Murder — the mens rea of homicide

Murder is a common law offence. Its actus reus is the unlawful killing of a human being under the Queen’s peace; its mens rea is traditionally described as malice aforethought. That expression is now misleading. It does not require malice in the ordinary sense, nor premeditation, nor any settled plan. In modern English law the mental element is an intention to kill or an intention to cause grievous bodily harm. The second limb — the so-called GBH rule — is the central moral and doctrinal controversy in this topic.

This week follows naturally from Weeks 1 and 2. You already have the tools of conduct, causation, omissions, intention, recklessness, and negligence. Murder is the first major offence in which those tools meet political gravity: the label is the most serious in the criminal law; the sentence is mandatory life imprisonment; and yet the mens rea can be satisfied by an intention to cause serious injury where death was neither desired nor foreseen as likely. That asymmetry between offence label and mental element is a recurrent theme in Cambridge supervision essays.

The topic has two distinct but connected questions. First, what state of mind is sufficient for murder? The answer is straightforward at the level of authority: intention to kill or to cause really serious injury. Foresight of consequences is not itself intention, though it may be evidence from which a jury may find intention. Secondly, how should a jury be directed where the defendant did not desire death or serious injury, but foresaw it as virtually certain? The answer is the Woollin formulation: in rare cases where a simple direction is insufficient, the jury may find intention if death or serious bodily harm was a virtual certainty, barring unforeseen intervention, and the defendant appreciated that fact.

The exam danger is to conflate three matters which must be kept separate: intention as a substantive mens rea; foresight as evidence of intention; and recklessness as a different mens rea altogether. The law of murder has moved away from objective tests and from mere probability. It has not, however, adopted the view that foresight of virtual certainty automatically is intention as a matter of law. Matthews and Alleyne confirms that Woollin is evidential, not definitional.

For Part IA purposes, the examiner will expect accurate statement of doctrine, but the better answer goes further. It will explain why Vickers and Cunningham make the GBH rule orthodox, why Moloney, Hancock, Nedrick and Woollin are best understood as a judicial retreat from over-expansive foresight liability, and why academic criticism of murder’s mens rea cannot be resolved simply by saying that serious injury is serious culpability. Murder is a useful topic precisely because the black-letter law is compact but the normative stakes are substantial.

The phrase malice aforethought is a historical survival. It originally gathered together a range of culpable mental states in homicide. Over time, it came to signify the mental element distinguishing murder from manslaughter, but its language became increasingly unsuitable for modern doctrine. The word malice does not mean spite or hostility; aforethought does not mean pre-planning. A sudden killing in a spontaneous fight may be murder if the defendant intended to kill or to cause grievous bodily harm. Conversely, a killing committed with dislike but without the relevant intent is not murder.

A central historical feature was constructive malice. At common law a killing in the course of committing another felony could be murder even without proof that the defendant intended death or serious injury. That doctrine treated the mens rea of the underlying felony as supplying the malice for homicide. It was abolished by section 1 of the Homicide Act 1957. The significance of that provision is often underestimated. It marks an important statutory insistence that murder requires its own mental element. A defendant does not commit murder merely because death occurs during another offence; the prosecution must prove the same malice aforethought as would be required outside that context.

The next historical controversy concerned the content of that malice. R v Vickers settled, for English law, that an intention to cause grievous bodily harm is sufficient for murder. The defendant broke into a shop, attacked an elderly woman, and she died. The Court of Criminal Appeal rejected the argument that an intention to kill was necessary. The rule was later affirmed by the House of Lords in Cunningham. This is not constructive malice, because the mental element is directed to the victim’s body rather than merely to another offence; but it is constructive in a broader moral sense, because the defendant is treated as a murderer despite lacking an intention to kill.

A separate line of development concerned oblique intention. Direct intention is comparatively easy: D acts in order to bring about death or serious injury. The difficult case is where D pursues another aim, knowing that death or serious injury is highly likely or inevitable as a side-effect. Early modern cases oscillated between probability, high probability, and virtual certainty. DPP v Smith adopted an objective approach, asking what a reasonable person would have foreseen as the natural and probable result. That approach was widely criticised and reversed by statute for murder by section 8 of the Criminal Justice Act 1967, which made clear that intention and foresight are not to be inferred merely because a consequence is natural and probable, but must be decided by reference to all the evidence.

Hyam illustrates the dangers of treating foresight of probable consequences as equivalent to intention. Moloney attempted to restore conceptual discipline by insisting that foresight is not intention. Hancock and Shankland then corrected the under-inclusiveness of Moloney’s jury guidance by emphasising probability: the greater the probability of a consequence, the more likely it is that it was foreseen, and the more likely it is that it was intended. Nedrick introduced the virtual certainty test. Woollin adopted it, with the important substitution of find for infer, and Matthews and Alleyne later treated the Woollin direction as evidential rather than as a substantive definition.

The modern law of murder’s mens rea is therefore the product of two histories. One is the survival and consolidation of the GBH rule. The other is the refinement of oblique intention away from objective foresight and mere probability towards a narrow virtual certainty direction. The first history arguably broadens murder liability; the second narrows it.

The mens rea of murder has three core propositions.

First, murder requires an intention to kill or an intention to cause grievous bodily harm. Grievous bodily harm means really serious harm, though D need not know that lawyers would classify the injury in that way. The jury must be satisfied that D intended to cause injury of that seriousness. It is not enough that D intended some harm, nor that D was reckless as to serious harm. The GBH limb is firmly established by Vickers and Cunningham. It applies even if D did not foresee death as probable, or indeed did not think about death at all. It is the intention to cause serious injury which supplies the malice for murder.

Secondly, intention must be distinguished from motive, desire, foresight and recklessness. Motive explains why D acted; intention identifies the consequence D aimed at or the consequence which the jury is entitled to treat as intended in the narrow oblique category. A mercy killer may act from compassion yet intend death. A terrorist who plants a bomb to destroy property may not desire the death of bystanders, but may appreciate that death is virtually certain. Recklessness, by contrast, involves conscious risk-taking. A person who foresees death as possible or probable and unreasonably runs that risk may be highly culpable, but that mental state is not, without more, the mens rea of murder.

Thirdly, the Woollin direction applies to oblique intention. In the ordinary case the trial judge need say little more than that intention bears its ordinary meaning. Where the defendant denies that death or serious injury was his purpose, but the evidence suggests that such a consequence was inevitable or virtually inevitable, the jury may be directed that they are not entitled to find intention unless they are sure that death or serious bodily harm was a virtual certainty, barring unforeseen intervention, and that D appreciated that. The test has two limbs: an objective limb about the consequence and a subjective limb about D’s appreciation. Both are demanding.

The cases fall into two groups: the GBH rule and oblique intention.

Vickers is the foundation of the modern GBH rule. The defendant broke into premises, attacked the victim, and she died. The Court of Criminal Appeal held that an intention to cause grievous bodily harm was sufficient for malice aforethought. The importance of Vickers lies less in its facts than in its conceptual move: murder is not limited to intentional killing. The law treats an intention to inflict really serious injury as sufficiently culpable for the murder label if death results.

Cunningham later confirmed that proposition at the highest level. The defendant struck the victim with a chair during a robbery; the victim died. The House of Lords rejected the argument that death had to be foreseen. Intention to cause serious harm was sufficient. Cunningham is therefore the answer to any problem question in which D intended only to disable, wound, or beat V severely, but V died. The examiner will expect recognition that, subject to proof of really serious injury and causation, murder may be made out.

The oblique intention line begins with instability. Hyam involved a defendant who set fire to a house, apparently to frighten a rival, and children died. The House of Lords treated foresight of a high probability of death or serious injury as sufficient. The case is difficult because it appears to blur intention and recklessness. It is now best treated as superseded by later authority, though it remains historically significant.

Moloney was an attempt at correction. Lord Bridge insisted that foresight of consequences is not intention; it is evidence from which intention may be inferred. The jury guidance proposed in Moloney asked whether death or serious injury was a natural consequence of D’s act and whether D foresaw it as such. That guidance was too weak, because natural consequence might include consequences less than virtually certain.

Hancock and Shankland refined the position by stressing probability. The defendants dropped a concrete block from a bridge during a miners’ strike; a taxi driver was killed. The House of Lords held that the Moloney guidelines were misleading if they omitted reference to the degree of probability. The greater the probability of a consequence, the more likely it was foreseen and, if foreseen, the more likely it was intended. Probability was evidentially central, but still not equivalent to intention.

Nedrick supplied the decisive formulation. The Court of Appeal held that in rare cases the jury should not infer intention unless death or serious injury was a virtual certainty and D appreciated that. Woollin adopted that approach after a father threw his baby towards a pram or cot and the child died. Lord Steyn substituted find for infer, producing debate over whether foresight of virtual certainty became a substantive definition of intention. Matthews and Alleyne later restored orthodoxy by treating Woollin as a rule of evidence.

Together these cases produce a careful but uneasy settlement. Vickers and Cunningham broaden murder by allowing GBH intent. Nedrick and Woollin narrow intention by excluding mere foresight of probability. Matthews preserves the conceptual distinction between foresight and intention. In an exam, the best answers do not recite these cases chronologically; they use them to show the tension between culpability, fair labelling, and workable jury direction.

See practice questions section below.

See further reading section below.