House of Lords · Unanimous: Lords Bingham, Nicholls, Hoffmann, Hutton and Rodger
Facts
Each claimant had worked for two or more employers at different times and had been exposed to asbestos dust by each in breach of duty. Each later contracted mesothelioma, a cancer of the pleura which can be caused by a single fibre of asbestos but which develops many years after exposure. Medical science could not establish which exposure, or which fibre, had caused the disease, because the disease is caused by a single event and cannot be traced to one employer. The claimants sued the surviving employers.
History
The claims failed at first instance and in the Court of Appeal, because the claimants could not satisfy the ordinary but for test of causation against any single employer. The House of Lords allowed the claimants' appeals.
Issue
Whether a claimant who has been exposed to asbestos in breach of duty by more than one employer, and who has contracted mesothelioma, can establish causation against any of them where it is impossible to prove which exposure caused the disease.
Held
Appeals allowed. In the special circumstances of the case, where each employer was in breach of duty, the claimant had contracted the disease the risk of which had been created by the breach, and the science could not show which employer's breach had caused it, it was sufficient that each employer had materially increased the risk of the disease, and each was liable.
Ratio
- Where a claimant suffers a disease after being exposed to a single causative agent by successive employers in breach of duty, and cannot prove on the balance of probabilities which exposure caused it, a defendant who materially increased the risk of that disease is treated as having caused the damage.
- The exception to the ordinary but for test is a narrow one: it applies where the defendants were each in breach of a duty to protect the claimant from that very risk, the damage resulted from the risk created, and it is impossible to prove which breach caused the damage.
The judgments
Lord Bingham of Cornhill (lead)
Held that injustice would result from applying the but for test in this situation, since the claimants would be denied a remedy though each defendant had breached a duty designed to protect them from the very harm that occurred. The case should be treated as an exception, not as a general principle, and he relied on earlier authority that treated a material contribution to the risk as sufficient.
Lord Hoffmann (concurring)
Held that the conventional causal requirement was unattainable and that it was open to the court to formulate a different test of causation in a case of this kind, one that asked whether the defendant had materially contributed to the risk of the disease. It was fair to impose liability because the employers had acted in breach of a duty and the claimants had suffered the harm that duty was meant to prevent.
Lord Rodger of Earlsferry (concurring)
Reviewed comparative law and the earlier cases and concluded that a rule imposing liability on those who materially increased the risk was justified where proof of causation was impossible and the duty was one that protected against the very risk.
Lord Nicholls of Birkenhead (concurring)
Agreed that the claimants should recover and that the principle should be confined to cases in which the breaches of duty by successive employers were of the same kind and the claimant had suffered the injury, so that it was fair to treat risk creation as the basis of liability.
Lord Hutton (concurring)
Agreed with the reasoning and result and regarded the earlier authorities on contribution to risk as supporting an exception to the ordinary rule of causation.
Key reasoning
- The ordinary rule that a claimant must prove causation on the balance of probabilities would deny a remedy to everyone who contracted mesothelioma after exposure by more than one employer, although each employer had done wrong.
- Earlier decisions in which a defendant who made a material contribution to the risk or to the injury was held liable showed that the courts had in some contexts departed from a strict but for approach.
- The result is fair because each defendant was in breach of duty towards the claimant, and a defendant who has exposed the claimant to the risk cannot complain at being held liable for the risk it created.
Obiter
- Statements about the possibility that the principle might apply to other diseases or to other kinds of situation where proof of causation is impossible, which were left open and were later developed by the courts.
Where it fits in the module
In Tort (FHS), on Factual causation, Fairchild is the leading authority on departures from the but for test. It is read with McGhee v National Coal Board [1973] 1 WLR 1, Wardlaw v Bonnington Castings Ltd [1956] AC 613, Barker v Corus UK Ltd [2006] UKHL 20, the Compensation Act 2006, section 3, Sienkiewicz v Greif (UK) Ltd [2011] UKSC 10 and International Energy Group Ltd v Zurich Insurance plc UK Branch [2015] UKSC 33. Candidates must explain the limits of the exception, and whether it rests on a justification of risk creation as damage, a presumption of causation, or a pragmatic policy of doing justice to claimants who have been wronged and are blameless.
Criticisms
- The decision redefines the harm so as to turn the creation of a risk into damage, which departs from the principle that a claimant must prove that the defendant caused the actual injury, and it blurs the line between cause and risk.
- Holding each employer liable for the whole of the harm, even though only one could have caused it, may impose a disproportionate burden on a defendant who may have been blameless in fact, and the later apportionment approach in Barker shows the discomfort.
- The exception was described as narrow, but its criteria are not fixed, and later cases had to decide which diseases and which situations were within it, leading to uncertainty.
- The result depended on a policy judgment that the claimants should win, and the House did not identify a principled criterion to separate cases where policy should override the but for test from those where it should not.
Written with AI (claude-sonnet-5.5 (Claude Code writer)) on 1 November 2026, reviewed for errors by a second model (claude-sonnet-5.5 (Claude Code checker)). Read the case itself at [2002] UKHL 22, [2003] 1 AC 32. Spotted an error? Tell us.