[After examining the evidence and stating his view that the overflow substantially contributed to the damage sustained by both pursuers]—It remains to consider the questions of law raised on behalf of the appellants.
The question of the liability incurred by any person who interferes with a natural water-course was considered in the Court of Session in the case of Kerr v. Earl of Orkney . In that case a dam had been constructed on a stream for the purpose of collecting water, and Lord Justice-Clerk Hope makes the following observations as to the extent of the liability for damage occasioned by the escape of such water:—
“Although we did not require any answer from the respondent upon the general point of Lord Orkney's liability for the consequences of his dam bursting from a violent fall of rain, yet I think it right to state the general principle on which the view of the Court is founded. That principle is—that if a person chooses upon a stream to make a great operation for collecting and damming up the water for whatever purpose, he is bound, as the necessary condition of such an operation, to accomplish his object in such a way as to protect all persons lower down the stream from all danger: He must secure them against danger. It is not sufficient that he took all the pains which were thought at the time necessary and sufficient. They were exposed to no danger before the operation. He creates the danger, and he must secure them against danger, so as to make them as safe notwithstanding his dam as they were before. It is no defence in such a case to allege the dam would have stood against all ordinary rains—it gave way in an extraordinary and unprecedented fall of rain, which could not be expected. The dam must be made perfect against all extraordinary falls of rain—else the protection is not afforded against the operation which the party must accomplish. An extraordinary fall of rain is a matter which, in our climate, cannot be called a damnum fatale—supposing the doctrine so denoted by that term to be applicable, generally speaking, to a dam for collecting water. And the experience of the last fifteen years has shown that the increased drainage of the country brings down in heavy rains the whole water in a very short space of time, and therefore in floods of a weight, and power, and force of water quite unknown in former times. But against such a state of things the party forming such dams must completely provide, so as to secure safety to those lower down the stream. When an operation is made which involves great risk to the safety of life and of property, the condition on which alone that can be allowed which causes such risk is complete protection. A dam that gives way in a night's rain is not such as the maker was bound to erect. The fact that it gives way is a proof that his obligation was not fulfilled, and that the protection was not afforded which he was bound to provide.
“What shall be considered a damnum fatale in such a case I need not inquire, but of this I am very clear, that a great fall of rain and consequent accumulation and weight of water is not a damnum fatale which exempts the proprietor from liability for the failure of his operation—for it is against such accumulation and weight of water that he is bound to provide.”
In my opinion the Lord Justice-Clerk in that passage correctly stated the law of Scotland, and it received approval in your Lordships' House when the Earl of Orkney's case came under consideration in Tennent v. Earl of Glasgow . In that case the defender had substituted a wall for a hedge as a defence for his property. A stream burst its banks at a point above the wall, and the water descending was dammed up by the wall, which after a time gave way and considerable damage was done by the accumulated water to the lands of an inferior proprietor. In giving judgment Lord Westbury says: “My Lords, this case differs very much from those which have been cited and relied upon at the bar. If anything be done by an individual which interferes with natural occurrences, such as, for example, in Lord Orkney's case, throwing a dam across the course of a stream, it is undoubtedly the duty of that individual so to construct the work as to provide in an efficient manner, not only against usual occurrences and ordinary state of things, but also to provide against things which are unusual and extraordinary. And, therefore, the decision of the Court in the Earl of Orkney's case, where a dam gave way, was property referable to that circumstance.” Lord Chelmsford says: “This case is not at all like the case of Lord Orkney—that is, the case with respect to the dam, because there, as I have already intimated, the stream before the erection of the dam flowed harmlessly to the pursuer's mill. Lord Orkney erected a dam, by which he obstructed and headed up the course of the water. He was bound, therefore, under those circumstances—interfering with the stream, and with another person's right over the stream—to provide against every contingency. And although it was an extraordinary flood in that case which occasioned the bursting of the dam, it was one which he ought to have provided against. He ought to have made the dam capable of resisting any force which might be directed against it.”
These authorities justify the view of the law propounded by Professor Rankine in his work on the Law of Land Ownership in Scotland (4th ed., p. 376): “The sound view seems to be that even in case of an unprecedented disaster the person who constructs an opus manufactum on the course of a stream or diverts its flow will be liable in damages, provided the injured proprietor can show—(1) that the opus has not been fortified by prescription; and (2) that but for it the phenomena would have passed him scathless.” This passage, in my opinion, expresses the true view of the law applicable to this case.
The appellants contend that they are not responsible, as the injury was in the nature of damnum fatale. What amounts to damnum fatale? Its definition is given by Lord Westbury in Tennent v. Earl of Glasgow : “Under these circumstances, my Lords, what has occurred is one of those things which do not involve any legal liability—what are denominated in the law of Scotland damnum fatale occurrences—circumstances which no human foresight can provide against, and of which human prudence is not bound to recognise the possibility, and which when they do occur, therefore, are calamities that do not involve the obligation of paying for the consequences that may result from them.” Lord Cockburn expressed the same idea in a picturesque phrase used by him in Samuel v. Edinburgh and Glasgow Railway Co. when he said: “I think he is bound to provide against the ordinary operations of nature, but not against her miracles.”
In my opinion the appellants have entirely failed to establish any defence on this ground. It is true that the flood was of extraordinary violence, but floods of extraordinary violence must be anticipated as likely to take place from time to time. It is the duty of any one who interferes with the course of a stream to see that the works which he substitutes for the channel provided by nature are adequate to carry off the water brought down even by extraordinary rainfall, and if damage results from the deficiency of the substitute which he has provided for the natural channel he will be liable. Such damage is not in the nature of damnum fatale, but is the direct result of the obstruction of a natural water-course by the defenders' works followed by heavy rain.
Reliance was placed by the appellants upon the case of Nichols v. Marsland . In that case it was decided that, if the escape of water from a reservoir was due to the act of God, the person maintaining the reservoir is not liable. As Mellish, L.J., put it: “If, indeed, the damages were occasioned by the act of the party without more—as where a man accumulates water on his own land, but, owing to the peculiar nature or condition of the soil, the water escapes and does damage to his neighbour—the case of Rylands v. Fletcher establishes that he must be held liable. “The Lord Justice then goes on to decide that, if the bursting of the reservoir is due to the act of God, the liability to pay damages does not arise. Nichols v. Marsland had been tried by a jury, and the finding of the jury is thus stated by Mellish, L.J.: “The remaining question is, Did the defendant make out that the escape of the water was owing to the act of God? Now the jury have distinctly found, not only that there was no negligence in the construction or the maintenance of the reservoirs, but that the flood was so great that it could not reasonably have been anticipated, although, if it had been anticipated, the effect might have been prevented; and this seems to us in substance a finding that the escape of the water was owing to the act of God. However great the flood had been, if it had not been greater than floods that had happened before and might be expected to occur again, the defendant might not have made out that she was free from fault; but we think she ought not to be held liable because she did not prevent the effect of an extraordinary act of nature, which she could not anticipate.”
Two observations arise upon this case of Nichols v. Marsland .
The first is that the case is dealt with in the argument and judgments with reference merely to the accumulation of water in a reservoir. There is no reference to the fact that the course of a natural stream had been interfered with. The operations which had in fact been carried out are described at p. 256 of the report of the case in L. R., 10 Ex., as follows:—“A natural stream called Bagbrook, which rose in higher lands, ran through the defendant's grounds, and after leaving them flowed under the four county bridges in question. After entering the defendant's grounds the stream was diverted and dammed up by an artificial embankment into a pool of three acres in area, called ‘the upper pool,’ from which it escaped over a weir in the embankment, and was again similarly dammed up by an artificial embankment into the ‘middle pool,’ which was between one and two acres in area. Escaping over a weir in the embankment, it was again dammed up into ‘the lower pool,’ which was between eight and nine acres in area, and from which the stream escaped into its natural and original course. “This decision, having reference merely to the storage of water as in Rylands v. Fletcher, does not affect the question of liability for interference with the course of a natural stream as laid down in the authorities cited above.
Secondly, the jury had found that the damage was occasioned by the act of God, and on p. 6 of the report in 2 Ex. D. there is this note: “The question whether the rule should be made absolute for a new trial, on the ground that the verdict was against the evidence, was reserved for future discussion, if the plaintiff should desire it.” It does not appear that this question was ever again brought up for discussion in the Exchequer Chamber.
In the case now under appeal the Lord Ordinary found, and, in my opinion, rightly found, that the flood could not be regarded as in the nature of damnum fatale, and that the appellants in constructing the culvert ought to have foreseen the possibility of such an occurrence and to have provided against it. In my opinion both the appeals fail upon all points, and should be dismissed with costs.
[After stating that he agreed with the Lord Chancellor in his view of the evidence]—As to the appellants being in fact responsible for the alteration of the bed of the stream, which made what happened possible, there is no dispute. The only question that remains is whether the responsibility in fact entails a responsibility in law.
I think I am making an accurate statement when I say that the case of Kerr v. Earl of Orkney has been since its date considered by Scottish lawyers to have been well decided, and it will from henceforth enjoy the approval of the noble Lord on the woolsack, and I believe of the other noble Lords who have taken part in this appeal. Mr Constable in his address, which was equally admirable for its force and its moderation, felt that he was pressed by that case, and argued that, though the decision itself was right, the dicta in it must be regarded as modified by what had since been decided, and notably by the cases of Nichols v. Marsland and Fletcher v. Smith.
Nichols v. Marsland was, as his Lordship has pointed out, decided upon the footing of the verdict of the jury, which, as construed by the Court, amounted to a direct finding that the occurrence in question was an act of God, which is the exact equivalent of the expression used in the Scotch cases damnum fatale. Lord Justice-Clerk Hope, in Kerr v. Earl of Orkney, expressly saved the case of damnum fatale, adding that whatever might be a damnum fatale, an extraordinary fall of rain in the climate of Scotland could not be so considered. But, further, what I think makes it clear that the doctrine of act of God or damnum fatale, which was what was given effect to in Nichols v. Marsland, did not in any way weaken the authority of Kerr v. Earl of Orkney is the way in which that case was considered and treated in a subsequent case in your Lordships' House, namely, Tennent v. Earl of Glasgow . In that case the Earl of. Glasgow had built a wall along a road where a hedge had been. There was a burn which ran parallel to the road at a distance of about a quarter of a mile. The burn eventually entered beneath the road by a conduit, and an opening had been made in the wall to allow of the burn entering the conduit. There was an extraordinary fall of rain and the burn burst its banks at a place where there was a bend, invaded the road at a place far above the entrance of the conduit, and formed an accumulation behind the wall, through which it eventually burst and caused the damage complained of. Lord Westbury, L.C., spoke as follows: “If anything be done by an individual which interferes with natural occurrences, such as, for example, in Lord Orkney's case, throwing a dam across the course of a stream, it is undoubtedly the duty of that individual so to construct the work as to provide in an efficient manner, not only against usual occurrences and ordinary state of things, but also to provide against things which are unusual and extraordinary. And, therefore, the decision of the Court, in the Earl of Orkney's case, where a dam gave way, was properly referable to that circumstance. … But there was nothing which the noble defender was bound to guard against in the building of the wall along the public road … nor was the wall erected for the purpose of interfering with anything like that which has been called at the bar the course of nature. … Under these circumstances, my Lords, what has occurred is one of those things which do not involve any legal liability—what are denominated in the law of Scotland damnum fatale occurrences—circumstances which no human foresight can provide against, and of which human prudence is not bound to recognise the possibility, and which, when they do occur, therefore, are calamities that do not involve the obligation of paying for the consequences that may result from them.” And in an earlier part of his judgment he had said: “It might have been a very material thing in this case, if the injury, or the wrong as I should rather call it, sustained by the appellant could have been shown to be caused by a state of circumstances directly occasioned by the building of the wall by the noble defender over the conduit, and along the parish road, because it is clear that the natural course of the stream was down the parish road, and that the conduit provided a means of carrying the water beneath the parish road.”
It is clear that a case decided by the Inner House of the Court of Session, and afterwards approved by your Lordships' House in another Scotch case, cannot as authority be overruled or modified by a decision of the English Exchequer Chamber. But the truth is that, once it is recognised that Nichols v. Marsland proceeded on the verdict of the jury, there is no inconsistency between that case and the case of Kerr v. Earl of Orkney .
As regards the case of Fletcher v. Smith, there was again a finding of the jury which precludes any effect of it as a decision against the case of Kerr, for the jury found that the substituted watercourse was not as efficient as the old. The appellants, however, pinned their faith to the preference expressed by Lord Penzance (though he expressly declined to give a positive opinion on the matter) for the second as opposed to the third of the questions he put. These questions were: “Secondly, were they (the defendants) bound (as they, for their own convenience, were making a new and artificial watercourse) to construct it in such a manner that it would be capable of conveying off the water that might flow into it from all such floods and rainfalls as might reasonably be anticipated to happen in that locality? Or, thirdly, were they bound to make provisions for any such quantities of water as might possibly be discharged into it from any mere rainfall, however heavy, however unusual, and however contrary to all previous experience?” Now, the second proposition, as contrasted with the third, is really of no assistance to the appellants unless it is possible to extract from the phrases used a definition of what is and what is not a damnum fatale. The appellants argue that, applying Lord Penzance's test, if they can show that this rainfall was much in excess of what had been previously observed in Greenock, that is enough. I do not think that you can rightly confine your view to Greenock alone. No one can say that such rainfall was unprecedented in Scotland; and I think the appellants were bound to consider that some day Greenock might be subjected to the same rainfall as other places in Scotland had been subjected to. With deference to Lord Penzance, I think that there is no clear-cut choice in law between his two propositions, but that it always comes to a question of fact whether such-and-such an occurrence was a damnum fatale; and I hold a clear opinion that this flood was not.
I agree with the Lord Chancellor that the law is accurately stated by Professor Rankine in his book. I may perhaps add that the expression “fortified by prescription,” does not, I think, mean that the work is protected by the actual prescription statutes, but that by analogy (as such analogy has been applied in the case of servitudes) the existence of a state of things for the period of the long prescription may serve to prevent any person alleging that another state of things was the true state of nature.
The appeals, in my judgment, should be dismissed.
Lord Shaw of Dunfermline.—[After stating that he agreed with the Lord Chancellor in his view of the facts]—This being as stated, the law of the case appears to be in no way doubtful. I have never known the law of Scotland as stated in the judgment of the Lord Justice-Clerk Hope in the Earl of Orkney's case, to be questioned. On the contrary, it has been, since its date, accepted as sound. And I think it right to add, being in this fortified by the opinion of the noble and learned Lord Chancellor, that I know of no decided authority for the proposition that there is any difference on this topic between the law of England and that of Scotland.
A person making an operation for collecting and damming up the water of a stream must so work as to make proprietors or occupants on a lower level as secure against injury as they would have been had nature not been interfered with. And this is so although the water accumulated suddenly, or the rainfall was extraordinary or even unprecedented in quantity. These are the general propositions of the law. While if any help as to Scotch climatic conditions might be sought, one would get that also from the observation made by Lord Justice-Clerk Hope—and, by the way, plainly applicable in the present case: “An extraordinary fall of rain … in our climate cannot be called a damnum fatale—supposing the doctrine so denoted by that term to be applicable, generally speaking, to a dam for collecting water.”
No doubt whatsoever is thrown upon these doctrines by Nichols v. Marsland . A perusal of the judgments and procedure therein shows that it was held by a jury's finding that the disaster did, as a matter of fact, occur by a damnum fatale. I cannot, I confess, view the case as wholly satisfactory; but its conclusion was reached undoubtedly and solely by the road of settled fact—an affirmance of damnum fatale.
Such an affirmance has not been made in the present case, and, in my opinion, on its merits as well as on the guide to a proper view thereof as expressed in the outstanding authority of Kerr v. Earl of Orkney, which I have cited, such an affirmance could not be made. These occurrences arose from a heavy, it may be an extraordinary, and it may be an unprecedented, spate. That spate would have passed away harmlessly but for the appellants' operations. These operations, however, converted it into a source of harm and damage, and the appellants are thus answerable therefor.
In the words of Lord Chelmsford in Tennent—a case which in substance entirely approved of the principle of Kerr v. Earl of Orkney —“He was bound, therefore, under those circumstances—interfering with the stream, and with another person's right over the stream—to provide against every contingency. And although it was an extraordinary flood in that case which occasioned the bursting of the dam, it was one which he ought to have provided against.”
It is accordingly quite unnecessary to go into the doctrine of damnum fatale in general. I am not entirely satisfied that that expression or the equivalent expression, “the act of God,” will ever be capable of complete, exact, and unassailable definition. The nearest approach which the law has made to that point is in the judgment of Lord Westbury, L.C., in Tennent. Further, I may be allowed to express the doubt whether expressions such as those used by Lord Cockburn in Samuel v. Edinburgh and Glasgow Railway Co. as to nature's “miracles” do anything to clarify, or indeed whether they do not confuse, the issue. And I am quite clear that when in Potter v. Hamilton and Strathaven Railway Co. Lord Ardmillan supplemented his citation from Lord Westbury's judgment in Tennent's case by the observation “A party who makes a new work is bound to protect those on a lower level from extraordinary as well as ordinary accumulations of water, provided they be not such as to amount to an unprecedented event, so improbable and unnatural as could not have been reasonably anticipated,” such a gloss is not warranted by law. Its effect might be to whittle away and undermine an affirmation of the law which, without it, would be, as it was meant to be and is, broad and firm.
Lord Parker of Waddington (read by Lord Shaw of Dunfermline).—I agree. With regard to the facts, I cannot find any valid reason for dissenting from the findings of the Lord Ordinary, and Kerr v. Earl of Orkney, approved by this House in Tennent v. Earl of Glasgow, is undoubtedly the governing authority. I do not understand that the Lord Justice-Clerk in the former case intended to decide that the Scottish doctrine of damnum fatale could never have any application in cases such as that with which he was dealing, but merely that the facts before him disclosed no such damnum. If this be so, Kerr v. Earl of Orkney is not in conflict with the English authorities. Rylands v. Fletcher saved the question whether the act of God might not have afforded a defence, and this question was answered in the affirmative in Nichols v. Marsland, in which the act of God had been established by the finding of the jury, though I have some doubt whether that finding was correct. With regard to Fletcher v. Smith, it decides nothing, but I think the House was inclined to accept the view of the law which had been taken in Nichols v. Marsland, though it is true that Lord Penzance's alternatives are not very clearly stated.
Lord Wrenbury.—[After stating his opinion that the defenders' operation in filling up the natural alveus of the stream was the direct cause of the flood-water finding its way down the Inverkip Road]—The other matter upon which I will add a word “is as to the law. Numerous cases have been cited, beginning in England with Rylands v. Fletcher, and in Scotland with Samuel v. Edinburgh and Glasgow Railway Co. and Kerr v. Earl of Orkney. But in none of these was the question one as to liability for the consequences resulting from works in alveo fluminis whereby the natural alveus was filled up and the flow of water under the force of gravity thrown into a new channel at a new and higher level. The effect of the Corporation's works was that, except in so far as their culvert sufficed to take and took water coming from the westward, the Inverkip Road was substituted for the V of the valley, and became the channel by which all that water had to be drained away. In such a case the Corporation is responsible, I conceive, for resultant damage howsoever arising. The responsibility to provide a substituted channel is not limited to providing a channel sufficient to meet all demands which might reasonably be anticipated, or even all demands (in excess of the ordinary) short of the act of God. The Corporation must provide a substituted channel which will be equally efficient, happen what will. Assuming an act of God, such as a flood, wholly unprecedented, the damage in such a case results not from the act of God, but from the act of man in that he failed to provide (as there was before) a channel sufficient to meet the contingency of the act of God. But for the act of man there would have been no damage from the act of God.
The case is not that of a man who has brought a wild beast upon his land and has effectually chained it, and the chain has been broken by the act of God. That was Nichols v. Marsland . It is a case in which the act of God (if there was one) brought the wild beast, and, but for the act of man, there was a safe exit for the wild beast and it would have gone away and there would have been no injury. The act of man consisted in closing the exit, which had it remained would have rendered the advent of the wild beast harmless. To construct a reservoir on your own land is a lawful act. To close or divert the natural line of flow so as to render it less efficient is not. It has never been held that in such a case there is not liability.
Upon the facts in the Glasgow and South-Western case I do not add anything. I am satisfied that, whether their wall fell before or after the overflow of the pond, the damage resulted from the fact that the Corporation had made the Inverkip Road a sort of substituted alveus fluminis, and that the wreckage of the garage, the consequent blocking of the Glasgow and South-Western culvert, and the resultant fall of the wall, are due to that state of things.
Upon the Caledonian case I have felt much more difficulty. When the Glasgow and South-Western wall fell at 11.40 or 11.45 there was opened to the flood a new channel of ample capacity and at a much lower level, namely the Glasgow and South-Western tunnel down to the Prince's Pier. There was, I think, upon the evidence, very considerable means of access from the Inverkip Road to that new channel. The respondents have an arduous task to maintain that the fall of the Caledonian wall some half-hour later, at 12.10, was due to water coming from the westward down the Inverkip Road and not to flood-water, of which there was plenty, reaching the station square from other directions. But as your Lordships are satisfied that the evidence is sufficient to support the Caledonian case, I do not take it upon myself to differ from your Lordships' conclusion in that case.