Professional Documents
Culture Documents
The Wagon Mound No. 2 (1967)
The Wagon Mound No. 2 (1967)
ICLR: Appeal Cases/1967/Volume 1/OVERSEAS TANKSHIP (U.K.) LTD. APPELLANT AND THE MILLER
STEAMSHIP CO. PTY. AND ANOTHER RESPONDENTS ON APPEAL FROM THE SUPREME COURT OF
NEW SOUTH WALES - [1967] 1 A.C. 617
[PRIVY COUNCIL]
OVERSEAS TANKSHIP (U.K.) LTD. APPELLANT AND THE MILLER STEAMSHIP CO.
PTY. AND ANOTHER RESPONDENTS ON APPEAL FROM THE SUPREME COURT
OF NEW SOUTH WALES
1966 Feb. 28; March 1, 2, 3, 7, 8, 9, 10, 14, 15, 16, 17; May 25
LORD REID, LORD MORRIS OF BORTH-Y-GEST, LORD PEARCE, LORD WILBERFORCE and LORD
PEARSON.
Australia - New South Wales - Negligence - Oil spilled on harbour waters - Fire - Foreseeability of
consequences test of liability - Whether justifiable to neglect real risk of small magnitude - Whether
reasonable man would neglect risk.
Australia - New South Wales - Nuisance - Public - Oil spilled on harbour waters - Fire - Whether negligence
essential element in nuisance - Whether foreseeability essential element of liability in nuisance.
Damages - Remoteness - Polemis rule - Damage directly caused by negligent act - Oil spilled on harBour
waters - Fire - Foreseeability - Whether justifiable to neglect real risk of small magnitude.
Negligence - Duty of care to whom? - Foreseeability of injury - Real risk of small magnitude - Whether
justifiable to neglect - Oil spilled on harbour waters - Fire - Foreseeability of consequences test of liability -
Whether reasonable man would neglect risk.
Nuisance - Public - Oil spilled on harbour waters - Fire - Whether negligence essential element - Whether
foreseeability essential element in liability in nuisance.
The respondents had two vessels undergoing repairs at a wharf in Sydney Harbour. The owners of the wharf
in the course of doing the repairs were carrying out oxy-acetylene welding and cutting, work which was apt to
cause pieces of hot metal to fly off and fall into the sea. The appellant was charterer by demise of a vessel,
the Wagon Mound, which in the early hours of October 30, 1951, had been taking bunkering oil from another
nearby wharf. By the carelessness of her engineers a large quantity of that oil overflowed from the Wagon
Mound on to the surface of the water and drifted to and accumulated round the former wharf and the
respondents' two vessels. On November 1, 1951, at about 2 p.m., that oil was set alight. The fire spread
rapidly and caused extensive damage to that wharf and to the respondents' two vessels.
Page 2
The respondents brought this action against the appellant for damages suing alternatively in nuisance and
negligence. Walsh J.,
in the Supreme Court of New South Wales, accepted that the most probable explanation of the fire was that
a hot piece of metal fell on some object supporting a piece of inflammable material in the oil-covered water
which was ignited. He found that the damage to the respondents' vessels was "not reasonably foreseeable
by those for whose acts the appellant was responsible." Further findings were that reasonable people in the
position of officers of the Wagon Mound would regard furnace oil as very difficult to ignite upon water and
that if they had given attention to the risk of fire from the spillage they would have regarded it "as a
possibility, but one which could become an actuality only in very exceptional circumstances." Having made
these findings, Walsh J. held that liability in nuisance did not depend on forseeability and that the appellant
was liable in nuisance but not in negligence. Judgment was accordingly given for the respondents on the
claim based upon nuisance and for the appellant on the claim based on negligence. On appeal by the
appellant against the judgment based on nuisance, and by the respondents, by cross-appeal, on the issue of
negligence:-
Held, (1) that creating a danger to persons or property in navigable waters (equivalent to a highway) fell in
the class of nuisance in which foreseeability was an essential element in determining liability and that it was
not sufficient that the injury suffered by the respondents' vessels was the direct result of the nuisance if that
injury was in the relevant sense unforeseeable (post, pp. 639G - 640B, D).
Dictum of Lord Denning M.R. in Morton v. Wheeler (C.A., No. 33 of 1956, January 31, 1956 unreported)
applied.
Dicta of Devlin J. in Farrell v. John Mowlem & Co. Ltd. [1954] 1 Lloyd's Rep. 437, 440 not followed.
Per curiam. Negligence is not an essential element in nuisance. But although negligence may not be
necessary, fault of some kind is almost always necessary and fault generally involves foreseeability (post,
639C, F).
(2) That, on the facts, a reasonable man having the knowledge and experience to be expected of the
appellant's chief engineer would have known that there was a real risk of the oil on the water catching fire
and the fact that the risk was small did not in the circumstances justify no steps being taken to eliminate it
(post, pp. 643A-C, 644A-C). Accordingly both the appeal on the claim based on nuisance and the
cross-appeal upon the claim based on negligence would be allowed.
Bolton v. Stone [1951] A.C. 850; [1950] 1 T.L.R. 977; [1951] 1 All E.R. 1078, H.L.(E.) explained.
Overseas Tankship (U.K.) Ltd. v. Morts Dock & Engineering Co. Ltd. (The Wagon Mound) [1961] A.C. 388;
[1961] 2 W.L.R. 126; [1961] 1 All E.R. 404, P.C. distinguished.
Per curiam. It does not follow that, no matter what the circumstances may be, it is justifiable to neglect a risk
of small magnitude. A reasonable man would only neglect such a risk if he had some valid reason for doing
Page 3
considerable expense to eliminate the risk. He would weigh the risk against the difficulty of eliminating it
(post, p. 642E).
APPEAL (No. 7 of 1964) from the judgment of Walsh J.1 (October 10, 1963), in the Supreme Court
of New South Wales in commercial cases Nos. 3000 and 3001 of 1955. The actions out of which
this consolidated appeal and cross-appeal arose were brought by the respondents, Miller
Steamship Co. Pty. Ltd. and R. W. Miller & Co. Pty. Ltd., against the appellant, Overseas Tankship
(U.K.) Ltd., in respect of damage by fire suffered on November 1, 1951, by two ships, one owned
by the first respondent and the other owned by the second respondent. Walsh J. held1 that the
appellant was liable in nuisance, but not liable in negligence. He entered a verdict for the first
respondent in the sum of £80,000 and a verdict for the second respondent in the sum of £1,000
and judgment was entered accordingly. The appellant appealed against Walsh J.'s decision based
on nuisance and the respondents appealed against his decision based on negligence.
The facts are fully stated in the judgment of the Judicial Committee.
1966. Feb. 28, Mar. 1, 2, 3, 7, 8, 9, 10, 14, Mark Littman Q.C. and Brian Davenport for the appellant on the
appeal. There is a general principle of civil liability applicable to contract and tort that the doer of the damage
must suffer. If that principle was carried through to its logical consequence the doer would be liable for all the
results of his wrongful act. The law has not gone so far. The law imposes a limitation on the liability on the
basis that it is unfair to hold the wrongdoer liable for the consequences which are highly unpredictable and to
hold the wrongdoer responsible for what was not his fault because it was unforeseeable. The question is
"where is the line to be drawn?" The possible test is to draw the line between those consequences which are
direct and immediate and those which are remote. The test to determine the existence of a public nuisance is
whether the defendant has done an act from which it was reasonably foreseen that physical injury to the
public would result: Harrold v. Watney2; Castle v. St. Augustine's Links3; Dollman v. Hillman. Ltd.4; Read v.
Lyons.5
Walsh J. was in error in holding that the appropriate tests for determination of the question of remoteness of
damage was whether
5 [1947] A.C. 156, 170-171; 62 T.L.R. 646; [1946] 2 All E.R. 471.
the damage was a "direct" consequence of the appellant's act. In public nuisance as in other torts the
appropriate test is whether or not damage of the type in fact suffered was reasonably foreseeable.
Damages depend on the same principle throughout the law of contract and tort: The Notting Hill6; The
Argentino7. The overwhelming weight of authority is that there is a general principle governing contract and
tort that only those consequences that are foreseeable in the ordinary course of nature would be regarded as
not too remote for liability in damages. Such differences as there are in the cases on contract and on tort are
not fundamental. In the cases on contract, "natural" consequences is synonymous with "probable" or
"foreseeable." In the cases on tort the same words mean something different, "natural" means according to
the course of nature, that is, involving "hindsight" and not "foresight." The basis of the appellant's case for the
proposition that the test of remoteness of damage in public nuisance is whether it was foreseeable is Sharp
v. Powell,8 where judgment was given for the defendant in an action in public nuisance on the ground that
"the injury was not of such a character as the defendant could have contemplated as the ordinary or likely
consequence to result from his permitting his van to be washed in the public street." In Sharp v. Powell8
there was no causal connection between that which was done in breach of the statute, namely, washing the
car in the street, and that which caused the injury to the plaintiff, namely, the ice on the road. The ice on the
road was caused by a blocked drain of which the defendant had no knowledge, direct or assumed; the ice on
the road was not the direct consequence of washing the car. To be responsible for the nuisance the
defendant has either to create the nuisance or to continue the nuisance after it is already there, by conduct
which he should have foreseen would continue the nuisance. Sharp v. Powell8 was really a case of
negligence. The relevance of the case to the present issue is that no one concerned thought there was any
difference between nuisance and negligence. The close relation between nuisance and negligence is shown
by Pearson v. Cox9 and Clark v. Chambers.10
The authorities illustrative of the proper measure of damages speak of "direct" or "natural" as a test for
remoteness of damage using those words as having the equivalent meaning of
"foreseeable." The principle is thereby recognized that the rule governing remoteness of damage in tort is the
same as that prescribed for breach of contract by the first rule in Hadley v. Baxendale.11 The two rules of
Hadley v. Baxendale11 are not two separate rules but come under the umbrella of reasonable foreseeability.
Frequently, in the cases "direct" is used indistinguishably from "natural" and "probable": see Rigby v.
Hewitt12; Greenland v. Chaplin13; Lynch v. Knight14; Hoey v. Felton15; Cox v. Burbridge16; Fletcher v.
Rylands17; Smith v. Green18; Pearson v. Cox19; Clark v. Chambers20; The City of Lincoln21; In re London,
Tilbury and Southend Rly. Co.22; Hardaker v. Idle District Council23; Harrold v. Watney24; Bradley v.
Wallaces Ltd.25; Theyer v. Purnell26; Weld-Blundell v. Stephens27; Victoria Laundry (Windsor) Ltd. v.
Newman Industries Coulson & Co. Ltd.28; Cory & Son Ltd. v. France Fenwick & Co. Ltd.29; Wormald v.
Cole30; Hall v. Pimm31; The Metagama32; The Edison33; Owners of Dredger Liesbosch v. Owners of
Steamship Edison34; The Arpad35; Haynes v. Harwood.36
The concept of breach by the defendant of the duty he owes to the plaintiff is, in general, fundamental to an
action in tort for damages for physical injury, whether to the person or to chattels. This concept is therefore
fundamental to a claim for damages for such injury whether the claim be framed in negligence or in
nuisance. This principle applies with particular force to actions brought in respect of acts or omissions done
by a user of a highway or by the owner or occupier of property adjoining a highway. It is the law that a
plaintiff cannot improve his position in a highway action by framing his case in public nuisance rather than in
negligence. Haley v. London Electricity Board37;
19 2 C.P.D. 369.
20 3 Q.B.D. 327.
28 [1949] 2 K.B. 528, 539; 65 T.L.R. 274; [1949] 1 All E.R. 997, C.A.
30 [1954] 1 Q.B. 614; [1954] 2 W.L.R. 613; [1954] 1 All E.R. 683.
37 [1965] A.C. 778; [1964] 3 W.L.R. 479; [1964] 3 All E.R. 185, H.L.
Southport Corporation v. Esso Petroleum Co. Ltd.38; Letang v. Cooper39; Parish v. Judd.40
Even if there is no general rule as to foreseeability in relation to all torts, it does apply to the tort of public
nuisance. In the tort of public nuisance foreseeability enters into the measure of damages. The present case
comes within the principle of Overseas Tankship (U.K.) Ltd. v. Morts Dock & Engineering Co. Ltd.41 (Wagon
Mound No. 1). Culpability or blame or fault is an ingredient in nuisance, but it is independent of foreseeability.
In Morton v. Wheeler42, Denning M.R. said that whether a state of affairs in or near a highway constituted a
danger depended on whether the injury might reasonably be foreseen. If the state of affairs was such that
injury might reasonably be anticipated to persons using the highway, it was a public nuisance. Morris L.J.
said that when it was being considered whether an existingstate of affairs constituted a nuisance, there must
be brought into play a measure of reasonable foresight, but that did not mean that there was insurance given
to the passer-by against the occurrence of any possible mishap; all probable occurrences and consequences
must be had in mind, with appreciation and remembrance of the experiences of life; there must be a
benevolent concern for the safety of the public: see too the judgment of Parker L.J. The Court of Appeal in
Morton v. Wheeler42 clearly held that the crucial test in the case of public nuisance is whether the danger is
foreseeable. All tort depends on fault of some kind or other but liability in public nuisance is limited to the
danger which could have been foreseen.
In Morton v. Wheeler42 Denning M.R. divided public nuisance into two broad categories (1) obstructions
other than obstructions which are also dangerous: such obstructions to be actionable must be "wrongful" and
they will not be wrongful if created in the exercise of a right and are reasonable in all the circumstances; (2)
acts alleged to be a public nuisance because they produce a risk of physical harm. In so far as the appellant
created a public nuisance, such nuisance did not obstruct the public's right of passage and the nuisance was
within the second and not the first category. On the facts found the only injury to the public which
Page 8
38 [1956] A.C. 218, 225, 226, 244; [1956] 2 W.L.R. 81; [1955] 3 All E.R. 864, H.L.(E.).
39 [1965] 1 Q.B. 232, 238, 240, 241, 242, 243; [1964] 3 W.L.R. 573; [1964] 3 All E.R. 929, C.A.
41 [1961] A.C. 388, 422-423; [1961] 2 W.L.R. 126; [1961] 1 All E.R. 404, P.C.
might reasonably be foreseen was injury by pollution, and was a nuisance by pollution only. The only
reasonably foreseeable physical injury to the public, including the respondents, resulting from the spillage of
the oil was physical injury by pollution and not physical injury by fire. As no damage by fire to the
respondents' vessels was reasonably foreseeable as a result of the spillage of oil, the appellant created no
public nuisance and broke no duty owed to the respondents. The fire damage to the respondents' vessels
was not a natural or foreseeable consequence of the spillage of oil and was too remote to be recoverable in
law.
Summarising the case-law the appellant advances four different principles, each narrower than the one
preceding. First: the measure of damage in contract and tort is governed by the same rule as to remoteness
of damage to be determined by the test of foreseeability. If there was to be a stricter test in torts of strict
liability, one would expect to find it used in the law of contract, where liability is strict. Damages in contract
cases are those which follow naturally and in the ordinary course of things. In contract cases "natural" has
been defined as "normal or like or probable in the ordinary course of things," that is, those which are
"foreseeable." Secondly: Before the Polemis heresy came into the law there was one general rule of law
applying throughout torts, namely, that the rules governing remoteness of damage were governed by
questions of foreseeability as set out in Salmond on Torts, 5th ed. (1920), para. 36. The cases illustrate the
rule cited in Salmond as applicable to all torts, including torts of strict liability. Until the Polemis heresy* there
was thought to be no difference between "natural," "direct" and "probable," all of which were related to the
test of foreseeability. Thirdly: the relevant principles of remoteness that apply to the tort of public nuisance
are those of foreseeability. Throughout the cases the test of foreseeability has been applied by the courts in
considering actionsin public nuisance arising from the creation of dangers, and it was never thought that
there was any difference between negligence and nuisance in this respect. In the cases concerning
obstructions the consideration was not whether the damage was foreseeable but whether what the
defendant did was unreasonable, and foresight was not in issue. This has been a common source of
confusion because no matter how carefully one may do an act, if foreseeable harm flows from it, one is liable
in nuisance for that foreseeable harm, if the defendant's act was unreasonable in
* In re Polemis & Furness, Withy & Co. [1921] 3 K.B. 560; 37 T.L.R. 940, C.A
Page 9
all the circumstances of the case and was a nuisance in law. Fourthly: in actions for physical damage
brought against users of the highway, negligence is a necessary ingredient of the cause of action. One
cannot improve one's position by framing one's cause of action in a tort other than negligence, because the
concept of a breach of duty is fundamental to an action for physical damage.
Desmond Ackner Q.C., W. P. Ash Q.C. (New South Wales) and Ralph Gibson for the respondents on
appeal. The civil cause of action on which the respondents' claim is founded involved proof merely of the fact
of a public nuisance having been committed with the direct causation thereby to the respondents of damage
of a substantial character: Benjamin v. Storr43; Campbell v. Paddington Corporation44; Fritz v. Hobson45;
Rickett v. Metropolitan Rly. Co.46; Rylands v. Fletcher47; Harper v. Haden & Sons Ltd.48; Cunard v. Antifyre
Ltd.49
The inadmissible nature of questions as to the foreseeability of damage is more readily apparent where the
public nuisance relied upon is one which stems from an act wrongful in itself rather than an act wrongful
merely because it is unreasonable. It is affirmatively established and found (1) that the respondents suffered
special or particular damage of the nature sufficient to found the civil cause of action; (2) that public nuisance
had been established on the evidence; and (3) that the damage was directly caused by the public nuisance.
On a set of facts similar to those in the present case the board accepted in the Wagon Mound (No. 1)50 that
the fire damage was directly caused by the discharge of the oil. It was only in the light of the direct but
unforeseeable nature of this damage that it became necessary and proper for the board to consider and
overrule the cases stemming from In re Polemis Furness, Withy & Co.51
The measure of damages is not the same in contract and in tort. In contract it is fundamentally different
because in contract the parties are concerned with commercial matters. In contract damages are not
recoverable for injury to feelings as in tort, where damages are recoverable for pain and suffering, and for
defamation. In contract "natural" cause of the damage indicates "probable," but it is different in tort. It is
difficult to follow the result in tort
unless "natural" means "direct": Smith v. Green52; Cory & Son Ltd. v. France Fenwick & Co. Ltd.53; The
Arpad54; Hall v. Pimm55.
Negligence is not an ingredient of nuisance. The same set of facts can give rise to an action for either
negligence or nuisance; nevertheless the two classes of action are distinct. A nuisance comes into existence
when the facts establish that there is knowledge or assumed knowledge of the existence of the nuisance and
the defendant takes no steps to avoid the nuisance. Where a state of affairs has been created which
constitutes an excessive interference with the rightsof the public there is liability in nuisance if the excessive
interference results in damage. That is determined by the actual state of affairs and not by foreseeability.
One looks at the state of affairs and all available data and one asks the relevant question, "have the rights of
the public been materially interfered with?" It is no answer to say that it would be clearly a public nuisance to
a top scientist but not to an ordinary man. Where a public nuisance originates off the highway and causes
sufficient substantial interference of the user of the highway all that has to be established is a degree of
inconvenience in excess of what the plaintiff should be called upon to tolerate, and it is not necessary to
establish negligence; all that is required in nuisance is that the defendant had knowledge, or must be
assumed to have had knowledge, of the state of affairs, directly or indirectly, created or continued by the
defendant. If the defendant's acts produce the nuisance, it is no defence that he took the utmost care to
prevent the result: Rapier v. London Tramways Co.56 On the facts of this case therefore the appellant was
guilty of a public nuisance in discharging the oil: Castle v. St. Augustine's Links57; Job Edwards Ltd. v.
Birmingham Navigations58; Liddle v. Yorkshire (North Riding) County Council59; Sedleigh-Denfield v.
O'Callaghan60; Dollman v. Hillman Ltd.61; Southport Corporation v. Esso Petroleum Ltd.62; Esso Petroleum
Ltd. v. Southport Corporation63; British Road Services v. Slater64.
52 1 C.P.D. 92.
60 [1940] A.C. 880, 894, 896-899; 56 T.L.R. 887; [1940] 3 All E.R. 349, H.L.
Cox v. Burbridge65; Pearson v. Cox66; Clark v. Chambers67; Hardaker v. Idle District Council68; and
Harrold v. Watney69 are not conclusive on the measure of damages in public nuisance. The existence of a
liability for nuisance is irrespective of negligence: Sedleigh-Denfield v. O'Callaghan70; Read v. J. Lyons &
Co. Ltd.71; Jacobs v. London County Council72; Everitt v. Martin73; Farrell v. John Mowlem & Co. Ltd.74;
Parish v. Judd.75
Morton v. Wheeler76 decided that the defendant had not created any nuisance, because he had not caused
directly or indirectly any appreciable risk in the danger potential of the highway. The motive or purpose with
which an act was done does not determine whether or not a nuisance has been created. If it be the probable
consequence of his act, he is answerable as if it were his actual object.
Foreseeability is not a relevant ingredient in public nuisance. The test is one of direct causation, that is, all
damage consequential from the public nuisance is recoverable. There was a creation of a public nuisance by
Page 12
the defendant by allowing the oil to escape, therefore they are liable for the consequences: Trevett v. Lee.77
[Reference was made to (1961) 77 Law Quarterly Review, 180; (1962-63) Cambridge Law Journal, p. 179;
Mayne on Damages, 12th ed. (1961), p. 79; Salmon on Torts, 5th ed. (1920), p. 136: Dudley v. Smith78;
Scott v. Shepherd79; The Marpesia80; Sneesby v. Lancashire & Yorkshire Rly. Co.81; River Wear
Commissioners v. Adamson82; Benjamin v. Storr83; Quinn v. Leatham84; Clinton v. J. Lyons & Co. Ltd.85;
The London86; The San Onofre87; Wringe v. Cohen88; Yorkshire Dale Steamship Co. v. Minister of War
Transport89; Minister of Pensions v. Chennell90; Stapley v. Gypsum Mines Ltd.91;
66 2 C.P.D. 369.
67 3 Q.B.D. 327.
88 [1940] 1 K.B. 229; 56 T.L.R. 101; [1939] 4 All E.R. 241, C.A.
89 [1942] A.C. 691, 703; 58 T.L.R. 263; [1942] 2 All E.R. 6, H.L.(E.).
91 [1953] A.C. 663, 681; [1953] 3 W.L.R. 279; [1953] 2 All E.R. 478. H.L.
Pigney v. Pointers Transport Services Ltd.92; Fowler v. Lanning93; Morts Dock & Engineering Co. Ltd. v.
Overseas Tankship (U.K.) Ltd.94; Hughes v. Lord Advocate.95]
Mark Littman Q.C. in reply. The true test does not rest on apportioning consequence in relation to the
degree of blame, but it rests on what is foreseeable as a natural consequence. That test relates nuisance to
the blame-worthiness of the wrongdoer and is a more practical test than the test of the doctrine of causation.
The line is drawn by treating as recoverable that kind of damage which is the consequence of a foreseeable
wrongful act. The question is not whether the board should make new law but to state the law as it was
before the error in In re Polemis & Furness Withy & Co.96
The discharge of the oil did not cause the fire, because there was an unpredictable link which was not
foreseeable; even though the damage was caused by the discharge of the oil the consequences which have
resulted from the discharge of the oil were not foreseeable. Culpability depends on foreseeability, therefore
there is no liability in public nuisance where foreseeability for that kind of damage did not exist.
The respondents have not established direct causation. The chain of causation is not the same as in In re
Polemis & Furness Withy & Co.96 If any one of the long chain of events which culminated in the fire can be
singled out as the dominate, or proximate, or direct cause of the fire, it was the resumption on October 30,
1951, and the continuation thereafter, of the oxy-acetylene cutting and welding operations at Sheerlegs
Wharf, notwithstanding the presence of the oil underneath and around the Wharf. In law the cutting and
welding operations, either by themselves or together with the other events which culminated in the fire,
constituted novus actus interveniens. The respondents' submission that "direct" adds nothing to "cause of
damage," that all damage caused is recoverable, and that "consequential" implies only causation, is contrary
to the case law. [Reference was made to Greenland v. Chaplin97; Rigby v. Hewitt98; Hoey v. Felton99;
Lynch v. Knight100; Cox v. Burbridge101; Fletcher v. Rylands102; Sharp v. Powell103;
93 [1959] 1 Q.B. 426; [1959] 2 W.L.R. 241; [1959] 1 All E.R. 290.
95 [1963] A.C. 837; [1963] 2 W.L.R. 779; [1963] 1 All E.R. 705, H.L.(Sc.).
97 5 Ex. 243.
98 5 Ex. 240.
99 11 C.B.N.S. 142.
Pearson v.Cox104; Clark v. Chambers105; The Notting Hill106; O'Connor v. Bank of New South Wales107;
The Argentino108; In re London, Tilbury & Southend Rly. Co.109; The City of Lincoln110; Hardaker v. Idle
District Council111; Harrold v. Watney112; Cory & Son Ltd. v. France Fenwick & Co. Ltd.113; Bradley v.
Wallaces Ltd.114; Theyer v. Purnell115; Weld-Blundell v. Stephens116; Haynes v. Harwood117; Read v.
Lyons118; Victoria Laundry (Windsor) Ltd. v. Newman Industries Coulson & Co. Ltd.119.]
Desmond Ackner Q.C. for the respondents on the cross-appeal. Wagon Mound (No. 1)120 was decided on
a basis and on evidence significantly different from the evidence in this case. In the earlier case it was the
welding operations of the employees of the plaintiff dock company which is fact ignited the oil, those
operations having been continued with the express approval of responsible officers of the plaintiff dock
company. In such circumstances the plaintiff dock company necessarily presented its case on the basis that
fuel oil on water did not create foreseeable fire risk. Had the plaintiff dock company in that case done
otherwise, it would have been met with a defence of contributory negligence which, in New South Wales,
then provided a complete defence, there being no relevant statute allowing apportionment of blame.
Accordingly, it was presented on the basis in which both parties joined, namely, that fuel oil on water did not
present a foreseeable risk of fire on the claim in negligence. The plaintiff set out to establish that furnace oil
was safe and that there was no foreseeability in relation to furnace oil being set on fire when spread on
water; the case put forward was that the plaintiff stopped the oxy-acetylene plant and themselves decided to
recommence the plant and that it was by allowing gasolene or petrol to escape that the fire was caused;
alternatively, that there was some highly inflammable type of oil in the tanks which came out with the oil, and
this made that which was uninflammable inflammable. So the plaintiff achieved part of what it set out to
achieve but failed to establish that there was
Page 16
another ingredient in the oil. The crucial finding in Wagon Mound (No. 1)120 was the finding that the plaintiff
could not reasonably be expected to have known that this oil was capable of being set afire when spread on
water. The crucial finding in the present case is in two parts - (1) that the officers of the Wagon Mound
"would regard furnace oil as very difficult to ignite upon water" - not that they would regard this as impossible;
(2) that their experience would probably have been "that this had very rarely happened"- not that they would
never have heard of a case where it had happened; and (3) that they would have regarded it as a "possibility,
but one which could become an actuality only in very exceptional circumstances" - not that they could not
reasonably be expected to have known that this oil was at all capable of being set afire when spread on
water. The respondents tendered evidence from two ship's engineers as well as from ship's masters and
other experts for the purpose of establishing the foreseeability of the risk of fire. The finding in Wagon Mound
(No. 1)120 was a finding of no foreseeability at all; but the finding here is that there was a recognisable risk,
but it was one that could be properly disregarded.
The inquiry into the foreseeability of the risk involves two steps. First, whether the risk, be it great or small, is
one which was in fact, or which ought to have been, recognised. Second, if there is a recognisable risk, the
inquiry must broaden into an evaluation of whether or not the actor could, consistently with his duty of
reasonable care to avoid injury to his neighbour, properly disregard the risk. The first step does not involve
any considerations of duty on the part of the actor to third persons but only involves a consideration of his
actual or imputed state of knowledge on the bare question as to the existence or absence of any element of
risk whatever. The second step involves the consideration that differing types of situations necessitate
differing degrees of imminence being imported, and the more serious the nature of the damage involved if
the foreseeable event should materialise, the lower will be the standard of likelihood recognised as requisite
to establish negligence. The determination of negligence involves a balancing of various factors - on the one
hand the gravity and imminence of the recognisable risk, and on the other hand the utility of the challenged
conduct. The question cannot be reduced to one of mathematical probability alone, because the seriousness
of the injury risk must at the very least be taken into account as well: Fleming on Torts (Australian), 3rd ed.
(1965), p. 118.
The impact of Bolton v. Stone121 has been misunderstood. That case held that the risk was so small that in
the circumstances a reasonable man would have been justified in disregarding it and taking no steps to
eliminate it. But it does not follow that no matter what the circumstances may be, it is justifiable to neglect a
risk of such a small magnitude. A reasonable man would only neglect such a risk if he had some valid reason
Page 18
for doing so, e.g., that it would involve considerable expense to eliminate the risk. He would weigh the risk
against the difficulty of eliminating it. Bolton v. Stone121 did not alter the general principle that a person must
be regarded as negligent if he does not take steps to eliminate a risk which he knows or ought to know is a
real risk and not a mere possibility which would never influence the mind of a reasonable man. Bolton v.
Stone121 gave effect to the qualification that it is justifiable not to take steps to eliminate a real risk if it is
small and if the circumstances are such that a reasonable man, careful of the safety of his neighbour, would
think it right to neglect it. Walsh J. decided in the respondents' favour that there was a recognisable risk, but
he evaluated the risk on a mathematical basis. On the facts as found it was the ship's engineer's foresight
that basically mattered; upon Walsh J.'s findings there was no justification whatever for discharging the oil
into Sydney harbour, and, if the ship's engineer had thought about the matter, there could have been no
question of balancing the advantages and disadvantages. It was both his duty and his interest to stop the
discharge immediately. "Foreseeability" involves reasonable care. A thing is deemed to be foreseen when it
would be foreseen by a man exercising reasonable judgment. The evidence which was ultimatelyaccepted
by Walsh J. in relation to the foreseeability of the damage falls within the limits ascertainable from the cases
as being those which mark out the field of foreseeability as sufficient to amount to negligence. The particular
events necessary to cause the risk to materialise into an actual fire need not themselves be shown to be
reasonably foreseen. It is not necessary for the plaintiff to show that the precise manner in which his injuries
were sustained was reasonably foreseen. It is sufficient ifit appears that injury to a class of persons of which
he was one might reasonably have been foreseen as a consequence. The evidence shows that the
discharge of so much oil onto the water must have taken a considerable time, and a vigilant ship's engineer
would have noticed the discharge at an early stage. Walsh J.'s findings show that
121 [1951] A.C. 850; [1951] 1 T.L.R. 977; [1951] 1 All E.R. 1078, H.L.
the ship's engineer ought to have known that it is possible to ignite this kind of oil on water and that this had
in fact happened before. The most that can be said to justify inaction is that he would have known that this
could only happen in very exceptional circumstances, but that does not mean that a reasonable man would
dismiss such a risk from his mind and do nothing when it was so easy to prevent it. If it is clear that the
reasonable man would have realised or foreseen and prevented the risk, then it must follow that the
appellants are liable in negligence. [Reference was made to Smith v. Leech Brain & Co.122; Hughes v. Lord
Advocate.123]
Mark Littman Q.C. for the appellant on the cross-appeal. In the very exceptional circumstances of this case
(1) there was no reasonable foreseeability of any real risk of fire in the mind of a reasonable person; and (2)
the risk of fire was so remote that the circumstances of this case bring in novus actus interveniens. If as to
the consequences that resulted from the spillage of oil there was a remote possibility of a fire, the defendants
are not liable, if in the very exceptional circumstances of this case a reasonable man would not have thought
that possibility sufficient. The amount of risk which a reasonable man would tolerate in the circumstances
would be very small. The consequences which resulted from the spillage of oil were of a type not reasonably
foreseen.
The tests applied to determine whether or not liability will attach are stated in the cases to be whether the
consequence of the act was "not reasonably probable," "extremely unlikely to happen," "a remote possibility
not likely to happen," "the probability very slight," "a real danger," "not a serious probability," "on the cards."
[Reference was made to Bolton v. Stone124; Fardon v. Harcourt-Rivington125; Glasgow Corporation v.
Muir126.] Walsh J. made very careful findings and they should not be disturbed.
Page 19
1966. May 25. The judgment of the Board was delivered by LORD REID.
This is an appeal from a judgment of Walsh J. dated October 10, 1963, in the Supreme Court of New South
Wales in commercial
122 [1962] 2 Q.B. 405; [1962] 2 W.L.R. 148; [1961] 3 All E.R. 1159.
126 [1943] A.C. 448; 59 T.L.R. 266; [1943] 2 All E.R. 44, H.L.(Sc.).
cases by which he awarded to the respondents sums of £80,000 and £1,000 in respect of damage from fire
sustained by their vessels Corrimal and Audrey D on November 1, 1951. These vessels were then at
Sheerlegs Wharf, Morts Bay, in Sydney Harbour undergoing repairs. The appellant was charterer by demise
of a vessel, the Wagon Mound, which in the early hours of October 30, 1951, had been taking in bunkering
oil from Caltex Wharf not far from Sheerlegs Wharf. By reason of carelessness of the Wagon Mound
engineers a large quantity of this oil overflowed from the Wagon Mound onto the surface of the water. Some
hours later much of the oil had drifted to and accumulated round Sheerlegs Wharf and the respondents'
vessels. About 2 p.m. on November 1 this oil was set alight: the fire spread rapidly and caused extensive
damage to the wharf and to the respondents' vessels.
An action was raised against the present appellant by the owners of Sheerlegs Wharf on the ground of
negligence. On appeal to the Board it was held that the plaintiffs were not entitled to recover on the ground
that it was not foreseeable that such oil on the surface of the water could be set alight: Overseas Tankship
(U.K.) Ltd. v. Morts Dock and Engineering Co. (The Wagon Mound (No. 1)).1 The issue of nuisance was also
raised but their Lordships did not deal with it: they remitted this issue to the Supreme Court and their
Lordships now understand that the matter was not pursued there in that case.
Page 20
In the present case the respondents sue alternatively in nuisance and negligence. Walsh J. had found in
their favour in nuisance but against them in negligence. Before their Lordships the appellant appeals against
his decision on nuisance and the respondents appeal against his decision on negligence.
Their Lordships are indebted to that learned judge for the full and careful survey of the evidence which is set
out in his judgment.2 Few of his findings of fact have been attacked, and their Lordships do not find it
necessary to set out or deal with the evidence at any length. But it is desirable to give some explanation of
how the fire started before setting out the learned judge's findings.
In the course of repairing the respondents' vessels the Morts Dock Co., the owners of Sheerlegs Wharf, were
carrying out oxyacetylene welding and cutting. This work was apt to cause pieces or drops of hot metal to fly
off and fall in the sea. So when their
1 [1961] A.C. 388; [1961] 2 W.L.R. 126; [1961] 1 All E.R. 404, P.C.
manager arrived on the morning of October 30 and saw the thick scum of oil round the wharf he was
apprehensive of fire danger and he stopped the work while he took advice. He consulted the manager of
Caltex wharf and after some further consultation he was assured that he was safe to proceed: so he did so,
and the repair work was carried on normally until the fire broke out on November 1. Oil of this character with
a flash point of 170øF. is extremely difficult to ignite in the open. But we now know that that is not impossible.
There is no certainty about how this oil was set alight, but the most probable explanation, accepted by Walsh
J., is that there was floating in the oil-covered water some object supporting a piece of inflammable material,
and that a hot piece of metal fell on it when it burned for a sufficient time to ignite the surrounding oil.
"(1) Reasonable people in the position of the officers of the Wagon Mound would regard the
furnace oil as very difficult to ignite upon water. (2) Their personal experience would probably
have been that this had very rarely happened. (3) If they had given attention to the risk of fire
from the spillage, they would have regarded it as a possibility, but one which could become an
actuality only in very exceptional circumstances. (4) They would have considered the chances
of the required exceptional circumstances happening whilst the oil remained spread on the
harbour waters as being remote. (5) I find that the occurrence of damage to the plaintiff's
property as a result of the spillage was not reasonably foreseeable by those for whose acts the
defendant would be responsible. (6) I find that the spillage of oil was brought about bythe
careless conduct of persons for whose acts the defendant would be responsible. (7) I find that
the spillage of oil was a cause of damage to the property of each of the plaintiffs. (8) Having
regard to those findings, and because of finding (5), I hold that the claim of each of the
plaintiffs. framed in negligence, fails."
Page 21
Having made these findings Walsh J. went on to consider the case in nuisance. There is no doubt that the
carelessness of the appellant's servants in letting this oil overflow did create a public nuisance by polluting
the waters of Sydney Harbour. And also there can be no doubt that anyone who suffered special damage
from that pollution would have had an action against the appellant. But the special damage sustained by the
respondents was caused, not by pollution, but by fire. So, having held in finding (5) that risk of fire was not
reasonably foreseeable, Walsh J. had to consider whether foreseeability has any place in the determination
of
[1967] 1 A.C. 617 Page 634
liability for damage caused by nuisance. He made an extensive survey of the case-law and said3 that the
principles which he found there
"suggest that a plaintiff may set up a case, depending upon the following steps. The defendant
has committed a 'wrongful' act in that it has created a public nuisance by polluting the harbour
waters with oil. As a result of the presence of that 'nuisance' (i.e., of the oil) the plaintiff has
suffered damage over and above that suffered by others. This gives the plaintiff an action,
subject only to proof that there is the requisite relationship between the presence of that
nuisance and the injury, so that it can be said that the injury suffered was direct. It matters not
that the injury was different in kind from a fouling of the ship by the polluted waters."
Then, coming to the words used by the judges in numerous cases of nuisance, he said that3
"by and large, the judgments are not expressed in terms of the concept of foreseeability. The
term used again and again is 'direct.' It is true that other expressions are also used, but one
does not find, in express terms, any testing of the matter by what the defendant might have
contemplated or might have foreseen"
"I do not find in the case law on nuisance, up to the time of The Wagon Mound5 decision any
authority for the view that liability depends on foreseeability."
Their Lordships must now make their own examination of the case-law. They find the most striking feature to
be the variety of words used: and that is not very surprising because in the great majority of cases the facts
were such that it made no difference whether the damage was said to be the direct or the natural or the
probable or foreseeable result of the nuisance. The word "natural" is found very often and it is peculiarly
ambiguous. It can and often does mean a result which one would naturally expect, i.e., which would not be
surprising: or it can mean the result at the end of a chain of causation unbroken by any conscious act, the
result produced by so-called natural laws however surprising or even unforeseeable in the particular case.
Another word frequently used is "probable." It is used with various shades of meaning. Sometimes it appears
to mean more probable than not, sometimes it appears to include events likely but not very likely to occur,
sometimes it has a still wider meaning and refers to
4 Ibid. 433.
events the chance of which is anything more than a bare possibility, and sometimes, when used in
conjunction with other adjectives, it appears to serve no purpose beyond rounding off a phrase.
Their Lordships must refer to a number of cases on which Walsh J. relied because they require that the
damage suffered by the plaintiff must be the direct or immediate result of the nuisance (generally obstruction
of a highway), and they make no reference to foreseeability or probability. But that is because they were
dealing with quite a different matter from measure of damages.
"by the common law of England, a person guilty of a public nuisance might be indicted; but, if
injury resulted to a private individual, other and greater than that which was common to all the
Queen's subjects, the person injured had his remedy by action" (per Brett J. in Benjamin v.
Storr6.
So the first step is to decide whether the plaintiff has suffered what may for brevity be called special damage.
The authorities on this matter are numerous and exceedingly difficult to reconcile. But one thing is clear.
There have been excluded from the category of special damage many cases where the damage suffered by
the plaintiff was clearly caused by the nuisance: it was not only foreseeable but probable and indeed the
inevitable result of the nuisance - the obstruction by the defendant of a highway giving access to the
plaintiff's premises. The words "direct and immediate" have often been used in determining whether the
damage caused by the nuisance is special damage. Benjamin v. Storr6 affords a good example. The
defendants' vans were constantly standing in the street outside the plaintiff's coffee house. They intercepted
the light to his windows so that he had to burn gas nearly all day, they obstructed access by his customers,
and the stench from the horses was highly objectionable. The damage caused to the plaintiff by this
obstruction of the highway was obvious, but that was not enough. Brett J. said7: "It is not enough for him to
show that he suffers the same inconvenience in the use of the highway as other people do." Then he cited
two cases in which the plaintiffs, who had clearly suffered damage as a result of obstruction, failed because
they were unable to show that they had suffered any injury other and different from that which was common
to all the rest of the public. And then he said8
7 Ibid. 406.
8 Ibid. 407.
"Other cases show that the injury to the individual must be direct, and not a mere consequential
injury; as, where one way is obstructed, but another (though possibly a less convenient one) is
left open; in such a case the private and particular injury has been held not to be sufficiently
direct to give a cause of action."
But he held that in the case before him there was "a particular, a direct and a substantial damage."
Such cases have nothing to do with measure of damages: they are dealing with the entirely different
question whether the damage caused to the plaintiff by the nuisance was other and different from the
damage caused by the nuisance to the rest of the public. When the word "direct" is used in determining that
question, its meaning or connotation appears to be narrower than when it is used in determining whether
damage is too remote, so their Lordships do not propose to deal further with cases determining what is and
what is not special damage. No one denies that the respondents have suffered special damage in this case
within the meaning of these authorities. The question is whether they can recover notwithstanding the finding
that it was not foreseeable.
Of the large number of cases cited in argument there were few in which there was separate consideration of
the proper measure of damages for nuisance. Many of the cases cited deal with the measure of damages for
breach of contract and their Lordships will later explain why they do not propose to examine these cases.
And a larger number were cases based purely on negligence in which there was no element of nuisance.
Their Lordships do not intend to examine these cases in detail. It has now been established by The Wagon
Mound (No. 1)9 and by Hughes v. Lord Advocate10 that in such cases damages can only be recovered if the
injury complained of was not only caused by the alleged negligence but was also an injury of a class or
character foreseeable as a possible result of it. So it would serve no useful purpose in this case to examine
the grounds of judgment in earlier cases of negligence. In so far as they are ambiguous they mustnow be
interpreted in light of these two cases: in so far as they exclude foreseeability they must be taken to be
disapproved: and in so far as they take account of foreseeability they do no more than amplify the grounds of
judgment in these two cases. The respondents can only succeed on this branch of the case by distinguishing
nuisance from negligence either because the authorities indicate that foreseeability
10 [1963] A.C. 837; [1963] 2 W.L.R. 779; [1963] 1 All E.R. 705.
In Sharp v. Powell11 the defendant's servant washed his van in a public street which was an offence. Owing
to a severe frost and a block in a drain the water did not get away but spread over the street some distance
away and became a sheet of ice on which the plaintiff's horse slipped and was injured. The case was laid in
nuisance. Bovill C.J.12 decided against the plaintiff because "the defendant could not reasonably be
expected to foresee that the water would accumulate and freeze at the spot where the accident happened."
Keating J. said13: "The damage in question, not being one which the defendant could fairly be expected to
Page 24
anticipate as likely to ensue from his act, is in my judgment too remote." And the judgment of Grove J. was to
the same effect. It may be that today the defendant's servant would be expected to be more wide awake and
observant but given the finding of fact regarding foreseeability the rest followed. Counsel for the appellant
argued that this was really a case of negligence. The relevance of the case to the present issue is that no
one concerned thought that there was any difference in this respect between nuisance and negligence.
The close relation between nuisance and negligence is shown by Pearson v. Cox.14 Workmen were working
in a house. "straightedge" was balanced on a plank: one of the men shook it and the tool fell on the plaintiff,
who was passing along the street. The case was laid in negligence but Bramwell L.J. dealt with it as a case
of nuisance. The ground of judgment was expressed by Brett L.J.15 "The accident was highly improbable
and a man need not guard against highly improbable accidents." Again it may be that a higher standard of
care is required today.
Clark v. Chambers16 an adjoining occupier placed chevaux de frise in the street to restrict passage.
Someone moved one of these barriers onto the pavement and at night a foot passenger came in contact with
it and was injured. Obstructing the street was clearly a nuisance. The main controversy was whether the
intervention of the stranger moving the barrier was novus actus interveniens but the plaintiff succeeded.
Again there was no suggestion of any difference between negligence and nuisance. Cockburn C.J. doubted
whether foreseeability came in at all but held that if it did this was foreseeable.
12 Ibid 259.
13 Ibid 261.
In Harrold v. Watney17 a child climbed onto a defective fence beside a road. It gave way and the child was
injured. The fence was held to be a nuisance and the action succeeded. Vaughan Williams L.J. said18:
"when asking one's self if the nuisance was the cause of the accident one gets a test in this
way: ought what the child did have been present to the mind of the person who created the
nuisance as a probable result of his act?"
Page 25
The only case cited where there is an express statement that liability does not depend on foreseeability is
Farrell v. John Mowlem & Co. Ltd.19 where the defendant had without justification laid a pipe across a
pavement and the plaintiff tripped over it and was injured. Devlin J. held this to be a nuisance. He said20: "I
think the law still is that any person who actually creates a nuisance is liable for it and for the consequences
which flow from it whether he is negligent or not." That is quite true, but then he added21: "It is no answer to
say I laid the pipe across the pavement but I did it quite carefully and I did not foresee and perhaps a
reasonable man would not have foreseen that anybody would be likely to trip over it." That case was before
The Wagon Mound (No. 1)22 and it may be that Devlin L.J. thought that the rule was the same in negligence:
or it may be that he thought that there was a different rule for nuisance. He cites no authority.
In their Lordships' judgment the cases point strongly to there being no difference as to the measure of
damages between nuisance and negligence but they are not conclusive. So it is desirable to consider the
question of principle.
The appellant's first argument was that damages depend on the same principles throughout the law of tort
and contract. This was stated emphatically by Lord Esher in The Notting Hill23 and in The Argentino24 and it
has often been repeated. But the matter has not been fully investigated recently. There has in recent times
been much development of the law of tort and developments in the law of contract may not have proceeded
on parallel lines. To give but one example, it is not obvious that the grounds of decision of the House of
Lords in Hughes v. Lord Advocate25 are consistent with the first rule in Hadley v. Baxendale26 as that rule is
uncommonly interpreted. But it is unnecessary to pursue this
18 Ibid. 325.
20 Ibid. 440.
21 Ibid. 440.
question in this case and therefore their Lordships do not intend to examine cases arising out of breach of
contract.
The next argument was that at all events the measure of damages is the same throughout the law of tort. But
there are many special features in various kinds of tort, and again their Lordships do not find it necessary to
make the extensive investigations which would be required before reaching a conclusion on this matter.
Comparing nuisance with negligence the main argument for the respondent was that in negligence
foreseeability is an essential element in determining liability and therefore it is logical that foreseeability
should also be an essential element in determining the amount of damages: but negligence is not an
essential element in determining liability for nuisance and therefore it is illogical to bring in foreseeability
when determining the amount of damages. It is quite true that negligence is not an essential element in
nuisance. Nuisance is a term used to cover a wide variety of tortious acts or omissions and in many
negligence in the narrow sense is not essential. An occupier may incur liability for the emission of noxious
fumes or noise although he has used the utmost care in building and using his premises. The amount of
fumes or noise which he can lawfully emit is a question of degree and he or his advisers may have
miscalculated what can be justified. Or he may deliberately obstruct the highway adjoininghis premises to a
greater degree than is permissible, hoping that no one will object. On the other hand the emission of fumes
or noise or the obstruction of the adjoining highway may often be the result of pure negligence on his part:
there are many cases (e.g., Dollman v. Hillman27 where precisely the same facts will establish liability both
in nuisance and in negligence. And although negligence may not be necessary, fault of some kind is almost
always necessary and fault generally involves foreseeability, e.g., in cases like Sedleigh-Denfield v.
O'Callaghan28 the fault is in failing to abate the nuisance of the existence of which the defender is or ought
to be aware as likely to cause damage to his neighbour. (Their Lordships express no opinion about cases
like Wringe v. Cohen29, on which neither counsel relied.) The present case is one of creating a danger to
persons or property in navigable waters (equivalent to a highway) and there it is admitted that fault is
essential - in this case the negligent discharge of the oil.
29 [1940] 1 K.B. 229; [1939] 4 All E.R. 241; 56 T.L.R. 101, C.A.
Page 27
"But how are we to determine whether a state of affairs in or near a highway is a danger? This
depends, I think, on whether injury may reasonably be foreseen. If you take all the cases in the
books, you will find that if the state of affairs is such that injury may reasonably be anticipated
to persons using the highway it is a public nuisance" (per Lord Denning M.R. in Morton v.
Wheeler30.
So in the class of nuisance which includes this case foreseeability is an essential element in determining
liability.
It could not be right to discriminate between different cases of nuisance so as to make foreseeability a
necessary element in determining damages in those cases where it is a necessary element in determining
liability, but not in others. So the choice is between it being a necessary element in all cases of nuisance or
in none. In their Lordships' judgment the similarities between nuisance and other forms of tort to which The
Wagon Mound (No. 1)31 applies far outweigh any differences, and they must therefore hold that the
judgment appealed from is wrong on this branch of the case. It is not sufficient that the injury suffered by the
respondents' vessels was the direct result of the nuisance if that injury was in the relevant sense
unforeseeable.
It is now necessary to turn to the respondents' submission that the trial judge was wrong in holding that
damage from fire was not reasonably foreseeable. In The Wagon Mound (No. 1)31 the finding on which the
Board proceeded was that of the trial judge: "the defendant did not know and could not reasonably be
expected to have known that [the oil] was capable of being set afire when spread on water." In the present
case the evidence led was substantially different from the evidence led in The Wagon Mound (No. 1)31 and
the findings of Walsh J. are significantly different. That is not due to there having been any failure by the
plaintiffs in The Wagon Mound (No. 1)31 in preparing and presenting their case. The plaintiffs there were no
doubt embarrassed by a difficulty which does not affect the present plaintiffs. The outbreak of the fire was
consequent on the act of the manager of the plaintiffs in The Wagon Mound (No. 1)31 in resuming
oxy-acetylene welding and cutting while the wharf was surrounded by this oil. So if the plaintiffs in the former
case had set out to prove that it was foreseeable by the engineers of the Wagon Mound31 that this oil could
be set alight, they might have had difficulty in parrying the reply that this must also have been foreseeable by
their manager. Then there would have been contributory negligence and at that
The crucial finding of Walsh J. in this case is in finding (5): that the damage was "not reasonably foreseeable
by those for whose acts the defendant would be responsible." That is not a primary finding of fact but an
Page 28
inference from the other findings, and it is clear from the learned judge's judgment that in drawing this
inference he was to a large extent influenced by his view of the law. The vital parts of the findings of fact
which have already been set out in full are (1) that the officers of the Wagon Mound31 "would regard furnace
oil as very difficult to ignite upon water" - not that they would regard this as impossible; (2) that their
experience would probably have been "that this had very rarely happened" - not that they would never have
heard of a case where it had happened, and (3) that they would have regarded it as a "possibility, but one
which could become an actuality only in very exceptional circumstances" - not, as in The Wagon Mound (No.
1),31 that they could not reasonably be expected to have known that this oil was capable of being set afire
when spread on water. The question which must now be determined is whether these differences between
the findings in the two cases do or do not lead to different results in law.
In The Wagon Mound (No. 1)31 the Board were not concerned with degrees of foreseeability because the
finding was that the fire was not foreseeable at all. So Lord Simonds had no cause to amplify the statement
that the "essential factor in determining liability is whether the damage is of such a kind as the reasonable
man should have foreseen."32 But here the findings show that some risk of fire would have been present to
the mind of a reasonable man in the shoes of the ship's chief engineer. So the first question must be what is
the precise meaning to be attached in this context to the words "foreseeable" and "reasonably foreseeable."
Before Bolton v. Stone33 the cases had fallen into two classes: (1) those where, before the event, the risk of
its happening would have been regarded as unreal either because the event would have been thought to be
physically impossible or because the possibility of its happening would have been regarded as so fantastic or
farfetched that no reasonable man would have paid any attention to it - "a mere possibility which would never
occur to the mind
32 Ibid. 426.
33 [1951] A.C. 850; [1951] 1 T.L.R. 977; [1951] 1 All E.R. 1078, H.L.
of a reasonable man "(per Lord Dunedin in Fardon v. Harcourt-Rivington34 - or (2) those where there was a
real and substantial risk or chance that something like the event which happens might occur, and then the
reasonable man would have taken the steps necessary to eliminate the risk.
Bolton v. Stone35 posed a new problem. There a member of a visiting team drove a cricket ball out of the
ground onto an unfrequented adjacent public road and it struck and severely injured a lady who happened to
be standing in the road. That it might happen that a ball would be driven onto this road could not have been
said to be a fantastic or far-fetched possibility: according to the evidence it had happened about six times in
28 years. And it could not have been said to be a far-fetched or fantastic possibility that such a ball would
strike someone in the road: people did pass along the road from time to time. So it couldnot have been said
that, on any ordinary meaning of the words, the fact that a ball might strike a person in the road was not
foreseeable or reasonably foreseeable - it was plainly foreseeable. But the chance of its happening in the
Page 29
foreseeable future wasinfinitesimal. A mathematician given the data could have worked out that it was only
likely to happen once in so many thousand years. The House of Lords held that the risk was so small that in
the circumstances a reasonable man would have been justified in disregarding it and taking no steps to
eliminate it.
But it does not follow that, no matter what the circumstances may be, it is justifiable to neglect a risk of such
a small magnitude. A reasonable man would only neglect such a risk if he had some valid reason for doing
so, e.g., that it would involve considerable expense to eliminate the risk. He would weigh the risk against the
difficulty of eliminating it. If the activity which caused the injury to Miss Stone had been an unlawful activity,
there can be little doubt but that Bolton v. Stone35 would have been decided differently. In their Lordships'
judgment Bolton v. Stone35 did not alter the general principle that a person must be regarded as negligent if
he does not take steps to eliminate a risk which he knows or ought to know is a real risk and not a mere
possibility which would never influence the mind of a reasonable man. What that decision did was to
recognise and give effect to the qualification that it is justifiable not to take steps to eliminate a real risk if it is
small and if the circumstances are such that a reasonable man,
careful of the safety of his neighbour, would think it right to neglect it.
In the present case there was no justification whatever for discharging the oil into Sydney Harbour. Not only
was it an offence to do so, but it involved considerable loss financially. If the ship's engineer had thought
about the matter, there could have been no question of balancing the advantages and disadvantages. From
every point of view it was both his duty and his interest to stop the discharge immediately.
It follows that in their Lordships' view the only question is whether a reasonable man having the knowledge
and experience to be expected of the chief engineer of the Wagon Mound would have known that there was
a real risk of the oil on the water catching fire in some way: if it did, serious damage to ships or other property
was not only foreseeable but very likely. Their Lordships do not dissent from the view of the trial judge that
the possibilities of damage36 "must be significant enough in a practical sense to require a reasonable man to
guard against them" but they think that he may have misdirected himself in saying
"there does seem to be a real practical difficulty, assuming that some risk of fire damage was
foreseeable, but not a high one, in making a factual judgment as to whether this risk was
sufficient to attract liability if damage should occur."
In this difficult chapter of the law decisions are not infrequently taken to apply to circumstances far removed
from the facts which gave rise to them and it would seem that here too much reliance has been placed on
some observations in Bolton v. Stone37 and similar observations in other cases.
Page 30
In their Lordships' view a properly qualified and alert chief engineer would have realised there was a real risk
here and they do not understand Walsh J. to deny that. But he appears to have held that if a real risk can
properly be described as remote it must then be held to be not reasonably foreseeable. That is a possible
interpretation of some of the authorities. But this is still an open question and on principle their Lordships
cannot accept this view. If a real risk is one which would occur to the mind of a reasonable man in the
position of the defendant's servant and which he would not brush aside as far-fetched, and if the criterion is
to be what that reasonable man would have done in the circumstances, then surely he would not neglect
such a risk if action to eliminate it
In the present case the evidence shows that the discharge of so much oil onto the water must have taken a
considerable time, and a vigilant ship's engineer would have noticed the discharge at an early stage. The
findings show that he ought to have known that it is possible to ignite this kind of oil on water, and that the
ship's engineer probably ought to have known that this had in fact happened before. The most that can be
said to justify inaction is that he would have known that this could only happen in very exceptional
circumstances. But that does not mean that a reasonable man would dismiss such a risk from his mind and
do nothing when it was so easy to prevent it. If it is clear that the reasonable man would have realised or
foreseen and prevented the risk, then it must follow that the appellant is liable in damages. The learned
judge found this a difficult case: he says that this matter is38 "one upon which different minds would come to
different conclusions." Taking a rather different view of the law from that of the judge, their Lordships must
hold that the respondents are entitled to succeed on this issue.
The judgment appealed from is in the form of a verdict in favour of the respondents upon the claim based
upon nuisance, a verdict in favour of the appellant on the claim based upon negligence, and a direction that
judgment be entered for the respondents in the sums of £80,000 and £1,000 respectively. The result of their
Lordships' findings is that the direction that judgment be entered for the respondents must stand but that the
appeal against the verdict in favour of the respondents and the cross-appeal against the verdict in favour of
the appellant must both be allowed.
Accordingly, their Lordships will humbly advise Her Majesty that the appeal and the cross-appeal should be
allowed and that the judgment for the respondents in the sums of £80,000 and £1,000 should be affirmed.
The appellant must pay two-thirds of the respondents' costs in the appeal and cross-appeal.