ALEX BOND

LAW 231 Law of Torts March 3-5, 2008

Introduction: Liability in tort

The word tort comes from the Latin word tortious and found its way into the English language as
a synonym for a ‘wrong’. A tort is neither a crime nor a contract.

Fleming: “A tort is an injury other than a breach of contract which the law will redress with
compensation”.
Introductory example:

There is a collision between two cars, A and B. A fails to give way at an intersection while B is
proceeding along at a leisurely pace. A’s failure to give way results in a collision with B.
Assume the consequences are as follows:
 A’s car suffers property damage of $2,000.
 B suffers a severe whiplash (neck) injury (personal injury) for which she had to incur
medical expenses and take time off work. B’s car was also damaged.
A traffic officer is called to the scene and it is clear that A’s failure to give way at the intersection
caused the accident. There is a witness.
So we ask what are the legal consequences of this accident?
 A receives a legal summons to attend the District Court, charged with a breach of the
Transport Regulations, failure to give way.
 These are criminal proceedings. Parties to the criminal prosecution are the State (usually
the Police in cases of traffic offences) and the accused, A. Punishment is regarded as the
primary purpose of the criminal law, deterrence to a lesser extent. The State exacts a
penalty from the wrongdoer to protect the interests of society at large.
 This however does not help B. B has suffered personal injury. Under the law of torts,
personal injury covers both physical and mental injury. Because of our accident
compensation scheme in New Zealand, victims of personal injury are unable to sue for
damages in a civil court in respect of their personal injury. New Zealand instead has an
ACC Act which provides compensation. There are a few exceptions to this however.
 The statutory benefits you could get under the ACC scheme include:
 a contribution to your medical expenses
 compensation for loss of income if off work (is called earnings compensation)
 a lump-sum payment of up to $120,000 for permanent physical impairment.
 rehabilitation, e.g. ACC will pay the cost of building wheelchair ramps up to
homes of those people who have become disabled.
 As A is at fault in this accident, he has to bear his own loss, so must pay for the costs to
repair his vehicle (that is unless he has insurance for his vehicle).
 B can sue for property damage, specifically the damage caused to her car. If B has first-
party insurance over her vehicle, B can make a claim under the policy and be paid the
cost of repair less the deductible or excess. B’s insurance company inherits B’s civil rights
against A, and B will make a claim against A saying A is legally responsible to pay the
amount of money paid out to B. If A is insured, then A’s insurance company will pay,
but A will have to pay the excess.
 But what if A or B is not insured? There are still legal rights to sue in the tort of
negligence. B himself can sue A saying A is legally responsible to pay the cost of
property damage. Negligence is basically the tort that relates to unintentional damage. A
did not deliberately ram B’s car, but was careless in obeying the rules of the road. A is at
fault.

1
 ALEX BOND
Distinction between a tort and a crime

A tort is a civil wrong as opposed to a criminal wrong. Tortious rights are enforced in civil
proceedings.

Some actions can give rise to more than one legal proceeding. The example on page 1 is one mere
example. A was charged for breach of the Transport Regulations. A was also able to be sued in
civil proceedings in the law of negligence. The same conduct can result in a criminal charge and a tort.
There are other examples of this, e.g. pushing someone, which is the crime of assault and the tort
of battery, or e.g. stealing a watch, which is the crime of theft and the tort of conversion.

Distinction between a tort and a contract

The purpose of contract law is protect a very important but single interest, the interest in having
promises performed, and if not performed, to have them enforced. Contract law is the bedrock of
the capitalist system, allows the operation of everyday commerce. Tort law does not protect a
single interest but a number of interests, e.g. our interest in being physically safe, not being
negligently injured. It exists to protect our person and our property interests. It can even protect
our interests in our reputations, e.g. the tort of defamation.

Functions or objectives of tort law

There are generally stated to be three primary objectives of tort law:

1. compensation
Essay exam questions are
2. corrective justice
often set on this topic.
3. deterrence

Compensation: Courts have stated that the essential purpose of tort law is to restore the plaintiff to
the position they would be in, had the tort never been committed: compensation. This can be
contrasted with the criminal which which exists to protect society’s interests at large. The award
of damages in tort law is meant to fully compensate the plaintiff for all his or her losses.

Corrective justice: Liability in negligence is bounded by the requirement that there be an initial
finding that the defendant is at fault in that objective sense. Negligence is a fault-based tort. This
requirement to find fault even in the subjective sense rests on this principle of corrective justice. It
is the idea that wrongdoers who injure other people through their fault bear a personal
responsibility to correct that fault or wrong. They should put the matter right by paying an award
of damages to restore – so far as money can do it – the victim to their former position. The balance
must be restored. This places the ethical principle of correcting wrongs at the forefront.

Deterrence: The theory here is that a finding of liability against a defendant in negligence and an
award of damages is expected to deter other similar conduct and injuries in the future. Deterrence
is meant to operate specifically on the defendant to discourage him or her from doing something
similar again in the future, and also generally as a guide to the conduct of others so that they can
adjust their behaviour. The tort action provides incentives to be careful.

2
 ALEX BOND
THE LAW OF NEGLIGENCE: BASIC NEGLIGENCE

Section 317(1) of the Injury Prevention, Rehabilitation and Compensation Act 2001 (“Injury Act”)
bars all civil claims for damages arising directly or indirectly out of personal injury or death where
the personal injury is caused by an accident, medical misadventure or an occupational disease.
That provision and similar versions have been a stable part of New Zealand legislation since the
first accident compensation legislation was enacted in 1974. The first Act was called the Accident
Compensation Act 1972.

Note that you can sue in civil proceedings for remedies other than damages, such as an injunction
(an order addressed to an individual restraining them from doing something that infringes
another’s legal rights). The Injury Act covers accidents but that term is interpreted quite liberally.
The term ‘injury’ is taken from the point-of-view of the victim who has suffered the injury. For
example, A’s personal injury caused by B’s battery is considered an ‘accident’ under the Act.

The tort of negligence is the most important of all torts, and is the tort one encounters most often
in practice. The problem with the law of negligence is that it lacks consistency. Judges have
different views as to what the ‘right’ answer is in some cases. Why? A court will ask itself in a
negligence case: is it fair, just and reasonable for there to be liability for negligence in this case?
When you have a fair/just/reasonable test in a tort, you will almost certainly have disagreement
amongst judges as to the limit of liability.

The essential elements of a negligence cause of action

1. The plaintiff must establish that the defendant owed the plaintiff a DUTY OF CARE.

 This issue focuses on the nature of the relationship between the plaintiff and the defendant.
It asks whether the plaintiff is one of a class of persons sufficiently close or proximate to
the defendant, so that the defendant is obliged in law to show care.
 This duty requirement exists because the potential scope of negligence is virtually
limitless. The law tries to lay a reasonable boundary to what you will be legally
responsible for. As Fleming said, the duty concern is a “control device” (a court uses it to
control what the defendant will and will not be liable for).
 If the defendant owes no duty of care to the plaintiff, the defendant can be as negligent as
he or she likes and not be legally responsible.

2. The plaintiff must show that there was a BREACH OF DUTY (negligent conduct).

 What is the applicable standard of care?

 The focus is on the defendant’s conduct. Has the defendant been careless or negligent?
Has the defendant fallen below the standard of care?
 The law compares the defendant’s conduct against the standard of a fictitious, reasonable
person: an objective test.

3. The breach must have caused the plaintiff injury or damage: CAUSATION.

 Proof of personal injury or damage has always been an essential ingredient of a cause of
action for negligence.

3
 ALEX BOND
4. DAMAGE AND REMOTENESS: The plaintiff must have suffered personal harm or injury
which attracts compensation in tort.

 The remoteness issue focuses on the type of harm that has occurred and asks whether that
harm is sufficiently proximate to attract compensation or too remote.
 The law has decided that to hold a defendant liable for all the consequences would be
disproportionate to his or her negligent conduct, i.e. it would be a burden of liability out
of scale. Ask: Was the damage a sufficiently proximate consequence of the defendant’s breach?
 The remoteness test is another "control device” on the extent of liability of a defendant.

All four elements are essential. To succeed the plaintiff bears the onus of proving them by
evidence to the balance of probabilities. The elements are interrelated. The concept of what is
reasonable pervades the whole area of negligence. The issue of reasonable foreseeability is
relevant in some way to each of the elements.

Defences to a negligence action

A defendant can defend themselves by:

1. Negativing one of the elements of the cause of action, i.e. persuading the court that one of the
four elements is not satisfied.

2. Raising an affirmative defence, such as:

 Voluntary assumption of risk: volenti non fit injuria
When the defendant pleads the plaintiff was volenti, the argument is that the plaintiff was
fully aware of the risk of harm and voluntarily agreed to run that risk, e.g. getting into the
car of someone you know is a blind drunk. If volenti succeeds it is an absolute defence
which negates the defendant’s liability altogether.
 Contributory negligence
Contributory negligence is the argument that ‘Yes, I, the defendant was negligent’ but that
the plaintiff was also negligent, and his own negligence caused his own injury (to some
extent). The Contributory Negligence Act 1927 provides that where the court finds both
the plaintiff and the defendant negligent, the court can fix the respective degrees of their
negligence, e.g. plaintiff 25% responsible, defendant 75% responsible.

4
 ALEX BOND
A. BREACH OF DUTY: SETTING THE STANDARD OF CARE

The standard of care is the primary means for determining whether the defendant was at fault:
smally negligent or careless.

Blyth v The Company of Proprietors of the Birmingham Waterworks (1851) 11 Ex 781

The defendant installed in the plaintiff’s house a fire plug with a safety valve designed to work in
stable temperatures. However during an exceptional frost the safety valve failed resulting a large
flood to the plaintiff’s house.

Negligence is famously defined in this case by Alderson B:

“Negligence is the omission to do something which a reasonable man, guided upon those
considerations would ordinarily regulate the conduct of human affairs, would do, or
doing something which a prudent and reasonable man would not do”.

Alderson B’s definition indicates negligence is not just a positive act, it can also be an omission. It
is failing to do something or take some precaution that a reasonable man would take in the
circumstances. A positive act is sometimes called misfeasance by the law, and an omission is
sometimes called nonfeasance by the law. Right from the beginning of Alderson B’s judgment it is
made clear to us that negligence can be both a positive act and an omission.

The standard of care is found in an abstract formula: what the reasonable man would do in the
circumstances. Why does the law use this abstract? It would be impossible for the courts to lay
down detailed codes of conduct for every human endeavour. To use a subjective test, i.e. what is
reasonable for you in your circumstances (tailored to the actual defendant), would be too variable
and no test at all. The courts invented this objective test measured by the standard of care of a man
of ordinary prudence.

Nettleship v Weston [1971] 2 QB 691

Lord Denning in this case quotes Lord MacMillan in Glasgow Corporation v Muir [1943] AC 448
(HL) as saying the standard of care “eliminates the personal equation and is independent of the
idiosyncrasies of the particular person whose conduct is in question”.

It is noted on the other hand that the standard cannot be too high, as that would be too harsh for
defendants. The courts have emphasised the reasonable person’s ordinariness. The reasonable
person has been described in some cases as “the man on the Clapham omnibus”, or more recently
as “the traveller on the London underground”. Fault in negligence is detemined by asking
whether the defendant’s conduct measured up to the standard of what the reasonable person
would or would not do in the circumstances.

A finding of negligence does not suggest any moral blameworthiness on the part of the
defendant, the courts suggest, because it is an objective standard. The theory is that fault has a
special meaning compared to the criminal meaning of fault which does emphasis the defendant’s
moral blameworthiness.

5
 ALEX BOND
The standard of care is set by the judge in the case. That is a question of law. Application of the
standard of care to the facts is question of fact. If there is a judge and a jury, which there seldom
is, applying the standard of care to the facts of the case is a question for the jury.

The reasonable person test has the appearance of a strictly objective test, independent of the court
or jury making a decision. But courts have sometimes quite frankly admitted that the reasonable
person test is a fiction, or a mere legal device.

McCarthy J said in McCarthy v Wellington City [1966] NZLR 481:

“The decision is made, in form, by adopting the familiar legal device of the reasonable man
— what such a man would anticipate; the fiction being that he would anticipate and guard
against only those matters which the Court considers reasonable and sufficiently proximate”.

In other words, the reasonable person is a guise, a fiction, a legal device. The reasonable person
foresees what the judge in question wants him or her to foresee. The reality is that it is the tryer of
fact – the judge or the jury – that stands behind the mask of the reasonable man. The reasonable
man test is a device used by the courts to make the courts appear as if at some distance.

Although the standard of care is that of the reasonable person, whether or not it has been attained
is determined in light of the particular circumstances of the case that the defendant found himself
in. The emphasis is on the particular circumstances of the case which gives the case its flexibility.
Those circumstances are infinitely variable.

On the facts of Blyth the defendant was absolved of liability because he had protected against
normal circumstances. It was not reasonably foreseeable that a flood would occur. But what if
there was a flood caused last year? Would there have been liability this time around? If you
change the facts you might change the result. Blyth suggests the question of negligence may have
something to do with the likelihood of the risk occurring.

The legal issue in Nettleship is whether the law will compromise its reasonable person test to take
account of the personal limitations of the defendant. In this case N was teaching his friend’s wife
W to drive, and on her third drive she froze, and drove into a lamp post, breaking N’s kneecap. N
sued W. Restating the issue: can the standard of care of the reasonable person be lowered because
W was a learner driver? (particularly when defendant knew of her inexperience).

The English Court of Appeal said the standard is that of the reasonably skilled and careful driver.
The fact that the defendant is a learner is irrelevant. She had to meet the standard of every other
driver on the road. There is no excuse or concession for amateurs. The court refused to lower the
standard of care to take account of inexperience.

Lord Denning stated the learner driver is morally not at fault, and is being held liable for failing
to meet the standard she is trying to learn (i.e. she cannot hope to attain the standard of the
reasonable driver because she is learning). But he acknowledged the other side of argument, that
everybody has to learn sometime.

Lord Denning says because W has compulsory third party insurance she should be held liable
regardless of moral fault. This is an instance of an extra-legal or policy factor influencing the
outcome of the case. It influences the legal issue of setting the standard of care in this case. The

6
 ALEX BOND
fact she will not bear all damage has led judges to set the standard of care at a high level so
plaintiffs receive compensation for their injury.

Lord Denning notes the law has begun to ask here “on whom should the risk of injury fall” as
opposed to “no liability without fault”. In other words he is saying there is really no fault on Mrs
W’s behalf, but he is still going to hold her liable so that the passenger, Mr N, gets compensation
for his injuries.

Megaw LJ accorded more weight to the disadvantages of making the standard of care dependent
on the relationship between the plaintiff and the defendant. He worried that doing such would
create a overly complex series of standards dependent on who the plaintiff actually was.

Impact of insurance on deterrence objective of tort law

Mrs’ W’s insurance company paid for the damages. In England at the time of this case there was
compulsory third party insurance. In New Zealand third party insurance is not compulsory. The
advent on insurance has had a huge impact on the tort of negligence. Frequently damages are
paid not by the actual defendant but by his or her insurance company. This is especially true in
respect of motor vehicle accidents. Professional defendants – e.g. accountants, lawyers – almost
always protect themselves against negligence actions by taking out professional emdemnity
insurance. Employers will take out insurance in respect of their own negligence. Insurance has the
effect that it shifts a large loss from the defendant and distributes it widely to all the policyholders
carrying insurance for this type of risk. Mrs W does not bear the loss here, the insurance company
does and it passes it on to all its policyholders. Insurance weakens tort law’s goal of deterrence.
The reality is that the effect of insurance in modern times blunts incentives to care. Insurance,
while incompatible with the deterrence objective of tort law, is consistent with the compensation
objective because it ensures that the plaintiff gets compensation for his loss. But what if Mrs W
had no money and no insurance? Mr W could not collect his damages reward.

In theory courts are not supposed to be told whether in fact the defendant carries insurance.
When insurance is referred to there is a mistrial. But in fact judges’ sense of this (that the
defendant is insured) is hugely influential. Courts will often speculate or assume that one or more
party is insured. It often paves the way for liability. Fleming described the factor of insurance as
being a “hidden persuader” because it tends to promote liability, especially so in the area of
compulsory insurance like motor vehicle insurance. The large crippling loss of the plaintiff is
transferred to the defendant, and then on to policyholders. This is sometimes referred to as the
‘loss-spreading’ effect of insurance; breaking up the loss amongst the community.

Sometimes the three objectives of tort law conflict with each other. In Nettleship we see the court
preferring the compensatory function of the civil action over the corrective justice function of
allocating the responsibility of the defendant based on fault.

Lord Denning in this case is responding to a demand that innocent victims be compensated.
Often judges do not explicitly acknowledge that the goals are in conflict. They do not try to
resolve the contradictions; they tend to appeal to the function of tort law which is relevant to or
promoted by the particular case.

In Megaw LJ’s judgment he refers to the Joyce case The Insurance Commissioner v Joyce (1948) 77
CLR 39, where the court held that the passenger’s awareness of the driver’s limitations may affect
that standard of care owed. Because the relationship between the parties may vary based on what

7
 ALEX BOND
the passengers knows about the driver’s limitations, the court said the standard of care may not
be the same in every case. The standard of care varies according to the knowledge the plaintiff has
of the defendant. The Court of Appeal in England expressly rejects the Joyce approach and refuses
to lower the standard of care. Megaw LJ said that the standard of care would be too variable
otherwise (as such a principle could not be confined to these facts, the court would essentially be
creating a complex series of standards dependent on who the plaintiff actually was: would create
‘unpredictability’ and ‘uncertainty’.

The passenger’s knowledge of the driver’s experience wasn’t completely irrelevant. The court said
it might mean volenti is established. It is extremely difficult to establish volenti or voluntary
assumption of risk nowadays, because if established it is a complete defence. This is why the
court prefers to apportion the damages between the two parties on the basis of contributory
negligence. Contributory negligence allows court to more carefully calibrate the outcome
according to its sense of who is to blame.

Another case on inexperience is Wilsher v Essex AHA [1987] 2 WLR 425 (CA). This case was
concerned with what the standard of care for training doctors in a hospital should be, i.e. whether
or not the law should take account of their inexperience.

Mullin v Richards [1998] 1 WLR 1304 (CA)

Will the court lower the standard of care to take account of age? Two 15-year-old girls were
fencing with plastic rulers. One ruler snapped and a plastic fragment entered the plaintiff’s eye,
causing the plaintiff to lose all sight in that eye. Which factors will the court take into account and
which will they not? The test was still an objective test of reasonable foreseeability, but the law
did not require the child to live up to the standard of the ordinary, prudent adult. It was an
objective test of the reasonable child. The fact that a child may be especially slow, forget or
inexperienced is disregarded.

The court held that the defendant was not guilty of negligence: a reasonable 15-year-old girl
would not foresee the outcome. A judgment call is involved in applying the objective test to the
facts; two reasonable people could disagree as to the right outcome. Why is the court prepared to
lower the standard of care for age but not for inexperience? Children lack maturity, and it takes
maturity to develop foresight, which is the test.

Cases tend to suggest that age is not taken into account in respect of driving, especially for a
defendant, but it is if you are a plaintiff for the purposes of contributory negligence.

Mansfield v Weetabix Ltd [1998] 1 WLR 1263 (CA)

In this case a lorry driver employed by the defendants was involve in a series of accidents in the
course of a 40-mile journey. The driver never was unaware that he suggered from malignant
insulinoma which prevented his brain from properly functioning. However he never completely
lost consciousness. We are concerned with whether the court will lower the standard of care to
take account of mental and physical disability.

The trial judge followed Roberts v Ramsbottom [1980] 1 WLR 823. The judge applied the criminal
law test for automatism and said only if you are an automaton that you are released from liability.
This plaintiff had some control and thus was still liable.

8
 ALEX BOND

The Court of Appeal said that the trial judge had erred in applying the criminal law test: “In my
judgment, criminal cases can only introduce confusion.” What was the standard of care? The standard
of care does take account of disability but not in every case. Disability will affect the standard of
care only if you are unaware of your disability. If you are aware and disregard your disability,
you look negligent and are at fault: in line with the corrective justice objective of tort law. If the
defendant is aware of their disability then they must take account of it and take corrective action.

Inexperience is irrelevant, but disability may be relevant. Why was disability an excuse in this
case but not inexperience in Nettleship? Both cases are about driving on the road. Are the opposite
outcomes reasonable? The cases aren’t really reconcilable. In both cases the court assumed the
plaintiff had insurance; the ‘hidden persuader’.

In Nettleship the court emphasised the objective of compensation, whereas here in Mansfield the
court is more sensitive to the corrective justice function of the law. The court thinks this defendant
is blameless, so it would unreasonable to visit fault on him. The stark contrast in these two cases,
is a good illustration of the fact that there is no right answer. The court furtherst up the hierarchy
is right – both are from the Court of Appeal, but Mansfield is more recent.

Billy Higgs v Baddeley [1950] NZLR 605

This is another (yet older) case concerning disability, enunciating a similar principle to Manfield.
The defendant, a driver travelling with a passenger, the plaintiff, went to overtake a slow-moving
milk lorry but drove into the back of lorry, injuring the plaintiff passenger. The driver stated as
evidence that something got into his eye (he thought it was a bee-sting) which caused him to
suddenly put his foot down on the accelerator.

Gresson J: “…A driver who suffers a sudden fainting to the extent of unconsciousness is not negligent,
unless, again, he was aware of the danger of such unconsciousness coming upon him. While unconscious,
he is physically incapable of taking care, and is not responsible for that which he is incapable of preventing.
In this case, it was not unconsciousness which befell the driver, it was a sudden and unavoidable onset of
pain, which completely distracted him from his duties as driver.”

Simply stated: if you suddenly lose consciousness, and you are unaware that you ran the risk of
doing so, the standard of care will be affected.

Goldman v Hargrave [1967] 1 AC 645 (PC)

In this Privy Council case the court was prepared to take account of the personal limitations of
the defendant. The facts concerned a farmer who had to deal with a fire on his land – which had
been caused by lightning striking his property – through no fault of his own. The Privy Council
took account of subjective factors, the defendant’s ‘individual means’.

Lord Wilberforce at pg 663: “…The standard ought to be to require of the occupier what it is reasonable
to expect of him in his individual circumstances. Thus, less must be expected of the infirm than of the able-
bodied: the owner of a small property where a hazard arises which threatens a neighbour with substantial
interests should not have to do so much as one with larger interests of his own at stake and greater
resources to protect them: if the small owner does what he can and promptly calls on his neighbour to

9
 ALEX BOND
provide additional resources, he may be held to have done his duty: he should not be liable unless it is
clearly provided that he could, and reasonably in his individual circumstance should, have done more”.

In other words, Lord Wilberforce is saying that courts should take caution in holding occupiers
who did not cause the danger themselves to a reasonable occupier standard. He made a distinction
between those occupiers who have sufficient resources to quite easily extinguish a fire, and those
who do not. The court will expect occupiers to behave reasonably given the resources they have
available to them.

Bolam v Friern Hospital Medical Committee [1957] 1 WLR 582

What about where the defendant has special skills or knowledge, such as a lawyer, doctor,
architect or builder? What is the standard of care in such a case? In general the standard of care
whilst still objective is the standard of conduct which ought to be obtained by persons ‘holding
themselves out’ as possessing that special standard.

The classic statement of the standard of care of a professional is stated in the case Bolam, which
concerned the standard of care of a doctor.

McNair J at 586:

“Where you get a situation which involves the use of some special skill or competence, then
the test whether there has been negligence or not is not the test of a man on the top of the
Clapham omnibus, because he has not got this special skill. The test is the standard of the
ordinary skilled man exercising and professing to have that special skill.”

Whitehouse v Jordan [1980] 1 All ER 650, [1981] 1 WLR 246 (HL)

This House of Lords case concerned the delivery of a baby after an attempted forceps delivery.
The baby was born seriously disabled. The plaintiff claimed the defendant had pulled too hard
and for too long. Prior to this Lord Denning in the Court of Appeal had said at p568:

“Every one of us every day gives a judgement that is found to be wrong. It may be an error of judgement
but it is not negligent. We must say and say firmly that in a professional man, an error of judgement is not
negligent….To test it, I would suggest that you ask the average competent practitioner…applying this test
I am clearly of the opinion that the defendant pulled too hard and for too long, but was not negligent…”.

Lord Denning was chastised for this statement in the House of Lords. Why? The line between an
error of judgment and negligence is too ambiguous. The House of Lords said it is a “false
antithesis” to say an error of judgment is not negligent. The House of Lords properly corrected
that. Lord Edmund Davies said at p257:

“To say that a surgeon committed an error of clinical judgment is wholly ambiguous, for, while some such
errors may be completely consistent with the due exercise of professional skill, other acts or omissions in the
course of exercising “clinical judgment” may be so glaringly below proper standards as to make a finding of
negligence inevitable”.

If it is an error of judgment which no reasonable practitioner would make, then it is negligent. If

on the other hand an error that a reasonably competent practitioner might have made, then it is

10
 ALEX BOND
not negligent: a non-negligent error of judgment. To label something an error of judgment does
not tell you whether something is negligent.

The courts have long warned in negligence against the danger of judging negligence with the
luxury and wisdom of hindsight. Whether or not someone has been sorely negligent, is judged
prospectively. The court is supposed to put itself in the position the defendant was in when the
defendant was acting. It is not judged retrospectively at the date of the court action, with full
knowledge of how the events worked out. It is a basic tenant of negligence law that in order to
judge fairly whether the defendant was negligent, the conduct must be judged in light of the
circumstances, facts known and prevailing standards at the time of the alleged negligence.

In Roe v Ministry of Health [1954] 2 All ER 131 Lord Denning said “We must not look at the 1947
accident with 1954 spectacles”. You have to place the conduct in its context in light of the facts and
knowledge reasonably possessed at the time, not with the ‘perfect vision’ afforded by hindsight.

Shakoor v Situ [2000] 1 WLR 410

The defendant was a practitioner of alternative Chinese medicine. There were some notices in
medical journals which warned of the risks of liver toxicity with certain concoxions of Chinese
herbal medicine. The victim had a very toxic reaction in his liver to the herbal remedy prescribed.
The plaintiff, this man’s widow, claimed the defendant should have been familiar with the
warnings laid down in the journals.

The issue in this case was what is the standard of care? The standard of care is not that of an
orthodox doctor, but the standard of a reasonable competent practitioner in traditional Chinese
herbal medicine (‘TCHM’). This practioner did not ‘hold himself out’ to be an orthodox medical
practitioner. The patient had rejected orthodox medicine and went to this practitioner of this
alternative medicine. The practioners will be judged against the standard of practioners “of that
art”. But is it the standard of a practioner of TCHM in Beijing? No. The events took place in
England. The standard of care will take account of fact that the defendant is practicing his ‘art’ in
a western country alongside orthodox medicine. You have to take account of the fact that you are
practicing your ‘art’ in the particular locality.

One of the implications of practicing in Britain was that adverse practice would be written up in
medical journals. The defendant had to take reasonable steps to check that there had not been an
adverse report in respect of the defendant’s TCHM. The court decided the equivalent of a
practitioner of Chinese medicine is a general practitioner, not a specialist. They would expect a
reasonable general practitioner and a reasonable Chinese practitioner to have the same level of
awareness of adverse reports. The court held a reasonable general practitioner would not have
noted these warnings because the adverse reports were in quite specialist medical journals. This is
how the court got around the difficult question.

If there is no ‘holding out’, the defendant will not have to comply with the higher standard.

Philps v William Whitely [1938] 1 All ER 566

This case concerned a jeweller who pierced ears and was sued in negligence when his patient’s
ears became very infected. The court held that a jeweller who pierces ears only has to measure up
to the standard of care of a jeweller of reasonable standard and skill. He does not have to measure

11
 ALEX BOND
up to the standards of sterility of a surgeon, because he ‘held himself out’ as a jeweller, not a
surgeon. If the jeweller had in some way ‘held himself out’ as a surgeon, which misled people to
think he was a surgeon, then by virtue of that ‘holding out’, the law would hold him to the higher
standards of a surgeon.
Lack of skill and experience is no excuse if the task demands greater skill and experience. In those
circumstances the reasonable person would summon specialist advice or assistance, and they will
be liable for not doing so; e.g. If the defendant is not a surgeon but undertakes to carry out
surgery other than in emergency circumstances, he or she will have to meet the standard of a
reasonable surgeon in the circumstances.

Whether specialist help and advice is required is a matter of fact to be determined by the judge or
jury. There are clear examples, like surgery, but there are other cases where courts have held that
the reasonable person isn’t obliged to seek advice or help. Wells v Cooper [1958] 2 QB 265
illustrates one example: minor household repairs, the reasonable person can be a ‘home
handyman’ and does not have to be held to the standard of a reasonable carpenter.

12
 ALEX BOND
BREACH OF DUTY: APPLICATION

We know what the standard of care is, we are now looking at the question of breach of the
standard of care. The courts use the terms ‘breach of duty’ and ‘breach of the standard of care’
interchangeably: both have the same meaning.

Bolton v Stone [1950] AC 850 (HL)

The plaintiff was walking down a street besides a cricket ground and was hit by a cricket ball that
soared over the boundaries. She was personally injured and sued for her injuries. In the 90 years
the ground had been in use no one had ever been hit, but the ball had been hit out of the ground
on 3 separate occasions over the past 28 years.

Did the House of Lords consider that the risk that this type of accident might occur was a
foreseeable risk? Yes. Once it happened once it was clear it could happen again. It is not
completely unforeseeable. Did Lord Reid quantify that risk as large? No. Lord Reid said the risk
was foreseeable but very small.

The issue raised in this appeal is whether the true test of breach of duty is foreseeability alone –
that is that the risk is foreseeable, no matter the degree of risk, probable or improbable – or is
something more than foreseeability? The test of negligence is not foreseeability alone. The
reasonable man will take account also of the size or degree of the risk and the probability of the risk
occurring. Foreseeability is a threshold requirement. But once the risk is foreseeable, then in
deciding how to act, the reasonable person must take into account the magnitude of that risk. Is it
a very small risk or is it a more likely risk?

Lord Reid at 865:

“I think that reasonable men do in fact take into account the degree of risk and do not act on
a bare possibility as they would if the risk were more substantial.”

He then formulates a test at 867:

“…The test to be applied here is whether the risk of damage to a person on the road was so
small that a reasonable man in the position of the appellants, considering the matter from the
point of view of safety, would have thought it right to refrain from taking steps to prevent the
danger.”

Another factor the reasonable person and a court will take into account is the seriousness of the
potential consequences if the risk eventuates, e.g. compare the injury that a tennis ball and a
cricket ball would result in.

The court held that the cricket club was not negligent, because even though the potential
consequences were big, the risk was so very small that the reasonable person would be justified in
ignoring the risk.

It has been suggested that there was a hidden policy factor in this case: the House of Lords’ love
of the game of cricket.

13
 ALEX BOND
Overseas Tankship (UK) Ltd v The Miller Steamship Co. Pty. [1967] AC 617 a.k.a. Wagon Mound No. 2

This is an Australian case taken to the Privy Council. The engineers on board the Wagon Mound
carelessly allowed a quantity of oil to overflow into Sydney Harbour. The oil spread to a wharf
where the Overseas Tankship company was welding. The result was damage to one of its ships.

There were two Wagon Mound cases, this case is no. 2:

 Wagon Mound No. 1: the plaintiff was the owner of the wharf which was damaged.
 Wagon Mound No. 2: the plaintiff was the owner of one of the ships moored at the wharf
which was also damaged.

The two cases have different outcomes. How can that be? In No. 1, the trial judge made a different
finding of fact, finding that it was completely unforeseeable that oil would catch alight if floating on
water. The plaintiff in that case wasn’t arguing it was foreseeable, because if it was found to be
foreseeable there would be a claim of contributory negligence.

In No .2, there is discussion in the House of Lords on whether or not it is foreseeable that oil might
catch alight on water. Because the owner of the ship wasn’t welding, the owner could run the
argument of foreseeability without potentially being contributorily negligent. The court said it
was “possible, but would only happen in exceptional circumstances”, i.e. a foreseeable but very small
risk. The case proceeds on the same basis as the cricket ball injuring someone in Bolton v Stone.

The different factual bases of the two cases explains the different outcomes in the cases. In No. 1
the plaintiff loses, in No. 2 the plaintiff wins.

A new problem was posed by Bolton v Stone. Before Bolton v Stone the law had been thought to
have drawn two categories:
1. Where the risk would have been regarded as unreal or unforeseeable either because it was
physical impossible or the risk would have been regarded by the reasonable person as so
‘fantastic’ or ‘far-fetched’.
2. Where the risk is real and substantial and a reasonable person has to take steps to
eliminate or minimise those risks.

Bolton v Stone seemed to add a third category:

3. Foreseeable but very small risks, which are reasonable to disregard.

Lord Reid qualifies what he said in Bolton v Stone. He says the decision is correct on its facts, but
that too much reliance has been placed on statements he made in it. He says, once there is a real
risk, even if the risk is very small, you aren’t necessarily justified in disregarding the risk.

“But it does not follow that, no matter what the circumstances may be, it is justifiable to neglect a risk of
such a small magnitude. A reasonable man would only neglect such a risk if he had some valid reason for
doing so, e.g., that it would involve considerable expense to eliminate the risk.”

He says you have to take steps to eliminate a foreseeable albeit very small risk if there is a valid
reason to do so. For example, if it is very cheap or very easy to minimise/eliminate the risk, even if
very small, you must do so. In Bolton v Stone there was a valid reason to disregard the risk: it was
virtually impossible to eliminate the risk short of stopping playing cricket on the ground
altogether (the House of Lords ignored the idea of building higher fences around the ground).

14
 ALEX BOND

So from this case we have another factor: the relative ease or difficulty of taking remedial steps.
How hard or easy would it be to eliminate or minimise the risk, before the damage occurs?

So far, what factors would a fictitious reasonable man weigh in the balance before deciding how
to act given a set of circumstances?
1. the degree or magnitude of the risk: is it small or it is more substantial?
2. the seriousness of the potential consequences if the risk eventuates
3. the ease or difficulty of taking remedial measures: if it is cheap to eliminate the risk, that is
going to push strongly in finding there was a breach.

In addition, it is to be noted as relevant that it was illegal to discharge oil into Sydney Harbour.
Just because something is a criminal offence doesn’t of itself follow that there will be a finding of
negligence, but it is another factor the courts will balance. There is no validity to the defendant’s
conduct. But in Bolton v Stone the defendants were undertaking a legitimate activity.

As obiter Lord Reid noted that if the activity that caused Miss Stone’s injury had been unlawful
there can be little doubt that Bolton v Stone would have been decided differently, i.e. the
legitimacy or utility of the defendant’s conduct is relevant.

Lord Reid has redrawn the categories. He distinguishes between the first category – ‘far fetched’
or ‘fantastic’ risks that no reasonable person would pay attention to – which you can safely
disregard. He calls the second category ‘real risks’, i.e. a risk that is reasonably foreseeable.

Summary: what factors would a fictitious reasonable man weigh in the balance before deciding
how to act given a set of circumstances?
1. the degree or magnitude of the risk: is it small or it is more substantial?
2. the seriousness of the potential consequences if the risk eventuates
3. the ease or difficulty of taking remedial measures
4. the social value or utility of the activity (e.g. is it a crime?)

In Wagon Mound No. 2, the defendants lost. Which of the factors influenced? No. 4: there was no
social utility to the defendant’s conduct (it was a criminal offence). Which of the factors pushed
away from liability? No. 1: It was a very small risk.

Paris v Stepheny BC [1951] AC 367

This House of Lords case illustrates the factor of the seriousness of the potential consequences.

The plaintiff was a mechanic, who was already blind in one eye. He was fully blinded when he
suffered damage to his good eye. A chip flew up and entered his good eye. He claimed against his
employer. The evidence was that it was not general practice in the industry at the time for
employees to wear goggles while carrying out work. The House of Lords held a breach in this
case. The reasonable and prudent employer would consider not only the risk of the accident, but
the seriousness of the consequences if the risk eventuated. i.e. here it was more of a risk to this
employee because he was already blind in one eye.

Watt v Hertforshire CC [1954] 1 WLR 835 (CA)

15
 ALEX BOND

This case illustrates factor of social value or utility of the defendant’s activity.
The plaintiff was employed as a firefighter by the defendant. An emergency call came through
that a woman was trapped under a heavy vehicle only 200-300 yards from the station. They
needed to transport a jack to prop up the vehicle to get her out. The jack was really heavy and
needed a vehicle to take it to the victim. The vehicle they normally used to carry that jack was
already out on a call. The officer in charge instructed them to carry the jack in another vehicle,
where it could not be strapped down. On the trip, the driver had to break suddenly. The jack
moved and all three men were thrown off balance and the plaintiff was injured by the jack.

He sued for his injuries, claiming this should not have been done. Lord Denning said you must
balance risk against the end to be achieved. In this case it was highly relevant that the end to be
achieved involved saving life and limb. The court held that the high social utility (saving a life)
meant that the small risk could be run. No liability for the defendant.

Note however that just because a defendant’s conduct is socially useful, that doesn’t confer an
immunity. It is always something to be weighed in the balance. It is a factor to be taken into
account. Even where there is a high degree of social utility to the defendant’s question, it is still a
question of weighing up the factors involved.

e.g. Police car chases when trying to apprehend an offender: so often the offender will be
travelling at high speeds and might plunge into an innocent person of the public/a property/etc.
There has been criticism that the police run too great a risk travelling at such high speeds,
endangering themselves, innocent members of the public and the offenders. Each case depends
on its particular facts, but at some point, the risk of injury outweighs the social benefit of the
defendant’s conduct.

United States v Carroll Towing Co 159 F.2d 169

Learned Hand J in this American case formulated a algebraic test to determine negligence.

Given P = probability of harm occurring

L = gravity/cost of injury
B = cost/burden of taking precautions

Negligent if:
PxL>B

In other words, if the probability of injury occurring and the seriousness of the potential injury
outweigh the cost to prevent it, the defendant is negligent: a cost-benefit analysis.

Learned Hand J does omit the social utility of the defendant’s conduct as a factor.

Hammond J acknowledged the usefulness of this formula, but also its main difficulties in the
Court of Appeal case Wilson & Horton v AG [1997] 2 NZLR 513:

 “…those variables in most cases are not capable of quantitative estimate”; and
 “…the law of negligence should not lie just in a search for the cheapest loss avoider”.

16
 ALEX BOND

In claims against professional defendants in particular, in determining the question of breach, the
court often attaches considerable importance to the expert evidence of other practitioners.

Bolitho v City and Hackey Health Authority [1997] 3 WLR 1151

In this case involving doctors, both the plaintiff and defendant brought along experts who gave
an opinion as to whether or not the defendant’s conduct was negligent. Such expert opinions are
intended to influence the court, and in fact they are often incredibly influential. Often the experts
will give evidence of common or accepted practice, i.e. what others would do in the industry.

This case is authority for the principle that the accepted professional practice of others in a similar
situation is relevant but not conclusive of a finding of negligence.

This case concerned a 2-year-old boy named Patrick who suffered severe brain damage after a
cardiac arrest resulting from respiratory failure, after the attending Dr Horn failed to attend on
two occasions after being summoned. Note however the doctor has admitted a breach/that she was
negligent.

The immediate issue was a question of causation: whether Patrick would have been intubated had
the breathing tube been put down his throat before the cardiac arrest occurred. Dr Horn’s
evidence was that if she had attended, she would not have intubated Patrick. The judge accepted
that evidence.

The second issue was whether any reasonably competent doctor have intubated Patrick if Dr
Horn had not. Note we are concerned with an omission here: nonfeasance.

So, there are two enquiries:

1. If she had attended, would she have put the tube down his throat to give him an airway?
(She said no)
2. If she would not have intubated him, would any reasonably competent practitioner have
done so in her position?

A total of eight distinguished doctors were called to give evidence: five experts on behalf of the
plaintiff/Patrick, and three on behalf of the defendant/Dr Horn. The evidence was to support one
way or the other the second enquiry. The two groups had completely opposed views or in the
words of Lord Browne-Wilkinson “diametrically opposed” views. The plaintiff’s experts all said that
intubation was the safe option and any reasonably competent doctor would have intubated
Patrick. The defendants’ experts all said it was entirely appropriate to watch and wait, as
intubation has its own risks, etc.

Lord Browne-Wilkinson held that it was not illogical and unreasonable to not intubate. He sees
the logical force of the plaintiff’s experts, but they disagree with each other, so he goes for
defendant’s view. He applied the Bolam case which he felt prevented him from preferring the
plaintiff’s evidence.

17
 ALEX BOND
The dictum of McNair J in Bolam quoted at 1156 (pg18) was: “I myself would prefer to put it this way,
that he is not guilty of negligence if he has acted in accordance with a practice accepted as proper by a
responsible body of medical men skilled in that particular art…Putting it the other way round, a man is not
negligent, if he is acting in accordance with such a practice, merely because there is a body of opinion who
would take a contrary view.”
The trial judge referred to Lord Scarman’s speech Maynard v West Midlands Regional Health
Authority [1984] 1 WLR 634, where Lord Scarman said:

“I have to say that a judge's 'preference' for one body of distinguished professional opinion to
another also professionally distinguished is not sufficient to establish negligence in a
practitioner whose actions have received the seal of approval of those whose opinions, truthfully
expressed, honestly held, were not preferred. If this was the real reason for the judge's finding,
he erred in law even though elsewhere in his judgment he stated the law correctly. For in the
realm of diagnosis and treatment negligence is not established by preferring one respectable
body of professional opinion to another. Failure to exercise the ordinary skill of a doctor (in the
appropriate speciality, if he be a specialist) is necessary."

It was subsequently suggested that this statement basically equated conformity with a practice
which was accepted in the profession with the absence of negligence. It was suggested that at its
highest, the Maynard case meant all a defendant doctor had to do to be absolved from liability was
to bring along one expert who would testify that, in their opinion, what the defendant doctor did
was something that others in the profession might do. If you could point out the defendant’s
conduct was practiced in the profession, even if many believed it was below the standard, this
would be enough to absolve defendant.

The plaintiff’s (Bolitho’s) lawyer argued that the Bolam test didn’t require the trial judge to accept
the views of one truthful body of professional opinion. He said ultimately it is not for doctors to
set the standard of care for negligence, but for the court.

There are some judicial statements which suggest that this Bolam test means that the practice of
medical profession is determinative of the issue of the standard of breach. Provided a doctor
conforms with an accepted practice, he or she cannot, as a matter of law, be found guilty of
negligence/must necessarily be absolved of liability. Where there is one accepted practice given by
the external evidence, provided that the defendant complies with one of them, he or she will be
excused. It doesn’t matter if there is a contrasting professional opinion. Plaintiffs would scarcely
win as a result.

This understanding in Bolam was rejected by the House of Lords in Bolitho. Lord Browne-
Wilkinson said that a doctor does not escape liability just by bringing expert evidence that the
medical treatment provided by the defendant was in line with sound medical practice. He instead
qualified that by saying the expert opinions must have a “logical basis” and be “responsible”,
“reasonable” or “respectable”. This includes that in coming to their views the experts must have
considered the comparative risks and benefits of adopting a particular medical practice.

The House of Lords said a court is not bound to conclude that a doctor may escape liability, just
because he or she leads evidence from a number of medical experts or even one medical expert,
who holds the view that what the defendant did accords with sound medical practice. A court is
not obliged to absolve a defendant just because his or her conduct conforms with what others do
in the profession. The court has to be satisfied that the expert opinion a defensible “logical basis”.

18
 ALEX BOND

The qualification is added by emphasising the words responsible, reasonable and respectable. The
word respectable was also emphasised in the passage from the Maynard case quoted above. The
court is responsible for setting the standard of care in negligence and if it can be demonstrated
that the opinion does not have a logical basis, the court can ignore such an opinion. Lord Browne-
Wilkinson noted however that this would only happen in “rare” cases.
An example where the courts court ignored expert evidence is Hucks v Cole [1993] 4 Med LR 393
(a 1968 case), where a doctor failed to prescribe penicillin to a patient. The court found the
defendant doctor negligent even though there was evidence that others in the profession would
have done the same thing – the court said the experts were adhering to outdated practice.

For a non-doctor example, see the case of Edward Wong Finance Co Ltd v Johnson Stokes & Master
[1984] 1 AC 296, a case involving solicitors in Hong Kong. The defendant solicitor payed over
money pursuant to a mortgage and did so ‘Hong Kong style’ on the basis of an undertaking
(promise) by the solicitor for the other side that they would hand over the executed security
documents. The defendant expected to get the signed mortgage documents, but instead the
solicitor for the other side absconded with the money. Was the defendant solicitor negligent? He
argued that everyone is Hong Kong did transactions in this way. The Privy Council found that
even though defendant’s conduct conformed with the universal practice, the defendant was
nevertheless negligent. As a result, the whole profession was negligent. Their practice lagged
behind that that the profession required. There was an obvious risk that the money could be
stolen by the other solicitor.

Lord Browne-Wilkinson went on to say: “As the quotation from Lord Scarman makes clear, it would be
wrong to allow such assessment to deteriorate into seeking to persuade the judge to prefer one of two views
both of which are capable of being logically supported. It is only where a judge can be satisfied that the body
of expert opinion cannot be logically supported at all that such opinion will not provide the bench mark by
reference to which the defendant's conduct falls to be assessed.”

The above makes it clear that this is no invitation for a court to look at divergent opinions where
both are capable of being logically supported and to prefer the one it likes best.

The defendant won on the facts of this case, but you could argue about the application of the test.

This decision was welcomed particularly in the field of medical negligence, because some courts
had interpreted the Bolam decision as a rule of law pursuant to which a court could not find a
doctor negligent so long as he conformed to some common medical practice.

Courts show very great deference (courteous regard) to expert opinion. In all cases what the
experts say is highly influential, but it is not conclusive.

19
 ALEX BOND
BREACH OF DUTY: PROOF OF BREACH, RES IPSA LOQUITUR

Res ipsa loquitur translates as “the thing speaks for itself”. This maxim allows the court to draw an
inference of negligence from the mere happening of an event or accident, without the benefit of
detailed evidence of cause and responsibility.

In other words, the mere happening of the event speaks negligence for itself.

1. When does res ipsa apply?

2. What is the effect of res ipsa?

When does res ipsa apply?

Scott v London and St Katherine’s Docks (1865) 3 H&C 596

The plaintiff was standing near the door of the defendant’s warehouse and was injured when
some bags of sugar fell on him. The plaintiff was unable to prove any specific acts of negligence
by the defendants, who themselves called no evidence. The trial judge directed the jury to find for
the defendants but on appeal the direction was overturned and a new trial was ordered.

Erle CJ explains why: “There must be reasonable evidence of negligence, but where the thing is shown to
be under the management of the defendant or his servants, and the accident is such as in the ordinary
course of things does not happen if those who have the management use proper care, it affords reasonable
evidence, in the absence of explanation by the defendants, that the accident arose from want of care.”

There must be reasonable evidence of negligence. The maxim will apply if the following two
conditions are satisfied, in the absence of explanation by the defendants.

1. The thing that inflicted the damage must be shown to be under the management of the
plaintiff or his servants
2. The accident or injury is such that in the ordinary course of things does happen without
negligence

Modern textbooks have recast these two conditions into three:

1. The thing that inflicted the damage must have been under the sole control or management
of the defendant
2. The event is such that it would not ordinarily happen without negligence
3. There must be no evidence that the plaintiff can adduce as to why or how the event took
place. If the plaintiff has some evidence albeit incomplete (if the facts are partially known),
then the courts have said that appeal to res ipsa is inappropriate.

If these three conditions are satisfied an inference of negligence can be brought in the absence of
an explanation from the defendant that the act was not negligent.

Why do we have this? Without it it would be unfair for plaintiffs who would not be able to prove
negligence. The facts are in the knowledge of the defendant, and the defendant is not going to
divulge those facts for fear of liability. As Allen Linden said, the purpose of res ipsa is to “smoke

20
 ALEX BOND
the defendant out”. The maxim is designed to get the defendant to volunteer facts if there are
facts which would exculpate him or her.

Res ipsa has been held to apply in several medical cases, e.g. where the defendant has been in an
operation under a general anaesthetic so in no position to prove what happened. In other words,
the plaintiff at a disadvantage in getting that evidence.

There are several medical negligence cases about retained instruments. The courts have held that
all the maxim res ipsa may support is an inference that one member of the surgical team was
negligent, not necessarily that it was the surgeon. That is because responsibility in all theatres is
divided between the operating surgeon and the nurse counting the swabs, etc. In the cases it is
not always clear whether the fault is the surgeon’s or the theatre nurse’s. If the latter it is
considered that the count nurse is an independent professional (thus the hospital could not be
sued on the basis of vicarious liability).

MacDonald v Pottinger [1953] NZLR 196

In this New Zealand case the court considered that proof that a pair of forceps found in the
plaintiff’s body “raises an inference of negligence against someone”, such that an explanation is
called for. Allowing res ipsa to raise an inference of negligence encourages the hospital to accept
liability vicariously as an employer.

What is the effect of res ipsa?

Hawke’s Bay Motor Co Ltd v Russell [1972] NZLR 542 (SC)

In this case the then Supreme Court discussed the extent of the res ipsa losquitur rule. There were
two different views of the rule.

First “stronger” view:

 If res ipsa losquitur is successfully invoked, it will reverse the normal burden
of proof. The defendant must rebut the inference of negligence drawn by the maxim, i.e.
must prove on the balance of probabilities that he or she were not negligent.
 The policy for this strong version of res ipsa is to motivate to the defendant
to divulge information which only he or she has access to.
 The defect to this view is that it puts the plaintiff who has no evidence of
negligence in a stronger position that a plaintiff who does. The plaintiff who does have
evidence faces the normal burden of proof, but the plaintiff without does not.

Second “weaker” view:

 Res ipsa loquitur allows the plaintiff to establish a prima facie case in
negligence and survive a non-suit (dismissal of the plaintiff’s case at the end of the
plaintiff’s side of the story). The ultimate burden of proof rests on the plaintiff throughout
the trial. But res ipsa only alters the burden of persuasion, sometimes called the evidential
burden. The defendant might feel obliged to give an explanation.
 Courts in the common law prefer this weaker view.

21
 ALEX BOND
 Beattie J of the Supreme Court preferred this second view over the first
view, as did the Magistrate in the first instance court.

“The burden of proof may often shift, or appear to shift in any evidentiary way, during the course of a
hearing but nevertheless the onus does not as a matter of law shift to the defendant. If, therefore, the
applicability or otherwise of the rule does not require affirmative disproof on the probabilities then, while the
plaintiff has shown beyond any shadow of any doubt that the defendant’s vehicle went on to its incorrect
side of the highway and struck the coach, nevertheless, … I cannot say on the probabilities that the plaintiff
has shown the defendant to be negligent.”

Ng Chun-Pui v Lee Cheun-Tat [1988] 2 HKLR 425 (PC)

The conflicting judicial opinions above were authoritatively settled in this Privy Council decision.
The Privy Council held that if properly invoked the maxim doesn’t shift the burden of proof to
the defendant to disprove negligence. All it does is enable a plaintiff to establish a prima facie case
of negligence sufficient to survive an argument by the defendant at the close of the plaintiff’s case
that there is no case to answer, i.e. non-suit.

22
 ALEX BOND
B. DUTY OF CARE [this is the first issue to consider and comes before standard of care]

This element focuses on the relationship between the parties: where the relationship is sufficiently
proximate such that the defendant owes the plaintiff a legal duty to avoid harming the plaintiff.
The starting point to the modern approach is found in Donoghue v Stevenson.

M’Alister (or Donoghue) v Stevenson [1932] AC 562

The appellant Mary Donoghue and a friend visited a café. The friend ordered icecream and
ginger beer, to make an icecream float. The friend poured some of the bottle into Donoghue’s
glass and Donoghue drank it. The friend poured out more of the contents of the opaque bottle into
the glass and the half decomposed snail came out. Mary Donoghue as a result suffered severe
gastro-enteritis. Mary Donoghue did not buy the ginger beer herself, her friend bought it for her.
Had she bought it for herself, there might have been a claim in contract against Minchella, the
store owner.

The main issue was whether there is a duty owed by a manufacturer to the ultimate consumer of
his products. The law at the time was established in the case Mullin v Barr [1929] SC 461, with the
same facts but a mouse instead of a snail. The claim in this case had been dismissed by the highest
court in Scotland, the Court of Session: ”in a case like the present, where the goods of the defenders are
widely distributed throughout Scotland, it would seem little short of outrageous to make them responsible
to members of the public for the condition of the contesnts of every bottle which issues from their works”.

Lord Buckmaster wrote a length dissenting judgment in Donoghue. His judgment is quite correct
in terms of precedent that the claim failed, as it did not fall within any existing category of duty.

According to his analysis, liability in tort existed in four exceptional classes of case:
1. The duty of the owner of an inherently dangerous chattel, e.g. a gun, to users of it
2. The duty of the manufacturer of a good known to the manufacturer to have a dangerous
defect, to consumers of that good
3. The duty of care of the owner of dangerous premises to visitors and invitees, an example is
the Heaven v Pender case
4. The duty of an individual belonging to a certain occupation known to the law as the
‘common callings’. ‘Common callings’ included the inn-keeper, a veterinary surgeon and a
carrier. The essential feature of a ‘common calling’ was that the services were generally
available to the public.

Apart from these exceptions, the liability of a manufacturer was seen as arising in contract alone.
Here the plaintiff was attempting to sue on a contract which she was not a party to. Lord
Buckmaster was also motivated by the fear of indeterminate liability if the duty was recognised in
this case. He said “if one step, why not fifty?”. This policy concern is known to the law as “the
floodgates”. It has been a very powerful policy factor especially in the law of negligence.

The majority considered that the case had nothing to do with contract law and the doctrine of
privity of contract, because the plaintiff was suing in respect of the plaintiff’s negligent conduct.
The majority took the view that whether the defendant happened to be acting in the course of a
contract was essentially immaterial.

23
 ALEX BOND
Lord Atkin said that in the law of negligence you had to bring your facts within a category of
liability previously recognised by precedent, otherwise your action would fail. The courts had
recognised and described several categories of liability. The problem was that these categories
were ‘one-off’ or ad hoc, lacking a common intention.

The courts began searching for a general principle that explained when a duty would be
recognised. There was a prior attempt to do by Lord Esher in Heaven v Pender. In that case the
plaintiff was employed as a ship’s painter and was injured when equipment supplied by the dock
owner collapsed on him. Liability was recognised within a pre-existing category, that of a duty of
care owed by an occupier to an invitee. Lord Esher said “Whenever one person is by circumstances in
such a position with regard to another that everyone of ordinary sense who did think would at once
recognise that if he did not use ordinary care or skill in his own conduct, he would cause danger or injury
to the person, property or other. A duty arises”. This was quite similar to Lord Atkin’s neighbour
principle. But Lord Essher was not supported by other members of the court in Heaven v Pender.

In the later case Le Lievre v Gould, Lord Essher himself qualified his principle. He added the
qualification of physical proximity or physical nearness to his statement. The plaintiff and the
defendant had to be physically proximate to each other.

In Donoghue Lord Atkin describes the categories of case and said “There must be some element
common to all the cases where liability is found to exist.” He searched for a general principle to explain
all the cases of duty, not limited to pre-existing categories.

His neighbour principle is based on the notion of fault. As he says liability in negligence “is based
on a general public sentiment of moral wrongdoing for which the offender must pay”. He draws the
principle from the Judeo-Christian ethic of love-thy-neighbour. The test is objective. It is a notion
of reasonable foreseeability or reasonable contemplation: what the defendant should reasonably
have foreseen or reasonably had in contemplation.

“The rule that you are to love your neighbour becomes in law, you must not injure your neighbour; and
the lawyer’s question–Who is my neighbour?–receives a restricted reply. You must take reasonable care
to avoid acts or omissions which you can reasonably foresee would be likely to injure your neighbour .
Who, then, in law is my neighbour? The answer seems to be–persons who are so closely and directly
affected by my act that I ought reasonably to have them in contemplation as being so affected when I am
directing my mind to the acts or omissions which are called in question”.

He discusses Lord Esher’s attempt in formulating a principle and his later qualification to that
principle. He said that the principle should not be tied to physical proximity, although physical
proximity may be a factor in bringing the plaintiff within your contemplation.

The majority held the existence of a contract between the defendant manufacturer and the third
party store owner did not prevent the defendant from owing a duty of care to the plaintiff in
relation to the performance of the contract.

Lord Buckmaster’s floodgate concerns, which were rejected by the majority, have been very
persuasive to the law in contentious issues of liability (see the later topic of Advanced Negligence),
for example, in cases of nervous shock or pure financial loss.

24
 ALEX BOND
It was critical to the case that the ginger bottle was made of opaque glass. The legal proposition
was that there was no reasonable opportunity of intermediate examination of the bottle. In these
circumstances the plaintiff could establish the necessary relationship of proximity. In the Mullin v
Barr case there was a clear bottle. If there is a reasonable opportunity of intermediate examination,
the relationship of proximity might be destroyed and there can be no duty of care. The defence of
intermediate examination is still raised sometimes, for example in the leaky homes litigation. It is
said that the owner could protect himself or herself by examining the house. That defence has
changed substantially since Donoghue where it was somewhat a complete defence. The issue tends
to feature nowadays under the rubric of contributory negligence. Why? Because courts do prefer
to apportion the loss between the parties. The courts see this as fairer.

The actual ratio of Donoghue v Stevenson is stated in the Hedley Byrne case.

“A manufacturer of products, which he sells in such a form as to show that he intends them to reach the
ultimate consumer in the form in which they left him with no reasonably possibility of intermediate
examination, and with the knowledge that the absence of reasonable care in the preparation or putting up
of the products will result in an injury to the consumer’s life or property, owes a duty to the consumer to
take that reasonable care.”

The ratio was confined to a manufacturer of products where there was no reasonably opportunity
for intermediate examination. The case has however been read much more widely. The neighbour
principle of reasonable foreseeability was capable of almost unlimited expansion. Lord MacMillan
stated in his concurring judgment “the categories of negligence are never closed. Since Donoghue, to
varying degrees, the courts have actually been concerned to find the limit of the neighbour
principle. Donoghue was a personal injury case. It has also been read as a case of positive act or
misfeasance case, rather than an omission or nonfeasance case.

In these circumstances the House of Lords held that the fact the defendant ought to have foreseen
harm to the plaintiff gave rise to a sufficiently proximate relationship to create a duty of care.

In cases of personal injury and positive conduct, the elements of reasonable foreseeability will
normally be sufficient for there to be a duty of care: an unproblematic category of case. Duty will
be imposed on the basis of reasonable foreseeability alone. In more contentious areas of the law,
for example where the loss is financial, where the loss is caused by omission, or where the
conduct consists of words rather than actions, the duty issue is not based merely on reasonable
foreseeability alone. The courts talk about duty based on proximity, and proximity has different
meanings in different categories of case, e.g. negligent misstatement/nonfeasance/financial loss.
“Proximity is a label empty of content”. The courts define content differently in different categories
of case. In contentious cases the courts are more likely to apply stringent tests, because of
concerns about “the floodgates”.

Hedley Byrne & Co. Ltd v Heller & Partners Ltd. [1964] AC 465 (HL)

This case developed the law of negligent misstatements. This was not a personal injury claim, but
a claim for financial loss. We are looking at it here to illustrate the development of the approach to
duty after Donoghue v Stevenson.

The plaintiffs wanted to know if they could extend credit to certain traders, so they sought a
reference from the defendant bank. The defendant bank gave a reference which was so carelessly

25
 ALEX BOND
phrased that it suggested the traders were creditworthy where in reality the traders were not. No
money was paid for the banker’s reference so there could be no claim in contract.

The first argument made by the defendant was that there was no duty of care to be careful in
speech and the plaintiff could not recover in financial loss. There had to be a contract or fiduciary
duty (fiduciary duty being a relationship of trust and dependence between the parties which
attracts liability between the parties) or financial loss flowing from physical damage. Donoghue
permits the law to be developed even though there is no precedent covering these facts. One can
open new lines of duties and one could so so in this case. Lord Devlin refers to the neighbour
principle as the principle of proximity.

The plaintiff’s argument was that the court should ignore previous cases, and just apply the
neighbour principle to the facts. Lord Devlin says this is not the way English law develops. “The
appellants (plaintiffs) in this argument tried to press Donoghue v Stevenson too hard”. You do not apply
the neighbour principle like the words of a statute. When the court encounters a new novel duty,
it look to existing precedent and see how the cases take the plaintiff. “As always in English law, the
first step in such an enquiry is to see how far the authorities have gone, for new categories in the law do not
spring into existence overnight”.

The court might say it is not prepared to recognise a new duty situation, as it would be too great a
step to take, or on the other hand it could say it is prepared to open up a new duty situation by
analogy to past cases.

There are extremes. Some courts reject that it is possible to open up new categories of duty. The
High Court of Australia quite controversially took this approach in Sullivan v Moody (2001) 183
ALR 404. It considered that in Australia there is no comprehensive test for determining the duty
issue in a novel case. In this jurisdiction it would look at previous cases in existing duty categories
such as negligent misstatement and pure economic loss, and would then identify the
characteristics common to duty categories, ask how essential they are and decide whether or not
to develop the law. The High Court of Australia rejected the English test for the duty of care in
Caparo. As a result, the Australian approach has come to be called ‘incrementalism’: the law as to
categories of duties of care grows incrementally. Kirby J dissented strongly, saying there should
be some test. In cases since Sullivan v Moody he has lamented the Australian approach.

Some courts have considered themselves very free of previous cases. An example is Cooke J, an
activist judge who was relatively unconstrained by precedent and keen to use the ‘Anns’ test.

What do the New Zealand courts do? New Zealand courts apply a general test for duty as a
framework but it is not determinative. They then examine the existing authorities closely to see
how far they go and whether they should be developed. They will examine the degree of analogy
with existing cases. The New Zealand Court of Appeal took this view for a long time, and it is
unlikely this will change since those judges have now been elevated to the Supreme Court.

Marx v Attorney-General [1974] 1 NZLR 164

The plaintiff was the wife of man negligently injured at work. Her husband as a result of his
injuries he manifested a hypersexual attitude towards her and on numerous occasions physically
and sexually assaulted her. She sues for her mental injuries.

26
 ALEX BOND
She argued that it was reasonably foreseeable to her husband’s employer that she might be
abused as a result of his physical injury. The defendant admitted a duty of care was owed to the
husband but denies a duty of care was owed to her.

The plaintiff argued that the court should open the “family circle”, i.e. allow the family of those
negligently injured to recover. McCarthy P talked about the function of the duty of care role. He
frankly says that the duty is a control device as to the extent of damage – its function is pragmatic:
to limit the range of complainants.

The plaintiff’s claim was considered a derivative claim rather than a direct injury claim. The court
referred to case called Bourhill v Young [1943] AC 92, the ‘Scottish fishwife’ case. Lord Wright in
that case said it had to be determined not merely whether the act was negligent against someone
else, but whether it was negligent vis-à-vis the plaintiff. If the plaintiff has a claim, it has to be a
wrong to herself. “She cannot build on a wrong to someone else”.

A similar principle cited here was stated in the famous case Palsgraf v Long Island Railroad Co 248
NY 339. The court recognised that the defendant guards might have been negligent to the
passenger who they knocked causing her fireworks to explode, but not to the plaintiff at the other
end of the platform who had a tile fall on her.

“The defendant’s conduct was not a wrong towards her merely because it was negligence towards someone
else. She must sue in her own right for a wrong personal to her and not as the vicarious beneficiary of a
breach of duty to another”.

The defendant may be under a duty, but that duty is not to the world at large. There is ‘no
negligence in the air’. The plaintiff must establish a breach of an independent duty owed to
himself or herself as a particular individual or member of a class, and cannot build her case on a
duty owed to someone else.

Home Office v Dorset Yacht Co. Ltd [1970] AC 1004

After Donoghue v Stevenson the law remained quite settled until the 1960s and 1970s when some of
great developments in negligence took place.

A party of Borstal boys were working as part of a training exercise on Brownsea Island, under the
supervision of three Borstal officers. The officers decided to go to bed, leaving the boys
unattended. The Borstal boys escaped, hijacked a yacht and because they didn’t know how to sail
a yacht, collided it with the plaintiff’s yacht and aboarded the plaintiff’s yacht causing damage.

The plaintiff yacht owner sued for the property damage to his yacht. The case came to the House of
Lords on a preliminary question as to whether a duty of care was owed. The Home Office is the
Borstal officers’ employer. It is being sued vicariously for the negligence of the Borstal officers.

Was it reasonably foreseeable that if the Borstal officers went off to bed, the plaintiff’s yacht might
have been damaged? It was seen as reasonably foreseeable. Lord Reid even described it as a
“likely consequence”.

The defendants argued that the Borstal officers owed no duty of care to a member of the public.
One of their reasons was that there was an absence of precedent to support such a duty category.

27
 ALEX BOND
Lord Reid said that at historically absence of precedent would have decisive, but not so anymore
because of the Donoghue principle.

“…The well-known passage in Lord Atkin’s speech should I think be regarded as a statement of principle. It
is not to be treated as if it were a statutory definition. It will require qualification in new circumstances.
But I think that the time has come when we can and should say that it ought to apply unless there is some
justification or valid explanation for its exclusion”.

The above passage was taken to more or less have created a presumption that the neighbour
principle would apply to determine the duty issue unless there was a valid reason not to apply it.
This was to change. It created very expansive liability, and tends to throw the burden of
establishing why the neighbour principle shouldn’t apply onto the defendant.

Lord Reid refers to two examples of categories of case where the neighbour principle won’t apply:
1. where the loss is economic
2. the duty to go to someone’s assistance when they are in distress, i.e. duty to rescue
[The classic example given is where someone is drowning in a pool. You could save them
without any danger and with minimal effort. In general, English law has said that while
there might be a moral duty to save them, there is no legal duty to do so, that is, in the
absence of a special relationship giving rise to a duty.]

This statement of Lord Reid’s is very well known and is the forerunner of the even more well
known but controversial (and overturned) statement of Lord Wilberforce in the Anns case.

Anns v Merton London Borough Council [1978] AC 728

This case is analogous to a leaky home case but concerns the foundations of a home. The
plaintiff’s house suffered cracking from inadequate foundations and sued for property loss. Lord
Wilberforce held liability. We are interested in Lord Wilberforce’s approach to duty: one of
principle not precedent.

The test is a two-step enquiry, developing from Lord Reid in Dorset:

1. Is there a prima facie or presumptive duty of care? i.e. asking whether there is a sufficient
relationship of proximity, seems to be based upon reasonable contemplation
2. The policy enquiry: is there any reason or policy, such as the floodgates, which may
negate the duty or cut down its scope?

This test came to be known as the Anns test. For a time this test was wildly popular, it was
applied across Britain and spread to Australia and New Zealand. It was grabbed with alacrity by
the Court of Appeal. Part of its attraction was that it was simple to apply. It was a principled
approach, that cut the law relatively free from precedent.

Doubts arose and the Anns test fell into disfavour. In a series of cases in Britain the House of
Lords pulled back from the test. Ultimately the Caparo three-stage test was substituted and the
Anns test was abandoned in England. The objection to Anns was that it was unduly expansive to
liability. The objection was that if you look at the first step, Lord Wilberforce appeared to equate a
sufficient relationship of proximity with reasonable contemplation. The test interpreted in this
way could be criticised as opening the way to virtually unfettered liability. In 1991 Murphy

28
 ALEX BOND
overruled Anns on the facts and the two-stage approach. In New Zealand after doubts about Anns
began to creep in, the courts said that in stage one of Anns, reasonable contemplation is only one
factor. In other words they qualified stage one of Anns. Once you reinterpret Anns to say a
sufficient relationship of proximity is not equated with reasonable contemplation, then you have
virtually arrived at the three stage test of Caparo.
Hill v Chief Constable of West Yorkshire [1989] 1 AC 53

The mother of the last victim of the Yorkshire Ripper sued as her deceased daughter’s personal
representative, claiming negligence on the part of the police in failing to apprehend him before he
killed her daughter. Her argument is that it was foreseeable that if not apprehended, the Ripper
would kill again. She relied on the Dorset case where the Borstal prison guards had been held
liable for property damage.

Did Lord Keith consider that foreseeability of injury to the plaintiff was sufficient to establish a
duty of care? He did not. Foreseeability is always necessary, a condition precedent.

He explains: “It has been said almost too frequently to require repetition that foreseeability of likely harm
is not in itself a sufficient test of liability in negligence. Some further ingredient is invariably needed to
establish the requisite proximity of the relationship between the plaintiff and defendant, and all the
circumstances of the case must be carefully considered and analysed in order to ascertain whether such an
ingredient is present. The nature of the ingredient will be found to vary in a number of different categories
of decided cases”.

In other words, in cases where a duty is argued for in a novel fact situation, some further
ingredient is necessary to establish the relationship of neighbourhood or proximity. Proximity is
the test for duty to some extent.

The difficulty with proximity is that it is a label which is empty of definite or fixed content. What
the necessary ingredient in addition to foreseeability is will vary from category to category. What
will be sufficient for a duty of care will vary according the category of case. It thus becomes
important to identify different categories of case and ask what in the particular category of case
makes up the necessary relationship.

What are some of the categories? Examples include negligent misstatements, pure economic loss
and nonfeasance. Hill falls within the category of nonfeasance. Nonfeasance has different
definitions. It is used to describe omissions but also, as in this case, it describes the situation
where the defendant doesn’t himself directly injure the plaintiff through his direct acts, but where
the negligence consists of failure to control a third party who is of full age and capacity from
inflicting injury on the plaintiff. Dorset is similarly a nonfeasance case. In Hill the mother is suing
the police for failing to prevent the Ripper from killing her daughter.

Nonfeasance is a problematic category of case. What is the additional ingredient? We will look at
this question later.

Caparo Industries PLC v Dickman [1990] 2 WLR 358

Lord Bridge in this case emphasised the inability of a single general test to be applied to
determine the duty issue. He approved the move back “in the direction of attaching greater

29
 ALEX BOND
significance to the more traditional categorisation of distinct and recognisable situations as guides to the
existence, the scope and the limits of the varied duties of care which the law imposes”.

He quoted with approval from Brennan J’s judgment in the Australian High Court case Sutherland
Shire Council v Heyman (1985) 60 ALR 1:

“It is preferable, in my view, that the law should develop novel categories of negligence incrementally and
by analogy with established categories, rather than by a massive extension of the prima facie duty of care
restrained only by indefinable ‘considerations which ought to negative, or to reduce or limit the scope of the
duty or the class of person to whom it is owed.’”

Brennan J in Sutherland was referring to Lord Wilberforce’s test in Anns, and he disparages it.
Brennan J’s statement has been frequently quoted.

‘Incrementalism’, the approach clearly favoured by Brennan J, was at its height in Sullivan v
Moody. But most courts don’t limit themselves to the incremental apporach. They start with a test
for duty but also look at the cases within that category and reason by analogy. The absence of a
duty isn’t fatal to that new fact situation, but there is a tendency to tread more carefully.

Lord Bridge’s test essentially replaces Lord Wilberforce’s two-stage test with a three-part test.

The test requires:

1. Foreseeability of harm
2. A sufficient relationship of proximity
3. That the situation should be one in which the court considers it fair, just and reasonable to
impose that duty on one party for the benefit of other

This is little more than a convenient framework to start analysing the duty situation. The precise
content of what is ‘fair, just and reasonable’ will change with the category of case. Canada
adopted this test, but Australia did not.

Instead of throwing out the Anns test as the House of Lords did, some courts simply reinterpreted
the first stage of the test. These courts said stage one was a sufficient relationship of proximity or
neighbourhood, not based purely on reasonable foreseeability alone as Lord Wilberforce had said,
but a composite enquiry encompassing the relationship and that foreseeability was just a mere
condition precedent. Cooke J stated (not here) that stage one was a “mere condition precedent”.

Once you reinterpret the Anns test in that way, it is indistinguishable from Caparo.

What is the New Zealand approach? Cooke J was inclined to an expansive activist approach to
negligence. He regarded the retraction from Anns by the House of Lords as dismay. The Court of
Appeal prefers to keep Anns but reinterpret the first stage.

Brown v Heathcote County Council [1986] 1 NZLR 76

Cooke P in this Court of Appeal case said the Anns approach was helpful and he described it as
having become indigenous to New Zealand. He stated that despite Anns being abandoned in
England, New Zealand should continue to ‘hew’ its own way.

30
 ALEX BOND

South Pacific Manufacturing Co v NZ Security Consultants & Investigations [1992] 2 NZLR 282

This case happened around the same time as Murphy v Brentwood DC [1991] 1 AC 398 (HL). The
House of Lords in that case declared Anns to be wrongly decided on its facts, but also on Lord
Wilberforce’s two-stage approach. Inevitably, in the present case it was argued that New Zealand
should also abandon the Anns test, in view of it being overruled in England.

A full bench sat on this case. Cooke P was on sabattical a year before this case and was infuriated
by the Murphy case. He wrote an article: Cooke R, “An Impossible Distinction” (1991) 107 LQR 46.
Now was his chance to transfer that article into law.

South Pacific Manufacturing states the approach for negligence for New Zealand. In subsequent
cases the Court of Appeal states the same principle as in South Pacific Manufacturing. South Pacific
Manufacturing is still cited frequently by lower courts.

What is the New Zealand approach? “Murphy should not lead to any changed approach to negligence
law in New Zealand”. Cooke P refers to the different approaches to duty – in Anns and the later
approach in Caparo and Murphy, characterised overall by ‘incrementalism’ – and then goes on to
affirm the Anns approach as the proper approach to the duty of care issue in NZ.

He cites from the Downsview case (First City Corporation Ltd v Downsview Nominees Ltd [1990] 3
NZLR 265), saying that the Anns approach has been taken in a long line of cases, there has been
no apparent dissatisfication with it and that it will not be departed from. Having said this he adds
a qualification to the Anns test. He approves the reinterpretation of stage one Anns as a composite
enquiry, involving foreseeability as a precondition but not sufficient for the purposes of a
sufficient relationship of proximity. It is the extract from Downsview that he approves of, where
Richardson J had said that the first stage of Anns “is not of course a simple question of foreseeability of
harm as between the parties and involves the degree of analogy with cases in which duties are already
established”.

The two stages are as follows:

1. A sufficient relationship of proximity
2. The policy inquiry – “strengthening factors can be weighed also”, not just policy factors
telling against a duty.

He says that the ultimate question is whether it is “ fair, just and reasonable” that there is a duty of
care, after satisfying the two stages above. But he stresses the approach is only a way of
organising your thinking. The same considerations should be weighed on either approach. He
considers the outcome should be the same on either approach. The outcome will be determined
by “judicial judgment”.

He summarises the duty issue as a ‘balancing exercise’: weighing the competing considerations for
and against a duty, organised in two broad fields of inquiry: proximity and policy, with the
ultimate question being it is fair, just and reasonable for the law to impose a duty?

What did he say about the label ‘incrementalism’? What is fair, just and reasonable for there to be
liability for one judge is a completely unwarranted extension of the law to another judge. He says
it all depends on where you stand. What is a small incremental development to one judge may be

31
 ALEX BOND
a giant jump to another. What about in terms of economic loss? It is not automatically excluded.
The fact that loss is purely economic is just a factor to be weighed in the balancing exercise. This is
stated because the English position in Murphy is much stricter: Murphy stated that in Britain pure
economic loss was excluded from recovery, outside of negligent misstatement cases.

J Stapleton, “Duty of Care Factors: A Selection from the Judicial Menus” in P Cane and J Stapleton
(eds), The Law of Obligations: Essays in Celebration of John Fleming (Oxford: Clarendon Press, 1998),
ch4, p59. (desk copy)

Keith Stanton, “Decision-making in the tort of negligence in the House of Lords” (2007) 15 Tort L
Rev 93. (to be put on desk copy)

Andrew Barker’s article “The Duty of Care and the Searched Certainty” [2003] NZLR 44

Rolls-Royce v Carter Holt Harvey [2005] 1 NZLR 324 (CA)

This case is important because it is a recent case from the Court of Appeal, the highest national
court at the time: a recent statement from a high-level court about the approach to duty of care. It
owes much to South Pacific Manufacturing, and indicates a continuity approach to the duty of care
question in New Zealand.

Carter Holt Harvey contracted with ECNZ, the predecessor of Genesis, and under that contract
Genenis was to get the co-generation plant designed and built. This was called the ‘co-generation
contract’. ECNZ entered into a contract with Rolls Royce, who was to design and build the plant.
This was called the ‘turnkey contract’. The plant that was built had many problems.

Because there was no direct contractual relationship between CC and RR, they only had an action
against RR in negligence. The contract with Genesis has limitation clauses lowering what Genesis
could be liable for.

At paras 58 and 59 Glazebrook J gives a summary of the New Zealand approach.

“The focus is on two broad fields of inquiry but these provide only a framework rather than a straitjacket.
The first area of inquiry is as to the degree of proximity or relationship between the parties. The second is
whether there are other wider policy considerations that tend to negative or restrict or strengthen the
existence of a duty in the particular case.”

“The inquiry into proximity is concerned with the nature of the relationship between the parties and is
more than a simple question of foreseeability. It involves consideration of the degree of analogy with cases
in which duties are already established.”

An important point made by Glazebrook J is that a balancing exercise is performed, weighing all
relevant factors for and against liability. Stage one of the Anns test was a prima facie presumption
of duty. That no longer exists here.

An important policy factor, noted at para 60, is the question of indeterminate liability. Glazebrook
J quotes Cardozo CJ in Ultramares Corporation v Touche, Niven & Co: “the Courts are concerned to
limit the risk of exposing defendants to ‘a liability in an indeterminate amount for an indeterminate time to
an indeterminate class’.”

32
 ALEX BOND

There were some new factors that Glazebrook J says the courts will take into account in deciding
whether a new duty should be recognised.

 Whether people in the plaintiff’s position are vulnerable, to a defendant with special skills.
[That will be a factor telling in favour of a duty of care].
 The fact that both CHH and RR were both equal, commercial parties looking after
themselves. [That will be a factor telling against recognising a duty].

Whether someone is vulnerable will depend on whether there are other remedies available for the
plaintiff. If the plaintiff’s only remedy is a duty of care recognised, that will be a factor in favour of
recognising a duty. On the other hand if the plaintiff has or had an alternative remedy, that is a
factor telling against recognition of a duty of care.

It was important in this case that there is an alternative remedy, there are limitation clauses in the
contract so CH can sue ECNZ for breach of contract. Because there was an alternative remedy, the
court didn’t recognise a new duty of care, even though an action of tort would be better. In this
case the contractual background was relevant to the question of duty, pointing away from the
recognition of a duty of care.

Surrounding statutory provisions especially for public authority defendants can often be relevant
as to whether there is a duty of care (see Price Waterhouse v Kwan case).

33
 ALEX BOND

Duty of Care in novel cases: SUMMARY

A. Modern British approach

1. Incrementalism is the term used to describe the modern British approach. There is less
reliance on a general principle of universal application, and greater significance attached
to decided cases as guides within recognised categories. The law develops incrementally.
2. BUT one judge’s natural, incremental step in developing law is another’s illegitimate leap.
Cooke J: an “unwarranted and irresponsible extension of the law”.
3. Caparo three-stage test:
(i) reasonable foreseeability of harm
(ii)a relationship of sufficient proximity, and;
(iii) fair, just and reasonable.
The problem with the Caparo test: proximity is a label ‘empty’ of content.

B. Modern New Zealand approach

1. The ultimate question: is it fair just and reasonable to impose a duty in the circumstances?
2. A balancing exercise, weighing factors for and against a duty.
3. There are two broad fields of inquiry (Anns)
(i) proximity (not just foreseeability), and;
(ii)policy considerations for and against, such as the floodgates, the need for a duty of
care so to provide deterrence/to improve standards of care/to compensate
deserving plaintiffs.
4. BUT this approach is just a framework. It shouldn’t affect the outcome because the same
factors are beingweighed.
5. Pure economic loss (outside negligent misstatements) is not fatal in New Zealand, but will
be a factor telling against a duty.

Think: Is there really much difference between two approaches?

There is no difference except maybe in terms of symantics.
The approach adopted should not affect the outcome.

C. Other Relevant Factors to Duty

1. Vulnerability of the plaintiff. Is the plaintiff a commercial corporation capable of looking

after self, or vulnerable vis-à-vis a professional defendant, for example?
2. Is there an alternative remedy the plaintiff has or could have provided for, for self, or is
duty in tort only remedy against defendant?
3. Statutory or contractual background. Should parties be left with their negotiated bargain?
Is a duty of care in tort inconsistent with a statute in the field?

34
 ALEX BOND
C. CAUSATION IN FACT

The third requirement in a cause of action. Unlike the tort of battery, negligence is a tort which
requires proof of damage. The tort of battery does not require you to prove personal injury. If you
negligently shove someone and there is no personal injury, that person cannot sue. Damage is an
essential element of liability. In addition the plaintiff must prove that the injury or damage is
causally linked to the breach of duty.

Traditionally, the causation (C) and remoteness (D) enquiries are regarded as separate enquries.
The causation enquiry is regarded as the more factual enquiry, whereas the remoteness enquiry
allows judicial evaluation to be taken into account. It is now more realised that the causation
enquiry is not a purely scientific or factual enquiry; elements of judicial judgment as to what is
reasonable come into play. This is visible in the Ambros case.

Barnett v Chelsea & Kensington Hospital Management Board [1969] 1 QB 428

This case is proof that causation can be a difficult element. Even if you prove negligence, a
plaintiff can miss out because causation cannot be proven.

Who bears the burden of proof? The plaintiff. In this case there was an argument made by the
plaintiff’s counsel, that once negligence is proven, the defendant has the burden in proof. An
effort to reverse the burden of proof. By and large this attempt has failed routinely; courts do not
like putting the burden of proof on the defendant.

Why was causation not established in this case? It was argued that the plaintiff would have died
anyway. An enquiry into the hypothetical? If there was no negligence, would the man have died?
This case is commonly referred to as the authority for the ‘but for’ test of causation. It poses the
question whether the plaintiff would have suffered the loss without (‘but for’) the defendant’s
negligence [But for the defendant’s negligence would the plaintiff still have suffered the loss]. If
he or she would not have suffered the loss, the defendant’s negligence is a cause of the loss, but
not the sole cause. If as in this case the loss would have arisen even without the defendant’s
wrongdoing, then it is not a cause of the plaintiff’s loss and does not give rise to legal liability. The
‘but for’ test tells us whether the defendant’s negligence is a cause.

Performance Cars Ltd v Abraham [1962] 1 QB 33 (CA)

‘But for’ causation was not established in this case because the plaintiff would have incurred the
cost of repainting the car in any event.

The ‘but for’ test is a threshold requirement or convenient starting point. The textbooks describe it
as a ‘preliminary filter’, because it filters out the causally irrelevant. After this the remoteness
enquiry (D) comes along. This is referred to as ‘causation in law’ as opposed to ‘causation in fact’.

What about the situation where two or more persons combine to create the same damage? They
are each a cause of the loss suffered by the plaintiff. The courts do not require that the defendant’s
conduct be the sole or even the main cause of the plaintiff’s loss. In the situation where there is
more than one cause, it is enough that the defendant’s conduct was a contributory cause. In the
Bonnington case, something is a cause providing it is not de minimis.

35
 ALEX BOND
In this case both the plaintiff and the defendant were fully liable, but could seek a contribution
from each other, e.g. two drivers collide causing harm to another innocent plaintiff, both drivers
are negligent, each driver is fully liable if his or her negligent conduct is a cause of the plaintiff’s
harm, and the court will fix their respective contributions.

There are always exceptions. Exceptional conduct may be held by the law to be a cause in fact
even though it does not satisfy the ‘but for’ test. The textbooks always refer to the ‘fire’ example or
the ‘shooting’ example.

‘FIRE’ EXAMPLE: an exception

A and B independently and negligently light fires, near to P’s property. Both fires spread and
damage P’s property. The fires combine. Even though the conduct of A or B alone does not
satisfy the ‘but for’ test, the courts say you don’t need to. Policy determines the result: it would
unjust for A to get off because of B and vice versa leaving P remedy-less.

Sometimes the harm is indivisible, i.e. the whole of the property burns down. A and B will
both be responsible for the whole loss. In comes cases it can be practicable to divise up the
defendant’s respective contributions. A and B are each responsible for the part he or she
caused. You have to clearly delineate each loss that was caused, e.g. A’s fire clearly burnt
down the east exterior of the building.

Bonnington Castings Ltd v Wardlaw [1956] AC 613 (HL)

In this case there were multiple causes: one tortious, one non-tortious. The plaintiff had inhaled
silica dust from the defendant’s grinders and the hammer and contracted a disease. The question
in the case was whether there was proof of causation? The lower courts had placed the onus on
the defendant to prove that the silica dust from the grinders did not cause the plaintiff’s disease.
On appeal, Lord Reid stated this was ‘erroneous’. It is clearly established that the legal onus to
prove the link remains on the plaintiff throughout.

The dust that came from the hammer was not the sole cause of his disease, but it is suggested the
greatest proportion came from the hammer, as his face was directly in front of the hammer. Some
dust, a minority of it it seems, came from the swing-grinders, the ‘tortious cause’. The proportion
that came from the swing-grinders was however not trivial.

Was causation established in this case? The court used the term ‘material cause’. The swing-
grinders contributed a portion of dust that was not negligible, and that was enough. It can be
difficult where tortious and non-tortious actions combine. In a case the defendant will argue that
the ‘but for’ test wouldn’t be satisfied, because the damage would have happened anyway. The
courts have tried to relieve the rigours of the ‘but for’ test for plaintiffs. They have been prepared
to modify the ‘but for’ test in this situation. The defendant’s act does not have to be the sole cause
but it must have materially contributed to the damage. The defendant could only escape if the
damage caused was so small to be negligible, de minimis.

36
 ALEX BOND
Accident Compensation Corporation v Ambros [2007] NZCA 304

Ambros shows us that satisfying the onus of proof of establishing the causal link can be very
difficult for plaintiffs.

Susan Ambros died because of a spontaneous coronary artery dissection (=where one’s arteries
tear). People usually die of this disorder. It is rarely diagnosed and fixed. One of the ways one can
contract this disorder is through pregnancy. Ambros was not diagnosed.

The issue is whether the failure to diagnose the disorder caused her death. North Shore Hospital
booked Ambros as an intermediate risk to go to Greenlane Hospital, but she died because her
vessels tore. The High Court considered that the treatment by her consultant was negligent.

‘Medical error’ had to be proven. ‘Medical error’ is defined in the statute as failure to achieve the
level of skill required in the circumstances [that definition has changed, you are no longer
required to prove medical error for negligence]. ACC argued that Ambros’ death was caused by
her underlying condition, the dissection of her cardiac artery. But Mr Ambros, the widower,
argued that there was a causal link between the negligence and her death.

Why was causation really difficult to prove in this case?

1. It may be very difficult to prove the patient’s injury or death was caused by the defendant’s
negligence rather than the patient’s worsening health condition. Sometimes the doctor is in a
better injury to know the condition than the plaintiff. The plaintiff is in no position to
establish a causative link because they do not have the knowledge of a doctor.
2. The plaintiff may be in no position to prove the causative link because of the current limits
of medical science. Perhaps medical science can only go so far to show that the defendant’s
negligence only caused an increase in risk. There may be limits to what medicine and
science is able to tell us, which prevents the plaintiff from proving a causative link.
3. There may be a problem where there is more than one causative agent at work. See Wilsher v
Essex Area HA [1988] AC 174 (HL), which involved a premature baby that suffered
blindness. A capita was inserted into a vein instead of an artery, and as a result he was given
five times the amount of oxygen he should have been given. Excess oxygen can cause
blindness in premature babies, but this baby also suffered from about 5 other conditions that
could also cause blindness. Even if there was a breach of duty, how do you know it was the
tortious injury that caused the damage?

In these cases it is traditional for the court to insist that the ‘but for’ test first be satisfied. The
Court of Appeal in Atkinson v ARCIC [2002] 1 NZLR 374 (CA) said this, as did the House of Lords
in Wilsher. Counsel for the plaintiff in Wilsher tried to argue for a reversal of the burden of proof.
They argued causation could be established on the basis that the act materially increased the risk
of the injury. Lord Bridge of the House of Lords rejected that submission.

The Atkinson case is not a well-thought out decision took a similar view and essentially followed
Wilsher. If you cannot prove it on the normal rules you fail. That was the law in New Zealand at
the time the Ambros case was argued.

37
 ALEX BOND
What had the judges in the High Court done? They had found causation, even though this was a
very rare condition on the basis of a new test of close proximity. They lowered the bar for the
plaintiff and said they would find causation established where there is a close temperate
proximity between…unless injury of death resulted. They did not say they were putting the onus
on the defendant but essentially they were. ACC was worried that this would greatly expand its
area of cover. In the Ambros case in the High Court did not even refer to the Atkinson case
although it was clearly in contrast with their decision (they probably didn’t know about it!).

This shows courts are sympathetic to plaintiffs who make an effort to argue causation where there
has been a breach of duty.

The Court of Appeal here asked whether Atkinson be departed from. It decided that Atkinson was
correctly decided, i.e. there is no reversal of the burden of proof. Atkinson said proof of an
increased risk was not causation.

Glazebrook J goes through the modified approach to causation in the industrial disease cases. She
discusses McGhee v National Coal Board [1973] 1 WLR 1 (HL), a controversial case about dermitis.
The majority of the House of Lords held that they would accept proof of a material increase in
risk as proof of sufficient causation. There are many arguments as to what is the proper
interpretation of McGhee. Some people read McGee was a case where the House of Lords were
prepared to accept a causative link, even though there were gaps in the evidence. Lord
Wilberforce said we should reverse the onus of proof, but this attempt failed.

In Fairchild some employees were exposed to asbestos in their conditions working for Employer
A, B and C. Their lung cancer was caused by exposure to asbestos in their working conditions but
it wasn’t known which employer’s fibre caused the asbestos. The House of Lords held that all
would be held liable. In the special circumstances of this case they would follow McGhee and say
causation could be proved on the basis there was an increased material risk. In some exceptional
categories of case, courts have been prepared to permit causation on the basis of a proof an
increase in material risk.

This has been tried in New Zealand, but the Court of Appeal in Atkinson said no.

The next attempt was to say ‘can we prove causation on the basis of loss of a chance’. In the case
of Greg v Scott the plaintiff wasn’t diagnosed in good time with cancer. The plaintiff’s chances of
successful treatment had gone down from 42% to 15%. The plaintiff can’t prove on the balance of
probabilities that if he had been treated earlier he would have survived because there was only
42% chance of surviving anyway. The plaintiff tries to argue that loss of a 27% chance. This is
simply another way of arguing an increase in material risk. The two dissenting Lords decided in
favour of the plaintiff. The other Lords said it makes logical sense, but for policy reasons will not
allow proof of injury on the basis of a loss of chance to constitute causation.

But in some commercial cases it is allowed. But the court in Ambros said they cannot extend these
commercial cases to the present case because of policy reasons, the floodgates.

There have been other attempts at ‘loss of a chance’ arguments elsewhere in common law
jurisdictions. Such attempts have been rejected by the Supreme Court of Canada. In Australia
there has been some success. In Ambros Glazebrook J discusses ‘loss of a chance’ and says it

38
 ALEX BOND
doesn’t apply under the ACC scheme, where the question is whether you have cover or not. There
is bound to be however a property damage or economic loss case in New Zealand.

She considers the Canadian approach and discusses a case called Snell v Farrell. The Supreme
Court of Canada said they would lower the bar for plaintiffs trying to prove causation.

What did Snell v Farrell hold? It held there has to be some evidence to draw an inference. There is
a shifting of the evidential burden as opposed to the legal burden. Glazebrook J draws a
distinction between the evidential and tactical burden. The tactical burden refers to the tactics of
litigation and if the plaintiff produces a cogent piece of evidence, there is an expectation of the
defendant that he or she will answer that piece of evidence. Sometimes called the ‘burden of
persuasion’. It is an expectation on the defendant that he or she will answer the plaintiff’s cogent
piece of evidence with his or her own piece of evidence if he or she possesses it.

The courts say that where there is breach of duty (negligence), if there is some evidence of a
causative link, then there is some kind of tactical burden on the defendant that a shifting of the
legal burden itself. The point was made in Snell v Farrell that in many medical negligence cases
the facts lie particularly within the knowledge of the defendant and the plaintiff is disadvantaged
from establishing that knowledge of causation. For that policy reason, very little evidence of the
causative link will be required before the court is justified in drawing an inference of causation
adverse to the defendant in the absence of evidence adduced by the defendant that that there was
no causative link.

This is just another device courts have used where there is a gap in the evidence of causation to
allow a negligence cause of action to be established. Para 59.

What is the New Zealand Court of Appeal’s attitude to drawing this inference? Are they in favour
of it or not? Of all the various devices to lower the bar for the plaintiffs, the Court of Appeal
indicates it will employ this device in New Zealand. The various others (such as the material
increase in risk and ‘loss of a chance’ principles) are shut off for various reasons. New Zealand
has taken the route of inferential reasoning. Where the evidence supports a possible connection
between the events and the injury, the court could be persuaded to fill the evidential gap by
drawing an inference of causation.

At para 69 Glazebrook J states: “We agree that the question of causation is one for the courts to decide
and that it could in some cases be decided in favour of a plaintiff even where the medical evidence is only
prepared to acknowledge a possible connection”. This could spread to other causation. Causation also
arises in criminal and contract law.

The court goes on to talk about the use of statistical evidence. That might be one way in which
some evidence of a causative link can be established, but statistical evidence does not tell you
anything about this patient, just trends in groups of patients. Sometimes evidence of a particular
plaintiff is enough.

The only time where there cannot be an inference of causation is where the evidence is against a
causative link. Close temporal proximity might be sufficient for a court to draw an inference.

So New Zealand courts will employ inferential reasoning:

 shifting of a tactical burden to the defendant

39
 ALEX BOND
 drawing an inference of causation where there is evidence of a possible causative link [a
question of law]

Causation in fact: SUMMARY

1. The “but for” test:

a. asks ‘would P have suffered the loss without (“but for”) D’s wrongdoing?’
b. is a threshold requirement which identifies a cause and eliminates the causally
irrelevant

2. Does not always apply, e.g. fire example, hunters’ example.

3. The legal burden to establish causation is on the plaintiff. Reversal of legal burden (i.e.
placing the burden on the defendant) is not accepted (Ambros).

4. Where there are multiples causes, the “but for” test is modified. Where the damage is
“indivisible”, P does not have to prove D’s wrongdoing is the sole or primary cause of
P’s harm. The test is whether D’s tortious conduct “materially contributed” to the injury
(Bonnington).

5. Exceptionally, courts have been prepared to treat “a material increase in risk” as

equivalent to a “material contribution to the damage”. This is not a general doctrine and
applicability under ACC is undecided (Wilsher, Atkinson, Ambros); Fairchild — industrial
exposure; facts exceptional.

6. Redefining the nature of the P’s loss as ‘loss of a chance’, and awarding discounted
damages not for the outcome but for a loss of a chance of gaining a benefit or of avoiding
an outcome? Permissible in some commercial cases of financial loss and some
jurisdictions in personal injury (e.g. NSW), but there is reluctance in personal injury
because of policy not principle (floodgates concerns in the UK). The ‘loss of a chance’
approach is not compatible with the ACC scheme in New Zealand.

7. In New Zealand causal uncertainty may be dealt with by the shifting of the evidential
(tactical) burden (onto the defendant) and the court drawing an inference of factual
causation where there is some evidence of a possible causal connection (Ambros).

40
 ALEX BOND
D. REMOTENESS OF DAMAGE OR CAUSATION IN LAW

This element concerns the question of how far legal responsibility for the consequences actually
caused by the breach stretches, given that an act of negligence can have all sorts of far reaching
and unforeseeable consequences. Is the link between the defendant’s conduct and the
consequences produced by that breach sufficiently proximate to be reasonable, as a matter of
policy, that the defendant should be liable for those consequences?

Overseas Tankship (UK) Ltd v Morts Dock & Engineering Co [1961] AC 388 a.k.a. Wagon Mound No. 1

Wagon Mound No. 1 has a different factual basis to Wagon Mound No. 2 because there was a finding
in this case that it was not reasonably foreseeable that the spilt oil would alight.

In the case Re Polemis the court held that a defendant is liable for all the consequences of his
negligence, even if they were unforeseeable, provided that they satisfied a test of directness (so
long as the consequences can be said to be a “direct” result of the negligence). But this test of
remoteness was said to be inherently vague. The Privy Council in Wagon Mound No. 1 preferred
the test of reasonable foreseeability. They said it all came down to current ideas of justice or
morality. They felt the Polemis test was not just or fair. They overruled Re Polemis.

Hughes v Lord Advocate [1963] AC 837

A post office employee whilst working knocked over a paraffin lamp into a manhole. The paraffin
escaped from the lamp. The plaintiff child entered the manhole to play, the lit wick of the lamp
ignited when it met the paraffin vapour and there was an explosion. The plaintiff sues for his
personal injuries.

The post office employees were negligent. They ought to have foreseen that children would be
attracted to this manhole, and having left the lamp in the mangole with a lit wick, they should
have known the lamp would have been a danger to the children.

Why had the lower courts dismissed the plaintiff’s claim for damages for his injuries? This was
such a bizarre combination of events which led to the explosion that the lower courts had said
that it was highly unforeseeable that it should happen in that way, and therefore, there was no
recovery. The lower court said some things were foreseeable, e.g. that the lamp would be played
with by children, and would be recoverable, but not this strange “concatenation of circumstances”.

Lord Guest’s judgment

Did Lord Guest in this case consider that the foreseeability test requires that all the details of the
accident leading up to the injury has to be reasonably foreseeable? No. The fact that the injury
was inflicted by this strange combination of circumstances did not matter. It is not necessary to
foresee the combination of events leading to the accident.

In this case the House of Lords qualified the Wagon Mound No 1 rule by saying it is enough that
injury by burning was of a type or kind of damage which was reasonably foreseeable. He defined
burning by an explosion and burning by liquid paraffin spilling and igniting as the same type or
kind of accident. Essentially, so long as the type or kind of accident or injury is reasonably
foreseeable, it doesn’t matter if all the precise circumstances are not.

41
 ALEX BOND

Hughes is authority for:

1. All the precise details of the circumstances leading up to the accident or injury do not have
to be reasonably foreseeable.
2. It is sufficient if the accident or injury is of a type or kind which is reasonably foreseeable
(and so here he equated injury by an explosion and burning as a result of liquid paraffin
being ignited as being of the same type or kind of damage).

Stephenson v Waite Tileman [1973] 1 NZLR 152 (CA)

The plaintiff was employed as a steeplejack by the defendant. A wire rope on the crane that he
used was in very poor condition. While working on the rope, he cut his hand, and as a result,
became very ill. His condition deteriorated and he was in a “pitiful” condition by the time the
case came before the court.

The court questioned how to reconcile the Wagon Mound No. 1 test of foreseeability with the
‘eggshell skull’ principle. What is the ‘eggshell skull’ principle? [If you have a skull made out of
eggshell, and someone negligently injures your head, and your skull caves in]. A tortfeasor had to
take his or her victim of negligence as they found them, even though the extent of the harm was
much worse because of the victim’s pre-existing weakness, e.g. by virtue of having a skull made
out of eggshell. The principle applied in personal injury cases in particular.

Todd calls the ‘eggshell skull’ principle the existing state rule: the tortfeasor is obliged to take the
victim in the existing physical, psychological or financial state they were in at the time of the tort .
You cannot complain if the damage is worse than anticipated by reason of that state.

Richmond J’s judgment

Richmond J says the rule in Wagon Mound No. 1 is that only damage of a kind that is reasonably
foreseeable is legally recoverable. The defendant is not liable for damage of a kind which is
unforeseeable, but, the defendant doesn’t have to foresee all the details of how the damage
occurs, such as an obscure combination of events leading up to the damage.

The remoteness rule of foreseeability has been qualified. Is the ‘eggshell skull’ principle and the
Wagon Mound No. 1 principle reconciliable? The judge says the ‘eggshell skull’ cases survived
Wagon Mound No. 1 and remain unaltered. The ‘eggshell skull’ principle was accepted in this case
as part of the law of New Zealand. Richmond J felt that there were a few problems reconciling the
two, but in the end it didn’t matter whether you saw the ‘eggshell skull’ principle as an exception
or as an adjusted application of the foreseeability principle.

Unforeseeable consequences are sometimes regarded as going to the extent rather than to the
kind of injury. In personal injury cases, all the defendant has to foresee is the initial kind of injury
or damage. It need not foresee its full extent. The plaintiff just has to demonstrate that the actual
consequences were caused by the initial injury.

“The decisions in England, Scotland, Ireland and Canada are of importance because in all of them the
Courts have limited the test of foreseeability to the initial kind of injury. The consequences, often
unforeseeable, of the initial injury have been treated as falling within the field of extent of the injury”.

42
 ALEX BOND
It becomes very important how different kinds or classes of injury are determined. On the duty of
care issue we have seen that courts tend to categorise the damage as bodily injury, physical
damage to the person, physical damage to property, financial loss, etc. When it comes to
remoteness, the categorisation of the initial kind or type of injury can be more specific than that.

We do not tend to litigate personal injury cases as a result of ACC. Courts in personal injury
jurisdictions rarely distinguish between different kinds or classes of injury to the human body for
remoteness purposes. Their reluctance to do so is well illustrated by the ‘eggshell skull’ principle.

In personal injury cases, particularly physical injury, there is rarely a good remoteness defence a
defendant can raise, because courts broadly define the initial type or kind of class the tortfeasor
has to foresee, for remoteness purposes. If some bodily injury is reasonably foreseeable, then you
can’t say the consequences are too remote.

Page v Smith [1996] 1 AC 155 (HL)

A nervous shock/mental injury case. The plaintiff was involved in a moderately serious car
accident, the fault of the other driver. No one was injured, but the plaintiff suffered a considerable
worsening of his pre-existing chronic fatigue syndrome. He suffered a serious relapse. The
defendant argued that in nervous shock cases, psychiatric shock injury is a different kind of injury
than physical injury to the person, and that this plaintiff’s mental injury was too remote.

He said that this accident was sufficiently dramatic or horrifying to make psychiatric injury
reasonably foreseeable. It all came down to how the House of Lords defined, narrowly or broadly,
the initial kind or type of injury. If you said that physical and psychiatric injury were different
classes, the defendant would win. If you said that they were within the same class, the plaintiff
would win.

Page v Smith is an illustration of how narrowly or broadly the court defines the type or class of
injury will determine whether the remoteness argument by the defendant will succeed or not. It
was held that mental and bodily injury were within the same class. The courts have considerable
latitude as to how they distinguish the category of liability. How they class them will determine
the remoteness issue. In personal injury cases there is a tendency to class things very broadly, so
the remoteness issue rarely arises (but in property damage cases they arise often).

Attorney-General v Geothermal Produce NZ Ltd [1987] 2 NZLR 348

The remoteness issue arose in this case because Geothermal argued that the loss of its rose crop
deprived it of the income it would have used to extend its operations. Its development plan was
to build three further greenhouses and stock them with roses.

This is a claim of loss of profits on the lost crop and of loss of future profits on the planned
greenhouses that had not yet been built. Geothermal’s argument is that it was financially sound
and on the verge of success. The defendant argued that Geothermal was actually financially
shaky, it had embarked on a risky venture and its claim for the lost profit on the damaged plants
and for the future profits was too remote: caused by its own impecuniosity. The defendant relies
on the Liesbosch case.

43
 ALEX BOND
In Liesbosch, which involved a contract, the plaintiff negligently sunk the plaintiff’s dredger. There
was a penalty in the contract under which the deposit was forfeited if any work was not
completed on time. The plaintiff was unable to raise money to buy another dredge so hired
another one at great expense from Italy. The plaintiff covered the costs but it also claimed for the
extra cost of hiring the substitute dredge until the court judgment, saying it could not afford to
buy a replacement one and it was reasonable to assume the extra costs to buy it.

If that claim was made today, the plaintiff probably would have succeeded. But the House of
Lords then rejected the claim on two bases:
1. The extra hire charge was caused by its own impecuniosity and not the defendant’s
negligence; and
2. even if that wasn’t the case, the claim for the extra hire was too remote and not recoverable.

Liesbosch principle: the damages for which a defendant is liable cannot be increased by reason of
the plaintiff’s own impecuniosity. Restated at para 51 of Lagden v O’Connor:“Losses due to the
plaintiff’s impecuniosity are unrecoverable”.

[Impecuniosity = having little or no money; penniless; poor; lack of financial means]

The defendant made an argument that Geothermal had a duty to mitigate its losses, by selling of
part of its land. There is a principle of contract and tort law that the plaintiff has a duty to
reasonably mitigate its losses (keep its losses to a minimum), and can’t make the defendant liable
for its failure to do so.

The question of fact is how far reasonable is it to expect a plaintiff to go?

McMullin J states at para 25: “the measure of damages in tort is the sum of money which will put the
part who has suffered a loss in the same position as he would have been in had he not sustained the wrong
for which he seeks compensation”. This shows the damages are restorative. The measure of damages
is completely different in contract. The primary measure in contract is to protect the plaintiff’s
expectation interest, what the plaintiff expected to get if the contract was fully performed. Damages
are forward-looking in contract, and backward-looking in tort.

He goes on to state the principle of remoteness in tort. “The injury for which damages are claimed
must be a foreseeable one…an injury will be foreseeable if the possibility of that kind of injury was
foreseeable; not necessarily the specific injury itself”. This is a concise summary of remoteness.

Lost profits caused by damage to the rose plants was held to be not too remote. What was the
reasoning? Reasons given on page 52.

What did the Court of Appeal say about the impecuniosity principle?

The Liesbosch principle didn’t succeed. The Court of Appeal essentially held that losses flowing
from impecuniosity are recoverable if reasonably foreseeable. The Court held that it was
reasonably foreseeable that if you damage the rose plants, that the plaintiff would suffer a
financial setback. Impecuniosity is within the normal principle of remoteness. The Court doesn’t
actually overrule Liesbosch, but says it must be read “in light of” the later Wagon Mound cases on
reasonable foreseeability. The test of reasonable foreseeability applies to the financial state of the

44
 ALEX BOND
victim in the same way as it applies to the plaintiff’s physical and mental state. The defendant
must take his or her victim as he finds him financially.

Even before Lagden, Liesbosch was more or less declared wrongly decided by the Court of Appeal.
The Court of Appeal felt it had been overtaken by subsequent case law.

Lagden v O’Connor [2003] UKHL 64

The Liesbosch principle was squarely at issue in this case. The defendant argued that the
additional fee charged to the motorist was a loss due to the plaintiff’s own impecuniosity, and
accordingly, within the Liesbosch principle, was too remote.

Liesbosch principle restated: “The rule is…that the damages for which the defendant is liable cannot be
increased by reason of the claimant’s impecuniosity”.

What particular principles of law had overtaken the Liesbosch principle? The House of Lords held
that Liesbosch was no longer good law. It had been overtaken by the mitigation principle but more
importantly the remoteness principle. Lord Hope says the Liesbosch principle couldn’t be
reconciled with the principle at para 61 “the wrongdoer must take his victim as he finds him”.

Lagden reflects the Geothermal decision.

45
 ALEX BOND

Remoteness of damage: SUMMARY

1. Remoteness concerns the extent of the consequences caused in fact by the breach (by the
negligence) for which the D ought to be legally responsible.

2. The test for remoteness is reasonable foreseeability of the kind or type of injury, not its
extent. It is sufficient if the accident or injury is of a type or kind which is reasonably
foreseeable.

3. It is not necessary for all the precise details of the circumstances leading up to the
accident to be reasonably foreseeable, i.e. allows a obscure combination of events.

4. Unpredictable consequences are recoverable, if they are of a kind which are reasonably
foreseeable.

5. Judicial policy is involved in defining the kind of injury to be reasonable foreseeability

(narrowly or broadly). A broad definition used in personal injury cases.

6. The “existing state” or “egg shell skull” principle — D takes his victim as he finds him.
This remains good law unaffected by the reasonable foreseeability test (is seen as either
an exception to or application of the reasonable foreseeability test). Applies to physical
and mental state of P before the tort.

7. The “existing state” rule applies also to existing financial states — the D must take her
victim as she finds him financially as well as physically. If a P’s financial weakness is
reasonably foreseeable, then the P is entitled to recover for extra expenses to reduce his
damages. Liesbosch departed from.

46
 ALEX BOND
E. OMISSIONS [nonfeasance] AND INTERVENING CAUSES [third party intervention]

The problem in these cases is the extent to which the defendant will be liable for the actions of a
third party human person. Sometimes that intervening person is an adult, a child, or the plaintiff
himself or herself. Sometimes those actions are criminal, sometimes tortious. When they will be a
novus actus interveniens breaking the chain of causation?

The plaintiff can sue the third party, but instead turns to the deep pocket of the government
department for example, to hold them liable for its failure to prevent the intervening cause.

Couch v Attorney General [2006] NZSC 66 (currently reserved in the Supreme Court)

This case concerns parolee William Bell who was placed on work experience in an RSA in
Panmure and killed some of his co-workers. Couch, the only survivor who was seriously injured,
and the husbands of one of the dead sued the probation service. They claimed there was a failure
to warn the co-workers of Bell’s history of violence.

These cases as Lord Reid explains in Dorset the remoteness issue concerns whether the injury is of
a kind that is reasonably foreseeable. But where human conduct intervenes, the remoteness
principle is different, because human conduct is unpredictable. Liability in negligence for the
conduct of others is limited.

In these cases the plaintiff is seeking to hold the defendant responsible not for damage the
defendant caused through his positive act but for failing to prevent a third party from causing
damage of the plaintiff through the third party’s own deliberate conduct. As Lord Diplock says in
Dorset they raise the issue of the extent to which in law, the defendant is required to be “my
brother’s keeper”. In other words, what is the extent to which the law will impose an affirmative
duty on the defendant to take positive steps to prevent harm to the plaintiff caused by the
deliberate wrongdoing or negligent wrongdoing of a third person?

This category of case is sometimes called nonfeasance. In some of the cases the courts approach this
issue as a question of duty of care. Does the defendant owe a duty of care to this plaintiff to
exercise reasonable care to prevent the third party from damaging the plaintiff or the plaintiff’s
property? Lord Diplock in Dorset approaches the case as one of duty of care. In other cases the
issue is dealt with as a question of causation, and it is asked, at what point do the acts of the third
party become a new cause/an intervening cause/a novus actus, so that the defendant’s negligence
is no longer regarded as the cause or a cause of the plaintiff’s injury.

Sometimes the damage caused by the intervening conduct is considered by the court to be too
remote to be recoverable. Lord Reid in Dorset initially treated this issue as a remoteness issue
rather than as a question of duty. He asked whether the Borstal boys’ actions in escaping and
damaging the plaintiff’s boat were too remote or not.

It is not clear which device judges will use. This is an uncertainty inherent in the common law.

47
 ALEX BOND
Dorset Yacht Co. v Home Office [1970] AC 1004

Lord Reid’s judgment

What is the test Lord Reid proposed for the third party intervention cases? If the actions of the
third party were a reasonably foreseeable possibility, that is not enough. The probability that the
third party did what they did had to be very likely, the natural and probable result. On the facts
he felt that the risk was the very kind of thing the officers should have seen as likely. The
defendants were still responsible.

Lord Diplock’s judgment

[Lecturer commentary] Lord Diplock’s judgment will be hopefully crawled over by the Supreme
Court in the Couch case. His judgment is a classic example of inductive reasoning.

First, Diplock looks at two cases, Ellis v Home Office and D’Arcy v Prison Commissioners. In these
two cases a duty of care was owed by the prison. Why was there liability in these cases? He
explains at page 1061 two characteristics:
1. A relationship between the defendant and the third party; and
2. A relationship between the defendant and the plaintiff
Both the plaintiff and the third party (the prisoner) were in the custody of the defendant prison
commissioner and had the power to control the acts of the third party and the plaintiff.

Diplock then asks whether there would be a duty of care owing to a member of the public whose
property or person is injured by a prisoner who escapes from custody? He refers to Smith v Leurs
and says that the general rule is that one man is under no duty to control another man to prevent
him doing damage to the third, unless there is a special relationship.

This judgment is regarded as inductive reasoning. He looks at analogous cases where there is a
duty and forms a set of characteristics. What characteristic is missing in the present case? Is it
necessary to extend the duty to cover this characteristic?

Why was there liability in this case? Diplock says there wouldn’t be a duty generally for an
escaped criminal or for offenders that can’t be caught. The risk is “shared by the public at large”.
You need a special relationship. What is that relationship founded on?

He says at page 1070: “There should be some relationship between the custodian and the person to whom
the duty is owed which exposes that person to a particular risk of damage in consequence of that escape
which is different in its incidence from the general risk of damage from criminal acts of others which he
shares with all members of the public”.

As a plaintiff you must show that you are at some special, higher risk of damage than the general
risk. What distinguished these boys was that there was a distinctive added risk incurred by
people in the geographical vicinity of the island. A duty might be founded on or owed to those
people who can single themselves out as being exposed to an added risk of harm. This could
apply to the Couch case, because Susan Couch worked with William Bell everyday.

48
 ALEX BOND

A ‘distinctive added risk’ is the term used in this case. The plaintiff did face a distinctive added risk
in that the boys in order to not be recaptured were likely to steal or damage property in the
vicinity. A test of physical proximity.

The duty cannot be based on reasonable foreseeability alone.

Think: was Lord Diplock’s distinction sensible or a merely arbitrary one? It may been seen as
really just a means of limiting liability/limiting the class of plaintiffs.

Lamb v London Borough of Camden [1981] QB 625 (HL)

Lord Denning’s judgment

Lord Denning talks about what he perceives to be the ‘truth’ in these cases. He rejected Lord
Reid’s test from Dorset, believing it would extend liability too far.

He says duty, remoteness and causation are all different legal devices used by the judges to draw
the line of liability, and are preferred by different courts in different situations. But the ultimate
question is one of policy for the judges to decide. It requires a value judgment. Is it fair and just to
impose liability?

Policy here acts as a ‘hidden persuader’. Policy can be as simple as what the judge thinks is the just
outcome. It can be the policy effects of whether there is liability or not, e.g. effect on insurance, the
‘floodgates’. Policy is not legal principle, it is the other factors that the court considers important
to the outcome. It is the weighing of policy considerations that determines the outcome in a large
number of negligence cases.

Lord Oliver’s judgment

Lord Oliver concurred with Lord Denning. However he does regard Lord Reid’s test a “a workable
and sensible one”, only he believes the degree of likelihood required should be made higher.

Lord Reid’s test is sometimes applied, but there is more of tendancy to treat this as a duty of care
question. We should be aware of Reid’s test and the dissention from it.

Hill v Chief Constable of West Yorkshire [1989] 1 AC 53

In Dorset there was a duty of care owed. But what was missing from Hill such that there was no
duty of care owed by the police? The plaintiff was not exposed to an exceptional risk of harm not
faced by the public. The class of potential plaintiffs was quite big which in policy did matter. The
key to the plaintiff is to put themselves in a small class of people at a special risk.

Smith v Littlewoods Org Ltd [1987] AC 241 (HL)

A group of children broke into an empty cinema and started a fire with rolls of disused film. The
fire broke out and damaged the neighbouring properties.

49
 ALEX BOND

Lord Goff’s judgment

Lord Goff does not analyse the issue in terms of remoteness but in terms of the duty of care. He
starts out by explaining why the law does not recognise a general duty to prevent other people
from harming the plaintiff (a duty to prevent harm from a deliberate action by a third party).
Why? The common law doesn’t impose a general duty to rescue. In law, not in morality, there is
no duty to prevent a third party from harming someone else. The traditional position of the
common law is the mere omission rule. There is no duty to rescue someone in danger or to take
positive steps to prevent harm or confer a benefit on another person, even if it would involve little
or no effort on your behalf – unless – that person is under a legal duty arising out of a special
relationship. [In New Zealand a number of legal duties to act have been codified in the Crimes
Act 1961, ss 151-157].

The underlying policy reason for this stance is essentially freedom of action, the concern to
maximise a person’s freedom of choice as to how to act. Not all legal systems take the same
approach. In the French civil code there would be liability if the danger to the victim was “grave,
imminent, constant… nécessitant une intervention immédiate" and the rescuer would not face any
“risque pour le prévenu ou pour un tiers”.

The situations where there will be liability is strictly limited. In the absence of a special
relationship giving the defendant some degree of control over the third party, there is nothing to
base the duty to intervene on.

Goff talks about the special situations in which a defendant may be liable for a third party’s
deliberate conduct.

1. Where the defendant causes or permits to be created a source of danger, e.g. fireworks
example. He says that liability under this category will be rare for the policy reason that
there are many common household items that are potentially a source of danger, and that
we do not want ordinary life to be hampered.
2. Where the defendant has knowledge or means of knowledge that a third party has created
or is creating a risk. He discusses the case of Goldman v Hargrave where a fire sprung up on
a property in Western Australia and where the farmer chose to let the fire burn out. The
wind changed and the fire spread and there was a property damage claim. The court said
the standard of care might require more from a financially well-resourced or physically
strong farmer as opposed to a poor or infirm farmer. If you are a large corporation with
substantial financial resources, you may have to place a security guard outside your
cinema, but the same would not be expected of a household.

Judgment was made for the defendant, the cinema. This case was not in the first category because
an empty cinema was not unusual. It was neither in the second category because no one had
contacted the defendant to tell them that children were breaking into the cinema.

Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360 (HL)

A remand prisoner waiting to be charged committed suicide. Police were aware that the man had
attempted suicide on two prior occasions.

50
 ALEX BOND

How is this case different to the previous cases? In this case the argument is made that the novus
actus interveniens is the plaintiff’s own action, so the defendant should have saved the plaintiff
from himself (compared to saving the plaintiff from a third party as in Dorset and Hill).

Lord Hoffmann’s judgment

Is it harder or easier to hold the defendant responsible for the plaintiff’s own action? If it difficult
to hold the defendant liable where the intervening act is by a third party, it is even more rare to
hold the defendant liable where the intervening act is of the plaintiff himself or herself.

Counsel’s first argument was that the real cause of the plaintiff’s death was his own deliberate act
of suicide, and not the defendant’s negligence. Counsel was arguing a causation theory. Lord
Hoffmann didn’t see the solution as through causation however, because the very reason you take
precaution in prison cells is to prevent this sort of thing happening.

Why was a duty here owed the plaintiff? Yes, and for two reasons.
1. The prison had physical and legal control over the man because he was in prison; and
2. There was a special danger that people in remand would harm or kill themselves (again,
an ‘added risk’).

51
 ALEX BOND

Omissions and intervening causes: SUMMARY

1. This is a special category of case where it is sought to hold D liable for damage caused
by a third party over whom D has no control or some measure of control

2. Traditional approach — no liability for failure to rescue in English law in the absence of
a “special relationship” between D and the third party party or D and P (the “general
omissions” rule; “nonfeasance”). Reason: societal interest in freedom of choice of action.

A “special relationship” arises:

 between D and the third party where
 there is an aspect of control over the third party.
 between D and P where
 there is a higher risk: you have to single yourself out as a small class of plaintiffs
at a distinctive, added risk

3. Different judicial approaches:

 A duty issue – what is the scope of the duty? (e.g. Lord Diplock in Dorset)
 A causation & novus actus issue
 A remoteness issue (Lord Reid in Dorset)
There approaches all are legal devices to limit the range of liability (per Denning LJ in
Lamb). Essentially the ultimate question is a value judgment as to whether the D should
be liable for the consequences. The pre-dominant approach is as a duty issue.
[have to go through all possible issues in applying the law in an exam]

4. As a remoteness or causation issue, some judges try to define a test describing the degree
of likelihood that third parties would by their conduct create a risk of damage, e.g. Lord
Reid in Dorset. Special definitions of reasonable foreseeability have been adopted (“very
likely to happen”, “almost an inevitability”).

5. As a duty issue, a duty arises where there are twin aspects of control (legal & physical)
over the third person, and P faces an exceptional ‘distinctive added risk’ of harm. P has
to single self out from risk faced by general public because of floodgates concerns.
(Dorset, Hill, Reeves).

6. Lord Goff in Smith v Littlewoods devised two categories of duty on occupiers of land:
1. D negligently causes or permits to be created a source of danger on his land, and it
is reasonably foreseeable that the third party may interfere with it, sparking off the
danger;
2. D had the knowledge or means of knowledge that the third party is creating a risk
and fails to take reasonable steps in his or her individual circumstances to prevent
the risk. [What is reasonable tailored to D’s individual circ’s (Goldman v Hargrave)].

52