The case of Palmer v Nightingale (decided by the TCC in November and published on Bailii a couple of weeks ago) was well timed as, in a sense, it was not unlike a classic mystery story. One that you would curl up with on a cold winter’s night. Fortunately no one was injured in the fire that gave rise to the case. The “victim” was a large Georgian house, which was badly damaged in an apparently spectacular conflagration – but who was the culprit?
The TCC had to cope with a variety of possible causes and there were several candidates, of which only one would lead to a guilty verdict (or, if you prefer, liability).
Palmer v Nightingale
The rather bizarre circumstances were that the defendant was a pest controller who had left bait blocks in a ceiling void not realising that they were combustible. In fact, he had no reason to think they were as the product sheets indicated otherwise, although subsequent tests showed that they were not what they seemed.
So liability was not found but, of more interest, were the court’s observations on causation if it had been (although these were of course obiter.)
The court went on to consider the proper test for causation and the result was an overwhelming rejection of the classic “Sherlock Holmes” test:
“How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?”
In other words, whenever the impossible has been eliminated, the least improbable of the remaining causes must be the operative cause.
This heresy was rejected by the House of Lords in the Popi M case in 1983 and cannot be correct because it ignores the need for the claimant to prove its case on the balance of probabilities. A cause that is unable to pass this test cannot be the basis of liability, even if it qualifies as the least improbable cause.
The practical effect of this approach is that the possible causes must be assessed and ranked in order of probability. The court can conclude that one is more probable than the others but even that cause must satisfy the balance of probability test. If that cause is a satisfactory basis for the defendant’s liability, then the claimant should succeed.
The mystery in Palmer v Nightingale was whether any of the potential causes were likely. The court concluded that the underlying cause of the fire was excessive heat generated by the claimant’s use of inappropriate halogen lamps. However, to support this finding, relevant combustible material had to be identified. There were four candidates, of which two were likely and two were unlikely. The court selected the claimant’s insulation as being the likeliest of the two causes in the frame. Of course, by this stage the claimant had already lost the case but he would have done so even if causation was still a live issue.
The “least improbable cause” theory is sometimes referred to as the “Sherlock Holmes” test. It follows the remark he made to Dr Watson in The Sign of the Four, but that’s all fiction…