Breach of duty

Breach of duty is the point at which negligence ceases to be a jurisdictional inquiry into relationships and becomes an evaluative inquiry into conduct. In Weeks 1 to 4 the central question was whether the defendant owed the claimant a duty of care, especially where the law is reluctant to impose affirmative obligations, duties to avoid psychiatric injury, duties for pure economic loss, or duties on public authorities. This week assumes that a duty exists. The question is then whether the defendant fell below the standard of care required by law.

The orthodox formulation is simple: the defendant is in breach if he failed to take such care as would have been taken by the reasonable person in the circumstances. That simplicity is deceptive. Breach is not merely factual. It is structured by normative judgments about the magnitude of the risk, the seriousness of the possible harm, the cost and practicability of precautions, the social value of the defendant’s activity, the defendant’s skill or profession, the claimant’s vulnerability, and the state of knowledge at the time of the accident. It is also affected, though not usually transformed, by statute.

The important point for Cambridge purposes is to avoid treating breach as a mechanical list of factors. The courts are not applying a mathematical calculus, although the language sometimes resembles one. They are asking whether reasonable care required a particular precaution in the circumstances. Bolton v Stone and Paris v Stepney BC are not inconsistent; they involve different assessments of probability and gravity. Latimer v AEC and Watt v Hertfordshire are not exemptions from liability; they are illustrations of reasonable response, cost, and emergency. Roe v Minister of Health does not make ignorance a defence in the abstract; it prevents hindsight from converting unknown risks into negligent conduct.

Nor should the student assimilate breach to duty. Duty identifies the class of obligations which the law is prepared to recognise. Breach asks whether the defendant conformed to the standard within an admitted obligation. That distinction matters in problem questions. A hospital may owe a duty to a patient, but the breach issue may turn on Bolam, Bolitho, and Montgomery. A school may owe a duty to a pupil, but breach depends on the reasonable supervision appropriate to age, activity, and danger. A local authority may owe a duty in a particular operational setting, but breach still requires proof that its conduct fell below the applicable standard.

For Part IA, breach is a high-yield topic because it tests legal technique rather than novelty. Good answers state the reasonable person standard, identify any special standard, apply the risk factors with precision, and avoid outcome reasoning. Excellent answers also recognise the policy undercurrents: negligence law must protect physical security without turning every accident into actionable fault. Breach is therefore the law’s chief instrument for separating misfortune from civil wrong.

The modern law of breach descends from nineteenth-century attempts to give legal content to negligence without reducing it to either strict liability or subjective blame. Blyth v Birmingham Waterworks Co remains the canonical starting point. Alderson B defined negligence by reference to the conduct of the reasonable and prudent person. This formulation accomplished two things. First, it made liability objective: the defendant’s personal clumsiness, timidity, ignorance, or inexperience would not ordinarily lower the standard. Secondly, it made liability contextual: the reasonable person acts upon the considerations which ordinarily regulate human affairs, so the inquiry is necessarily practical rather than abstract.

The early common law had to distinguish negligence from accident. Industrialisation, urbanisation, railways, factories, public utilities, and later motor vehicles created large numbers of accidental injuries. The courts were unwilling to compensate all loss. Breach therefore became a filter. It allowed the law to say that some risks are an unavoidable price of ordinary life, while others are culpably created or insufficiently controlled. The famous cases on cricket balls, factory floods, fire engines, and medical ampoules are not historical curiosities. They show how the law uses breach to allocate the burden of precaution.

A second historical movement is the emergence of specialised standards. The reasonable person is not always an undifferentiated adult. Professionals are judged by reference to the ordinary competent member of the profession. Drivers are judged by the standard of the reasonably competent driver, not by their own experience. Children are judged in a more age-sensitive way. This did not displace objectivity; it refined the reference class. Nettleship v Weston is the hard edge of the rule: a learner driver must meet the same standard as a reasonably competent driver. Mullin v Richards shows the softer edge: a child defendant is measured against the ordinary child of comparable age.

Medical negligence produced the most important doctrinal substructure. Bolam v Friern Hospital Management Committee gave strong weight to responsible professional practice. Bolitho v City and Hackney Health Authority then insisted that professional opinion must withstand logical analysis. Montgomery v Lanarkshire Health Board altered the position for disclosure of treatment risks: the standard is no longer controlled by medical paternalism but by the patient’s entitlement to be informed of material risks and reasonable alternatives. The result is a dual structure: diagnosis and treatment often remain within a modified Bolam approach; advice and consent are governed by Montgomery.

A third development is statutory anxiety about defensive practices and the so-called compensation culture. Section 1 of the Compensation Act 2006 permits courts to consider whether requiring a precaution would deter desirable activities. The Social Action, Responsibility and Heroism Act 2015 requires courts to have regard to social action, responsible conduct, and heroic intervention. These provisions are more rhetorical than revolutionary. They largely restate themes already present in cases such as Watt and Tomlinson: courts should not impose standards which drain socially useful activities of life, spontaneity, or reasonable risk.

The historical lesson is therefore not that breach has become increasingly claimant-friendly or defendant-friendly. It is that breach has become more candidly contextual. The reasonable person standard supplies continuity; the factors by which reasonable conduct is assessed supply flexibility. That combination explains both the power and the difficulty of the doctrine.

1. The standard is objective. The ordinary starting point is that the defendant must act as a reasonable person would have acted in the circumstances. The defendant is not excused because he was inexperienced, unusually nervous, poor at judgment, or personally unaware of a risk which a reasonable person would have appreciated. This is clearest in driving cases. In Nettleship v Weston a learner driver owed the same standard as a reasonably competent qualified driver. The policy is obvious: road users are entitled to a common level of safety, and loss is usually insurable. The same logic applies to many activities where the defendant voluntarily engages in conduct posing physical risks to others.

1. The standard is objective but contextual. Objectivity does not mean abstraction. The reasonable person is placed in the defendant’s circumstances, excluding merely personal deficiencies but including legally relevant circumstances. Thus a surgeon is judged as a reasonably competent surgeon; a rescuer acting in an emergency is judged in light of urgency; a schoolchild is judged by reference to an ordinarily prudent child of comparable age. The key is to identify the correct reference class. A poor answer says simply, ‘the defendant was inexperienced’. A good answer asks whether inexperience is legally relevant to the standard. Usually it is not; childhood may be.

1. Foreseeability of harm is necessary but not sufficient. A defendant is not negligent merely because harm was foreseeable. The question is whether the risk was sufficiently real that reasonable care required precautions. Bolton v Stone shows that a very small risk may reasonably be left unguarded. The cricket ground had a long history of only rare balls being hit out. By contrast, in The Wagon Mound (No 2) a small risk of serious fire could require precautions where the precaution was simple and the defendant had no good reason for failing to take it. The exam point is to avoid treating foreseeability as doing all the work. Breach requires assessment of foreseeable risk, not bare possibility.

1. Probability of harm matters. The more likely the harm, the more precautions reasonable care may demand. A remote risk may not require expensive, intrusive, or activity-stopping precautions. But probability is never assessed in isolation. A low probability risk of catastrophic injury may require action; a high probability risk of trivial inconvenience may not. The court’s inquiry is practical: what would a reasonable person do when confronted with this risk, in this activity, at this time?

The breach cases should be learned as a set of principles rather than isolated facts. Blyth supplies the general objective standard. It is the foundation case because it expresses negligence as a failure to act as a reasonable person would act. It also resists strict liability: the water company was not liable merely because its hydrant failed during an exceptional frost. The standard was reasonable foresight and reasonable precaution, not perfection.

Bolton v Stone introduces the probability limb. Cricket balls had occasionally been hit out of the ground, but very rarely. The House of Lords held that reasonable care did not require the club to take further precautions. The case is sometimes misunderstood as saying that infrequent accidents never found breach. That is wrong. Its force depends on the combination of low probability, ordinary recreational activity, and the precautions already in place. If the likelihood had been higher, or the likely injury more serious, or the precaution easier, the answer might have differed.

Paris v Stepney BC supplies the gravity principle. The employer knew the claimant had only one good eye. Goggles were not standardly required for all workers doing the task, but the risk to this claimant was more serious than usual. The case demonstrates that the reasonable person sometimes must adapt precautions to known individual vulnerability. It also helps distinguish breach from remoteness: the issue was not merely that the claimant’s eventual loss was more extensive; it was that the employer should have taken greater care before the accident.

Latimer v AEC is the leading case on practicability and proportionality. The employer had taken substantial steps after a flood by spreading sawdust, but some slippery patches remained. The claimant argued that the factory should have been closed. The House of Lords rejected this. Reasonable care did not demand the elimination of every residual risk at the cost of stopping operations. Students should treat Latimer carefully. It does not excuse cheap safety failures. It protects a defendant who has taken reasonable precautions from liability based solely on the fact that a further, more drastic precaution would have avoided the harm.

Watt v Hertfordshire CC illustrates emergency and social utility. Firefighters were transporting equipment needed to rescue a trapped woman. Because the ordinary vehicle was unavailable, they used another vehicle and a firefighter was injured. The court treated the emergency purpose as highly relevant. The case should not be converted into a broad emergency services immunity. The better reading is that reasonable care is calibrated to urgent circumstances and the value of rescue.

Roe v Minister of Health is indispensable on hindsight. Medical staff stored anaesthetic ampoules in a disinfectant solution, as was then standard. Invisible cracks allowed contamination, causing paralysis. The risk was not reasonably known at the time. Denning LJ’s reasoning insists that negligence must be judged by the knowledge reasonably available when the defendant acted, not by later scientific understanding.

Bolam and Bolitho form the professional negligence pair. Bolam states that a doctor is not negligent if acting in accordance with a responsible body of professional opinion. Bolitho qualifies this by requiring that the opinion have a logical basis. The courts defer to expertise, but not to irrational professional custom. Montgomery then marks the modern autonomy turn in medical advice. Disclosure of material risks is not governed simply by what doctors customarily disclose. Patients are entitled to information material to their decision, including reasonable alternatives. Together these cases show that breach is one doctrine, but its application varies sharply with context.

See practice questions section below.

See further reading section below.