To what difficulties had the use of a 'but-for' test of factual causation in the law of tort given rise? Have the courts resolved these difficulties in a satisfactory manner?

Tort Essay 2 Neda Eslamian

To what difficulties had the use of a ‘but-for’ test of factual causation in the law of tort given rise? Have the courts resolved these difficulties in a satisfactory manner?

Introduction

The purpose of the but-for test is to prove, as a matter of fact, that D’s action was a cause of the damage. In order to establish this we must ask; ‘but-for’ D’s negligence, would the damage have occurred anyway? The methodology behind it is the use of a balance of probabilities i.e. if the chance of the damage occurring anyway was over 50%, then there is no right to compensation. The simplicity of this test is not only an asset but also a fault. In certain situations complications arise. Often we are not so sure who is to blame for the damage, or whether the role that they played was a substantial factor in the resulting consequences. There are 3 problem areas that have caused extensive discussions, and in each individual area the court has dealt with the problems differently, sometimes producing a fair and satisfactory outcome, and at other times leaving the law completely inconsistent. These three problem areas are;

The problem of ‘loss of a chance’
The problem of ‘what should have been done’ versus ‘what would have been done’
The problem of multiple causation

(Each will be dealt with individually below).

The test was first demonstrated in Barnett v Chelsea & Kensington Hospital Management Committee where the hospital sent an ill man home having failed to realise that he was in fact poisoned. He later died. The pivotal question was; would he have died ‘but-for’ the negligence shown by the hospital, and the judge concluded in the affirmative; even if the deceased had been examined and treated with the proper care, the probability of him dying anyway was over 50%, and it would have been impossible to save his life by the time he arrived at the hospital.

The House of Lords’ decision in Hotson v East Berkshire Health Authority expanded on this, where Lord Bridge said that you cannot apportion damages; either you pass the test of the balance of probabilities or not i.e. is it over 50% likely that the negligence caused the damage? If so, then the claimant will be awarded his compensation. On the other hand, if it was under 50%, then the claimant is awarded nothing. It cannot be allowed that if it was 25% likely that it was the fault of the defendant, that 25% of the complete compensation can be awarded.

1) Problem of the ‘loss of a chance’

Furthermore, the latter case gave rise to the discussion of the ‘loss of a chance’. This can be viewed as a way of the claimant getting round the rigidity and all or nothing nature of the but-for test. This is argued as follows. The facts of this case were that a boy was taken to hospital after an accident, where the doctors failed to see any damage to the hip. The boy was sent home, but a few days later returned to the same hospital due to extreme pain. But it was too late for an operation at this point, which leads to him getting an illness resulting in the loss of mobility in his hip. The claimants argued that they are suing for the loss of a chance of a better medical outcome, which would have been achieved through a prompt diagnosis and correct medical treatment. This argument was rejected too. Lord Mackay said that the reason for this is that when the patient arrived at hospital it was too late already and so there was no loss of a chance to lose. But what if there is a case where there is no such certainty that it was too late – surely here there are legitimate grounds for compensation. My opinion is that the courts are reluctant to begin allowing the claimants to win on these grounds, because naturally once they win they expect some sort of compensation, but it is extremely difficult to give a monetary value to the chance which has been lost. It has been suggested that it could be done in the following way; if there was a 40% chance of preventing the loss of the mobility to the hip, then 40% of the damages that would have been awarded had it been the fault of the hospital that the mobility was lost. But if we do this, then we completely undermine the nature of the but-for test. Even if we did allow this mode of compensation, another difficulty arises. If it is more likely than not that the hospital is at fault, (regardless of if it is 51% likely, or 100%) then the claimant is awarded 100% of the damages. Whereas the person who had a 40% chance of preventing the loss only gets 40% of the damages.

One can therefore see why the courts are reluctant to begin allowing compensation for the loss of a chance, but I am of the opinion that there is deficiency in the court’s method; surely a defendant would rather get some compensation rather than none what-so-ever.

This is a preview of the whole essay

There is a little more leniency in commercial cases, like Allied Maples v Simmonds & Simmonds where the plaintiff sued the solicitors for a failure to advise regarding the deletion of a warranty, and this lead to great economic losses. The court decided that a substantial enough chance was lost to allow compensation. All that the plaintiff had to do was prove that he would have taken the advice, not that if the advice was given, it could have protected against the loss. Thus it has been accepted that where there has been an economic loss, and a duty of care was owed to the person that suffered the loss, then claims will be allowed for loss of a chance.

This result is quite disturbing since it appears as though the courts are giving greater protection to economic loss than personal injury, which is a morally inept outcome that should be amended. Thus I stand by my proposition that the courts should begin allowing compensation for the loss of a chance in non-commercial cases too, since if courts can guess the value of the loss of mobility in the hip, then there is no reason why they cannot value the loss of a chance.

2) The problem of ‘what should have been done’ versus ‘what would have been done’

Another problem arises in cases where the defendant alleges she would have done one thing, while the claimant alleges she should have done another. The but-for test here, though its application is not doubted, lacks a certain dimension. It simply stops at asking ‘would the damage have occurred but for the defendant’s negligence’. But in some cases the defendant goes on to say that even if they did not act negligently, then the outcome would have been the same as if they did act negligently. Thus the negligence would not have made a difference to the damages, and so the defendant would argue that there are no grounds for compensation. This occurred in Bolitho where the defendant’s negligence was that she did not respond to her pager in time, therefore a patient of hers died. The only way to save the patient was to insert a tube into her throat, which the defendant would not have done, even if she had responded in time to her pager. Where the but-for test would result in no compensation (since had she not been negligent, the damages would have occurred anyway), the House of Lords stated that this was not the end of the inquiry, and one should ask, what should she have done i.e. should she have inserted the tube?

The outcome here is a lot fairer on the defendant. It prevents an outcome of no compensation being awarded for a doctor of insufficient medical knowledge. Therefore the courts have resolved a potential difficulty in the single-faceted nature of the ‘but-for’ test.

3) Problem of multiple causation

This is where most of the complications arise, since we no longer have 2 different outcomes e.g. the damage being the result of the fall or the negligence of the medics. Now there are a variety of causes to the damage that occurred, the result of which was partly D’s fault.

In Smith New Court Securities Ltd v Scrimgeour Vickers Ltd doubt in the but-for test is expressly stated; ‘Although it often yields the right answer, [it] does not always do so’ – this demonstrates how the but-for test, though it is a useful basic rule, can lead to results that defy common sense. Take the hypothetical example that 2 negligently caused fires merge into one and burn the claimant’s house down. Following the but-for test, one negligent person that started the fire can argue that but-for his negligence the house would have been burnt down anyway due to the negligence of the other person that started the fire, as well as visa-versa, such that that the claimant is left with no right to compensation from either party. This is clearly unjust and the fault lies in the but-for test.

This complication also arises where there is a combination of the defendant’s negligent act and some other innocent cause e.g. in Bonnington v Castings Ltd v Wardlaw where the claimant contracted a lung disease from the combination of dust which the defendant had created in breach of safety regulations and other dust which was the inevitable accompaniment of the activity itself. The House of Lords could no longer ask whether the damage would have occurred but for the defendant’s negligence, since the answer to this is so abstract that it is impossible to tell; the disease was due to the accumulation of dust over a number of years and whether the innocent cause was more to blame or the negligent one was a question that literally could not be answered. So Lord Reid created a new test where the courts would ask; did the dust from the negligent source materially contribute to the illness? The meaning of ‘material’ here is very broad – anything outside de minimis was accepted. Thus in this case liability was upheld.

McGhee v National Coal Board was about a claimant that caught dermatitis and sued his employer. The negligent act on the part of the employers was to provide showers after work, thus he had to go home before he could wash off the dust he had been exposed to. Although medical evidence was able to prove that the dust was the cause of the illness, and that the delay in showering increased the likeliness of the illness, it did not say whether it was the prolonged exposure that caused the dermatitis or a single rogue particle. None-the-less the appeal was allowed on the grounds that there is no difference between materially contributing to the risk and materially contributing to the illness. This outcome compromises the orthodox requirement of causation set out in Bonnington.

Of the two decisions so far, my opinion is that McGhee was a more reasonable decision. Otherwise cases would begin to arise where the claimant would be awarded to compensation for lack of medical proof that the particular defendant was the cause. The obvious argument against this is that we are being too harsh on the defendants, but one must bear in mind the all important policy considerations; where the claimant is one person of unfortunate circumstance, the defendant is often a company, and thus is more likely to have the financial capacity to provide compensation. Furthermore the courts are not saying that automatically, where there is vagueness as to whom the caused the disease, then compensation will be provided. It is also saying that the negligence must have been a material contribution to the risk. Also if Bonnington was to be left as the authority, then there would be no way of resolving what Fleming calls ‘indeterminate defendant’ problem e.g. 2 people shoot towards X negligently and one hits him, but X cannot prove which one. Bonnington would grant no compensation to the claimant since we do not know who materially contributed to the risk, but McGhee would award substantial compensation since both shooters materially contributed to the risk.

Unfortunately Wilshire v Essex AHA overturned the decision in McGhee. The House of Lords said that it was insufficient to simply apply McGhee since we need to prove on the balance of probabilities that his negligence caused the damage. Furthermore they amended the decision in McGhee, in the sense that they effectively reinvented history and said that McGhee was an evidential question and that what they said regarding risk was just related to that particular case and that it was never meant to be generalised at all. This is obviously untrue, but never the less we are taken back to the original conservative-natured approach of Bonnington.

But the contradictions in decisions do not end there. In Fairchild, McGhee is resurrected. This case involved asbestos causing a disease where it was hard to tell whether it was a cumulative exposure to blame for the disease, or one rogue particle. The Court of Appeal simply took the conservative approach and said that the claimant had to prove who it was that caused the disease (i.e. which employer), and if he cannot, then there is to be no compensation. The House of Lords however, adopted the approach in McGhee and said that the defendants will be to blame if they have materially contributed to the risk.

One cannot leave the discussion here, since Fairchild has not overturned Wilshire at all. Thus we are left with 2 contradictory precedents. This is a serious deficiency in the law and is most definitely an unresolved problem that the courts have not dealt with satisfactorily.

Conclusion

One can see that this area of the law in general contains many problems and inconsistencies. The but-for test is too simple to deal with these problems.

When discussing the loss of a chance, we were left with authority that gave more favourable outcomes to commercial cases than personal injury cases. Obviously this does not reflect well on the courts, and so one must argue that here the law is definitely unsatisfactory. My opinion is that the apportioning of damages should be permitted for the loss of a chance as explained above since claimants will prefer some compensation over none.

Regarding situations where the defendant would have done something, even if they had not been negligent, that would have lead to the same damage that their negligence created, the courts have been fairer to the claimants. Their action will not fail where they have a doctor that would have made an incorrect or medically unethical decision had he not been negligent. Thus it can be said that the courts have resolved this particular difficulty in a satisfactory manner.

Finally, where there are cases of multiple causes, the law is left in its most serious state of dissatisfaction. In fact the outcome is such that we have two precedents, one of which is the complete antithesis of the other. There is no doubt, that quite objectively, the courts have not resolved any difficulty here.

Similar case arose in Fitzgerald v Lane