, then, any subsequent authority binding on this Court which prevents it, by the doctrine of precedent, from holding that the law of England, as laid down in the Sedleigh-Denfieid case, is extended by what the Judicial Committee of the Privy Council regarded as inevitable logic?

    I do not find it necessary to consider all the numerous cases (only a fraction, be it said, of the total number of such reported cases) to which we were referred in argument, in which Rylands v. Fletcher has been discussed in relation to "natural user" of land: what is and what is not natural user, and what are and what are not "things naturally on land". It is notorious that those cases involve apparent anomalies and grave difficulties of reconciliation one with another. As Viscount Simon, with no doubt studied moderation, said in Read v. Lyons [1947] AC 156, at page 166: "I confess to finding this test of 'non-natural' user (or of bringing on the land what was not naturally there, which is not the same test) difficult to apply". Viscount Simon went on to give as an example of his difficulty the fact that in the oft-quoted passage in Mr. Justice Blackburn's judgment in the Court of Exchequer Chamber in Rylands v. Fletcher, to which I have previously referred, that learned judge "treats cattle-trespass as an example of his generalization". Yet, says Viscount Simon, "The pasturing of cattle must be one of the most ordinary uses of land...." So the pasturing of cattle is said by implication not to be a natural user of land. Yet in Wilkins v. Leighton (1932) 2 Ch 106 at page 113, Mr. Justice Luxmoore held, citing as authority Lord Justice Scrutton in St. Anne's Well Brewery Co. Ltd. v. Roberts 140 Law Times 1, that "one of the most normal uses of land" (for the purposes of the Rylands v. Fletcher doctrine) " to put buildings on it". The pasturing of cattle on land is not a normal user. The building of houses on land is one of the most normal uses. So it is said. The anomalies, if not the absurdities, of the development of this supposed doctrine, are fully discussed in the late Professor Goodhart's article "Liability for Things Naturally on Land" in (1932) 4 Cambridge Law Journal, pages 13-33, a passage from which I have already cited. That article clearly was a powerful influence on the decision of this Court in Davey v. Harrow Corporation (1958) 1 QB 60, as was mentioned at page 73 by Lord Goddard, Chief Justice, who delivered the judgment of the court, consisting of himself, Lord Justice Jenkins and Lord Justice Morris. The same article is referred to as "a formative article" in the judgment in Goldman v. Hargrave [1967] 1 AC 645 at page 662/G.

    In Davey v. Harrow Corporation it was held that the encroachment on to a neighbour's land of roots or branches of trees, causing damage, gives the neighbour an action in nuisance; and that no distinction is to be drawn between trees which may have been self-sown and trees which were deliberately planted on the land. Contrast the decision with Rouse v. Gravelworks Ltd. (1940) 1 King's Bench 489. There the defendants had dug out gravel from their land, leaving a large hole adjacent to the boundary with the plaintiff's land. It was held by this Court - a two-judge court consisting of Lord Justice Slesser and Lord Justice Goddard - that the plaintiff's claim failed because the damage to his land was caused by "natural agencies". It would seem that the decision would have been different if the water which filled the hole left by the excavation of the gravel had been brought in by pumping or perhaps even by percolation emanating from outside the defendants' land and induced by the excavation to flow into that land. If so, I should have thought that few people would regard this as a satisfactory state of the law. It may, perhaps, be arguable that, following Rylands v. Fletcher, there is some special doctrine relating to the rights of landowners to dig out coal or gravel from their land as being a "natural user". If there is no such valid distinction, then, in my judgment, the decision in Rouse v. Gravelworks Ltd. cannot stand with the decision in Davey v. Harrow Corporation. In that event I have no hesitation in preferring the later decision as stating the law as it now is: subject to the proviso that the duty arising from a nuisance which is not brought about by human agency does not arise unless and until the defendant has, or ought to have had, knowledge of the existence of the defect and the danger thereby created.

    The decision of this Court in Thomas and Evans v. Mid-Rhondda Co-operative Society (1941) 1 King's Bench 381 is not to be explained on the basis of some special principle of "naturalness" applicable to the digging of minerals on one's land. That was a case of flooding from a river. But the decision itself, it appears to me, can be justified without any inconsistency with the Goldman v. Hargrave development of the law. There was nothing whatever which could be described as unreasonable, bearing in mind that an occasional flood in the river was foreseeable, in the defendants taking down an old protective wall on their own land in order to re-build it. The fact that a flood happened to come before the wall was completely re-built could not, on any view, of itself give rise to liability on the defendants. But the potential liability of an owner or occupier to his neighbours for the overflow of a stream from his land on to their land does, indeed, give rise to consideration as to possible injustice. To this I shall return later. It is not without significance that in at least two of the cases in which it has been held that there was no liability for nuisance because of the "natural user" doctrine, it was suggested that the position at least might have been different if the owner or occupier had had knowledge of the existence of the danger. See the decision of the Privy Council in Rickards v. Lothian [1913] AC 263, at pages 281 and 282; and Wilkins v. Leighton [1932] Ch 106, at page 114, where Mr. Justice Luxmoore said: "To make the occupier liable the plaintiff must prove that he had knowledge of the existence of the nuisance".

    Lord Wright in Sedleigh-Denfield[1940] AC 880 at page 907, referred to that judgment in Wilkins v. Leighton with approval, on the basis, as I think is clear from the context, that the defendant in that case did not know of the defect. In other words, it would not have been enough to provide a defence that the user was "natural", if the defendant had had knowledge of the danger.

    I should refer briefly to two other cases on which much reliance was placed by the defendants.

    The first is Giles v. Walker (1890) 24 Queen's Bench Division 656. That was not, as Mr. Justice O'Connor thought, a decision of this Court. Despite the fact that one of the two members of the court was Lord Esher, Master of the Rolls, it was a decision of the Queen's Bench Divisional Court. It is thus not binding on us. Giles v. Walker, the thistledown case, is discussed at length in Professor Goodhart's article. In Davey v. Harrow Corporation [1958] 1 Q.B. 60, at page 72, it was said in the judgment of the court: "We think such an action" (that is, Giles v. Walker) "today....might well be decided differently". We were referred to a very careful and helpful review of the authorities in the judgment of Mr. Justice McMullin in the New Zealand Supreme Court in French v. Auckland City Corporation (1974) 1 New Zealand Law Reports 340. Mr. Justice McMuilin declined to follow Giles v. Walker. I think he was right. I think this Court should overrule Giles v. Walker.

    The second case is Pontardawe R.D.C. v. Moore-Gwynn (1929) 1 Chancery 656. Again, though the actual decision may be supported on the basis suggested as a possibility in the judgment in Goldman v. Hargrave [1967] 1 AC 645 at page 663/G, I would hold that the reasoning in support of the decision is not now good law.

    Suppose that we are not bound by Rylands v. Fletcher or any other authority to hold in favour of the defendants where the nuisance arises solely from natural forces: but suppose also that we are not bound by the decision in Sedleigh-Denfield or other binding authority to hold that there is a duty on the defendants in a case such as the present. Ought we as a matter of policy to develop the law by holding that there is a duty in a case such as the present?

    If, as a result of the working of the forces of nature, there is, poised above my land, or above my house, a boulder or a rotten tree, which is liable to fall at any moment of the day or night, perhaps destroying my house, and perhaps killing or injuring me or members of my family, am I without remedy? (Of course the' standard of care required may be much higher where there is risk to life or limb as contrasted with mere risk to property, but can it be said that the duty exists in the one case and not in the other?). Must I, in such a case, if my protests to my neighbour go unheeded, sit and wait and hope that the worst will not befall? If it is said that I have in such circumstances a remedy of going on my neighbour's land to abate the nuisance, that would, or might, be an unsatisfactory remedy. But in any event, if there were such a right of abatement, it would, as counsel for the plaintiffs rightly contended, be because my neighbour owed me a duty. There is, I think, ample authority that, if I have a right of abatement, I have also a remedy in damages if the nuisance remains unabated and causes me damage or personal injury. That is what Lord Justice Scrutton said in the Job Edwards case (1924) 1 KB 359, with particular reference to Attorney-General v. Tod Heatley (1897) 1 Chancery 560. It is dealt with also in the speech of Viscount Maugham in the Sedleigh-Denfield case [1940] AC 880 at pages 893, 894, and in the speech of Lord Atkin at pages 899, 900.

    In the example which I have given above, I believe that few people would regard it as anything other than a grievous blot on the law if the law recognises the existence of no duty on the part of the owner or occupier. But take another example, at the other end of the scale, where it might be thought that there is, potentially, an equally serious injustice the other way. If a stream flows through A.'s land, A. being a small farmer, and there is a known danger that in times of heavy rainfall, because of the configuration of A.'s land and the nature of the stream's course and flow, there may be an overflow, which will pass beyond A.'s land and damage the property of A.'s neighbours, perhaps much wealthier neighbours. It may require expensive works, far beyond A.'s means, to prevent or even diminish the risk of such flooding. Is A. to be liable for all the loss that occurs when the flood comes, if he has not done the impossible and carried out these works at his own expense?

    In my judgment, there is, in the scope of the duty as explained in Goldman v. Hargrave, a removal, or at least a powerful amelioration, of the injustice which might otherwise be caused in such a case by the recognition of the duty of care. Because of that limitation on the scope of the duty, I would say that, as a matter of policy, the law ought to recognise such a duty of care.

    This leads on to the question of the scope of the duty. This is discussed, and the nature and extent of the duty is explained, in the judgment in Goldman v. Hargrave[1967] 1 AC 645 s 663, 664. The duty is a duty to do that which is reasonable in all the circumstances, and no more than what, if anything, is reasonable, to prevent or minimise the known risk of damage or injury to one's neighbour or to his property. The considerations with which the law is familiar are all to be taken into account in deciding whether there has been a breach of duty, and, if so, what that breach is, and whether it is causative of the damage in respect of which the claim is made. Thus, there will fall to be considered the extent of the risk; what, so far as reasonably can be foreseen, are the chances that anything untoward will happen or that any damage will be caused? What is to be foreseen as to the possible extent of the damage if the risk becomes a reality? Is it practicable to prevent, or to minimise, the happening of any damage? If it is practicable, how simple or how difficult are the measures which could be taken, how much and how lengthy work do they involve, and what is the probable cost of such works? Was there sufficient time for preventive action to have been taken, by persons acting reasonably in relation to the known risk, between the time when it became known to, or should have been realised by, the defendant, and the time when the damage occurred? Factors such as these, so far as they apply in a particular case, fall to be weighed in deciding whether the defendant's duty of care requires, or required, him to do anything, and, if so, what.

    There is a passage in this part of the judgment in Goldman v. Hargrave, defining the scope of the duty, which, on the one hand, is said to be likely, if accepted, to give rise to insuperable difficulties in its practical working: and, on the other hand, is said to provide a sensible and just limitation on the scope of the duty, avoiding the danger of substantial injustice being caused, even in exceptional cases, by the existence of the duty. The passage in question, 663/F of the report, reads as follows:

    "The owner of a small property where a hazard arises which threatens a neighbour with substantial interests should not have to do so much as one with larger interests of his own at stake and greater resources to protect them: if the small owner does what he can and promptly calls on his neighbour to provide additional resources, he may be held to have done his duty: he should not be liable unless it is clearly proved that he could, and reasonably in his individual circumstances should, have done more".

    It is then suggested in the judgment that the anticipated cost of the remedies in those cases may have justified the actual decisions against liability in the Job Edwards case and in the Pontardawe case.

    What are the submissions of the parties in this case as to the. exposition in the judgment in Goldman v. Hargrave of the scope of the duty?

    The plaintiffs accept, and invite us to adopt as being the law of England, the exposition, including the passage above cited.

    The submissions of the defendants on this aspect of the case are more complex. I said early in this judgment that there was a third proposition of the defendants which I would consider later. That third proposition, as it was originally presented to us, was that, if the defendants were wrong on their first two propositions, as to the existence of the duty, then, applying the principles of Goldman v. Hargrave as to the scope of the duty, Mr. Justice O'Connor ought to have held that the quantity and cost of the work required in this case, while not going beyond the financial or other capacities of the defendants, were greater than necessary to deal with the actual damage caused to the plaintiffs. But in his speech in reply counsel for the defendants on instructions withdrew that submission. So, as I understand it, if there be a duty and if its scope is as expounded in Goldman v. Hargrave, the defendants do not now challenge the judgment against them. They say, however, that there is no duty; and one of their arguments for saying that this Court, if free to do so, should refrain from holding that such a duty exists is that an essential element of such a duty, if it is to conform with the dictates of justice, is the factor of relative financial resources; but, they say, the introduction of that factor would make the law unworkable. However, if that submission is wrong, and if there is a duty, then the defendants accept that the scope of the duty should be as expounded in Goldman v. Hargrave. I hope that I have correctly expressed the substance of the defendants' submission.

    The difficulties which are foreseen, arising out of the passage which I have quoted, include unpredictability of the outcome of litigation, delay in reaching decisions, which in everyone's interests ought to be made promptly, as to protective measures to prevent damage, and the increased complexity, length and expense of litigation, if litigation is necessary. All this, and other disadvantages, would arise, it is suggested, because the parties and their advisers, before they could form a fair and confident view of their respective rights and liabilities, and before they could safely ask the court to decide these matters, whether finally or at an interlocutory hearing, would find it necessary, or at least desirable, to put themselves in a position to ascertain and compare the respective financial resources of the parties. This might involve detailed, embarrassing and prolonged investigation, even before the stage of discovery in an action.

    If I thought that that sort of result would be likely to follow - or to follow in a substantial number or proportion of cases where this duty comes in question - I should, at least, hesitate long before accepting that this factor could be regarded as a proper factor in deciding whether the duty had or had not been broken in a particular case. But I do not think that anything of that sort is contemplated by Goldman v. Hargrave, any more than the decision of the House of Lords in Herrington v. British Railways Board [1972] AC 877 contemplated, or leads to the possibility of, a detailed examination of the financial position of the defendant landowner or occupier who is sued for damages by a trespasser who has been injured while on the defendant's land. The extent of the defendant's duty, and the question whether he has or has not fulfilled that duty, may, it is clear as a matter of English law, depend on the defendant's financial resources. See the speech of Lord Reid at page 898/H of that report. I do not believe that there was any contemplation that in such a case there would be discovery of the defendant's bank accounts or any detailed examination of his financial resources.

    So here. The defendant's duty is to do that which it is reasonable for him to do. The criteria of reasonableness include, in respect of a duty of this nature, the factor of what the particular man - not the average man - can be expected to do, having regard, amongst other things, where a serious expenditure of money is required to eliminate or reduce the danger, to his means. Just as, where physical effort is required to avert an immediate danger, the defendant's age and physical condition may be relevant in deciding what is reasonable, so also logic and good sense require that, where the expenditure of money is required, the defendant's capacity to find the money is relevant. But this can only be in the way of a broad, and not a detailed, assessment; and, in arriving at a judgment on reasonableness, a similar broad assessment may be relevant in some cases as to the neighbour's capacity to protect himself from damage, whether by way of some form of barrier on his own land or by way of providing funds for expenditure on agreed works on the land of the defendant.

    Take, by way of example, the hypothetical instance which I gave earlier: the landowner through those land a stream flows. In rainy weather, it is known, the stream may flood and the flood may spread to the land of neighbours. If the risk is one which can readily be overcome or lessened - for example by reasonable steps on the part of the landowner to keep the stream free from blockage by flotsam or silt carried down, he will be in breach of duty if he does nothing or does too little. But if the only remedy is substantial and expensive works, then it might well be that the landowner would have discharged his duty by saying to his neighbours, who also know of the risk and who have asked him to do something about it, "You have my permission to come on to my land and to do agreed works at your expense"; or, it may be, "on the basis of a fair sharing of expense". In deciding whether the landowner had discharged his duty of care - if the question were thereafter to come before the courts - I do not think that, except perhaps in a most unusual case, there would be any question of discovery as to means of the plaintiff or the defendant, or evidence as to their respective resources. The question of reasonableness of what had been done or offered would fall to be decided on a broad basis, in which, on some occasions, there might be included an element of obvious discrepancy of financial resources. It may be that in some cases the introduction of this factor may give rise to difficulties to litigants and to their advisers and to the courts. But I believe that the difficulties are likely to turn out to be more theoretical than practical. I have not heard or seen anything to suggest that the principle laid down in Herrington v. British Railways Board has given rise to difficulties in trespasser cases. If and when problems do arise, they will have to be solved. I do not think that the existence of such potential difficulties justifies a refusal to accept as a part of the law of England the duty as laid down in Goldman v. Hargrave, including the whole of the exposition as to the scope of the duty.

    As I have said, no difficulty now arises in this present appeal as regards the application of the Goldman v. Hargrave scope of the duty, once it is held that the duty exists.

    I would dismiss the appeal.

    LORD JUSTICE SHAW (Judgment read by Lord Justice Cumming-Bruce): I must confess to substantial misgivings as to what is the course which the law of England has taken in regard to the liability of a landowner for a nuisance arising upon his land independently of the intervention of any human agency. In the end, having had the advantage of reading the judgment which has just been delivered I have come to the view that, if I may respectfully say so, the weight and direction of authority support the conclusions of Lord Justice Megaw as to the incidence of liability.

    One begins by recognising that the negative proposition for which Rylands v. Fletcher (1868) 3 House of Lords 330 is generally cited as authority was not a necessary part of the decision of that appeal. On the facts which were there admitted by the defendant landowner he had used his land for what was described by the Lord Chancellor as "a non-natural use". It was not, therefore, necessary to decide what the landowner's legal position would have been if the same injury had been occasioned to the plaintiff by a natural use of the land. Accordingly, as Lord Justice Megaw has pointed out, the opinion expressed by their Lordships that a landowner would not be liable for damage resulting from a natural use of the land was obiter and does not constitute a binding authority of the House of Lords for that proposition. Nonetheless it has long been accepted as stating the law and, having regard to its source, must be treated with respect. It does not appear to me to be necessarily weakened by the anomalous exception as to encroachment by trees. This may be a consequence of the obvious fact that trees though naturally on land are living things which grow and extend root and branch and are a readily recognisable and remediable source of damage by encroachment.

    However, as is clear from the citation in the judgment of Lord Justice Megaw of the passage in Professor Goodhart's article in the Cambridge Law Journal, the view that liability for nuisance does not extend to nuisances which arise solely from a natural use of land, has been subjected to formidable criticism. Elsewhere in that article the learned author referred to the decision of a Divisional Court in Company of Proprietors of Margate Pier and Harbour v. The Town Council of the Borough of Margate (1869) 20 Law Times Reports 564, as supporting the view that an owner could be liable for nuisances arising naturally on his property. In a special and limited sense this was so: but the decision was based on a special duty created by a local Act of Parliament. It does nothing to fortify the general principle for which Professor Good-hart contended. The logical argument developed in his article is nonetheless an impressive one without the assistance of any judicial decision.

    The underlying theory of this approach is the correlation of control and responsibility. As the owner of land is normally in the best position to obviate or to contain or to reduce the effect of nuisances arising naturally on his land, he should be primarily responsible for avoiding the consequences of such nuisances or for compensating those who suffer by their occurring. This principle is reflected in the judgment of Mr. Justice McMullin in the New Zealand case of French v. Auckland City Corporation (1974) 1 New Zealand Law Reports 340.

    There are, however, so it seems to me, powerful arguments the other way. Why should a nuisance which has its origin in some natural phenomenon and which manifests itself without any human intervention cast a liability upon a person who has no other connection with that nuisance than the title to the land on which it chances to originate? This view is fortified inasmuch as a title to land cannot be discarded or abandoned. Why should the owner of land in such a case be bound to protect his neighbour's property and person rather than that the neighbour should protect his interests against the potential danger?

    The old common law duty of a landowner on whose land there arose a nuisance from natural causes only, without any human intervention, was to afford a neighbour whose property or person was threatened by the nuisance a reasonable opportunity to abate that nuisance. This entailed: (1) that the landowner should on becoming aware of the nuisance give reasonable warning of it to his neighbour; (2) that the landowner should give to the neighbour such access to the land as was reasonably requisite to enable him to abate the nuisance.

    The principle was relatively clear in its application and served in broad terms to do justice between the parties concerned. The development of "the good neighbour" concept has however blurred the definition of rights and liabilities between persons who stand in such a relationship as may involve them in reciprocal rights and liabilities.

    It has culminated in the judgment of the Privy Council in Goldman v. Harerave [1967] AC 645. Though not binding on this Court, it is of great persuasive authority. It might have been possible to explain the result of that case in the context of its special facts, for the defendant had added his own ill-directed exertions to what the forces of nature had initiated. However, the judgment of the Board is in general terms and expounds a universal principle. It is for this reason that it has to be seen as powerful support for a development in the law relating to the liability of a landowner for nuisance. It is to be observed that the judgment was expressly stated to be founded on negligence (page 657/A) and it repudiated any necessity to resolve jurisprudential distinctions between a liability arising from a duty to take care on the one hand and on the other a liability not directly related to any duty of care but absolute in its character as in the case of what is understood as nuisance in the English common law. It is readily understandable that the scope of absolute liability must be bounded by relatively narrow limits, and that it is right to hesitate before adopting a principle which will have the effect of extending those limits in the law of nuisance. I do not for myself, if I may respectfully venture to say so, see how the difficulty is disposed of by transmuting a liability in nuisance (howsoever occasioned) into a duty to do what can reasonably be done in the circumstances of a particular case to prevent or to diminish the consequences of a nuisance. This formulation may, so it seems to me, create fresh problems, and the derivative problems may defy resolution.

    However, a judgment based on general principle and given by a tribunal so commanding of attention and esteem must, even though not authoritative in the courts of England, be regarded as relating to the law of England and as being an induction of that law.

    Seen in the light of Lord Justice Megaw's analysis and exposition of the long line of cases, the judgment in Goldman v. Hargrave may represent the climax of a movement in the law of England expanding that part of the law which relates to liability for nuisance.

    It is in this attitude of mind that I find myself in agreement with the conclusions which Lord Justice Megaw has stated.

    Accordingly, albeit with diffident reluctance I would dismiss the appeal.

    LORD JUSTICE CUMMING-BRUCE: I agree with the order proposed by Lord Justice Megaw, for the reasons stated in his judgment.

    (Appeal dismissed with costs. Leave to appeal granted, the Defendants through counsel undertaking not to ask in the House of Lords for an order that their costs be paid by any other party at any stage of the proceedings. Legal Aid taxation of Respondents' costs)

Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII