Introduction
In the law of England and Wales, the doctrine of causation is a fundamental concept that connects a defendant's conduct to a prohibited result. It is a necessary element that must be established by the prosecution in criminal law to secure a conviction for a result crime, and by a claimant in tort law to prove that the defendant’s breach of duty caused their loss. Without a proven causal link, there can be no liability. The law approaches causation through a two-stage test. First, the prosecution or claimant must establish ‘factual causation’. Second, they must also prove ‘legal causation’. This essay will explain these two stages, using key case law to illustrate how the principles are applied by the courts in both criminal and civil contexts. It will demonstrate that while factual causation is a relatively straightforward enquiry, legal causation involves a series of rules designed to determine whether it is fair to attribute the result to the defendant's actions.
Factual Causation: The 'But For' Test
The first step in establishing causation is to determine whether the defendant's actions were a factual cause of the outcome. The standard test for this is the 'but for' test. The question asked is: 'but for the defendant's conduct, would the result have occurred?' If the answer is yes (the result would have happened anyway), then the defendant’s conduct is not a factual cause, and there is no liability. If the answer is no, then factual causation is established, and the enquiry can proceed to the next stage of legal causation.
The application of this test can be clearly seen in the criminal case of R v White (1910). The defendant put cyanide in his mother's drink, intending to kill her. She was later found dead, but medical evidence showed that she had died from a heart attack, not from the poison. The 'but for' test was applied, and it was found that she would have died regardless of the defendant’s act. Therefore, his actions were not the factual cause of her death, and he could not be convicted of murder. He was, however, found guilty of attempted murder.
A similar principle is applied in the law of tort. In Barnett v Chelsea & Kensington Hospital Management Committee (1969), a man went to a hospital emergency department complaining of severe stomach pains and vomiting after drinking tea. The doctor on duty, who was himself unwell, refused to see him and sent him home, advising him to see his own doctor in the morning. The man died five hours later from arsenic poisoning. His widow sued the hospital for negligence. The court found that the hospital had breached its duty of care by not examining him. However, evidence showed that even if he had been examined and treated with all due care, he would have died anyway as the poison was already too far advanced. Applying the 'but for' test, the doctor's negligence was not the factual cause of the death, and the claim failed.
These cases demonstrate that the 'but for' test acts as a primary filter, eliminating cases where the defendant's conduct had no real effect on the outcome. However, the test can be difficult to apply in cases involving multiple potential causes, and the courts have sometimes adapted the test, for example, by asking whether the defendant's conduct made a 'material contribution' to the harm (Bonnington Castings Ltd v Wardlaw, 1956). Nonetheless, the 'but for' principle remains the starting point for any analysis of causation.
Legal Causation: Attributing Responsibility
Once factual causation is established, the prosecution or claimant must also prove legal causation. This stage is concerned with whether the defendant's conduct was a legally significant cause of the result. It is a more complex and normative enquiry than factual causation, involving policy considerations to ensure that blame is fairly attributed. For legal causation to be established, the defendant's act must be an 'operating and substantial' cause of the result (R v Smith, 1959). This involves two main considerations: the significance of the defendant's contribution and the absence of any intervening act that breaks the chain of causation.
The requirement that the cause be 'substantial' means that it must be more than a minimal or trivial cause (R v Kimsey, 1996). The defendant’s role does not need to be the only or even the main cause, but it must be a significant one.
More importantly, the defendant’s act must be the 'operating' cause at the time of the result. This means that the original act must still be a live factor. The chain of causation between the defendant’s initial act and the final result can be broken by a novus actus interveniens – a new intervening act. If such an act occurs, it will be considered the true cause of the result, and the defendant will be absolved of liability for that result. The law recognises three main types of intervening acts that may break the chain of causation.
Act of a Third Party
An act by a third party will only break the chain of causation if it is a free, deliberate, and informed act that is not a reasonably foreseeable consequence of the defendant's own actions. In R v Pagett (1983), the defendant, armed and resisting arrest, used his pregnant girlfriend as a human shield. He fired at the police, who returned fire, hitting and killing the girlfriend. The defendant was convicted of manslaughter. He argued that the police officers' act of firing the fatal shot broke the chain of causation. The Court of Appeal disagreed, holding that the police officers' actions were a reasonable and foreseeable response to the defendant’s own dangerous act. Their actions were not 'free and voluntary' but were a form of self-defence. Therefore, the chain of causation was not broken.
Act of the Victim
The victim's own actions can also, in some circumstances, break the chain of causation. This often arises in 'fright and flight' cases, where the victim is injured while trying to escape from a threat posed by the defendant. The test, established in R v Roberts (1971), is whether the victim’s reaction was a reasonably foreseeable consequence of the defendant's actions. In Roberts, a girl was injured when she jumped from a moving car to escape the defendant’s unwanted sexual advances. The court held that her reaction was not 'so daft' as to be unforeseeable, and therefore the chain of causation was not broken. However, if the victim's reaction is entirely disproportionate or unreasonable, as suggested in R v Williams and Davis (1992), it may be held to break the chain.
Furthermore, the law applies the 'thin skull' rule, which states that the defendant must take their victim as they find them. This means that if the victim has a pre-existing vulnerability, whether physical or psychological, that makes them more susceptible to harm, the defendant is still responsible for the full extent of the harm caused. In R v Blaue (1975), a young woman, a Jehovah's Witness, was stabbed by the defendant. She required a blood transfusion to save her life but refused it due to her religious beliefs and subsequently died. The defendant argued that her refusal was an unreasonable act that broke the chain of causation. The Court of Appeal rejected this argument, stating that the defendant had to take his victim not just as he found her in body, but also in mind, including her religious beliefs. Her refusal did not break the chain of causation.
Medical Treatment
In cases where a victim receives medical treatment for injuries inflicted by the defendant, it is very rare for that treatment to break the chain of causation. The courts are reluctant to allow defendants to escape liability by blaming medical professionals who are trying to save a life. The leading principle comes from R v Smith (1959), where a soldier was stabbed by another soldier. On the way to the medical station, he was dropped twice and then received treatment described as 'thoroughly bad', which likely contributed to his death. The court held that the original wound was still an 'operating and substantial cause' of death, and the medical treatment, however poor, did not break the chain of causation.
The later case of R v Cheshire (1991) confirmed this position. The victim was shot by the defendant and required a tracheotomy as part of his treatment. Weeks later, after the original wounds had largely healed, he died from complications arising from the tracheotomy. The court held that even though the medical complication was the immediate cause of death, the defendant was still liable. The chain of causation would only be broken if the medical treatment was so independent of the defendant's act and so potent in causing death that the defendant's acts are rendered insignificant. This is a very high threshold, and as such, negligent medical treatment will not generally break the chain of causation unless it is exceptionally poor.
Conclusion
In conclusion, the legal concept of causation is a two-part process that is essential for attributing legal responsibility in both criminal and civil law. The first stage, factual causation, is determined by the 'but for' test, a relatively simple factual enquiry. If this is satisfied, the court moves to the more complex and evaluative stage of legal causation. This second stage ensures that the defendant's act was a 'substantial and operating' cause of the result. The key function of legal causation is to consider whether a subsequent event, a novus actus interveniens, is sufficient to break the chain of causation. The case law shows that the courts have developed specific rules regarding acts of third parties, the victim, and medical professionals, generally setting a high bar for the chain to be broken. This ensures that defendants cannot easily evade responsibility for the consequences of their actions, reflecting the law's underlying purpose of achieving a just and fair outcome.
References
Barnett v Chelsea & Kensington Hospital Management Committee [1969] 1 QB 428.
Bonnington Castings Ltd v Wardlaw [1956] AC 613.
R v Blaue [1975] 1 WLR 1411.
R v Cheshire [1991] 1 WLR 844.
R v Kimsey [1996] Crim LR 35.
R v Pagett (1983) 76 Cr App R 279.
R v Roberts (1971) 56 Cr App R 95.
R v Smith [1959] 2 QB 35.
R v White [1910] 2 KB 124.
R v Williams and Davis [1992] 1 WLR 380.
