Omissions and the Mysterious Public Veil: Consideration of the “Immunity” of Public Authorities in Negligence Claims

In Yuen Kun Yeu v A-G of Hong Kong, Lord Keith famously stated that it

would be quite preposterous to impose ‘liability in negligence on the part of one who sees another about to walk off a cliff with his head in the air and

forbears to shout a warning’.3 In short, while society does encourage kind and

selfless behaviour to help others, the law does not punish those who had the

opportunity but have chosen not to become ‘good Samaritans’. In fact, it does

not make much sense to impose such liability anyway. Lord Hoffmann, in

Stovin v Wise, stated that it would be unreasonable if such a duty to prevent

harm to others existed, as an innumerate group of people would become liable

for each harm suffered.4

While that rationale is reasonable when applied to defendants belonging to

the general public, this article questions whether that reasoning can subsist

when applied to the context of emergency services and the armed forces. In

doing so, this article will briefly explore the aims of tort law itself before

examining whether the genesis and development of the ‘public immunity’

reasoning remains justifiable and logical today. Following which, this article

will analyse the latest Supreme Court decisions, such as Smith v Ministry of

Defence5 and Michael v Chief Constable of South Wales6

, and raises pertinent

questions pertaining to the judicial reasoning behind the courts’ decisions in

denying liability on emergency services and the armed forces, when they

could have actively prevented harm from occurring because they were

furnished with more resources and information than the average layman, but

had failed to do so.

Lord Dyson, in Jones v Kaney7

, explains that the entire system of law

predicates on the fundamental notion of addressing wrongful behaviour. A

wrong, in the context of tort law, is simply a breach of a duty owed to

someone else which results in harm to another. Harm refers to any physical or

mental damage, or a material and tangible detriment or loss to a person.

Without a system that lies on such a ‘cornerstone’ of justice, the public

perception of the law as an infallible system of fairness and impartiality would

be severely undermined.8 The law, therefore, takes a much stricter stance on

imposing liability on the wrongdoer, and providing a remedy to the victim.

There are a few popular theories of tort. While they differ slightly on what

is the ultimate goal that the system of tort law achieves, they agree on the

fundamental concept that harms which result from another’s wrongdoing should be remedied by law. Some writers believe that tort law provides a

structure that provides compensation for harms caused. Lord Dyson, writing

extra-judicially this time, argues that if a party has harmed another through

foreseeable actions, and there is a sufficient degree of proximity between the

two parties, the wrongdoer should compensate the victim.9 In doing so, the

compensation model provides an element of deterrence; there is an economic

disincentive for people to commit the same wrongs. On the other hand, several

writers believe that tort law is just about righting a wrong between two parties.

Weinrib highlights the exclusive nature of the relationship between the party

which inflicted harm and the victim and argues that tort law is only about the

imposition of liability on one party and the distribution of remedies to the

other.10 Stevens, however, believes that wrongs in tort law are defined so

because everybody has an inherent right not to be treated in a certain

manner.11 However, the fundamental point of agreement is that tort law does

provide a structure to remedy such wrongs, and especially harm, which occur

in society.

However, the courts have always relied on a controversial line of reasoning

when public authorities, including emergency services and armed forces, are

sued for negligence, and that is of the defence of public ‘immunity’. The

source of this infamous controversy arose from Hill v CC of West Yorkshire12,

which involved the murder of a young woman by the ‘Yorkshire Ripper’.

There were two strands of reasoning from Lord Keith’s decision to not hold

the police liable for failing to save the deceased. Firstly, unlike Home Office v

Dorset Yacht13, there was no special relationship between the police and the

deceased. It cannot be said that there is sufficient legal proximity between the

police and a potential victim of a crime.14 Secondly, as a matter of public

policy, the police are immune from actions for negligence. Imposing liability

for negligence would result in the police acting defensively, and that would

adversely affect the effectiveness of the police when they carry out investigation and supress crime.15 Henceforth, the Hill ‘immunity’ shall refer

to Lord Keith’s second strand of reasoning.

Nonetheless, this ‘immunity’ is qualified and only applies for cases

regarding omissions; the police continue to remain liable for negligence if

harm was directly caused or contributed by their actions. For example, in

Informer v Chief Constable, the Court held that the police owed a duty of care

to the claimant because they had recruited him as a covert human intelligence

source; by placing him in a position of risk, the law dictates that the police

cannot deny their responsibility for the claimant’s physical safety and wellbeing.16

Development of the ‘Immunity’ of Public Authorities

However, the greatest problem with the Hill ‘immunity’ principle was the way

in which the court applied the principle outside of the facts present in Hill.

The original case concerned two blameworthy parties: the Yorkshire Ripper

and the negligent police force. McIvor makes a salient point: Hill was a

unique case concerning third party liability, which is a unique species within

the law of omissions liability. In such cases, the courts are slow to impose

liability of negligence on the middleman because there was always a more

blameworthy party, i.e. the third party.17 In Caparo v Dickman, Lord Bridge

emphasised the principle that the law of negligence should develop

incrementally. When considering cases with different facts, the courts should

be careful to apply rules from previous analogous cases.18 However, this sort

of judicial scrutiny was lacking in two cases which has effectively expanded

the ambit of the Hill principle. In X v Bedfordshire CC, when Lord BrowneWilkinson endorsed the Hill reasoning and extended the ‘immunity’ of the

police forces to ‘all those charged by Parliament with the task of protecting

others’.19 This means that the ambit of the principle has been expanded to

include all public authorities, including emergency services and the armed

forces. Again, in Brooks v Commissioner of Police of the Metropolis20, which

concerned the negligence of the police during a racist attack on the claimant,

the court merely applied the Hill principle. In both cases, the House of Lords

merely applied the Hill principle without considering the individual factual

contexts, and failed to provide cogent justification why the Hill principle should extend outside third party liability situations. It is argued that this

broad-brush approach adopted in both cases, in contradistinction to the

scrutiny advocated in Caparo, in failing to take into account the unique

factual situation in Hill is strong indication that the public ‘immunity’ should

not be blindly applied in all negligence cases concerning public authorities

today.

In addition to arguing that the extension of the Hill principle outside third

party liability situations was questionable, the following parts in the article

will argue that the underlying justification for the principle itself is flawed.

Further, if the principle of defensive practice cannot subsist as a reasonable

argument, then there should be reason to redefine the limits of the ‘conflict of

duties’ defence.

The ‘Defensive Practice’ Argument Is Flawed

A strong indication of the fallibility of the ‘defensive practice’ argument is

that this policy consideration has not been consistently applied through case

law. In Dorset Yacht, Lord Reid explicitly denied the existence of such fears

of the suppression of police efficiency in the presence of legal duties.21 This

argument was denied again in Phelps v Hillingdon LBC, where Lord Clyde

was doubtful that such an attitude of fearful defensive practice will exist in

real life.22 In Barrett v Enfield LBC, the contrary was even suggested; the

potential existence of liability would actually enhance the overall standard

among public authority employees.23 Perhaps the reason for the court’s

inconsistency lies in the fact that the defensive practice argument is very

conjectural in nature. Fairgrieve, in a case commentary of Smith v MOD

(analysis below), points out that the defence is premised on ‘anecdote and

assumption’, and there is no empirical evidence to suggest that there would be

a culture of defensive practice.24 There is a strong reason for agreeing with the

conclusion drawn by Fairgrieve; in Hill, Lord Keith merely pointed to the

possibility that the police might act defensive if liability was imposed on

them25, and there have been no following cases which lists out evidence that

the imposition of liability will generate a culture of defensive practice.

Additionally, the defensive practice argument has been denied in Canada. In

Hill v Hamilton-Wentworth Regional Police Services Board, the Canadian courts refused to recognise an investigatory immunity for the police, explicitly

ruling out that there would be a culture of defensive practice.26

There is a further problem with the application of the defensive practice

argument. In the latest decisions concerning the emergency services and the

armed forces, it seems like the defensive practice mentality has taken such

prevalence in the minds of the highest courts that they dismiss any cases of

negligence concerning public authorities without analysing each case for its

merits. This has led to conflicting decisions between the Court of Appeal, the

House of Lords and the Supreme Court.

In the Court of Appeal decision of Smith v Chief Constable of Sussex

Police27, the court followed the earlier decision of Swinney v Chief Constable

of Northumbria28 and argued that, in contrast to the claimant in Hill, Smith’s

claim did not depend on his status as a member of the general public facing a

common risk; rather, because he was a witness for the police, an analogous

position to those occupied by informants, he should be entitled to a measure of

protection.29 However, when the case was considered again in the combined

appeal of Van Colle v Chief Constable of Hertfordshire, the House of Lords

adopted a blunt utilitarian rationale and dismissed the appeal on the basis that

the defensive practice argument effective overwrote any claim against the

police for negligence, regardless of the status of the claimant.30 While

applying the Hill principle, several of their Lordships mentioned obiter that

they were bound by precedent to deny any liability on the part of the police,

but noted their dissatisfaction with the outcome by stating that the law was in

need of reform.31

In the recent armed forces decision of Smith v MOD32, the judicial

confusion about the correct weight of the Hill principle was readily apparent

in the close 4:3 decision of the Supreme Court. Lord Hope, representing the

majority, stated that the general source of public ‘immunity’ from Hill has

been extended and applied to situations involving the armed forces in

Mulcahy v MOD33. However, a decent proportion of his Lordship’s judgment

was dedicated to proving that the scope of ‘combat immunity’, a relative of

the Hill ‘immunity’ principle, should be curtailed to include only actions taken

in the course of actual or imminent armed conflict.34 It is for that reason that

the Challenger claims could be permitted because they concerned a preparatory stage in the operations, whereas the Ellis claims concerned a

phase where there was an imminent threat of war.35 Lord Mance, one of the

dissenting judges, disagreed on this point. He argued that it was impossible to

distinguish between decisions made during pre-combat preparation and

decision made during actual combat.36 Lord Mance’s argument is logical; the

entire operation of the armed forces when it is engaging in war is geared

towards the goal of combat. It is arbitrary to draw lines between activities that

are ‘preparatory’ and ‘combat’ in nature, because all activities seek to achieve

the purpose of warfare.

Furthermore, Lords Mance and Carnwarth note the problem of giving the

Hill principle too much weight relative to the other factors would result in a

situation where the judges cannot see the wood for the trees, for doing so

would be engaging in a ‘dilemma’.37 Lord Mance states that it is incorrect to

determine common law liability based on the position which the defendant

occupies and whether he falls within the ambit of the Hill principle, but rather

the determination of liability involves the balancing exercise between the

considerations of harm and any public policy considerations, one of which is

combat immunity.38 McIvor makes a similar point in her article regarding

police liability: the courts are too preoccupied with the ambit of any public

authority ‘immunity’, and will only consider a duty of care as an exception.39

There is therefore strong authority to suggest that there too much judicial

weight is given to the Hill principle. Even in the original case of Hill v Chief

Constable of West Yorkshire itself, it was raised as a public policy exception

to deny liability. The problematic arrangement of perceiving the Hill

immunity as a prima facie assumption that all public authorities will not be

liable for their negligence leads to decisions where the courts may fail to

impose any duty on the individual merits of the case. Whilst Lord Rodger’s

statement that not all harms suffered can be remedied by law is correct, the

heavy-handed utilitarian reasoning adopted by the courts, when giving the

immunity a higher priority, is submitted to be an incorrect form of reasoning.

Failure to appreciate the individual merits of the case before considering the

exceptions is argued to be a betrayal of the first priority of the law to seek to

rectify wrongs.

The correct approach was adopted in the latest Supreme Court decision of

Michael v Chief Constable of South Wales.

40 Lord Toulson, representing the

majority, was at pains to state that the denial of liability was not because of the existence of police immunity41, but rather because there was a lack of

sufficient factual evidence to create an exception under the general omission

principle.42 It is argued that this approach is correct: the courts should only

consider any public policy considerations only after the examination of the

general law of omissions. However, Lord Toulson seems to suggest that

previous decisions were recognised on this correct approach.43 It is argued that

this is a flawed proposition. The authorities analysed above which already

prove that the court denied liability for public authorities because they

considered the public ‘immunity’ as a rule, rather than as an exception. In

addition, in Kent v Griffiths, liability was imposed on the ambulance services

because their duties were thought to be equivalent of that to doctors and

nurses, persons acting in a private capacity, and not that of the police and fire

services.44 In doing so, Lord Woolf implied that those occupying public office

would not be found liable if they were found in a similar position because of

the nature of their occupation. This is another salient example whereby the

courts have denied liability because of the existence of public immunity.

The ‘Conflict of Duties’ Argument Should Take Into Consideration the

Flawed ‘Immunity’ Principle

It is submitted that Lord Toulson’s approach in Michael is correct: the default

position under omissions liability is an absence of liability, but there should be

a balancing exercise of policy factors to determine if an exception can be

created to find liability, and one of the possible policy considerations is the

public ‘immunity’ principle. However, as argued above, the ‘immunity’

principle is flawed in substance, and should therefore not be given much

judicial weight. As a result, the balancing exercise should not always be

skewed towards the denial of liability when considering cases concerning

public bodies.

The ‘conflict of duties’ argument is a defence against the finding of

liability on the basis that the action to save or protect the claimant and

conflicts with the public duty of the defendant to protect a larger group of

persons. In two of the three appeals in Capital Counties v Hampshire County

Council, the Court of Appeal stated that the fire services were not liable for

causing damage to the claimants’ property because the process of

extinguishing the fire included damage to neighbouring properties.45 The recognition of the private duty of care would undermine the defendant

authority’s primary public duties to prevent fire damage to all properties.

The element that the recognition of the claimed duty of care will

undermine another duty of care owed a larger group of people is key to the

‘conflict of duties’ argument. It follows that if the claimed duty of care does

not conflict against another larger duty of care, the ‘conflict of duties’ will not

be a reason to find an absence of liability. In addition, if the non-conflicting

private duty of care owed to the claimant falls within one of the recognised

exceptions to the omissions principle, it is logical that the defendant will owe

the claimant a duty of care.

In Kent v Griffiths, the court held that the ambulance services were liable

for the negligence of sending an ambulance on time. They stated that there is

no conflict of duties between the claimant and the larger public. Rather,

because only an identified individual called the emergency hotline, the only

person who could be adversely affected for the failure of the ambulance

services to respond on time is the dialler himself.46

However, both the Court of Appeal and Supreme Court denied liability for

the police in the case of Michael v Chief Constable of South Wales. In that

case, it was the failure of the police to respond on time to an emergency call

which resulted in the tragic death of Miss Michael. Lord Toulson

distinguished the case from Kent v Griffiths on the basis that the call handler

in Kent had misled the deceased by assuring them that the ambulance would

arrive shortly. On the other hand, this element of ‘assurance’ was absent in the

facts of Michael. It was taken that the police had not assumed any

responsibility for the lack of its actions against Michael, and therefore an

exception cannot be created to impose liability on the negligence of the

police.47 It is submitted that this is an arbitrary distinction. Regardless whether

it is the police, which are taken to occupy public office, or the ambulance

service, which is perceived to be in the private sphere, the fact remains that

both are emergency services that respond to calls from the public. It makes

sense that people have a certain common perception that the emergency

services will respond quickly. Potential victims call emergency hotlines in the

most desperate time of need, and it often represents their lack of alternatives

available to them. The reliance on emergency services represents all they can

do to seek help. It would be unfair to the public if it the law recognises that the

emergency services could effectively shirk off its duties to respond quickly by

omitting any reassurances in its call.

The Liability Principle Can Be the Solution to the Existing Problem

In Van Colle, Lord Bingham attempted to create the ‘liability principle’ that

addressed this apparent lacuna in the law. He wrote that if a specific member

of the public were to give a public authority employee sufficient credible

evidence that a third party was creating a specific and imminent threat to his

life and physical safety, the public authority then owes that specific member

of the public a duty to assess the threat, and if required, take reasonable steps

to prevent the threat altogether.48 However, the Supreme Court in Van Colle,

and both the Court of Appeal and the Supreme Court in Michael refused to

endorse the ‘liability principle. Lord Toulson in Michael reiterates the

objections to the principle from Van Colle, stating that the principle creates an

unacceptable arbitrary distinction between people who call the emergency

services, and others who do not. He goes further in stating that there are

insufficient elements of a unique relationship between the caller and the

emergency services encapsulated in the principle, therefore the failure to

protect the claimant, which is part of its larger duty to society, does not

automatically give rise to a private cause of action.49

It is argued that Lord Toulson’s analysis was faulty in two respects.

Firstly, Lord Kerr, representing the minority, presents a refined version of

Lord Bingham’s ‘liability principle’; if a person delivered personalized

information about an imminent attack to anyone, the police is said to owe a

duty of care to that intended victim because they are furnished with specific

information and therefore have the capability to prevent the attack. The

quintessential element of specific and personalised information to the intended

victim, to be found in the factual matrix of each case, is key to establishing the

degree of proximity between the police and the victim, which was lacking in

Lord Bingham’s principle.50 Lord Toulson dismisses this argument on the

basis that the test for proximity is circular and provides no useful

measurement for determining liability in law.51 Lord Kerr concedes its

circular nature, but the entire test for proximity, in of itself, is already circular,

and yet it continues to be used by the courts today for the test of negligence.

This is an indication that the circular test still retains utilitarian value for

determining liability52, so there should be no reason to dismiss Lord Kerr’s

refined ‘liability principle’ on this point.

Secondly, while Lord Toulson agreed with the underlying rationale of the

‘liability principles’ from the submissions and the minority, he argued that the

proposition that acceptance of the principle would create an improvement in the efficiency of public authorities is speculative at best. The only significant

undesirable public policy eventuality which Lord Toulson proposes is that the

police would suffer significant financial ramifications,53 and this is submitted

to be a flimsy defence for the argument for public authority liability in such

situations. It may be the case that Lord Kerr’s test will create arbitrary

distinctions between people who dial the emergency hotline and people who

do not, but this should not prevent the law from recognising liability on the

police, or any public authority, where their blatant omissions have resulted in

tragic, but entirely avoidable, consequences.54 When authorities have been

furnished with specific information, and therefore could have done something

to prevent the harm, it seems ironic that the law is bound by its own rules and

is unable to provide a remedy.55 The salient point must be reiterated: when the

caller has done all he can to furnish information to the emergency services to

prevent the occurrence of harm during a desperate time, it seems unfair that

the authorities will not be found liable for their negligence simply because

they have not done anything. They occupy a specific role in society which

carries the common perception that they will respond in urgent situations,

however the law only attaches liability when the authorities ‘reassures’ the

caller. In other situations, they would not be found liable. This is an

unacceptable state of judicial reasoning.

Under the Human Rights Act 1998 (HRA 1998), individuals can rely on

specific rights under the European Convention in litigation against other legal

entities, including public authorities. In the latest decisions concerning the

emergency services and the armed forces, the common law claim for

negligence was always tried in conjunction with an Article 2 ECHR claim,

which concerns the right to life. Therefore, the question is whether the

common law action in negligence should develop in tandem the statutory

action for breach of ECHR rights under HRA 1998. This is especially so since

section 2 HRA 1998 states that the courts cannot act incompatibly with the

Convention or the jurisprudence from the European Court of Human Rights.

While Osman v UK is commonly remembered for the Strasbourg court’s

misunderstanding of the English law of negligence, its test for the liability of

public authorities remains important: in third party liability situations, there

should be a factual inquiry to determine if the authorities had known of a real and immediate threat, and whether they could have reasonably been expected

to prevent that risk.56 This presented a new and distinct test of liability from

the common law of negligence. However, the Supreme Court in Van Colle

was determined to emphasise that the findings under common law will not be

undone by any claims under HRA 1998, thereby implying that there are two

distinct tests for the law of negligence.57 McIvor writes that the result of this is

that subsequent cases would usually fail for both claims.58 It is argued that her

reaction was rather excessive: it is plain from the latest decisions of Smith and

Michael, that the HRA claims have succeeded when the common law claims

were dismissed.

Lady Hale, in the minority in Michael, argued for the proposition that the

common law duties for public authorities should develop in tandem with the

obligations of public bodies under HRA 1998. Lord Toulson, however, rejects

this call for pro-harmonization between the two distinct causes of actions

because of the different rationales giving rise to each action.59 It is argued that

Lord Toulson’s approach of differentiating the two tests of liability makes

sense. The entire bases of the two types of laws are already different; the

common law has a duty to remedy wrongs, whereas Convention claims are

intended to maintain the minimum standard of basic freedoms. There is,

therefore, no sound reason for creating a single test for two distinctly different

kinds of claims.

Lord Toulson is correct that any form of reform in this area of law,

whether to address the questionable weight of the ‘immunity’ principle or to

collate the two tests for liability, should be left to the decision of Parliament.60

It is not for the common law to determine it on its own accord without

introducing a host of uncertain rules through extensive jurisprudence and

different interpretations by different judges. Most importantly, interference by

the common law runs the risk of infringing the sanctity offered by the

separation of legislative and judiciary powers.

However, it seems that all hopes of leaving reform in the hands of the

Government may be futile. In 2008, the Law Commission published a

Consultation Paper detailing that the underlying rationale of the tort of

negligence to ‘provide compensation for those who suffer loss as a result of

the negligence of others’ should apply equally regardless whether the

defendant is a public individual or private body.61 Unfortunately, in a

subsequent paper published in May 2010, the Law Commission announced its intentions to abandon any reform efforts in this area of law because ‘the key

stakeholder – Government – was firmly opposed’ to its recommendations.62 In

Mohammed v Home Office, Sedley LJ commented directly on this episode and

stated: ‘it is a troubling comment on the functioning of the separation of

powers that the state’s independent law reform advisory body has had to

abandon a project affecting the liability of government to govern principally

because the control exercised by government over Parliament would frustrate

any reform, however wise or necessary, which would make government’s life

more difficult’.63 It is an unfortunate and lamentable reality that any reform

proposed by the Law Commission or within this article itself would not be

implemented in the foreseeable future.

This article is not intended to pin blame on those who willingly join the civil

service in the noble pursuit of protecting and saving lives. For example, it

would be blatantly unfair and unreasonable to impose liability on hospitals if

patients pass away due to natural causes. However, this article attempts to

point out that in certain cases whereby it is clear that the public authority in

question has fallen short of its mandatory operating protocol, which leads to

the harm or death of innocent individuals who have called for their help, the

courts tend to obfuscate the judicial process of determining their liability. In

other words, it has been highlighted in this article that the courts have failed to

distinguish between situations where harm and death is not of the fault of the

public authorities, and other situations where their blatant omissions have

caused the harm and death of innocent individuals. It is argued that the fact

that public authorities have, for an uncomfortable period of time, been given a

blanket immunity which manifests great social injustice. The result in Michael

v Chief Constable of South Wales64 is just one of the many decisions which

illustrate the awkward jurisprudence of not having any authority to account

for their mistakes. It is argued that it is incorrect to attach judicial weight to a

principle, and consequently deny any liability, on the speculative basis that the

public bodies might drop in efficiency in the future. This article recognises the

efforts of our everyday heroes which often go unnoticed, but the law should

not forget that these heroes are also mortals which err. The point which this

articles tries to emphasise is this: if society cannot rely on the public authorities for responding promptly at its most crucial juncture, who else can

society turn to in its most dire time of need?