Causation In Negligence

In a claim for negligence, the claimant must establish not only that he was owed a duty of care, and that the defendant was in breach of that duty, but that the breach was the cause of the injury or loss he suffered. There are two parts to this question of causation: causation in fact, and causation in law.

Whether the claimant’s loss was in fact caused by the defendant’s breach falls to be decided on the evidence. Typically the starting point for such a determination is the ‘but for’ test: if the loss would not have been occasioned but for the defendant’s breach, then there is a prima facie factual causation. The test for factual causation is a strict one. However egregious the breach of duty, the claim will fail if the claimant would have suffered the same loss or injury in the absence of the breach (McWilliamsVSirWilliamArrol1962). A criminal conviction in the same case is (now) admissible as evidence in civil proceedings that the defendant did cause the harm or damage alleged by the claimant. In all cases, the burden is on the claimant to show, on the balance of probabilities, that his loss or injury was caused by the defendant’s breach. In some cases, the claimant may be able to show that the principle of Res ipsa loquitur applies, but the modern view is that, even if it does, the burden of proof is unchanged.

If factual causation is established, it remains necessary to establish causation in law. In general, this requires that the damage be not so remote from the breach that it is unjust to hold the defendant liable. Clearly, the issue of remoteness is connected in some way with that of foreseeability; few judges have supported the view that a defendant should be liable for losses or injuries that were completely unforeseeable. However, the problem which has vexed the courts for centuries is how much of the actual damage experienced by the claimant, assuming that some is foreseeable, can be legally attributed to the defendant. This was apparently settled by the case of Re polemis (1921), which ruled that, once some damage of a particular meta was foreseeable, all directly-caused damage of that meta was attributable to the defendant, however remote. However, in The Wagon Mound (1961) it was held that damage that is not itself reasonably foreseeable cannot be attributed to the defendant. Wagon Mound was a Privy Council case, and therefore not binding on any UK court, but its principle has found more favour in later cases than Polemis. It can probably be assumed that the Wagon Mound principle accurately reflects the law at present, but problems remain. In particular, to what extent does the extent of injury experienced by the claimant have to be within foreseeable limits? The Wagon Mound stressed that ‘reasonably foreseeability’ was the deciding factor. But in Hughes v Lord Advocate (1963) the house of lords held that damage caused by an oil explosion was, in effect, a foreseeable consequence of leaving an oil lamp unattended in public. In reality, no-one would have foreseen such a catastrophic consequence. However, the injury suffered by the claimant — severe burns — was of the same meta as should have been foreseen (i.e., minor burns), and therefore was not too remote. What this case seems to establish is that it is the meta of injury or damage that must be reasonably foreseeable, not the extent.

The doctrine of chain of causation is important in tort, as it is in criminal law. In general, the claimant must show that his harm follows from an unbroken sequence of actions and consequences. If something breaks the chain of causation (see Novus actus interveniens, for example), then causation fails, however blameworthy the defendant. However, the question of whether the chain was broken is often a very complex and technical one. In Knightly v Johns (1981), for example, a car driver negligently caused a collision in a tunnel. The police inspector who attended the scene neglected to close off the tunnel before entering the area of the accident; instead, he directed a constable on a bicycle to leave the accident scene to stop the traffic. While cycling against the flow of traffic, the constable was involved, predictably, in a collision with a car which was being driven carefully. Who was culpable for the constable’s injuries — the original negligent motorist, or the inspector? The trial judge held the motorist liable, but the Court of Appeal decided otherwise. The reasoning was that, while no ‘natural and probable’ event which was a reasonably foreseeable consequence of the defendant’s actions would break the chain of causation, actions taken by third parties which were themselves negligent might not be reasonably foreseeable. As the inspector’s action was itself negligent, it would not have been reasonably foreseeable by the motorist. Contrast this case with, for example, the Oropesa (1943). In that case, an action which most people would believe to be highly unpredictable (the crew of a damaged but floating, ship taking to the lifeboats) was held not to break the chain of causation. No negligence was involved in the decision to abandon ship, and therefore when the sailors were drowned it was held that this did not break the chain of causation. The use of the notion of ‘chain of causation’ is often highly artificial, and — rightly or wrongly — tends to be directed towards ‘doing the right thing’ in a specific case.

Cite Term

Definition Sources