Court:

House of Lords, Judicial Committee, 3 May 2006

Topics:

Definition of the ambit of the exception to the normal balance of probabilities test of causation established by the House of Lords in Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22; [2003] 1 AC 32 ("the Fairchild exception").

Whether liability for damages under the Fairchild exception should be limited to reflect their contribution to the claimant's injury.

Cases Referred to:

Chaplin v Hicks [1911] 2 KB 786

Bonnington Castings v Wardlaw [1956] AC 613

Kitchen v Royal Air Force Association [1958] 1 WLR 563

Dingle v Asociated Newspapers [1961] 2 QB 162

McGhee v National Coal Board [1973] 1 WLR 1

Thompson v Smiths Shiprepairers [1984] QB 405

Hotson v Berkshire Area Health Authority [1987] AC 750

Wilsher v Essex Area Health Authority [1988] AC 1074

Rahman v Arearose Ltd [2001] QB 351

Hatton v Sutherland [2002] 2 All ER 1

Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22; [2003] 1 AC

Gregg v Scott [2005] UKHL 2; [2005] 2 AC 176

Facts:

Three appeals from decisions of the Court of Appeal were heard together. In the first case a workman, who had died of mesothelioma, had been exposed to asbestos during two periods of employment (one of the employers being the defendant/appellant) and one period of self-employment. The trial judge and the Court of Appeal held the defendant jointly and severally liable with the other employer and reduced the damages by 20% for the claimant's contributory negligence in failing to take care for his own safety whilst self-employed.

In the second and third cases the solvent defendants (two out of four employers and five out of many employers respectively) had been found jointly and severally liable for all damage and loss.

Legal Questions:

What are the limits of the Fairchild exception?

What is the damage caused to the claimant in a case within that exception: is it the disease (leading to joint and several liability for all losses) or creation of the risk of contracting the disease (leading to an apportionment corresponding to the proportion of the risk created?

Decision:

What are the limits of the Fairchild exception?

The exception will apply, in addition to the circumstances of the Fairchild case itself, (i) where not all sources of exposure to the causative agent were tortious (and there is no requirement that a non-tortious source of exposure was created by a person whose conduct was also a tortious source of exposure) and (ii) notwithstanding the fact that the claimant's conduct was also a source of exposure but only provided that the impossibility of proving that the defendant caused the damage arises out of "the existence of another potential causative agent which operated in the same way".

What is the damage caused to the claimant in a case within that exception?

Lord Rodger dissenting, following the decision of the majority in Fairchild, the damage sufficient to found liability was the creation of the risk of damage, even where the injury itself would be regarded as indivisible damage. The claimant should recover from each defendant damages proportionate to the risk attributable to its breach of duty.

Comments:

This decision emphasises that the Fairchild exception is to be kept within narrow bounds rather than be available as a springboard to the general imposition of liability for loss of a chance. The antipathy of the House to such a doctrine revealed in Gregg is maintained. The control mechanism in this case for the restriction of Fairchild, despite the finding that the creation of the risk of damage suffices to found liability, is the requirement that the impossibility of proof arises out of the presence of another potential cause of the damage that operates in the same way. This picks up Lord Rodger's requirement in Fairchild (at paragraph 118-119) that "the claimant must prove that his injury was caused, if not by exactly the same agency as was involved in the defendant's wrongdoing, at least by an agency that operated in substantially the same way". This requirement permits the development of liability into the fact patterns involving third (or first) parties and an admixture of tortious and non-tortious exposure without major extension of the ambit o fliability.

Thus the claimant in Wilsher would still lose - because the other four (endemic) potential causes of his injury did not operate in the same way as the doctor's negligent administration of excessive oxygen. Their Lordships were at pains to stress that the exception would have no general application to clinical negligence claims or to cases where it is impossible to tell whether cancer was been caused by exposure to asbestos or to tobacco.

It is hard to see further questions on the limits of the Fairchild exception being raised save for the possibility of semantic litigation on the question whether agents operated in "substantially the same way". Both this case and Fairchild concern injury caused by asbestos. The opinions do not expressly deal with the question whether the decisions are confined to asbestos litigation or of general application. Their Lordships give examples of claims to which the doctrine does not apply but no explicit definition of those to which it does. This may provoke further litigation to resolve the uncertainty.

The speeches on the apportionment issue reveal a curious difference between the majority who confidently assert that the basis for the decision in Fairchild was that the actionable damage was the creation of a material risk of mesothelioma and Lord Rodger who asserts with passion (in paragraphs 80-87) that the decision in Fairchild was based on proof of a material contribution to the injury sustained. Given the 4:1 majority, it may be that this question will not arise for the debate in the foreseeable future. The approach of the majority will have a far-reaching effect changing, as it does, the gist of the tort in a manner that is likely to give rise to instability. Mr Gregg might very well have welcomed compensation in proportion to the health risk created by the delay in referring him to a cancer specialist. The justification for the Fairchild special case does not obviously trump the unfairness of other fact patterns.

Once it is accepted that liability is for creation of a risk of injury, it is a short, logical and arguably fair step to impose proportionate liability on each defendant. Baroness Hale proposes that the rule imposing liability in solidum for an indivisible injury stems from conventional causation rules: once they are modified there is no illogic in permitting the creator of a risk of that injury to pay only in proportion to the risk created. The assessment of the proportion recoverable from any defendant will be undertaken by the trial judge but the basis for it will have to be decided on a case by case basis. Lord Scott suggests length of tortious exposure by each defendant and possibly the exact type of agent used and the "intensity of the exposure". It is clear that this will give rise to complexity and, unless a working arrangement is swiftly put in place, a raft of hitherto unnecessary litigation. It would be ironic, if the English courts introduced a variant of the market share doctrine familiar in some US jurisdictions by the back door.

This will introduce great uncertainty and expense into litigation that has hitherto been run with great efficiency. The difference will of course be that the costs risk and the risk of the insolvency of a putative contributor to the damages pot will fall upon the claimant rather than the defendant sued and found jointly and severally liable. The decision has provoked loud protests from the labour lobby and the Prime Minister announced on 13 June that he was looking for an opportunity to change the law to overrule the effect of the judgment. Given the definitional and other uncertainties outlined above, this may call for the greatest resourcefulness available from the drafters of any amending legislation.

Original Decision: Click here for original judgment

© Product Liability Forum, Duncan Fairgrieve, and contributors