Introduction to Tort Law

View with images and charts

Introduction to Tort Law


Negligence as a tort is a breach of a legal duty to take care which causes damage to the claimant that is not too remote.

Duty of care:

In an attempt to find a general test to determine whether a duty of care exists Lord Atkin in Donoghue v Stevenson set out – a duty is owed to persons who are so closely and directly affected by defendant’s act/omission that he ought reasonably to have them in contemplation as being so affected. He described such people as defendant’s ‘neighbors’, accordingly his formulation came to be known as the neighbor principle.

There were many developments in the law of negligence in the years following that decision. These led Lord Wilberforce to redefine the neighbour principle. He turned it into a two-stage test in Anns v Merton LBC.

The first question was, whether there was a sufficient relationship of proximity or neighbourhood such that in the reasonable contemplation of the defendant carelessness on his part may be likely to cause damage to the claimant. If so, a prima facie duty of care arose. The second question was, whether there were any considerations which ought to negative or reduce or limit the scope of the duty of care, or the class of persons to whom it was owed.

However, this test came under criticism in the following years as being too expansive and indeed the Anns case was itself overruled in 1991.

The basic test for a duty of care is now the one set down in Caparo v Dickman. The claimant has to show three things if there is to be a duty of care –

It was reasonably foreseeable that the person in the claimant’s position would be injured. There was sufficient proximity between the parties. It is just, fair and reasonable to impose liability.

Note, this test is to be applied only when the court is presented with a new factual situation which is significantly different from, or wider in scope than, any of the previously decided cases.

Where a duty of care has been imposed in an earlier analogous case a court will be more inclined to impose a duty on the facts of the case before it unless it is satisfied that the analogy is not properly drawn. Conversely, where the precedents are against the recognition of a duty of care a court will be reluctant to impose a duty unless it considers that the earlier cases are not directly in point; or it is convinced that reasoning incrementally and by reference to the precedents, an extension of the scope of liability is justified. Thus, it appears that the tripartite test could be ousted in circumstances where an incremental step beyond existing authorities could be taken.

However, the courts have developed more detailed and more restrictive rules applying in certain type of cases –

Where the damage caused is psychiatric, rather than physical.

Where the damage caused is purely economic loss.

Where the damage was caused by a failure to act.

Where the damage was caused by a third party.

Where the defendant falls within a range of groups who have become subject to special rules on policy grounds.

Scope of the duty of care: Even if the defendant owed a duty of care to some people, there remains the question of whether the particular claimant was within the scope of that duty. In order for a duty to exist, it must be reasonably foreseeable that damage or injury would be caused to the particular claimant in the case, or to a class of people to which he or she belongs, rather than just to people in general – (Bourhill v Young, Palsgraf v Long Island).


A person breaches his duty of care if he does something which a reasonable person would not do, or if he does not do something which a reasonable person would do – (Blyth v Birmningham). Accordingly, if the defendant causes loss or injury to the claimant, but is able to show that he acted in a way that a reasonable person would have acted, no liability will arise – (Al-Sam v Atkins).

Lord Macmillan in Glasgow v Muir clarified – ‘the standard of foresight of the reasonable man is an impersonal test. It eliminates the personal equation and is independent of the idiosyncrasies of the particular person whose conduct is in question’. However, Lord Macmillan did consider the test contains a subjective element in it. According to him, the standard of the reasonable man needs to be applied, to the facts of the case and to what the reasonable man in the position of the particular defendant ought to have foreseen.

As the test is objective, the particular defendant’s own characteristics are usually ignored. However, there are a limited number of situations in which special characteristics of the defendant will be taken into account.

Professionals and special skill: Where someone holds himself out as having special skills or qualification, he is required to show the skill normally possessed by person doing that work – (Bolam v Friern: it was held that, a doctor failing to diagnose a disease cannot excuse himself by showing that he acted to the best of his skill, if a reasonable doctor would have diagnosed it). In Nettleship v Weston, it was irrelevant that the learner driver defendant was doing as well as she could, since a reasonable driver in her position would have done better.

It is immaterial if defendant does not in fact have that skill but yet engages himself in conduct usually done by person having that skill, the standard demanded is still that of those who actually do possess that skill – (Adams v Rhymney Valley). However, where someone has not held himself out as having special skill but who does in fact have that skill, he will not be in breach if his conduct falls below the standard of a reasonable person with that skill – (Wooldridge v Sumner).

In Shakoor v Situ, it was held that a practitioner of Chinese herbal medicine does not have to meet the standard of skill and care of a reasonably competent practitioner of orthodox medicine, he only have to meet the standard applicable to his practice; but he does have to take account of the reports relating to his practice in orthodox journals.

Interestingly in Phillips v Whiteley, it was argued that a jeweler who had pierced the claimant’s ears have to show the care of a reasonable surgeon. The court rejected this argument and held that, he would be judged by the standard of a reasonably competent jeweler carrying out that particular task.

Medical negligence: In assessing the standard of care to be expected in areas where the defendant is exercising special skill or knowledge, the courts have accepted that within a profession or trade there may be differences of opinion as to the best techniques and procedures in any situation. This type of situation frequently arises in case of medical practice.

Faced with a particular patient, one doctor might recommend surgery, but another might recommend treatment with drugs. The court do not insist that one of these approached must be right and the other wrong. What has to be shown is that, the defendant has acted in a way that would be supported by ‘a body of respectable medical opinion’. This is known as the Bolam test, since it was set out in that case. Note, it is no objection that only a small number of doctors would have supported the defendant’s conduct – (Defreitas v O’Brien).

It is then for then for the court to decide whether the medical experts opinion is reasonable, considering whether they had weighed up the risks and benefits and had a logical basis for their conclusion – (as explained in Bolitho v City and Hackney HA). In Bolitho it is suggested that, the fact that medical experts held a particular view, would in fact demonstrate its reasonableness; and it would only be in very rare cases that a court would reject such a view as unreasonable. Thus, it will rarely be necessary for a court to find that the views held by a competent medical expert are unreasonable, but it is nevertheless possible.

Duty to explain: Previously, it was sufficient to show that the standard of disclosure practiced by a doctor in respect of potential risk of damage associated to his treatment conformed to a practice of disclosure sanctioned by responsible medical opinion, as seen in Sidaway v Bethlam. This decision has been criticised for not recognizing a patient’s right to make informed consent and determining the sufficiency of disclosure from the doctor’s point of view. A different approach was adopted by the Australian court in Rogers v Whittaker – the standard of disclosure to be met is the one, which a reasonable patient would want to know.

Accordingly, the HL in Chester v Afshar stated, doctors not only have a duty to take reasonable steps to make sure their advice is right, but also to explain the advantages and disadvantages of a particular treatment, unless there is a good reason for not doing so.

Chester is also an interesting case on causation. In this case, the claimant did not show that she would probably (i.e. balance of probabilities) never have had the operation had the information been given; but she did show that she would have taken her time and consulted friends and therefore would not have had the actual operation on the particular day that she did have it. Surprisingly, this was held sufficient to establish causation.

In most cases, it will be that the individual doctor (or nurse, etc) is negligent, and that the health authority or private hospital is vicariously liable. However, in some cases it may be appropriate to consider arguing whether the health authority or hospital is negligent, e.g. by entrusting a procedure to an inappropriately junior doctor, or by overworking its staff so that they are too tired and make mistakes.

Illness: Where the defendant’s conduct is affected by some kind of illness or other external factor, this will be taken into account while assessing the standard of care to be met – (Mansfield v Weetabix: Here the driver of a lorry was suffering from a disease which on the day in question caused hypoglycemic state (a condition in which the blood sugar falls so low that the brain’s efficiency becomes temporarily impaired). This affected his driving, with the result that he crashed into the claimant’s shop. The driver did not know that his ability to drive was impaired, and there was evidence that he would not have continued to drive if he had known. The CA said that the standard by which he should be measured was that of a reasonably competent driver who was unaware that he suffered from a condition which impaired his ability to drive; on this basis he was found not to be negligent. But the result would have been different if the driver had known that his ability to drive was impaired).

Knowledge: The defendant must be judged in the light of the state of scientific, technological or other expert knowledge which was available to him at the time of the alleged breach, rather than any latest advancement – (Roe v Minister of Health: medical witness in the case said that until the man’s accident occurred, keeping the ampoules in disinfectant was a standard procedure and there was no way of knowing that it was dangerous; it was only the injuries to the claimant that had revealed the risk. Therefore, the defendant was held not liable).

Special knowledge concerning the claimant: Where the defendant knows or ought to have known about particular characteristics of the claimant which increase the risk of harm, then more will be required of the defendant – (Haley v London Electricity Board: Here the HL held that a body conducting operations on a city highway should foresee that blind persons would walk along the pavement, and that it owes a duty to take those precautions reasonably necessary to protect them from harm. On the facts it was held liable although a person with normal visi8on would not have been injured in consequence of its operations).

Children: Where the defendant is a child, the standard of care is that of an ordinary careful and reasonable child of that age – (Mullin v Richards).

Some factors extraneous to the defendant is also considered to set the standard of care to be met in a particular case:

Likelihood of harm: Although harm to someone may be foreseeable, the risk of that harm being inflicted may be so unlikely that the defendant will not be required to take any precautions against it happening – (Bolton v Stone).

In Bolton v Stone, Miss Stone was hit by a cricket ball struck from a cricket ground surrounded by a fence 17 feet above the level of the square. The batsman was 80 yards away. The ball was only the 6th in about 30 years to be hit out of the ground. The HL held that there had been no breach of duty by the club in allowing cricket to be played without taking further precautions.

Lord Radcliffe stressed two points: first, the fact that the ball had to clear the fence which itself was a remote possibility and, secondly, that having cleared the fence the ball would then have to strike a passer-by, an even more remoter possibility. Lord Oakley said ‘an ordinarily careful man does not take precautions against every foreseeable risk…… life would be almost impossible if he were to attempt to take precautions against every risk’.

By contrast in Hilder v Associated Portland, the risk of injury to a road user from a football being kicked from a patch of open land was much greater, thus the defendant responsible for the land was held to be in breach of a duty of care.

Gravity of potential harm: Where the consequence of the harm which is likely to be done if the risk materialises is serious, then more will be required of the defendant – (Paris v Stepney BC: The claimant, who had one eye, was employed as a mechanic in the defendants’ garage. Part of his job involved welding. It was not normal to supply goggles to men involved in such work. A piece of metal flew into the claimant’s eye with the result that he became completely blind. The defendants were held liable, although they would not have been liable to person with normal sight. The greater risk to the claimant meant the greater precautions than normal, had to be taken).

Practicality of protection: If the cost of eliminating the risk is out of proportion to the benefit obtained by its elimination, then it will not generally be negligent to fail to eliminate the risk – (Latimer v AEC Ltd: Flooding had occurred in a factory owned by the defendants, following an unusually heavy rainfall. This had left patches of the floor very slippery. The defendants had covered some of the wet areas with sawdust, but had not had enough to cover all of them. The claimant, a factory employee, was injured after slipping on the uncovered area, and sued alleging that the defendants had not taken sufficient precautions; in view of the danger, they should have closed the factory. The HL agreed that the only way to eradicate the danger was to close the factory, but held that given the level of risk, particularly bearing in mind that the slippery patches were clearly visible, such an onerous precaution would be out of proportion. The defendants were held not liable).

Potential benefit of the risk taking: Where the defendant takes some risks, his risk taking may be outweighed by the potential benefit his conduct is to achieve, then he will not be negligent by taking the risk – (Watt v Hertfordshire CC: The claimant was a firefighter. He was among others called to the scene of an accident where a woman was trapped under a car; a heavy jack was needed to rescue her. The vehicle in which the fire officers travelled to the scene was not designed to carry the jack, and the claimant injured when it slipped. He sued his employers, but the court held that the risk taken in transporting the jack was outweighed by the need to get there quickly in order to save the woman’s life).

It is important to appreciate that an emergency does not exonerate the defendant from displaying any level of care at all, it merely reduces the standard demanded – (Ward v London CC).


In order to establish negligence, it must be proved that the defendant’s breach of duty actually caused the damage suffered by the claimant, and that the damage caused was not too remote.

Causation is relevant to all torts in which proof of damage is essential, though very often it is discussed in the context of negligence.

The basic rule of causation is set out as ‘but for’ test. The test may be stated as: ‘if the damage would not have happened but for a particular fault, then that fault is the cause of the damage; if it would have happened just the same, fault or no fault, the fault is not the cause of the damage’ (per Denning LJ in Cork v Kirby). The burden of proof lies on the claimant. He has to show on the balance of probabilities that the damage would not have occurred but for the defendant’s breach of duty.

The application of this test can be seen in Barnett v Kingston and Chelsea Hospital: The defendant breached his duty towards the claimant by failing to examine him. However, he was not held liable, as there was evidence that, even if he had been examined it was too late for any treatment to save him.

In Bolitho v City and Hackney HA, the defendant was in breach of her duty by failing to attend a patient, but there was evidence that the action she in fact would have taken if she had attended, would not have saved the patient. Accordingly, she was not held liable.

In Mcwilliams v Sir William Arrol, the defendant employer failed to provide safety equipment for his employee, but there was evidence that the deceased employee would not have used it even if it had been provided. Accordingly, the defendant was not held liable.

Interestigly, in Chester v Afshar, the surgeon had advised the claimant to undergo surgery but in breach of duty failed to advice her of the risk. The claimant did not show that she would probably never have had the operation, but she did show that she would have taken her time and consulted friends and therefore would not have had the actual operation on the particular day that she did have it. The HL held that she had therefore established that the breach of duty was a cause of her injury.

The HL accepted that it was very difficult to prove causation on conventional principles, and said that this was a case where legal policy required a judge to decide whether justice required the normal approach to causation to be modified. In this case it did. To find otherwise would mean that only those claimants who could categorically say that they would not have had the surgery, would benefit from the existence of the duty of care; whereas those who needed time to think or more advice would not. This would leave the duty of care useless where it was needed most. On policy grounds therefore, the test of causation was satisfied and the claimant won her case.

In light of this Chester can be seen as a case, where the HL ha defectively abolished any meaningful requirement for factual causation. But the CA has since been keen to emphasize in cases like White v Paul (Chester inapplicable to solicitor’s negligent advice), Beary v Pallmall (Chester inapplicable to negligent advice regarding financial matters) that the result in Chester was an extraordinary one reached on policy grounds and it is confined to its own limited context: that is, cases involving a negligent failure to warn a patient of the dangers associated with a particular medical treatment or procedure. Whether, if the opportunity arises, the HL will be as keen to confine Chester v Afshar in this way remains to be seen. But for the present, it would appear wise to treat the decision as exceptional, and one that does not lay down any general rule capable of universal application.


In a number of situations the application of the ‘but for’ test would lead to outcomes that would be absurd or arguably unjust. The courts have in some of these cases been prepared to abandon or modify that test.

Various causes of same damage: Where there are multiple causes of the same damage, the application of ‘but for’ test produces an absurd result; as it will be possible in each cases to argue that the damage would still have occurred in the absence of the one.

In Baker v Willoughby, the claimant was knocked down by the defendant while he was crossing the road. As a result of the accident the claimant suffered injuries to his left leg, which caused him some pain and discomfort and also detrimentally affected his earning power. The claimant took up a new job after the accident, but while he was at work he was shot in the same leg during an armed robbery. As a result of this second incident the claimant’s left leg had to be amputated.

The defendant argued that his liability only extended to the point at which the armed robbery occurred, the damage he has caused would have been done in anyway (in addition to the additional injury) by the robbers. (Note, we are concerned with continuing liability for the consequence of the original injury, and not with liability for the additional consequences of the second injury; this will be discussed under remoteness). Thus, according to the ‘but for’ test, the defendant’s conduct should no longer be treated as a cause of the damage.

But, the HL in finding him liable refused to apply the ‘but for’ test on two grounds –

The loss of earnings was a permanent state of affairs and had resulted from the original injury. The armed robbery and amputation of the leg had not altered this fact even though the eventual damage was different and worse.

Had the claimant sued the robbers, the robbers would have been able to take advantage of the rule that a tortfeasor takes his victim as he finds him, and so would only have been liable for the damage which they had caused to an already damaged leg. So, to refuse the claimant a remedy in respect of the damage done by the defendant would be to leave the claimant under-compensated because he would not be able to recover in respect of that damage from the robbers.

So, to do justice on the facts of the case, the defendant was held liable for the loss resulting from the original accident, the additional damage caused by the robbers was discounted.

A different result was however reached by the HL in the case of Jobling v Associated Dairies. Here, the claimant suffered injury to his back at work in 1973. This injury was caused by the defendants’ breach of statutory duty and the effect of the injury was to compel the claimant to take a less strenuous job, with the result that his earning capacity was reduced by 50%. Before the action against the defendants was tried in 1979, it was discovered in 1976 that the claimant was suffering from a crippling back disease named myelopathy. The effect of the latter disease was to render him totally unfit for work by the end of 1976. There was no connection between the myelopathy and the defendants’ negligence. It was held that the defendants were not liable for any loss of earnings suffered by the claimant after 1976 when the myelopathy rendered the claimant totally unfit for work.

The HL was critical of, but did not overrule, the earlier decision in Baker. Baker was criticised on the ground that it took no account of the vicissitudes of life. Thus, the House appeared to suggest that the contraction of another independent disease would count as a vicissitude and would have the effect of reducing the damages payable, but that the commission of another tor was not a vicissitude of life and so would not reduce the damages payable. But, Evans pointed out, ‘it is just as much a contingency of life that one might get knocked down by a negligently driven car as it is that one might suffer some disabling illness’. So, this distinction is hardly a satisfactory one.

Uncertainty of the facts: Where there are multiple causes each contributing to cause the onset of a disease, but uncertainty as to the proportion of contribution each made; causation is established if the claimant can show that the defendant’s breach of duty materially contributed to the disease being caused – (Bonnington v Wardlaw). In this scenario, the ‘but for’ test becomes ineffective since the claimant will not be able to show on the balance of probability because of the uncertainty, that he would not be have suffered the harm but for any of those causes.

This principle was later extended in McGhee v NCB. Where there are various possible causes of a damage but uncertainty as to which cause is the culprit, causation is made out if the claimant can show that the defendant’s breach of duty had materially increased the risk of the injury occurring. This is another situation where ‘but for’ test would appear ineffective as it would be impossible to say that the damage probably would not have occurred but for the defendant’s breach of duty or it would probably still have happened without the defendant’s breach of duty.

Lord Wilberforce justified the McGhee conclusion on the ground that the defendants, by their negligence, had created the risk of a particular damage occurring and when that risk became reality they were not allowed, on grounds of policy to hide behind the evidential difficulty of showing what had caused the damage to the claimant.

But, in Wilsher v Essex AHA, the HL declined to follow this principle and insisted for strict ‘but for’ test. The HL found itself faced with this issue once again in Fairchild v Glenhaven, where it favoured the McGhee approach, but accepted that Wilsher was correctly decided on its own facts as there were a number of possible agents involved. It appears that, Fairchild principle applies only where there is one noxious agent involved. Thus, these two lines of authority can be summarized as –

Where there are various possible causes each generating same agent that is likely to cause a damage on its own, but uncertainty as to which cause is the culprit, causation is established if the defendant’s breach of duty had materially increased the risk of the damage occurring – (Fairchild, McGhee).

Where there are various possible causes generating several agents that are likely to cause a damage on their own, but uncertainty as to which cause is the culprit, causation can only be proved if the claimant can show on the balance of probabilities that the defendant’s breach of duty caused the injury – (Wilsher).

It should be noted that, there remains no convincing answer as to why a case involving several agents should be treated differently from a case involving single agent. Fairchild was revisited in Barker v Corus where it was held that, a defendant may still be liable even though the other exposures were non-tortious, by natural cause, or by the act of the claimant himself.


Damages may be denied even where the claimant is able to establish a causal connection between the breach of duty and damage, on the ground that the damage is too remote.

Initially, the test of remoteness of damage in the tort of negligence was said to be – whether the damage was the direct consequence of the breach of duty – as laid down in Re Polemis. Modern test is the one set out in The Wagon Mound (No.1) – Defendant is liable for damage only if it was the foreseeable consequence of the breach of duty. It is sufficient if the injury is of the type that could be foreseen, even if it came about in an unexpected way – (Jolley v Sutton LBC); or was much more severe than expected – (Hughes v Lord Advocate).

New & intervening causes: Where the damage caused by a torfeasor is worsened by an intervening event, which is afforded the stage to occur by the original tort, particular problem of remoteness arises. Clearly, the intervening event would not have occurred resulting in additional damage, had the original tort not been committed. Therefore, the additional damage is in a factual sense caused by the original tort. But, the question is, is it recoverable from the original tortfesor? This additional damage cannot be recovered from the original tortfeasor where the intervening event renders them too remote, by breaking the chain of causation.

Intervening events may take one of three forms –

Natural events: An act of nature will break the chain of causation where it is unforeseeable and independent of the defendant’s negligence – (Carslogie v Royal Norwegian Government).

Acts of third party: Act of third party will break the chain of causation if it is ‘something unwarrantable, a new cause that disturbs the sequence of events and something that can described as either unreasonable, or extraneous or extrinxic’ – (per Lord Wright in The Oropesa). Thus, the defendant will remain liable if the act of third party is not truly independent of the defendant’s negligence as was the case in Oropesa.

But even where the conduct of the third party is necessitated by the initial negligence of the defendant, it is still possible that the subsequent act will break the chain of causation as shown by Rahman v Arearose. It was held in Rahman v Arearose that, negligent medical treatment which intervenes between the breach and the damage will be treated as breaking the chain of causation, if it is serious and amounts to a completely inappropriate response to the patient’s condition.

In Knightley v Johns, Stepenson LJ stated ‘negligent conduct is more likely to break the chain of causation than conduct which is not’. He stated that the courts sought refuge in ‘common sense rather than logic on the facts and circumstances of each case’.

In Knightly v Johns, the defendant’s negligent driving caused an accident which had the effect of blocking a busy tunnel. The second defendant, a police officer, forgot to seal off the tunnel and, in breach of police standing orders, sent the claimant, a police motor cyclist, back into the tunnel against the flow of traffic. While travelling through the tunnel, the claimant was injured when he was involved in a collision with another driver. It was held that, the second defendant was negligent in not closing the tunnel when the initial accident occurred and so the first defendant was not liable for the injuries sustained by the claimant.

On the other hand, a different result was reached in the case of Rouse v Squires. The first defendant through his negligent driving caused an accident in which his lorry ‘jack-knifed’ across the road. A few minutes later the second defendant negligently collided with the vehicle involved in the first accident and killed the claimant. It was held that the first defendant’s negligence was an operative cause of the claimant’s death and so he was held to be 25% responsible for the claimant’s death. Here, the second accident was a natural and probable consequence of the first accident and was foreseeable; therefore it did not amount to break in the chain of causation.

Where the third party act consists of deliberate wrongful conduct, chain of causation will be broken only if it was something very likely to happen – (Home Office v Dorset Yacht), a mere foreseeable possibility is not enough – (Smith v Littlewoods).

Acts of the claimant: Where the intervening event is an unreasonable conduct perpetrated by the claimant himself, this will break the chain of causation – (McKew v Holland).

In Mckew v Holland, defendant’s negligence caused injury to claimant’s leg. Claimant later broke his ankle attempting, while still suffering from the effects of the first injury, to descend s steep staircase unaided. Claimant’s imprudent and unreasonable conduct constituted a fresh and separate cause of the second injury. Defendant was only liable for the initial injury.

By contrast, in Wieland v Cyril Lord Carpets, the claimant suffered neck injuries and had to wear a collar in consequence of the defendants’ negligence. She later fell downstairs because, as a result of the initial injury and the neck collar, she could not use her bifocal lenses with her usual skill. Her further injury was found to be attributable to the defendants’ original negligence. There was no unreasonable conduct on her part that could be taken to constitute a novus actus interveniens breaking the chain of causation.

Even if the claimant’s act may be viewed as unreasonable it will not break the chain of causation, where the defendant could foresee such an act and was under a duty to take care to prevent it occurring. Thus, in Reeves v Commissioner of Police, the fact that a prisoner was a known suicide risk combined with the fact that the defendants were under a duty to take care to prevent that suicide meant that the defendants were causally responsible when the prisoner did in fact commit suicide. By contrast, in Grieves v FT Everard where claimant’s anxiety about the onset of a long-term asbestos-related disease was foreseeable, it was held that there was no duty to take reasonable care to prevent claimant suffering such anxiety.


Where the claimant’s injuries are psychiatric and not physical, damages are recoverable only exceptionally. In the past, the courts were reluctant to award damages for psychiatric injury caused by the negligence (as apparent in Victoria Railways v Coultas) for various reasons. They include the difficulties of putting a monetary value on such harm, the risk of fictitious claims and excessive litigations, and the problems of proving the link between the defendant’s negligence and the injury to the claimant. But, such claims are now entertained but are subject to a number of restrictions.

The first restriction is that, damages are available only for a recognised psychiatric illness; and not for grief, distress, sorrow etc (This was the reason why damages is not awarded in Nichols v Rushton). In exam questions, this should be assumed as satisfied.

Claimants who can prove such injury can only claim in negligence if they can establish that they are owed a duty of care by the defendant to avoid inflicting psychiatric injury (and of course that the defendant’s negligence actually caused the injury). The other restrictions apply in establishing this duty of care. Different sets of rules apply for different categories of claimant; that means, whether the claimant is a primary victim or a secondary one.

Primary victims are one who suffers psychiatric injury after being physically injured in an accident, or being put in danger of physical injury, or reasonably believing himself to be in danger of physical injury. In primary victim cases, reasonable foreseeability of physical injury to the claimant is sufficient to establish the duty of care in respect of the psychiatric injury, it is not necessary to show psychiatric injury as such was foreseeable – (Page v Smith). Although affirmed by the HL in Simmons v British Steel, Lord Neuberger in Corr v IBC Vehicles described Page as a ‘somewhat controversial decision’. In light of Page, damages for psychiatric injury will be recoverable even where the psychiatric damage is suffered because of some susceptibility of the claimant to shock as long as physical injury is foreseen.

Secondary victims are one who are not put in danger of physical injury to themselves, but suffers psychiatric injury as a result of witnessing such injury to others. To succeed as a secondary victim following conditions deriving from McLoughlin v O’Brian and subsequently refined in Alcock v CC of South Yorkshire should be satisfied.

It was reasonably foreseeable that the claimant would suffer psychiatric injury, as his relationship of love and affection with the primary victim was sufficiently close.

So long as a bystander of normal fortitude would be likely to suffer psychiatric injury, it does not matter that, that psychiatric injury is made more serious by some characteristics personal to the claimant. But, if the psychiatric injury would not have occurred at all to someone without the claimant’s particular susceptibility, there will be no claim. Only some psychiatric injury has to be foreseeable, its precise form or severity need not be foreseen.

In cases of parents, children, and spouses – there relationship with the primary victim is presumed to be sufficiently close, though evidence might be brought to rebut the presumption. In all other cases, it must be established. Claims by brothers, sisters and brother-in-law are failed in Alcock while the claim on the part of a fiancée was allowed.

Curiously in Alcock, Lord Keith, Lord Ackner and Lord Oliver went so far as to say that the claim of a bystander was not ‘entirely excluded’ and they suggested that a claim may be brought by a bystander where a catastrophe occurs very close to him, which is particularly horrific. However, this expansive approach to the class of claimants has not been followed in a later CA case, named McFarlane v Caledonia. It was held that, a bystander to a horrific event could not recover on the grounds that, the basis of the decision on Alcock was that the test of proximity was not simply reasonable foreseeability but that there must be a sufficiently close tie of love and affection between the claimant and the victim. To extend that duty to those who have no such connection, would be to base the test on reasonable foreseeability alone.

But surprisingly, the necessary relationship can be with the claimant’s property – (Attia v British Gas plc (with house), Owens v Liverpool Corp (with favourite pet)). In Attia, the CA did not say what factors would be considered in deciding whether the psychiatric injury was foreseeable, but presumably it is the closeness of the relationship between the claimant and the ‘thing’ damaged.

Claimant’s proximity to the accident or its aftermath was sufficiently close both in terms of time and space. This means, the claimant must normally either witness the accident himself or come upon its immediate aftermath.

It is not clear what really constitutes ‘immediate aftermath’. In McLoghlin claimant’s arrival at the hospital after two hours of the accident and finding her family, was held to come within the meaning of immediate aftermath. While, in Alcock claimants’ arrival at the mortuary after some eight or nine hours to identify their relatives who had been the victim of Hillsborough tragedy was held insufficient to come within ‘immediate aftermath’.

One factor may be the state of the primary victim, and the purpose of the visit. In McLoughlin, the primary victims were still in the same state as they were after the accident. In Alcock, the claimants’ went to the mortuary simply to identify the body. Lord Jauncey opined that, this was very different from the case of a relative who goes within a short time of the accident to rescue or comfort the victim.

However, a more generous approach to this proximity requirement is apparent in recent cases. The CA in Galli-Atkinson v Seghal held that, aftermath of an accident can be made up of a number of components. The HL in W v Essex CC declined to strike out a parent’s claim on the basis that the parents had only been told of the abuse, without not actually witnessing it or its immediate aftermath. Although, a full discussion of this point must await the full trial of W v Essex CC (as it was only a striking-out case), it might be that, in certain circumstances, the strict time requirement that have been set in Alcock and McLoughlin might be relaxed.

The claimant suffered psychiatric injury as a result (an issue of causation) of sudden shock caused by seeing or hearing the accident or its immediate aftermath.

In Sion v Hampstead HA, the claimant had developed a stress related psychiatric illness as a result of watching his son slowly die intensive care as a result of negligent medical treatment. It was held that as the father’s psychiatric injury had not been caused by a sudden shock, he could not recover damages for it.

A contrasting case is North Glamorgan NHS Trust v Walters. Here the claimant was the mother of a baby boy, who died after receiving negligent treatment for which the defendants were responsible. The baby was ill in hospital. Unknown to his mother at the time, the hospital had misdiagnosed his illness. She woke up to find him choking and coughing blood, and was told by the doctors that he was having a fit, but that he was very unlikely to have suffered any serious damage. Later that day, he was transferred to another hospital, where she was told – correctly – that he had in fact suffered severe brain damage and was in a coma; she was asked to consider switching off his life support machine. She and her husband agreed to this on the following day.

The events caused her to suffer a psychiatric illness, but the hospital argued that they were not liable for this as it was not caused by a sudden shock, but by a sequence of events that took place over 36 hours. The CA disagreed: it said that the ‘horrifying event’ could be made up of a series of events, in this case witnessing the fit, hearing the news that her son was brain damaged after being told that he was not, and then watching him die. Each had their own immediate impact, and could be distinguished from cases where psychiatric injury was caused by a gradual realization that a child was dying.

In many cases, causation will be difficult to prove, since in addition to the required shock, claimants will have experienced the grief of bereavement, which could equally well have caused their psychiatric injury. However, Vernon v Bosley (No.1) made it clear – provided psychiatric injury is partly caused by shock, the claimant can recover even if there were other causes.

As rightly rejected in Alcock a person watching simultaneous television would normally have no claim as the broadcasting guidelines prevent the showing of suffering by recognisable individuals. If such pictures were shown, then the transmission would normally be regarded as a novus actus interveniens. However, there may be cases where viewing simultaneous television may be treated as equivalent to sight and sound of the accident. An example was given of a televised hot air balloon event with children in the balloon, which suddenly burst into flames.

The claimant must suffer psychiatric injury as a result of directly hearing or seeing the accident or its immediate aftermath. There can be no liability where the claimant is merely told about the accident by a third party – (although in Ravenscroft & Hevican such a claim was entertained by the High Court, these were disapproved in Alcock and accordingly Ravenscroft has been overruled by the CA; but no appeal has been heard in Hevican yet). But, the HL in W v Essex CC declined to strike out a parent’s claim on the basis that the parents had only been told of the abuse, without not actually witnessing it or its immediate aftermath. However, if a duty is to be recognised on such facts, this is best done by reference to the local authority’s assumption of responsibility to the claimants in placing the foster child with them, and not by dilution of the general requirement.

One further limitation in this area was added in Greatorex v Greatorex. Where the defendant himself is the primary victim he cannot owe a duty to secondary victim – (Greatorex v Greatorex). In this case, the defendant caused himself serious injury by negligent driving. The claimant, his father, went to the scene as a member of the rescue service and suffered post-traumatic stress disorder as a result. He sued his son, in practice he was not suing his son, but the Motor Insurance Bureau, who compensate victims of uninsured drivers; had the son been insured, the defendant would have been his insurance company. But, the claim was dismissed on the ground that imposing a duty on people to take care of themselves so as not to shock others would impinge on the right of self-determination.

The need for reform in respect of secondary victim’s claim for psychiatric injury, has been identified by the Law Commission in its Report, Liability for Psychiatric Illness in 1998. Their chief recommendations are –

To retain the requirement of a close tie of love and affection to the primary victim in the case of secondary victims.

To remove the requirements for secondary victims to show proximity in time and space, and that the event has been witnessed by the claimant’s own unaided senses.

That the injury should be accepted even where not caused by sudden shock.

The proposals seem to be much fairer. However, it is not clear whether or not there is any likelihood of them becoming law. In light of recent response to Law Commission proposals it may be unlikely, at least in the near future.

Other cases – Rescuers: At one time it was thought that a rescuer who suffered psychiatric injury was the result of participating in distressing scenes could recover damages on policy grounds. This was the position taken in Chadwick v BRB. The policy behind this was that such selfless behavior should be encouraged and supported, and therefore should not be subject to stricter rules than those of ordinary personal injury.

But the HL in White v CC of South Yorkshire held that, in order for a rescuer to be able to recover for psychiatric damage some physical damage had to be foreseeable, so that he can claim as primary victim. He cannot claim as secondary victim as he will have no ties of love and affection with the primary victims. Thus, rescuers are not treated as a special category of secondary victims; and they are subject to normal rules on secondary victim as stated in Alcock.

Two main reasons were given for the ruling –

Once rescuer includes those who help without putting themselves in any physical danger, the line between rescuers and bystanders may become difficult to draw.

Police officers’ conditions of service (in White the rescuers were police officers) provided for them to be compensated in other ways for the psychiatric injury they had suffered. But this does not explain why volunteer rescuers should be treated in the same way as professional ones.

Other cases – Employees: The second argument made by the claimant in White was that, they were owed a duty of care as employees of the party whose negligence caused the shocking event.

The House stated that the employer’s duty to employees was not a separate tort with its own rules, but an aspect of the law of negligence, and therefore subject to the normal rules of negligence. This meant that where a type of injury was subject to special restrictions on when a duty of care would exist, these rules applied where the injury was caused by an employer to an employee, just as they would normally.

An attempt to widen employers’ liability for psychiatric injury caused by a shocking event was firmly rejected by the HL in French v CC of Sussex Police. Employees are therefore only able to claim for psychiatric injury caused by a shocking event where they can satisfy the rules on claims by secondary victims, or where they can be considered primary victims.

Other cases – Unwitting agent: Where the claimant through defendant’s negligence becomes the agent of another person’s injury or reasonably believe that he is about to do so, and suffers psychiatric injury as a result – he can recover damages if he can prove sufficiently close, both in terms of time and space, proximity to the accident or its immediate aftermath.

The leading authority is Dooley V Cammell. Here the claimant was operating a crane at the docks where he worked, when through no fault of his, it dropped a load into the hold of the ship being unloaded. He successfully claimed for psychiatric injury caused through fearing for the safety of a colleague working below. In Hunter v BCC the claim was rightly rejected as the claimant could not show the necessary proximity, he had reacted only to what he was told by a third party.

Other cases – Assumption of responsibility: Should a defendant be liable for causing psychiatric injury by carelessly passing on wrong information, or by passing on correct information in a carelessly insensitive way – is far from clear. However, there may emerge a principle in the light of few isolated cases with no clear principle (as yet) such as Allin v City & Hackney H, that a defendant should be liable if there is an assumption of responsibility to protect the claimant against psychiatric injury or if there is an ongoing relationship between the parties that entails such responsibility.


Defamation has been succinctly defined by the leading tort expert Professor Winfield: ‘Defamation is the publication of a statement which tends to lower a person in the estimation of right-thinking members of society generally, or which tends to make them shun or avoid that person’.

There are two types of defamation, libel and slander:

Libel covers statement made in some permanent form: this usually means printed or written words, but also covers film, pictures, statues and effigies. By statute it includes radio broadcasts (Sec. 1 Defamation Act 1952) and the performance of plays (Sec. 4(1) Theatres Act 1968).

Slander applies to statement made in a transitory form, such as spoken words or gestures.

But, the distinction is not always straight-forward. In Youssoupoff v MGM Pictures, the court had to consider the status of the spoken words in a permanent form, in the context of a film. Slesser LJ held that the film pictures, being ‘a permanent matter to be seen by the eye’ could be regarded as libels. He also found the speech to be libel, but not simply because it too was permanent: in his view, it amounted to libel only because it was ‘part of one complex, common exhibition’. To him, it was because the sound recording was ‘ancillary’ to the visual image that it could be regarded as libel rather than slender: it was ‘an ancillary circumstance…… …….explaining that which is to be seen’.

Thus, where a sound recording exists independently of any visual image, Slesser LJ’s reasoning suggests that the liability would arise only in slander. Yet, this ignores the fact that although the words are merely heard, they are permanently recorded which, it is submitted, suggests that they should be considered libel.

Libel is actionable per se (means on its own or alone sufficient ground for bringing an action), but slander is only actionable where the claimant can show that he has suffered special damage; that is not too remote. However, there are few exceptional situations where the claimant is not required to prove special damage for slander. The exceptional cases are as follows –

Where the imputation is that the claimant has committed a criminal offence punishable with imprisonment (as seen in Hellvig v Mitchell),

Where the imputation is of un-chastity or adultery on the part of a woman or girl (Sec. 1 Slander of Women Act 1891, as seen in Kerr v Kennedy),

Where it is imputed that the claimant is infected with a contagious or infectious disease likely to prevent others associating with the claimant (as seen in Taylor v Hall),

Where the words impute unfitness, dishonesty or incompetence on the part of the claimant in relation to any offence, profession, calling, trade or business (Sec. 2 Defamation Act 1952).

ELEMENTS OF DEFAMATION: To succeed in an action for defamation, the claimant must prove three things –

The statement is defamatory,

The statement refers to the claimant, and

The statement has been published to a third party.

Note, in the case of a libel merely committing the tort without a defence is sufficient for liability, but with slander a defendant will only be liable if the defamation has caused the claimant special damage (subject to the exceptions discussed above).


To be defamatory, a statement must ‘tend to lower the claimant in the estimation of right thinking members of society generally’ – (Sim v Stretch), or lead to the claimant being shunned or avoided – (Youssoupoff v MGM Pictures). In the more recent case of Berkoff v Burchill, in which the claimant was described as ‘hideously ugly’, it was held by the CA that, although insults that did not diminish a person’s standing are not defamatory, a statement can be defamatory if it held up the claimant to contempt, scorn or ridicule, or tended to exclude him from society, even if the statement did not impute disgraceful conduct or any lack of business or professional skill.

In Byrne v Deane, the claimant was a member of a golf club, whose owners illegally kept gambling machines on the premises. Someone reported them to the police, and afterwards a poem was posted up in the club, implying that the claimant had been the informant. He sued, and won the original case, but on appeal the courts held that the suggestion was not defamatory, because a right-thinking member of society would not think less well of someone for telling the police about criminal activity. It did not matter that the claimant would be less well thought of by his fellow club members. So, the fact that, a section of the public with which the claimant has the closest contact thinks less of him is not enough, if the views are not shared by right-thinking members of the society.

The intention of the defendant in making the statement is wholly irrelevant – (Cassidy v Daily Mirror), although this may help to support a defence of unintentional defamation.

The judge has to be satisfied that the words are capable of being defamatory. The jury will then decide whether the words are in fact defamatory.

Defamation is not confined to direct attacks on the claimant’s reputation. To protect the claimant’s reputation defamation may also include implied or veiled attacks, which are generally known as ‘innuendo’. There are two types of innuendo: true (or legal) and false (or popular).

A true innuendo is one where the statement contains nothing of a defamatory nature, but when combined with some special facts and circumstances known to the readers of the publication appears defamatory. In Tolley v Fry, the claimant was the English amateur golf champion and he was featured, without giving his consent, on a poster advertising the defendant’s chocolate bar. The text of the poster compared the excellence of the chocolate bar with the excellence of the claimant’s swing. The claimant alleged that this constituted an innuendo, because it implied that he had agreed to feature in the poster for personal gain and that he had flouted the rules relating to his amateur status. It was held that the poster was capable of being defamatory and the jury indeed found that he had been defamed.

Another example is Cassidy v Daily Mirror. Here, a picture was published which showed Mr. Cassidy with a young lady under a heading which stated that she and Mr. Cassidy had just announced their engagement to be married. The claimant was Mr. Cassidy’s wife, and this was generally known to be the case even though they led separate lives. She was unable to prove that several people believed, as a result of the publication, that she was ‘living in sin’ with Mr. Cassidy; a serious problem for her in the 1920s. It was held that the words were capable of being defamatory and, once a jury considered the issue, it was decided that the defendant was liable.

A false innuendo is one where a reasonable person guided by general knowledge considers the statement to be defamatory from the natural and ordinary meaning of the words. In Lewis v Daily Telegraph, a paragraph in the newspaper had stated that the Fraud Squad were investigating the affairs of a company and its chairman, Mr. Lewis. This was in fact true, but it was claimed that the paragraph contained an innuendo: that the company was being operated in a fraudulent and dishonest way. The HL stated that a reasonable person might infer from the paragraph that the company and Lewis were suspected of fraud, but would not assume that Lewis and the company were guilty of such conduct. As there was no evidence of some special facts to support the claim as a true innuendo, it was rejected by the HL.

The court will not allow a claimant to point a particular sentence in isolation, but will examine the statement in its whole context. Accordingly, in Charleston v NGN where two well known actors had their faces superimposed in pornographic photographs in a newspaper, and the accompanying text stated that they were innocent and unknowing participants, no action in defamation lay. It was argued that a significant proportion of readers skimming through the newspaper would only read the headlines and look at the photographs, and would come to the wrong conclusions. The court refused to approach the case in this way. ‘Defamatory’ was judged by the standard of the ordinary reasonable person, who would have taken the trouble to discover what the article was about. It was therefore irrelevant that the newspaper may have some readers who only read the headlines.

In Charleston v NGN, the HL held that, an explanation can neutralize defamatory statements only if it is obvious to the reader. So, the decision of Charleston would have been different had the explanatory texts were given further down the article or on a continuation page.


A claimant will only be able to bring an action for defamation if he or she can show that the words in the statement referred to him or her. Otherwise, the claimant would not have standing to sue.

This is not a problem where the claimant is individually named, for example, ‘David cannot be trusted’. Difficulty arises when the defendant makes general statements – such as, ‘politicians cannot be trusted’; or does not name the claimant expressly – such as, ‘the man with blue eyes cannot be trusted’. The claimant will have to prove that an ordinary reasonable person with knowledge of all the circumstances would recognise that the person referred to is the claimant.

As with the question of defamatory meaning; the defendant need not have intended the statement to refer to the claimant – (Hulton v Jones, Newstead v London Express). Accordingly, where a defamatory statement is intended to refer to a fictitious character – (Hulton v Jones), or to someone other than the claimant – (Newstead v London Express), the defendant will be liable for defamation of the claimant, if a reasonable person with knowledge of all the circumstances would think the statement referred to the claimant.

Whether cases like these can now withstand the passage of the Human Rights Act 1998 is highly questionable, as there is arguably an infringement of the Art-10 right to freedom of expression. In O’Shea v MGM Ltd. (2001) an advert for an adults-only internet site contained a picture of a well known glamour model. It was alleged that the picture was look a li