Please note that this site is now an archive site and is no longer being updated. Updates ceased in May 2003.

For current information please go to



Negligence: Duty of Care

Ian Fraser

The story of how English courts treat claims, in negligence, that D has caused P to suffer psychological (mental/ emotional/ nervous) harm, is a useful illustration of common-law evolution, as well as a potentially important source of liability. This evolution has occurred in phases, driven by conceptions of the purpose of the negligence tort (how much suffering should be alleviated by judge-made law?) as well as the judicial reception of changes in scientific thinking (here, psychology).

Through the 1970s, there were two well-established points, & a set of suggestive lower-court cases. The points well established were that psych harm was compensable if ‘consequent upon’ physical injury (exactly like economic loss before Hedley-Byrne), and that psych harm was acceptable as a link in causation to physical injury (as in Bourhill, and Wilkinson; also legislated in 1934). Otherwise such harm was to be presumed non-compensable, for much the same reasons as pure economic loss was. The cases featured –

Dulieu v White  [1901] KB 269: damages for ‘nervous shock’ in the form of fear;
Hambrook v Stokes  [1925]1 KB 141: same when fear is for P’s baby;
Chadwick v BRB  [1967]1 WLR 912: damages when P incurs shock during rescue attempt.

One such case, Mt Isa Mines v Pusey (1970)125 CLR 383, was especially influential (although Australian, it was in Australia's highest court). P, foreman at D’s mine, saw and tried to help two fellow workers who had just been literally fried by a massive short-circuit. They died soon after; he became depressed and eventually developed schizophrenia. The HC decided this was compensable harm – a “recognised psychiatric illness”, as opposed to emotional distress – and that the mine’s common-law duty of care should include preventing shocks which might cause it. The theoretical basis was foreseeability. But the opinions frankly acknowledge doubts about exactly what should be 4Cable (any harm or psych harm, to this P or a ‘reasonable’ one), and about the significance of this P’s status (as rescuer and as D’s employee).

Then came the first landmark, McLoughlin v O’Brian [1983]1 AC 410. Lawyers knew ‘nervous shock’ was an important liability, but the field was confused (note the range of decisions among the four judges below the HL). McLoughlin was intended to set things in a new order (just as the contemporaneous Anns was). Like Anns, it went too far, in the judgment of subsequent courts; unlike Anns, it did succeed in setting the terms of subsequent developments.

In McLoughlin, two judges declare that ‘policy’ is for Parliament: courts should follow principle, because although policy considerations may be necessary or prudent, they lead to arbitrary rules – which are never ‘just’ (Scarman, Bridge). Two others deny that, asserting that policy is and always has been a factor in how the common law is developed (Edmund-Davies, Russell). This disagreement remains the dynamic within the law to the present day. In McLoughlin, however, it was academic; all agreed on what the instant result should be, and seemed to support the fullest opinion, that of Wilberforce (the ‘leading opinion’).

Declaring himself guided by “most people’s sense of justice”, and by logic (in taking the next step from the analogies of precedents), Wilberforce allowed the negligence duty to cover a mother’s psych illness brought on by seeing her injured husband & children in hospital. This duty, covering people who suffer because D caused injury to a third party, is subject to “policy” limits (as is implied, he says, by the term “ought” in Atkin’s neighbour principle). Why such limits are necessary, he doesn’t clearly say: it’s not ‘floodgates’ (slighted as an argument by all opinions). It may be fairness to Ds. The limits are described only vaguely, the P here being clearly within them. They are –

In the 1980s ‘nervous shock’ became an important part of personal-injury legal practice, apparently to the HL’s surprise. But every accident victim is someone’s child – most of them spouses and/or parents too – and industry and motor vehicles provide plenty of shocking accidents. During this time, too, psychologists’ practice expanded greatly in emotional and affective disorders, recognising and treating depression, anxiety, and stress as ‘psychiatric illness’.

If psychologists agree that learning of a son’s death can injure his mother regardless of her “proximity” to the event, surely we must admit that a reasonable D would foresee even unusual degrees of devastation by grief in some mothers? (Ravenscroft v Rederiaktiebolaget [1991]3 All ER 73 (QB), the judge bemoaning the uncertainty of this ‘policy’-driven law but hoping that if he’s lost his way on the sea of common-law, it’s not for being “too wet”.) If psych damage is 4Cable when P sees a person injured by D’s negligence, why not when P sees her house destroyed? (Attia v British Gas [1987]3 All ER 455 (CA)).

In 1989, at the Hillsborough football stadium, during an FA Cup semi-final, a gate was opened that should have been shut and hordes of fans pressed into a fenced-in enclosure which was already full of people. 95 people were crushed to death, 400 injured the same way, the whole thing took place in front of thousands of other fans – and on live TV – and it was all due to one bad decision by the police commander (a decision having nothing to do with investigating crime). That’s an awful lot of nervous shock ... and the first HL case arising from this horror, Alcock v CC S Yorkshire [1991]4 All ER 907, set a new landmark for claims for psychological injury: moving back and narrowing the goal-posts.

Clearly basing themselves on ‘policy’, and clearly meaning by that closing the floodgates, the Lords adapted McLoughlin for “secondary victims” of shock like the various relatives of the Hillsborough dead. There still must be 'recognized psych illness', as one “control mechanism”. The other three become –

(What adaptation of a previous case does this remind you of?) None of the Ps qualified.

The law here, as elsewhere, had shifted against plaintiffs. Nonetheless one important step was taken for Ps in the 1990s. In Page v Smith [1995]2 All ER 736 the HL allowed a claim by the “primary victim” of a traffic accident, in which the injury was ‘chronic fatigue syndrome’. No-one knew what caused this CFS, so it could not even be categorised as physical or mental – and the HL said that didn’t matter. It was finally time to abolish that old-fashioned distinction anyway. Injury is injury. If any is 4Cable (as of course it always is in a traffic accident), then all is covered. Primary victims of psych damage thus recovered on the usual 4Cability/proximity/F & R basis of duty – aided by developments in psych theory (cf Vernon v Bosley (No 1) [1997]1 All ER 577: “pathological grief disorder” counts as 'illness').

Exactly who was a primary victim and who a secondary one was not clear, however. Employees are already owed a duty by their employers to maintain a safe workplace, for instance; does that mean they are primary victims when injured psychologically by seeing other people injured? Rescuers have always been considered a special case – their physical injuries deemed 4Cable no matter how unlikely – does that mean they are ‘primary’ when shocked by their rescuees’ injuries?

At the close of the century, the HL dealt with these points in another Hillsborough case. This time it was the police officers suing, men & women who had suffered psych illness from the shock they experienced while trying to help the people crushed and trampled.

The CA had allowed the suits, even though it meant the police could recover where the relatives had got nothing. This, in White v CC S Yorkshire [1999]1 All ER 1, the HL refused to follow: “the man on the Underground” would be “confused” (i.e., appalled). The police were owed no such duty.

If claiming as secondary victims, the police like anyone else have to meet the Alcock conditions. If they claim as primary victims, then like anyone else they have to have been exposed to physical danger. (And if that conflicts with Page, then Page is wrong; the physical/mental distinction may be bad science, and arbitrary law, but it’s a "rational" protection of the public purse.) Neither being employees on the job, nor being rescuers, changes the rules for recognising a duty. Perhaps it should, on principle, but that would deeply offend the public: "and the search for principle was called off in Alcock” (Hoffman); this law may be “a patchwork of awkward distinctions...but we can only say, ‘Thus far and no further’” (Steyn).

White does seem a final abandonment of judicial development of this law (as Jones takes it to be). But – as with any dramatic dicta – remember the context. The Hillsborough disaster was extraordinary; more importantly, these cases were widely publicised. And the spectacle of grieving relatives getting nothing while, ten years later, the courts order payment to distressed professionals employed by the defendant (because they were working for the defendant) would hardly have improved the law's image.

A couple of years on, things are indeed developing, as we’ll see next week in HL decisions on governmental duties of care. The physical/mental distinction has again been held to be obsolete (Barrett v Enfield [1999]3 All ER 193: a very different factual context, however), and the Alcock conditions of suddenness of shock and proximity in time & space have been doubted (W v Essex CC [2000]2 All ER 237).

pixgrn.jpg (632 bytes)

[About the School of Law] | [Courses Offered] | [Course Materials] | [Staff]
[Pacific Law Materials] | [Journal of South Pacific Law]
[Emalus Campus Library] | [Recommended Internet Links]
[Latest Additions]  | [Search the Site]

© University of the South Pacific 1998-2002

If you have any comments, suggestions or difficulties with using this web site please email
Robynne Blake
, Internet Project Manager, The School of Law, The University of the South Pacific or fax: (678) 27785

Last Update: Monday, June 02, 2003 08:20