Actus Reus Lecture

Elements of an Offence

In order for an individual to be found liable for any offence, three elements must be satisfied.

  • There must be guilty conduct by the defendant (actus reus)
  • The defendant must have a guilty state of mind (mens rea)
  • There must be no valid defence

In later chapters it will be seen that sometimes a defendant can be guilty of an offence without a guilty state of mind (strict liability offence, for example), but in every case, the actus reus must be proved before a conviction can occur.

Although in most cases a defendant is required to do something in order to satisfy the actus reus, situations where this is not the case will be discussed below. Additionally, in certain circumstances, the actus reus requires an additional factor to exist or some specific consequence to follow in order for the actus reus to be established. Under section 18 of the Offences Against the Person Act 1861, for example, the defendant’s actions must wound or cause grievous bodily harm to the victim. The act of assaulting the victim will not be sufficient to satisfy the actus reus without this particular consequence. The distinction is described as being a conduct crime or a result crime.

A further example can be found under section 1(1) of the Criminal Damage Act 1971 which provides:

A person who, without lawful excuse destroys or damages property belonging to another intending to destroy or damage any such property or being reckless as to whether any such property would be destroyed or damaged shall be guilty of an offence.

The actus reus of the offence is the destruction or damage of property which belongs to somebody other than the defendant. It is the result of the defendant’s actions that satisfy the actus reus rather than the actions themselves, with the additional circumstance being that the property destroyed or damaged belongs to somebody other than the defendant. In other words, all three elements/types of actus reus are included.

  • Conduct - some action that results in damage to property;
  • Circumstance - the property belongs to somebody other than the defendant;
  • Result - the property is damaged or destroyed.

If this offence is contrasted with the offence of rape, section 1 of the Sexual Offences Act 2003 provides that:

  1. A person (A) commits an offence if -
    1. he intentionally penetrates the vagina, anus or mouth of another person (B) with his penis,
    2. B does not consent to the penetration, and
    3. A does not reasonably believe that B consents.

All that is required for the actus reus to be satisfied in this context, is that the penetration with the penis occurred, the victim did not consent and the defendant. It is therefore the conduct of the defendant that satisfies the actus reus, no particular result is necessary.

Examination Consideration

In all circumstances, it is essential that all elements of the actus reus are satisfied. This includes the defendant’s conduct, the existence of certain circumstances and/or a particular result. All of the elements for each offence must be worked through methodically on the basis of the definition of the offence. This approach applied to every offence and in whatever circumstances the offences arises. It is not a matter that is limited to questions on actus reus.

Liability for Omissions

In most cases, a defendant will be required to act in order to satisfy the actus reus of an offence; the defendant must do something. There is no general requirement under English and Welsh law for a person to act. If, for example, a person sees another drowning and can easily save them by throwing into the water a nearby buoyancy aid, the person decides not to throw in the aid and the drowning person dies, the person who could have affected the rescue will not be guilty of an offence. The potential rescuer is not obliged to act.

There are however, several circumstances where a duty to act is imposed and where omission satisfies the actus reus of an offence.

Statutory Duty

There are numerous statutory provisions that require an individual to act in a particular manner and which, if the individual does not act in that manner, cause the individual to be criminally liable. Under section 7(4) of the Road Traffic Act 1972, for example, it is an offence to fail to provide a specimen of breath when required to do so by a police officer. Section 170 of the Road Traffic Act 1988 makes it an offence to fail to stop after a road traffic accident. It is also an offence to fail to keep proper accounts under the provisions of the Companies Act 2006 or the Value Added Tax Act 1994.

Examination Consideration

It is useful to learn a few of these examples (there are numerous more), because questions on liability for omission are common.

Duty Arising from a Special Relationship

Care or Control of Children

Section 1 of the Children and Young Persons Act 1933 makes it an offence for a parent or any other person over the age of 16 years who has responsibility for a child under the age of 16 years to wilfully neglect the child. This specifically includes failing to provide adequate food, clothing or medical care. Whilst this is a statutory provision and could easily fall within the previous section, it has links to liability that falls outside the express wording of the Act. Under this provision, a parent will be liable if they fail to save their child from drowning in the example set out above. Therefore, the term neglect relates to the care of the child in a general sense.

Assumption of Care for Another

The Children and Young Persons Act 1933 does not extend beyond those of the age of 16 years and therefore no direct obligation is placed on a parent to care for an independent 18-year-old child (R v Shepherd (1862) 9 Cox CC 123). However, two important judgments highlight the approach that is taken in respect of where care is assumed over an individual who is over the age of 16.

Case in Focus: R v Instan [1893] 1 QB 450

In Instan the defendant lived with her aunt who became ill with gangrene. The aunt was unable to feed herself or to seek medical help personally. The defendant did not feed the aunt, nor did she seek any medical help for her, despite remaining living in the aunt’s house and eating the aunt’s food. The defendant was convicted of the aunt’s manslaughter. It was considered that the defendant’s neglect of the aunt had facilitated the aunt’s death.

Whilst the relationship between the aunt and niece differs from parent and child, it could be suggested that the blood relationship contributed to the decision in Instan. However, this limitation was removed in the subsequent judgment in R v Stone and Dodinson.

Case in Focus: R v Stone and Dobinson [1977] QB 354

The two defendants were not married, but lived together as a couple when they took into their house Stone’s sister, Fanny. Fanny suffered from anorexia nervosa, but was initially able to care for herself when she went to live with Stone and Dobinson. However, when her condition worsened and she became unable to care for herself, Stone and Dobinson did not obtain medical care for her and Fanny subsequently died. Both Stone and Dobinson were found guilty of Fanny’s manslaughter.

The reasoning behind this judgment addresses a number of the factors considered above. In the first instance, Stone’s obligation to care for his sister could be considered to be borne out of the blood relationship between them. This approach could not apply to Dobinson. Her liability for Fanny’s death, arose because she had previously taken limited steps to care to Fanny. The facts that these steps were taken, imposed a duty on the defendants to ensure that the steps were ultimately adequate. Failing to meet this requirement rendered the defendant’s liable.

Examination Consideration

It is important to know the facts of these two judgments because they demonstrate clearly the effect of assuming care for another. Think about the distinction between the two and whether liability would have arisen in Instan without the blood relationship. Consider whether it might have been better for Dobinson to have ignored Fanny completely. Would she have been found liable if she had made no attempt to assist her?

Contractual Duties

A person may be found criminally liable if they fail to comply with a contractual duty they have to another. The most commonly cited example of this approach is found in R v Pittwood (1902) 19 TLR 37. In this case, the defendant was employed to operate the gates at a level crossing. He failed to close the gates when a train was approaching and a cart was struck by the train killing one of the carters. Pittwood was convicted of manslaughter. Although the position is described as being based on Pittwood’s breach of contractual duty, the victim was not a party to the contract and therefore, it was a breach of Pittwood’s duty to the crossing users that founded his liability. The contractual element arising out of the fact that this duty arose from his contractual obligation to manage the crossing. A contractual duty to do a particular thing, imposes a duty to do that thing in order to prevent harm to particular individuals. A doctor, for example, has a contractual duty to the hospital where they work, but this duty requires them to act to prevent harm to patients.

Duty Arising out of a Danger of One’s Own Making

Where a person creates a dangerous situation, they may be placed under an obligation to take reasonable steps to remove the danger and therefore, may be criminally liable if they do not do so. In R v Miller [1983] 2 AC 161, the defendant was squatting in a building and fell asleep on a mattress whilst smoking a cigarette. He woke up to find that the mattress was smouldering, but rather than attempting to put the fire out, he simply moved to another room thereby allowing the fire to spread. The defendant’s conviction for arson was not based on the fact that he deliberately started the fire, but rather that once he had started it, he failed to take steps to prevent its spread. It was the failure to act that gave rise to the actus reus of the offence.

In R v Evans [2009] EWCA Crim 650, it was held that the approach in Miller could be taken a stage further where it was held that an individual could be liable in gross negligence manslaughter when they unlawfully supply a victim with a dangerous drug and then fail to obtain help for the victim when it becomes clear that the victim has fallen ill as a result of taking the drug. It is however necessary for it to be shown that the supply of the drugs is a contributory cause of the death (see below).

Medical Treatment

In ordinary circumstances, as was set out above, doctors are required to act in order to provide their patients with proper medical treatment. This requirement is removed where a patient with the required level of capacity specifically refuses treatment. In certain limited circumstances, the court will allow doctors to remove medical treatment where it is considered to be in the patient’s best interests.

In Airedale National Health Service Trust v Bland [1993] AC 789, doctors sought the court’s permission to withdraw feeding from a patient who was in a persistent vegetative state, but who was able to breathe unaided. The court granted permission, but drew a careful distinction between actions that could be considered omissions, such as failing to feed the patient, and those that could be considered positive acts, such as administering a fatal drug. The latter will always give rise to liability, the former may not, in certain circumstances.

Offences where Omission Never Gives Rise to Liability

Offences that specifically require a positive act can clearly not be committed by omission. On this basis, it seems that offences relating to assaults cannot be committed by omission. Neither can theft, burglary or rape, because each requires the defendant to do something in order to satisfy the actus reus. It is not possible to commit rape, without penetrating with a penis. Although it could be suggested that failing to remove a penis if consent is withdrawn could constitute an omission, it is probable that this would be considered a continued act of penetration, in the same way that failing to remove a car parked on a foot is considered a continued act of assault rather than an omission to remove the car (Fagan v Metropolitan Police Commissioner [1969] 1 QB 439).

Summary on Liability for Omission

  • There is no general liability for omission.
  • Liability will arise if:
    • A statutory duty to act exists;
    • A special relationship exists between the parties;
    • The defendant has adopted a caring role towards the victim;
    • A contractual duty to act exists;
    • The defendant has created a dangerous situation.

Examination Consideration

Questions relating to liability for omissions are common. It is necessary therefore to know all of the circumstances where liability may arise and be aware of the links between them. A doctor carrying out an operation may have a contractual duty to act, but may also have assumed care of the patient and may also be under a duty to remove a danger of their own making. A particular approach is often not clear.

Actus Reus Must be Voluntary

A defendant can only be found to satisfy the actus reus of an offence if they act voluntarily. This does not mean that the defendant must have deliberately done a particular thing. In other words, the actus reus of assault will be satisfied if the defendant punches a victim where the only though process is the decision to knock the victim out - it is not necessary for the defendant to actively think ‘I will clench my fist, raise my arm’ etc. However, if a defendant trips, puts out their hand to save themselves and accidentally hits another person, knocking them out, there will be no actus reus. This is because the defendant’s actions were out of their control. The distinction is one between an act that the defendant does and one that happens to the defendant. Only the former can satisfy the actus reus of an offence.


Automatism is distinct from acts that simply happen to a defendant because they can be considered acts that are done by the defendant, but that are out of the defendant’s control. The full definition will be considered in a subsequent chapter in relation to the similar and connected defence of insanity. It is important to note at this stage however, that where automatism is successfully pleaded, the defendant is acquitted because they will not be considered to have satisfied the actus reus of the offence.


In order for a defendant to be found liable for an offence, it is necessary that the defendant’s actions cause the harm. In the example above, a person was considered to satisfy the actus reus of assault if they punch another person. However, what would be the situation if the victim jumped aside to avoid the punch and fell down a steep slope hitting their head and subsequently dying from their injuries? Would the defendant be liable for the death? In other words, could the defendant be considered to have caused the victim’s death? In law it is the causing of the result that is relevant, not necessarily act that lead to it and therefore it is this that causation addresses.

Causation falls into two categories: factual causation and legal causation - both must be satisfied in order for a defendant to be found liable.

Factual Causation

Factual causation is often alternatively known as ‘but for’ causation. The question that is asked is whether ‘but for’ the defendant’s actions, the harm to the victim would have occurred. In R v White [1910] 2 KB 124, the defendant laced his mother’s bedtime drink with potassium cyanide. When the mother was found dead the following morning, the defendant was charged with her murder. It subsequently came to light however, that the defendant’s mother had actually had very little of the drink and that her death was coincidental. The result was that the defendant could not be considered to have caused the death and was only guilty of attempted murder.

Although in White, factual causation was absent, it can be seen that in many circumstances it would easily be found. The victim in the example would still be alive if the punch had not been thrown. Indeed, it is possible that a chain of events could flow from a single act of a defendant. Suppose in the punch example, the victim was collected by ambulance and taken to hospital, but on the way to the hospital, the ambulance crashed into a school bus killing several of the children on the bus. But for the defendant’s punch, the children would still be alive. Could it be considered that the defendant caused the death of the children?

Multiple Causes

A defendant may be guilty of causing something to happen if his conduct was not the only cause of it, even if his conduct alone was not sufficient to cause the harm to occur (R v Warburton [2006] EWCA Crim 627). Although the courts have been a little unwilling to suggest the extent of the legal cause that must be attributable to a particular defendant, it was suggested in R v Hennigan [1971] 3 All ER 133, that a 20% contribution would be sufficient and in R v Notman [1994] Crim LR 518, that anything above a minimal contribution would suffice. Therefore, if two individuals are considered to have caused the harm to a victim in a factual sense, both may be liable.

Legal Causation

It is not difficult to see that factual causation is extremely broad and that a defendant can be found to be the factual cause of a victim’s harm even if the harm is far removed from the defendant’s actions. Legal causation has the effect of limiting this potentially broad liability.

There are three specific elements that must be satisfied in order for legal causation to be satisfied:

  • The cause must be substantial to the extent that it is more than slight or trifling (R v Kimsey ([1996] Crim LR 35));
  • The Defendant must to to some extent blameworthy in respect of the harm (R v Dalloway (1847) 2 Cox CC 273);
  • The defendant’s actions must be operating on the victim at the time that the liability arises (R v Pagett (1983) 76 Cr App R 279).

These elements may not be present and the defendant may avoid liability even if found to have factually caused the victim’s harm in two circumstances. These can be categorised as:

  • some act of another person that intervenes between the defendant’s conduct and the end result;
  • or some event which occurs between the defendant’s conduct and the end result.

These circumstances are often described collectively as a novus actus interveniens.

Novus Actus Interveniens

Acts by Third Parties

If a third party intervenes in the events that result in the harm to the victim, the effect of these actions may be sufficient to break the chain of causation between the defendant and the victim’s harm. However, there are several limitations that arise in this respect.

The first of these is that, where both parties’ actions can be considered collectively to result in the harm suffered by the victim, both potential defendants are considered to have caused the harm. It does not matter that one defendant’s actions are the main cause of the harm and one defendant’s are not. All that is necessary, is that the defendant can be considered to a contributed significantly to the harm (R v Pagett (1983) 76 Cr App R 279). It is important to note that is most areas of the criminal law, and certainly in this circumstance, the word significant simply means more than minimally (R v Cheshire [1991] 3 All ER 670). Therefore, if one defendant’s actions have a more than minimal effect on the victim, they will be found to have caused the victims harm.

Again, the punch example is helpful here. If defendant A throws a punch at the victim and this time connects with the punch and the victim is knocked to the ground. Following this, defendant B repeatedly kicks the victim in the head causing serious injuries from which the victim dies, both defendants will be considered to have caused the victim’s death, even though the actions of defendant A are on the face of it less serious than defendant B.

In practice, the acts of a third party will only break the chain of causation between the defendant and the victim’s harm if they are such as to render the defendant’s actions non-operable on the harm. In other words, if the defendant’s act can still be considered an operating and substantial (more than minimal) cause of the victim’s harm, the chain of causation will not be broken (R v Malcherek and Steel [1981] 2 All ER 422).

The final consideration on this point, is that even if the intervening acts are on the face of it sufficient to break the chain of causation, causation will not be broken unless the actions of the third party are free, deliberate and informed. In R v Pagett, the defendant used his former girlfriend as a human shield against the police whilst shooting at the police. The police returned fire and unfortunately killed the girlfriend. Pagett suggested that it was the actions of the police that had caused the death, not his own actions. It was held however, that the police had little choice but to defend themselves and therefore, their actions were not freely undertaken. The result was that it was Pagett who had caused the death because of his actions, the police shots did not break the chain of causation. What appears to be fundamental in this context, is that an act of a third party will not break the chain of causation if it is an action that is foreseen or foreseeable. In Pagett, the defendant ought to have foreseen the police actions and therefore could not suggest that the chain of causation was broken.

It can be seen that the occasions where the acts of a third party will break a chain of causation are relatively limited. If there is any link between the actions of the defendant and the ultimate harm that results, the chain of causation is unlikely to be broken.

Examination Consideration

Questions on causation are common. You may be required to consider a factual scenario or you may be asked to discuss the relative elements. It is important that you are aware how all of the elements link and that you will need to consider the reasoning in all of the judgments in order to either apply the judgments to a set of facts or consider whether the approach taken is a reasonable one.

Medical Intervention

Where a victim is subject to an assault, it is foreseeable that they may need medical treatment, and it is also foreseeable that this medical treatment may be given wrongly or negligently. This means that it is extremely unlikely that poor medical treatment will have the effect of breaking the chain of causation. Three judgments highlight this point.

In R v Jordan (1956) 40 Cr App R 152, the defendant was admitted to hospital with stab wounds. When the victim died 8 days later, his wounds had largely healed and it was held that it was medical treatment subsequent to the healing of the wounds that caused the death, rather than the wounds themselves. The wounds were not an operating and substantial cause of the death.

In R v Smith [1959] 2 QB 35 on broadly similar facts to Jordan the defendant was unable to rely on the intervening medical acts, even though these were found to be extremely poor, because in this context, the victim’s wounds were still operating and substantial at the time of the death.

The difficulty in reconciling these two judgments and in particular reconciling Jordan with everything that has been discussed before in respect of causation is clear. In Smith it was stated that the chain of causation would only be broken if ‘it can be said that the original wounding is merely a setting in which another cause operates … only if the second cause is so overwhelming as to make the original wound merely part of history can it be said that the death does not flow from the wound’. Whilst it could be suggested that this approach might be valid where the faulty treatment relates to a completely different illness, it seems difficult to suggest that treatment, however overwhelmingly bad, if a direct result of the initial injury, should render the original injury less than substantial. Nevertheless, this is a distinction that must be drawn.

The threshold was narrowed in R v Cheshire [1991] 1 WLR 844, where it was held that the chain of causation will be broken by poor medical treatment if the treatment is ‘… so independent of [the defendant’s] acts, and in itself so potent in causing death, that [the court] regard[s] the contribution made by [the defendant’s] acts as insignificant’.

The effect of medical treatment on causation must be considered very carefully and the remaining effect of the initial wound must be looked at in detail. It is only safe to say that the chain of causation will only be broken where the original injury is insignificant in the face of the poor medical treatment. If any remnant of the original injury remains, it is unlikely that even very poor medical treatment will mean that a defendant is not found to have caused a victim’s harm.

Examination Consideration

The potential for questions in this area is clear. You must be aware of the distinction between Jordan and Smith for practical questions. Also, consider whether you think the Jordan or Smith approach is better. Is it possible for them to be reconciled?

Acts of the Victim

The general position in respect of acts of victims is that an individual should be held responsible for acts that they undertake freely. In R v Kennedy(No 2)[2007] UKHL 38, for example, it was held that a person who provided another will a syringe containing heroin could not be held liable when the victim injected themselves with it. This was because the act of injection was carried out freely and therefore broke the chain of causation from the act of providing the drug. The defendant would have only been liable if the drug was jointly administered, meaning the conduct was not solely the victims.

This approach does not apply where the acts of a victim cannot be considered to have been carried out freely. The most common situation where this principle is applied is where a victim is attempting to escape from a defendant. In R v Williams and Davies [1992] 1 WLR 380, the victim died after jumping from a car in an attempt to escape from the defendants who were trying to rob him. In this judgment it was held that two considerations should be made. In the first instance, the question should be asked as to whether it was reasonably foreseeable that some kind of harm was likely to result to the victim from the actions of the defendant. Secondly, whether the victim’s action, in the face of the defendant’s behaviour, was within a range of responses that might be expected of a victim in that situation.

Unfortunately, even this approach is slightly unclear. This is because in R v Roberts(1971) 56 Cr App R 95, it was held that where a victim’s actions were so daft that no reasonable person could foresee them, the chain of causation would be broken. Whereas, in Williams and Davies it was held that the particular characteristics of the victim should be taken into account, including whether they were of a particularly nervous disposition or whether they may have acted in panic.

The Eggshell Skull Rule

The position mentioned directly above in relation to the potentially nervous disposition of a defendant, reflects the general rule that a defendant must take their victim as they find them. In R v Hayward (1908) 21 Cox CC 692, the defendant was found liable for his wife’s death from a pre-existing heart condition when he threatened her causing her to suffer a heart attack. The approach was taken further in R v Blaue [1975] 1 WLR 1411 where the defendant stabbed a woman who was a Jehovah’s Witness. As a result of her religious beliefs the victim refused a blood transfusion, which would have saved her life. The defendant was, however, found liable for her death on the basis that the religious beliefs of the victim could not be found to interfere with causation.

The only limited exception to the eggshell skull rule is that where a victim dies of fright from heart failure, for example, the jury must be directed as to the dangerousness of the defendant’s behaviour in the absence of the particular characteristic of the victim. In other words, a jury cannot be told of the specific medical condition until it decides that the defendant’s actions are inherently dangerous to the victim (R v Dawson(1985) 81 Cr App R 150).

Acts of God

An act of God is defined as freak natural phenomenon, so unpredictable that it ought to excuse the defendant of all liability (Southern Water Authority v Pegrum [1989] Crim LR 442). Such an act must be completely unpredictable and therefore seasonal weather will not fall within the category unless it is extreme in nature.  Therefore, where a victim is punched and falls into a fast flowing river caused by heavy winter rain, it is unlikely that the weather would be considered an act of God sufficient to break the chain of causation. If the same victim is punched, falls and is then struck by lightning, the lightning will be considered an act of God and the defendant will not be found to have caused the death.

Summary of the Principles of Causation

  • In most cases causation is a simple question of fact.
  • The defendant’s actions do not need to be the sole or main cause of the victim’s harm. The defendant’s acts must simply contribute (more than minimally) to the harm.
  • Where difficulty arises, both factual and legal causation must be satisfied.
  • Factual causation - ‘but for’ the defendant’s actions, would the victim’s harm occurred?
  • Legal causation - was the defendant’s action an operating and substantial cause of the harm? If yes, the defendant caused the harm.
  • Consider whether an intervening act occurred. If so was it a reasonably foreseeable act? If yes, the defendant caused the victim’s harm.
  • Medical treatment will only break the chain of causation if it is so poor as to render the victim’s injuries insignificant.
  • A defendant must take their victim as they find them (eggshell skull rule).

To export a reference to this article please select a referencing style below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.