In Re Polemis (1921) and The Wagon Mound (No. 1) (1961)

Every negligence answer has three limbs — duty, breach, damage. Remoteness sits at the far end of the third limb. A defendant has been careless; the carelessness has set off a chain of events; somewhere along that chain the law says "stop, you need not pay for this." The question is where the law draws that line, and these two cases are the two rival answers.

Polemis said: draw the line at directness. Once you are negligent at all, you pay for everything your carelessness directly produced, however astonishing. The Wagon Mound said: draw the line at reasonable foreseeability of the kind of damage. You pay only for the sort of harm a reasonable person in your position would have had in mind.

For a Mumbai LL.B. paper these two are almost always asked together — "Discuss the rules of remoteness of damage with the help of decided cases" is a standard 15-mark question, and the examiner wants the two tests, the reasons the Privy Council gave for abandoning the first, and the qualifications that have since grown up around the second. What follows takes them in that order, because the second case cannot be understood except as an answer to the first.

---

Polemis: a plank, a spark, and a ship

The Polemis dispute began as a commercial arbitration, which is why the case has its odd name. Messrs Polemis and Boyazides were Greek shipowners. They had chartered their vessel to Furness, Withy & Co. The charterparty contained the usual exception clause for fire, and it was that clause, rather than any pleading in tort, which sent the parties to arbitration.

The ship was carrying a cargo that included benzine and petrol in cases. On the voyage some of the cases leaked, and the hold filled with petrol vapour. Nobody at the time appreciated the significance of this.

At Casablanca the charterers employed Arab stevedores to discharge the cargo. To get at the cargo the stevedores had rigged a temporary staging of planks across the hatchway. While a sling of cases was being hoisted, one of the stevedores was careless with the rope; the sling caught a plank of the staging, and the plank fell into the hold.

The fall of a heavy plank into a hold is an ordinary sort of accident with ordinary sorts of consequences — a dented case, a bruised workman, a hole in the ceiling of the hold. What actually happened was that as the plank struck something on its way down it produced a spark. The spark met the petrol vapour. The vapour ignited, and the ship — the whole ship — was destroyed by fire.

The arbitrators found three things, and the whole case turns on the combination of them. First, that dropping the plank was negligent. Second, that some damage to the ship from dropping a plank was reasonably foreseeable. Third, that the spark, and therefore the fire, were not reasonably foreseeable. They awarded the owners £196,165 — the equivalent of about twenty months' hire. Sankey J. upheld the award, and the charterers appealed.

The Wagon Mound: oil in Sydney Harbour

Thirty years later, and on the other side of the world, the same problem arose out of a very similar set of facts.

On 30 October 1951 the Wagon Mound, a vessel under the appellants' control, was taking on bunkering oil at the Caltex wharf in Sydney Harbour. Through carelessness a large quantity of furnace oil was allowed to spill into the water. The ship then sailed, leaving the oil behind.

The oil spread across the harbour to the respondents' wharf, some 200 yards away. Morts Dock & Engineering were ship repairers, and at their wharf the vessel Corrimal was under repair. The repairs involved oxy-acetylene welding and cutting.

The respondents' manager, seeing the oil on the water, did the responsible thing: he stopped the welding and made enquiries as to whether it was safe to continue. He was told — and the assurance accorded with the best scientific opinion of the day — that furnace oil spread on water could not be set alight. Satisfied, he directed that work resume, with instructions that precautions be taken against inflammable material falling into the water.

For two days nothing happened. Then, on 1 November, a piece of molten metal or slag fell from the welding and landed on a piece of debris — a rag or piece of cotton waste floating in the oil — which acted as a wick. That burned long enough and hot enough to raise the oil above its flash point. The oil caught fire, and the respondents' wharf was destroyed.

Kinsella J. in the Supreme Court of New South Wales made the finding that decided the case: the destruction of the wharf by fire was a direct but unforeseeable consequence of spilling the oil, though some damage by fouling of the slipway might have been anticipated. That is Polemis on all fours. Applying Polemis, he gave judgment for the wharf owners; the Full Court affirmed; the appellants took the case to the Privy Council.

---

1. Where a defendant's careless act is negligent because some damage to the plaintiff was foreseeable, is the defendant liable for further damage of a kind that was not foreseeable, provided it was directly caused?
2. Put the other way: is the test of remoteness in negligence directness of causation, or reasonable foreseeability of the kind of damage suffered?
3. (Wagon Mound) Should Polemis be treated as good law?

---

In Polemis, counsel for the charterers tried to draw a distinction that has since become the orthodoxy. He conceded that a negligent defendant cannot complain that he did not foresee the extent of the damage — you take your victim's losses as you find them. But, he argued, the defendant is entitled to say that he could not have foreseen the type of damage. Dropping a plank risks impact damage; it does not risk incineration.

Bankes L.J. rejected this in a single sentence that is worth remembering: "I do not think that the distinction can be admitted." Once the act is negligent and the damage directly follows, the defendant's powers of anticipation are, in his words, irrelevant.

In The Wagon Mound, the appellants did not try to distinguish Polemis; on the facts as found they could not. They attacked it head-on, arguing that it was inconsistent with principle, with Donoghue v. Stevenson, and with the House of Lords' own approach in Bourhill v. Young and Liesbosch Dredger v. Edison. The respondents relied on forty years of Polemis being cited without disapproval, and on the common-sense observation of Manning J. in the Full Court that the man in the street would unhesitatingly assign the cause of the fire to the spillage of the oil.

---

Polemis: the appeal was dismissed; the award stood. Damages were not too remote. Warrington L.J. gave the cleanest statement of the resulting rule:

"The presence or absence of reasonable anticipation of damage determines the legal quality of the act as negligent or innocent. If it be thus determined to be negligent, then the question whether particular damages are recoverable depends only on the answer to the question whether they are the direct consequence of the act."

Scrutton L.J. put the same point in the words that are usually quoted in Indian textbooks:

"Once the act is negligent, the fact that its exact operation was not foreseen is immaterial."

The Wagon Mound: the appeal was allowed. Viscount Simonds, for the Board:

"The authority of Polemis has been severely shaken though lip service has from time to time been paid to it. In their Lordships' opinion it should no longer be regarded as good law."

And the sentence every examiner is waiting for:

"For it does not seem consonant with current ideas of justice or morality that for an act of negligence, however slight or venial, which results in some trivial foreseeable damage the actor should be liable for all consequences however unforeseeable and however grave, so long as they can be said to be 'direct.'"

The claim in negligence therefore failed. The Board remitted the alternative claim in nuisance to the Full Court, expressly saying nothing about it — a point students often miss.

---

The ratio of The Wagon Mound (No. 1) is that a defendant in negligence is liable only for damage of a kind that a reasonable man in his position would have foreseen; unforeseeable damage is too remote, even if it is the direct physical consequence of the negligent act, and even if the defendant is liable for some other, foreseeable damage arising from the same act.

A caution on precedent. The Wagon Mound is a decision of the Privy Council on appeal from Australia. Strictly it did not, and could not, overrule a decision of the English Court of Appeal. Its authority rests on the fact that English courts have consistently followed it since, and that the House of Lords adopted its reasoning in Hughes v. Lord Advocate and later cases. In India the Privy Council's advice has no binding force after 1950 either, but Indian courts have adopted the foreseeability test as a matter of principle. So the correct thing to write is that Polemis has been "disapproved" or "displaced", not "overruled".

Polemis vs Wagon Mound — comparison table

| | Re Polemis (1921) CA | The Wagon Mound (No. 1) (1961) PC | |---|---|---| | Test of remoteness | Directness of consequence | Reasonable foreseeability of the kind of damage | | Role of foreseeability | Relevant only to whether the act was negligent at all; irrelevant thereafter | Relevant twice — to breach, and again to remoteness | | Facts | Plank dropped into hold → spark → petrol vapour ignited → whole ship destroyed | Oil spilled during bunkering → floated to wharf → ignited by welding → wharf destroyed | | Foreseeable damage | Impact damage to ship, cargo or workmen | Fouling of the slipway | | Actual damage | Destruction by fire | Destruction by fire | | Result | Defendant liable for the whole loss | Defendant not liable in negligence | | Underlying philosophy | The wrongdoer, not the innocent plaintiff, should bear the risk of the unexpected | Liability should match moral culpability; a man answers for what he ought to have foreseen | | Practical difficulty | Forces courts into "never-ending and insoluble problems of causation" | Invites argument about how narrowly or broadly the "kind" of damage is described | | Status today | Disapproved; not good law | Good law, subject to the qualifications below |

---

Viscount Simonds' argument runs on four legs, and it is worth separating them because a good answer reproduces the structure, not just the famous sentence.

First, the moral argument. Liability in negligence rests, as Lord Atkin said in Donoghue v. Stevenson, on "a general public sentiment of moral wrongdoing for which the offender must pay." A man is blamed for failing to foresee. It follows that what he is blamed for should be measured by what he could have foreseen. To make him pay for the unforeseeable is to punish him for something that was, from his point of view, an accident.

Second, the practical argument. "Direct" is not a workable word. Viscount Simonds quoted Pollock — the lawyer "cannot afford to adventure himself with philosophers in the logical and metaphysical controversies that beset the idea of cause" — and pointed out that Polemis forces exactly that adventure, with all the tangles of novus actus interveniens attached. Foreseeability, by contrast, "corresponds with the common conscience of mankind."

Third, the argument from the cases. He tested Polemis by observing it in operation. In Liesbosch Dredger v. S.S. Edison the House of Lords had the perfect opportunity to endorse or bury Polemis, and did neither — it distinguished the case on the ground that the injuries there were the "immediate physical consequences" of the negligence. Viscount Simonds observed drily that it is not easy to understand why such a distinction should be drawn, nor where the line is to be drawn. Denning L.J.'s attempt in Roe v. Minister of Health to explain the rule with a further new word ("precipitating") was, he said, a loyal effort to enforce an unworkable rule.

Fourth, the argument from logic. This is the subtlest part and the part students most often skip. Manning J. in the Full Court had said that the man in the street would unhesitatingly assign the fire to the spillage. Viscount Simonds replied:

"After the event even a fool is wise. But it is not the hindsight of a fool; it is the foresight of the reasonable man which alone can determine responsibility."

He then destroyed the "once negligent, always liable" idea with an example. Suppose B's careless act causes similar unforeseeable damage to both A and C, but foreseeable damage to A only. On the Polemis approach B pays A for everything and pays C nothing — for no reason except an accident of what else happened to A. "A system of law which would hold B liable to A but not to C for the similar damage suffered by each of them could not easily be defended." Each claim, he said, "rests on its own bottom."

The clinching move is his insistence that there is "no such thing as liability in the air", just as there is no such thing as negligence in the air. You cannot ask whether B is liable and then ask for what damage. The liability is in respect of that damage and no other.

---

1. The Wagon Mound (No. 2) (1967) — the magnitude-of-risk qualification. Overseas Tankship (U.K.) Ltd. v. The Miller Steamship Co. Pty arose from the very same fire, but was brought by the owners of the ships being repaired at the wharf rather than by the wharf owners. This time the evidence was led differently, and the finding was that the ship's chief engineer knew, or ought to have known, that furnace oil on water could be ignited — it was possible, though very unlikely.

Lord Reid held the defendants liable. A reasonable man does not brush aside a risk merely because it is small. He weighs the magnitude of the risk against the difficulty and cost of eliminating it. Here the risk, though slight, could have been eliminated at no cost and with positive advantage — the oil was valuable and there was no justification for discharging it at all. Rule to remember: a very small risk is foreseeable in law if there was no good reason for running it. The two Wagon Mound cases are not inconsistent; they differ on the findings of fact about foreseeability.

2. Hughes v. Lord Advocate (1963) HL — kind of harm, not manner of occurrence. Post Office workmen opened a manhole in an Edinburgh street, covered it with a tent, and at teatime left it unguarded with lighted paraffin warning lamps around it. Two boys, one aged eight, took a lamp into the tent. The lamp went into the hole; there was a violent explosion, and the boy fell in and was severely burned.

The defenders argued Wagon Mound: burning by a spilt lamp was foreseeable, but an explosion was not. The House of Lords rejected the argument. Lord Pearce:

"The accident was but a variant of the foreseeable... There was thus an unexpected manifestation of the apprehended physical dangers."

And Lord Reid pointed out that the appellant's injuries "were mainly caused by burns, and it cannot be said that injuries from burns were unforeseeable."

The principle that emerges — and it is the single most useful sentence in this whole topic — is that the kind of harm must be foreseeable; the precise manner of its occurrence, and its extent, need not be.

3. Doughty v. Turner Manufacturing Co. (1964) CA — the contrast. An asbestos cement cover was accidentally knocked into a cauldron of molten sodium cyanide at about 800°C. Nothing happened at first; a minute or two later a chemical reaction between the cover and the heat caused an eruption of the molten liquid, which burned a workman standing nearby. The foreseeable risk was of a splash at the moment of immersion. The eruption was held to be damage of a different kind, and the claim failed.

Read together, Hughes and Doughty show that everything depends on how broadly the court describes the "kind" of damage. Describe it as "burning" and Doughty is wrongly decided; describe it as "splashing" and Hughes is wrongly decided. This is the central criticism of the Wagon Mound test, and a strong exam answer says so.

4. Smith v. Leech Brain & Co. Ltd. (1962) — the egg-shell skull rule survives. A galvaniser was splashed on the lip by molten metal. The burn was minor, but the tissue at that spot was in a pre-malignant condition; cancer developed and he died three years later. Lord Parker C.J. held the employers liable for the death. His reasoning is the key to reconciling the doctrine with The Wagon Mound:

"The test is not whether these employers could reasonably have foreseen that a burn would cause cancer and that he would die. The question is whether these employers could reasonably foresee the type of injury he suffered, namely, the burn."

Once the kind of injury is foreseeable, the extent is governed by the old rule that a tortfeasor takes his victim as he finds him — talem qualem. So a thin skull, a weak heart, a haemophiliac's blood, or (in Smith) a pre-cancerous lip is the defendant's bad luck. The maxim is: foreseeability governs the kind of damage; the egg-shell skull rule governs its extent.

5. The Indian position. India follows The Wagon Mound. There is no Supreme Court decision that squarely adopts it as a ratio in a two-party negligence claim, but the reasonable-foreseeability approach runs through the Indian cases, and Polemis is treated in Indian courts as discredited. In Jay Laxmi Salt Works (P) Ltd. v. State of Gujarat (1994) the Supreme Court, dealing with damage from a State-built bund, discussed foreseeability and reasonableness as the controlling ideas in remoteness. High Courts hearing motor accident and occupier's liability claims routinely apply the Wagon Mound / Hughes combination.

Two Indian qualifications are worth adding. First, in cases of absolute liability under M.C. Mehta v. Union of India (1987) the enterprise is liable for all harm resulting from the hazardous activity, and foreseeability is not a defence at all — so remoteness in the Wagon Mound sense simply does not bite. Second, in claims under the Motor Vehicles Act 1988, particularly under the no-fault provision now in s. 164, the statutory scheme replaces the common-law enquiry; the tribunal is concerned with whether the death or injury arose out of the use of a motor vehicle, not with what a reasonable driver would have foreseen.

---