What Is Negligence? When Carelessness Becomes a Legal Wrong

Before Donoghue v Stevenson, English law recognised duties of care only in particular relationships: innkeeper and guest, carrier and passenger, and so on. Lord Atkin's judgment in the snail case replaced the list with a principle. You must take reasonable care to avoid acts or omissions which you can reasonably foresee would be likely to injure your neighbour, and your neighbour is anyone so closely and directly affected by your act that you ought to have them in mind. A manufacturer who sends out a sealed bottle that no one will inspect before it is drunk owes a duty to the person who drinks it.

The case never established whether there was a snail. The House of Lords decided only that if the facts were as Mrs Donoghue said, she had a claim, and the manufacturer settled before trial. The principle it laid down has been applied since to doctors, drivers, employers, builders, solicitors, councils and the makers of everything.

A claimant in negligence has to prove four things, and a case fails if any one of them is missing:

• •Duty: the defendant owed the claimant a duty of care, which for physical injury is now assumed in most everyday situations
• •Breach: the defendant fell below the standard of a reasonable person in their position, whether by doing something a careful person would not or by failing to do something they would
• •Causation: the breach caused the loss, usually tested by asking whether the harm would have happened but for the defendant's conduct
• •Damage: the claimant suffered a recognised kind of harm, such as injury, property damage or, within limits, financial loss

The standard is objective. The question is not whether the defendant did their best but whether they did what a reasonable person would have done, and the reasonable person is neither perfect nor careless. A learner driver is judged by the standard of a competent driver, which seems harsh until one asks who should bear the cost of a learner's mistakes: the learner, who chose to drive, or the pedestrian, who did not. A professional is judged by the standard of a competent member of that profession, so a surgeon is asked whether a responsible body of surgeons would have acted the same way.

Courts weigh the likelihood of harm, how serious it would be, how costly it would have been to prevent, and the social value of what the defendant was doing. A cricket club whose balls cleared the fence a handful of times in thirty years was not negligent when one finally hit a passer-by; a factory that knew its floor was slippery and did nothing was.

The neighbour principle would make everyone liable to everyone if it were not limited, and most of the argument in negligence law since 1932 has been about where to stop. Pure financial loss, where nobody is hurt and nothing is damaged but money is lost, is recoverable only in narrower circumstances, because a careless statement could otherwise ruin an unlimited number of people who relied on it. Psychiatric harm to bystanders is limited to close relatives who witnessed the event or its immediate aftermath. And the law is reluctant to impose a duty to rescue: in most common-law countries a passer-by who watches a stranger drown commits no legal wrong, however great the moral one.

The remedy is damages, money to put the claimant as nearly as possible in the position they would have been in, and it is reduced where the claimant's own carelessness contributed to the harm. Someone who was not wearing a seat belt recovers less; someone who accepted a lift from a driver they knew to be drunk recovers less again.

Negligence is the law most people are most likely to meet. Road accidents, medical mistakes, injuries at work, faulty products and slips in shops are all negligence claims, and the standards the courts set feed back into how carefully drivers, doctors and manufacturers behave, because the cost of carelessness is charged to those who could have prevented it. Critics say it produces a compensation culture and defensive practice; defenders say that a rule requiring people to take reasonable care for their neighbours is about the least a legal system can ask.

Proving that someone was careless is not enough, and two well-known cases show why. In Barnett v Chelsea and Kensington Hospital Management Committee, decided in 1969, three night watchmen went to a hospital after drinking tea that made them vomit. The doctor on duty did not examine them and sent them home, and one later died of arsenic poisoning. The hospital had clearly breached its duty. But the evidence showed that even with prompt treatment the man would almost certainly have died, so the claim failed on causation: the carelessness did not make the difference.

The second limit is remoteness. In the Wagon Mound case of 1961, oil spilled carelessly from a ship in Sydney Harbour spread to a wharf where welding was going on. Nobody thought the oil could catch fire on water, but it did, and the wharf burned. The Privy Council held that the ship's operators were liable only for damage of a kind a reasonable person could have foreseen, and fire of that kind was not foreseeable at the time. Together the cases show that negligence asks not only whether someone was careless, but whether that carelessness caused a harm the law will let you recover for.

In 1947 the American judge Learned Hand turned the weighing of risk into a formula. In United States v Carroll Towing, about a barge that broke loose in New York harbour and sank with nobody aboard, he suggested that a defendant is careless if the burden of a precaution is less than the probability of harm multiplied by the size of the loss. Suppose a guard rail costs 500 pounds a year, and without it there is a one in a hundred chance each year of an accident causing 100,000 pounds of harm. The expected loss is 1,000 pounds a year, so leaving out the rail is unreasonable.

English courts do not use the formula openly, and putting prices on injuries is uncomfortable, but its logic runs through their reasoning. In Paris v Stepney Borough Council, a garage worker who was already blind in one eye was not given goggles, and a chip of metal blinded his good eye. Goggles were not standard for the job, but the House of Lords held that his employer should have supplied them to him in particular, because for a man with one eye the loss from an injury was far greater. In the language of the formula, the same small burden was now outweighed by a much larger possible harm.

Medical negligence uses a special version of the reasonable person test. In the English case of Bolam in 1957, the court held that a doctor is not negligent if they acted in a way accepted as proper by a responsible body of medical opinion, even if other doctors would have done things differently. Judging a doctor by what competent colleagues would accept, rather than by what a judge thinks best, became known as the Bolam test.

The law has since moved on in one important area. In 2015, in the case of Montgomery, the UK Supreme Court held that when it comes to warning patients about risks, the question is not what doctors typically disclose, but what a reasonable patient would want to know about material risks and alternatives. Consent became something shaped around the patient rather than medical custom.

Negligence is liability for harm caused by failing to take the care a reasonable person would have taken towards someone one ought to have had in mind, a principle established by the ginger-beer case of 1932. A claimant must show a duty, a breach of it, causation and recognised damage, the standard is objective, and the courts limit the duty where extending it would make liability unbounded.