The defence of necessity

What is it?

A defendant seeks to rely on a necessity-type defence when he argues that he should not be convicted of a criminal offence because his conduct was justified, and it was justified because he did more good than harm in acting as he did.

Defences that are not necessity defences

Given this definition, two defences that are often categorised as being instances of an overarching defence of ‘necessity’ actually having nothing to do with necessity, properly understood.

(1) Duress

The defence of duress – which is available to a defendant who acted under the pressure created by the need to avoid death or serious injury to himself or to someone for whom he is responsible – cannot be analysed as a necessity defence because a defendant seeking to raise a defence of duress is not seeking to justify his conduct. Instead he is seeking to excuse his conduct by arguing, in essence, that ‘I know what I did was wrong, but I acted as I did under pressure that would have led a normal person to act in the same way as I did. So it’s unfair to find me guilty of committing an offence here because had you been in the same position as me, you would have done the same as me. So it’s just an arbitrary accident of fate that had led me rather than you to end up being charged with this offence, and we shouldn’t allow arbitrary accidents of fate to determine people’s criminal guilt.’

(2) Self-defence

The defence of self-defence is often analysed as being a form of necessity defence because: (1) a defendant who seeks to rely on the defence of self-defence is seeking to show that his conduct was justified, just as he would be doing if he were seeking to rely on a defence of necessity; and (2) a defendant who seeks to rely on the defence of self-defence will not be allowed to rely on the defence unless he used reasonable force to defend himself given the facts as he believed them to be, just as a defendant who was seeking to rely on a necessity defence would not be allowed to unless he could show that his actions were reasonable, all things considered.

However, the better analysis seems to be that in a self-defence case, the defendant is arguing that his conduct was justified not because he did more good than harm in using force against his attacker, but because his attacker had no right that the defendant not use force against him. For example, suppose that Anna is being raped by David when Anna manages to reach for a knife and stabs David to death. Most people agree that Anna is entitled to be acquitted of murder here: she will be able to take advantage of a defence of self-defence. But that is not because she did more good than harm in stabbing David to death. Whether or not that was the case is irrelevant. Anna’s conduct here was justified because David had no right that Anna not stab him to death, given what he was doing to Anna. By raping Anna, David forfeited the right he would have normally had against her that she not subject him to lethal force.

(Note that this analysis does not work to explain cases like R v Gladstone Williams (1984) where V was not actually attacking D, but D used force against V because she honestly believed V was a threat to her; in such a case, it is hard to say that V had no right that D not use force against him. Gladstone Williams must be explained on some other basis, such as that D’s conduct can be excused. But a problem with this is that in an excuse case, the defendant has to show that his conduct was reasonable, so as to allow the defendant to argue that ‘You would have done the same as me had you been in my position’ – and the effect of Gladstone Williams is to acquit a defendant who has acted on the unreasonable belief that he or she is being attacked.)

Statutory defences that look like necessity defences

There are some statutory defences that, it could be argued, look like forms of necessity defences:

(1) Damage to property

Under s 5(2) of the Criminal Damage Act 1971, a defendant who is charged with causing criminal damage to another’s property will have a defence if he acted as he did in order to protect other property from being damaged, and he acted reasonably in that belief. The classic example where this defence would apply where a fire is raging down a row of houses; the fire services will be authorised to pull down a house in the middle of the row so as to create a fire break and stop the fire spreading any further. By pulling down the house in the middle of the row, the fire fighters have done more good than harm and are entitled to justify their conduct on that basis.

(2) Refugees

Under s 31 of the Immigration and Asylum Act 1999, a defendant who has been charged with committing various offences in order to enter the UK will have a defence if he came to the UK directly from a country ‘where his life or freedom was threatened’ and promptly presented himself to the authorities and made a claim for asylum. A defendant who prevented himself being killed or imprisoned by entering the UK under a false passport, or by smuggling himself across the UK’s borders, will have done more good than harm and is entitled to justify his conduct on that basis.

(3) Crime stoppers

Section 3(1) of the Criminal Law Act 1967 says that ‘A person may use such force as is reasonable in the circumstances in the prevention of crime, or in effecting or assisting in the lawful arrest of offenders or suspected offenders or of persons unlawfully at large.’ Provisions of the Police and Criminal Evidence Act 1984 confer numerous powers on the police to do things that would otherwise be criminal offences, such as stopping and searching someone (s 1), seizing property (s 19), arresting someone (s 24) and detaining someone in custody (Part IV). Such provisions can most easily be explained on the basis that the sort of conduct covered by these provisions is in the public interest and is therefore justified.

Necessity under the common law

It seems that the courts do not accept that a defendant who has been charged with a criminal offence can ever take advantage of a defence of necessity under the common law. So unless some statute provides the defendant with a necessity-type defence, it will not do the defendant any good to plead, ‘In acting as I did, I did more good than harm.’ Such a plea will fall on deaf ears.

This may seem a controversial claim, given the many cases in which the courts have begun to recognise the existence of a general defence of necessity and set out the situations in which such a defence will be available to a defendant. However, in all those cases (except two) the courts consistently run together the defence of necessity with the defence of duress of circumstances. They do this so consistently that when the courts talk of a defendant being allowed to take advantage of a defence of ‘necessity’ it seems that they are really talking about when a defendant will be allowed to take advantage of a defence of duress of circumstances.

Consider, for example, the following quote from Simon Brown J in R v Martin (1989) (driving while disqualified):

‘First, English law does, in extreme circumstances, recognise a defence of necessity. Most commonly this defence arises as duress, that is pressure upon the accused’s will from the wrongful threats or violence of another. Equally, however, it can arise from other objective dangers threatening the accused or others. Arising thus it is conveniently called “duress of circumstances”.

‘Secondly, the defence is available only if, from an objective standpoint, the accused can be said to be acting reasonably and proportionately in order to avoid a threat of death or serious injury.

‘Thirdly, assuming the defence to be open to the accused on his account of the facts, the issue should be left to the jury, who should be directed to determine these two questions: first, was the accused, or may he have been, impelled to act as he did because as a result of what he reasonably believed to be the situation he had good cause to fear that otherwise death or serious physical injury would result? Second, if so, may a sober person of reasonable firmness, sharing the characteristics of the accused, have responded to that situation by acting as the accused acted? If the answer to both those questions was yes, then the jury would acquit: the defence of necessity would have been established.’

This passage was endorsed in R v Abdul-Hussain (1998) (hijacking of plane) as providing ‘the clearest and most authoritative guide to the relevant principles’ as to when a defence of necessity would be available. (Which dictum was in turn approved in R v Shayler (2001) (breach of Official Secrets Act 1989) and R v S (2012) (child abduction).) But it is clear that Simon Brown J was not talking about necessity at all – in the sense in which I have defined that defence – but was rather talking about when a defendant will be able to take advantage of a defence of duress (either duress by threats or duress of circumstances). As we have already observed, duress is not necessity. As Brooke LJ observed in In Re A (2001) (separation of conjoined twins), ‘In cases of pure necessity the actor’s mind is not [as it is in duress cases] irresistibly overborne by external pressures. The claim is that his or her conduct was not harmful because on a choice of two evils the choice of avoiding the greater harm was justified.’

Despite this, the courts have continued to identify the defence of necessity with the defence of duress of circumstances. In R v Conway (1989) (reckless driving), Woolf LJ held that:

‘necessity can only be a defence to a charge of reckless driving where the facts establish “duress of circumstances,”… i.e. where the defendant was constrained by circumstances to drive as he did to avoid death or serious bodily harm to himself or some other person.  …Whether “duress of circumstances” is called “duress” or “necessity” does not matter. What is important is that, whatever it is called, it is subject to the same limitations as [the] “do this or else” species of duress.’

Twelve years on, in R v Shayler (2001, CA), Lord Woolf CJ was still of the same mind. Dismissing arguments that necessity and duress were distinct in that necessity is a justification and duress is an excuse, he said: ‘…the distinction between duress of circumstances and necessity has, correctly, been by and large ignored or blurred by the courts… the law has tended to treat duress of circumstances and necessity as one and the same’ (at [55]).

The two cases that might genuinely be said to have recognised the existence of a defence at necessity under the common law are Gillick v West Norfolk and Wisbech AHA (1986) and In Re A (2001). Let’s look at each of these in turn.

In Gillick, the issue was whether a doctor who prescribed contraception to a girl who under 16 would be guilty of an offence under s 28 of the Sexual Offences Act 1956 of causing or encouraging unlawful sexual intercourse ‘of a girl under the age of sixteen for whom he is responsible’. (Note that this offence was repealed by the Sexual Offences Act 2003, s 10 of which makes it an offence to cause or incite someone who is under 16 to engage in a sexual activity.) The majority (Lords Fraser, Scarman and Bridge) took the view that a doctor may well commit an offence under s 28 if he prescribed contraception to an under 16 year old girl ‘with the intention of facilitating her having unlawful sexual intercourse’, but a doctor could not be said to have such an intention if he was honestly prescribing the contraception ‘for the maintenance or restoration of [the girl’s] health.’ (Both quotes from Lord Scarman’s judgment.) The minority (Lords Templeman and Brandon) disagreed, holding that if a doctor prescribed contraception to a girl who was under 16, he would be ‘promoting, encouraging or facilitating the having of sexual intercourse’ (per Lord Brandon) and as a result would commit an offence under s 28. None of the judgments in Gillick say anything about necessity: they were purely concerned with the question of how s 28 of the 1956 Act should be interpreted. So there is nothing in Gillick that can be said to support the existence of a defence of necessity under the common law.

That leaves In Re A, in which the Court of Appeal had to decide whether it would be lawful to carry out an operation to separate two conjoined twins, Jodie and Mary, even though such an operation would inevitably result in the death of one of the twins, Mary. There was no possibility of arguing that the defence of duress of circumstances could apply to make the operation lawful: (1) duress cannot be excuse murder (a defendant cannot claim ‘you would have done the same as me’ when what he has done is intentionally kill someone else); (2) as Robert Walker LJ observed ‘The doctors are not faced with any threat to themselves, but they are faced with the anxious dilemma of trying to perform their professional duties which they owe to their two infant patients.’ So In Re A provided an excellent test case for whether the Court of Appeal were willing to recognise a defence of necessity independent of a defence of duress of circumstances.

All three judges in the Court of Appeal held that the doctors would not be committing a crime in operating to separate the conjoined twins.

Ward LJ held that the operation would be justified on the grounds of self-defence, as Mary – who was dependent on being supplied with blood from Jodie’s heart to stay alive, with the result that Jodie’s heart was destined to fail if she remained joined to Mary – posed an ‘unjust’ threat to Jodie’s life.

Brooke LJ held that the operation could be justified on the ground of necessity. He held that ‘there are three necessary requirements for the application of the doctrine of necessity: (i) the act is needed to avoid inevitable and irreparable evil; (ii) no more should be done than is reasonably necessary for the purpose to be achieved; (iii) the evil inflicted must not be disproportionate to the evil avoided.’ He thought that all three conditions were satisfied in this case. It is not, however, clear that Brooke LJ thought that satisfying conditions (i) – (iii) would be sufficient to give rise to a defence of necessity. A few paragraphs before setting out these conditions, he held that there were two objections to recognising a defence of necessity in a case where a defendant was charged with committing a criminal offence: (1) that it was beyond the competence of the courts to assess whether the operation would do more good than harm; (2) that allowing a defence of necessity would result in the court authorising someone to act immorally. Brooke LJ thought that neither objection applied In Re A. On (1), it was obvious whether or not the operation would do more good than harm: dealing with this question did not require the courts to put a comparative value on Jodie and Mary’s lives. If the operation were not carried out, both Jodie and Mary would die. If it were carried out, Jodie would live and Mary would die. Given this, it was obvious what was the better option. On (2), it was not clear that carrying out the operation would be an immoral act, so it was not clear that by allowing the operation to be carried out the court would be ‘marking an absolute divorce of law from morality’.

Robert Walker LJ seemed to take the view that the operation did not even approach the threshold of criminality that would require the doctors to rely on a defence to justify or excuse their actions. He held that the doctors carrying out the operation could not be said to have intended to kill Mary (so they would not have the mens rea for murder), and could be said to be acting in Mary’s best interests in carrying out the operation as ‘The operation would give her, even in death, bodily integrity as a human being’ (so they would not be breaching the duty of care they owed Mary as her doctors in carrying out the operation and could not therefore be held guilty of gross negligence manslaughter).

So – of the three judges who decided In Re A, only one (Brooke LJ) rested his decision on the basis that the doctors could take advantage of a defence of necessity which was genuinely distinct from a defence of duress of circumstances. And Brooke LJ’s judgment remains the only English authority that there exists a defence of necessity at common law that is distinct from a defence of duress of circumstances. Given this, it seems fair enough to say that the English law does not currently recognise that a defendant who has been charged with a criminal offence can ever rely on a defence of necessity to justify his conduct in the absence of some specific statutory provision which allows him to raise a necessity-type defence to the charge.

Why are the courts so unwilling to recognise a free-standing defence of necessity?

There seem to be four reasons for the lack of authority in favour of the proposition that a defendant is entitled to be acquitted of a criminal charge if he can show that he did more good than harm in acting as he did.

(1) The definition of offences

There are two ways in which the definition of an offence can exclude a defendant from seeking to rely on a defence of necessity to justify his conduct.

First, the definition of an offence may make it unnecessary for the defendant to rely on a defence of necessity to justify his conduct, because the facts which the defendant might have wanted to rely on as giving him a defence of necessity will also allow him to argue that he does not commit the actus reus of the offence with which he is charged, or that he did not commit the actus reus of the offence with the requisite mens rea. For example, in R v Abdul-Hussain (1998), Rose LJ observed that ‘if Anne Frank had stolen a car to escape from Amsterdam and been charged with theft, the tenets of English law would not have denied her a defence of duress of circumstances, on the ground that she should have waited for the Gestapo’s knock on the door [before attempting to escape in a stolen car].’ But in such a case Anne Frank would not have needed to rely on a defence to avoid being convicted of theft. She would have been able to argue that in the circumstances, her appropriation of the car was not dishonest – with the result that she did not commit the actus reus or mens rea of theft (depending on how you classify the requirement of dishonesty in theft, as going to the actus reus or the mens rea of theft).

Secondly, where an offence has been created under a statute, it might be impossible for the defendant to rely on a defence of necessity to justify his conduct because it will be clear from the definition of the offence that Parliament did not intend such a defendant to be allowed to escape conviction by pleading that his actions were in the public interest. The most obvious example of this is the Official Secrets Act 1989. The Act does not provide any defence to someone who claims that they were acting in the public interest in making public secrets ‘relating to security or intelligence’ that they came across by virtue of their position as ‘a member of the security and intelligence services’ or as a government official who has been notified that they are bound by the Act. Given this, the courts would be flouting the will of Parliament if they were to allow such a defendant a defence of necessity under the common law; and they so ruled in the case of R v Shayler (2003, HL). For similar reasons, the appeal in R v Quayle (2005) – where the defendant attempted to justify their possession of marijuana on the basis of ‘medical necessity’ – was doomed to fail. The Misuse of Drugs Act 1971 provides no defence to those who possess marijuana for medical purposes. Given the extreme likelihood that this was a deliberate decision on the part of Parliament, it would have been a violation of Parliamentary sovereignty for the courts to hold that the defendants in Quayle were entitled to a defence of necessity. It was different in DPP v Pipe (2012), where the defendant was charged with breaking the speed limit when he drove his son – who had broken his leg playing football – to hospital. Allowing him to rely on a defence of necessity (here, a synonym for ‘duress of circumstances’) did not involve the courts in violating the will of Parliament in setting speed limits for motorists. Parliament – or the ministers exercising powers to set speed limits delegated to them by Parliament – could not be expected to have had circumstances such as those in DPP v Pipe in mind in setting speed limits.

(2) Rights

Most common law offences exist to protect individual’s rights that other people not treat them in certain ways (kill them, rape them, take their property, batter them, and so on). The fact that English law recognises that people have rights reflects the individualistic ethos of English law, under which greater priority is placed on protecting the freedom of the individual than the welfare of the community. Rights give people a veto power over what can be done to them in the name of the public interest. In Ronald Dworkin’s vivid phrase, ‘rights are trumps’.

In contrast, the existence of necessity defences reflect a more communitarian ethos, under which greater priority is placed on promoting the welfare or the community than on protecting the freedom of the individual. If ‘I did more good than harm’ were always a defence to being charged with committing a criminal offence, then people would have no rights. All people would have were interests that would be taken into account in determining what the right thing to do would be. So if V’s interests in not being killed by D were outweighed by the community’s interests in seeing that V was killed by D, then the law would permit V to be killed by D. It would not be possible anymore to say that V had a right not to be killed by D because killing V might be allowed, depending on the circumstances.

Of course, if the balance of interests indicated that V should not be killed, then we could say in that situation that V has a right not to be killed by D. But that statement, that V has a right not to be killed by D, would simply represent the conclusion of a chain of reasoning: ‘It would do more harm than good to kill V, so V should not be killed, so V has a right not to be killed by D.’ The statement that V has a right not to be killed by D would not – as it does in a community which adopts an individualistic ethos – form the premise of a chain of reasoning: ‘V has a right not to be killed by D, so it would be wrong for D to kill V even though killing V would, on balance, do more good than harm.’

It follows that so long as English law adopts an individualistic ethos, it cannot recognise the existence of necessity defences to crimes which involve violating an individual’s rights. Recognising the existence of such defences would be fatal to English law’s individualistic ethos, in the same way that kryptonite is fatal to Supeman: they cannot co-exist in the same space. English law could not at one and the same time recognise that people have rights and recognise that a defendant could justify treating someone in a particular way by arguing ‘I did more good than harm.’ It has to be one or the other; not both. Once we grasp this, we can see that arguments about whether the law should recognise a defence of necessity (in the sense that it has been defined in this essay, a justification based on the plea ‘I did more good than harm’) are really debates about whether the law should be influenced by an individualistic ethos or by a communitarian ethos. And the fashionability of communitarian ideas among academics explains why so many academics are in favour of the criminal law recognising a fully-fledged defence of necessity.

(3) Judgment

Even if English law were to adopt a more communitarian ethos – and many people would say that it has been in the process of doing that since the 1970s – the courts would still have good reason to hesitate over whether or not to allow defendants to rely on a necessity defence, at least in cases where they are not excluded from doing so by the terms of the offence with which they have been charged. In a case where a defendant argued ‘My conduct was justified because I did more good than harm’, it will often be very difficult for the courts to judge whether or not the defendant’s argument is made out. The problem is not really that the courts lack the expertise to weigh the goods and harms that were produced by the defendant’s conduct, but more that no one could weigh those goods and harms because no one can tell what those goods and harms are. For example, in R v Dudley and Stephens (1884), the court held that the defendant seaman – who had abandoned ship in the middle of a storm and put to sea in an open boat with a cabin boy – could not justify killing the cabin boy in order to eat his body on the basis that had they not done so, they would have all starved to death. Lord Coleridge CJ pointed out that even if it were the case (which was not actually certain, as they might have been picked up any day by a passing ship; or they might not have been picked up at all, ever) that the people in the open boat had a choice between ‘One of us dies, and the rest of us live’ and ‘We all die’, that could not justify killing the cabin boy as opposed to any of the other people in the boat. Killing the cabin boy could only be justified if that was the least worst option. But:

‘Who is to be the judge of this sort of necessity? By what measure is the comparative value of lives to be measured ? Is it to be strength, or intellect, or what? …In this case the weakest, the youngest, the most unresisting, was chosen. Was it more necessary to kill him than one of the grown men? The answer must be “No”…’

None of us can really tell what the consequences of our actions will be. Given this, asking the courts to determine whether the defendant’s actions did more good than harm is asking them to do the impossible. Even in a case like In Re A, where it seems clear-cut that going ahead with the operation would do more good than harm, it is not impossible to imagine futures where not carrying out the operation would have been the better option: for example, a future where Jodie grew up to become a mass murderer, or a future where Jodie became so depressed at owing her life to the death of her twin that she was never able to function normally or enjoy life. And it is not impossible to imagine twists on those futures that would allow us to say that, in fact, carrying out the operation was actually still the better option, where the tragedies resulting from Jodie’s existence had outcomes that worked to offset the initial suffering involved in those tragedies.

(4) Disorder

The final reason for being cautious about recognising the existence of necessity defences, even in a legal system that is fully committed to adopting a communitarian ethos, is the potential for disorder that the existence of such defences creates. The criminal law will speak with a more uncertain voice, and provide less of a constraint on people’s actions, if people know that their conduct will not be attract a criminal sanction if they can establish that they did more good than harm in acting as they did. This consideration played some part in the decision of the court in R v Dudley and Stephens to deny a necessity defence to the defendants in that case. Lord Coleridge CJ warned of the ‘awful danger’ involved in allowing that ‘the temptation to murder’ could give rise to a defence to a charge of murder:

‘such a principle once admitted might be made the legal cloak for unbridled passion and atrocious crime. There is no safe path for judges to tread but to ascertain the law to the best of their ability and to declare it according to their judgment… [A judge] has no right to declare temptation to be an excuse, though he might himself have yielded to it, nor allow compassion for the criminal to change or weaken in any manner the legal definition of the crime.’

Almost a hundred years later, in Southwark London Borough Council v Williams (1971), Lord Denning MR advanced similar reasons for refusing to find that squatting in abandoned premises could be justified on the grounds of necessity:

‘The doctrine [that private property may be sacrificed to avoid ‘great and imminent danger’] must…be carefully circumscribed. Else necessity would open the door to many an excuse. It was for this reason that it was not admitted in Reg. v. Dudley and Stephens (1884) 14 Q.B.D. 273, where the three shipwrecked sailors, in extreme despair, killed the cabin boy and ate him to save their own lives. They were held guilty of murder. The killing was not justified by necessity. Similarly, when a man, who is starving, enters a house and takes food in order to keep himself alive. Our English law does not admit the defence of necessity. It holds him guilty of larceny. Lord Hale said that “if a person, being under necessity for want of victuals, or clothes, shall upon that account clandestinely, and animo furandi, steal another man’s food, it is felony…”: Hale, Pleas of Crown, i. 54. The reason is because, if hunger were once allowed to be an excuse for stealing, it would open a way through which all kinds of disorder and lawlessness would pass. So here. If homelessness were once admitted as a defence to trespass, no one’s house could be safe. Necessity would open a door which no man could shut. It would not only be those in extreme need who would enter. There would be others who would imagine that they were in need, or would invent a need, so as to gain entry. Each man would say his need was greater than the next man’s. The plea would be an excuse for all sorts of wrongdoing. So the courts must, for the sake of law and order, take a firm stand. They must refuse to admit the plea of necessity to the hungry and the homeless: and trust that their distress will be relieved by the charitable and the good.’

Conclusion

It seems that the current state of law on when a defence of necessity (properly understood as a justification resting on the argument ‘I did more good than harm’) will be available is justified. If necessity-type defences are to be recognised under the law, it would be better if they were created by Parliament, rather than the courts. Giving effect to statutory necessity defences would pose no threat to the doctrine of Parliamentary sovereignty. Moreover, Parliament is in a better position than the courts to judge when acting in a particular way can be justified on the basis that acting in that way did more good than harm, and to set out what factors are to be taken into account in reaching that conclusion. Further, the statutory necessity-type defences are less likely to be abused or misused as their scope should be clearly delineated and understood in advance; something that is not possible when the courts undertake to sketch out on a case-by-case basis when they will permit a defendant to rely on a necessity defence.

Introduction to criminal law

Introduction

The criminal law performs two primary functions. The first – what I will call the ‘core’ function – is to protect our rights by punishing those who deliberately violate them. The criminal law performs this core function when it punishes murderers, rapists and thieves. All these categories of people deliberately violate other people’s rights.  The second function  – what I will call the ‘secondary’ function – is to deter people from acting in ways that are contrary to the public interest. The criminal law performs this secondary function when it punishes polluters, drug dealers and public officials who act corruptly. None of these categories of people can be said to violate other people’s rights; but they do act in ways that are contrary to the public interest.

The language that I’ve used to describe these two functions of the criminal law might be taken to suggest that the vast majority of criminal offences are concerned with protecting our rights, and that these offences are orbited by a small number of criminal offences that are concerned to deter people from acting in an anti-social way. In fact, the opposite is true. It has been estimated that there are about 8,000 different criminal offences that may be committed in England and Wales. (That number is rapidly increasing all the time – more than 700 different criminal offences were created between 1997 and 2003.) No more than 100 of those offences are concerned with protecting our rights. I’ll call these offences ‘rights-based’ offences. All of the other criminal offences recognised in England and Wales exist to deter people from acting in ways that are contrary to the public interest. I’ll call these offences ‘public interest-based’ offences. Despite this disparity in the number of rights-based offences and public interest-based offences, almost all of the offences that a law student will study when he or she studies ‘Criminal Law’ will be rights-based in nature. He or she will spend virtually no time looking at offences that exist to deter people from acting in ways that are contrary to the public interest.

In order to establish that a defendant has committed a rights-based offence, it has to be shown that the defendant has violated someone else’s rights, and that he did so deliberately. In the language customarily used by criminal lawyers, it has to be shown that the defendant committed an actus reus (a wrongful act) with a mens rea (a wrongful mind). In contrast, in order to establish that a defendant has committed a public interest-based offence, it will always be necessary to show that the defendant acted in a way that was contrary to the public interest, but it will almost never be necessary to show that the defendant acted deliberately. For example, suppose the law said that a company that has polluted a river will only be guilty of an offence if it polluted it deliberately. Such a rule would not do anything to encourage a company to take safety precautions to ensure it did not pollute a nearby river. It would know that if it failed to take such precautions with the result that it polluted the river, it would face no criminal punishment as the pollution was not deliberate. What if the law said that a company that has polluted a river carelessly will be guilty of an offence? Such a rule would almost certainly encourage our company to take some precautions to ensure that it did not pollute the river, but it would probably do no more than the bare minimum to see that it did not pollute the river. It would bet that doing no more than the bare minimum would be enough to escape being successfully prosecuted for carelessly polluting the river. The only rule which would encourage the company to do as much as it could to avoid polluting the river would be one which said that a company that has polluted a river will be guilty of an offence if it could possibly have done something to avoid polluting the river. Under this strict liability rule, the company would know that unless it could prove that it had done everything possible to avoid polluting the river, every time it polluted the river, it would face a fine. It’s for this reason that most public interest-based offences do not require it to be shown that the defendant acted with any kind of mens rea. Instead, most public interest-based offences are strict liability in nature, holding the defendant liable unless it can be shown that the defendant did everything possible to avoid doing what the defendant did.

The meaning of intention

The most serious criminal offence recognised in English law is, of course, murder. This is a rights-based offence, protecting a person’s right that others not unjustly deprive them of their life. A will commit the offence of murder if:

(i) A’s actions cause another to die; and

(ii) when A performed those actions he had an intent to kill; and

(iii) A had no lawful justification or excuse for acting as he did.

(Note that this definition is not strictly accurate: to convict A of murder it will be enough to show that when A acted as he did, he had an intent to cause someone to suffer really serious bodily harm. However, to keep things simple this point about the law will be disregarded in the discussion below.)

So, for example, suppose that I attend a political rally carrying a gun. I intend to use the gun to kill the main speaker at the rally. A policewoman spots that I have a gun. She sidles up to me and attempts to take the gun away from me. In the ensuing struggle, the gun goes off accidentally and the policewoman is killed. Am I guilty of murder? Yes I am. Running through each of the elements that have to be made out before I can be found guilty of murder:

(i) My bringing a gun to the rally caused the policewoman to die – had I not brought the gun to the rally, she would not have died.

(ii) When I took the gun to the rally, I intended to kill the main speaker at the rally. (Note that it does not have to be shown that I intended to kill the very person who died as a result of my actions.)

(iii) I had no lawful justification or excuse for bringing the gun to the rally.

Now, let’s consider a different case. I need to raise a lot of money very urgently. I own a passenger plane that is insured for £20m. I plant a bomb on the plane just before it is due to take off. My idea is that when the bomb goes off, the plane will blow up and I will be able to collect the insurance on the plane. The bomb explodes 30 minutes after the plane took off, and all the passengers and crew on the plane are killed. Am I guilty of murder? There is no doubt that my action in planting a bomb on the plane has caused the deaths of the passengers and the crew, and there is equally no doubt that I had no justification or excuse for doing what I did. The real difficulty is in establishing that I had the mens rea for murder. When I planted the bomb, can I say that I intended to kill someone?

In principle, the answer is ‘No’. I intend to bring about a certain outcome if it is my aim or purpose to bring about that outcome. When I planted the bomb, my aim or purpose was not to kill anyone. Suppose that in the case we are considering, by some miracle no one was killed or seriously injured as a result of the plane blowing up. In such a case, would I have snapped my fingers and thought, ‘Confound it! My plans didn’t work out! The plane was blown up but no one was killed! That’s not what I wanted!’ Of course I wouldn’t have. I probably would have thought, ‘Thank goodness no one was killed. Now I can collect the insurance money on my plane with a clear conscience.’ So it was not my aim or purpose to kill anyone when I planted the bomb. It follows that, in principle, I should be acquitted of murder. When I planted the bomb I did not have an intent to kill.

However, this would not be regarded by most people as a satisfactory result. In this case, I have exhibited a callous disregard for the passengers’ and crew’s rights not to be unjustly deprived of their lives. As a result, most people would regard me as being just as bad as someone who acts with a clear intention to kill, such as an assassin. If I am just as bad as an assassin, people will think that I should be convicted of murder, just like the assassin. To achieve this result, the courts have twisted the definition of when someone can be said to have had an intent to kill. The courts say that a defendant will have had an intent to kill if: (1) he or she acted with the aim or purpose of killing someone; or (2) when he or she acted, he or she knew that it was virtually certain that someone would be killed as a result of his or her actions. Adopting this definition allows me to be convicted of murder in the case we are considering. When I planted the bomb on the plane I knew it was virtually certain that the passengers and crew on the plane would be killed as a result – so under the courts’ definition of intention I can be said to have had an intent to kill when I planted the bomb on the plane. And if I can be said to have had an intent to kill, then I am guilty of murder.

A good result? In this particular case, yes. However, while it is very common for the courts to twist legal rules – redefining their terms – in order to achieve the results that they want to achieve in particular cases, manipulating legal rules in this way is rarely a good idea in the long run. Consider whether the offence of murder has been committed in the following situation. You and I are mountaineering. We are attached by a rope. You are below me on the mountain. You lose your footing on the mountain and slip into a deep crevasse. I manage to keep my place on the mountain, but we are now stuck. The only thing stopping you from falling to your death is the rope that attaches you to me. I don’t have the strength to haul you out of the crevasse, and you cannot get any kind of footing that would allow you to climb out of the crevasse. Night is approaching and we will both freeze on the mountain if we remain where we are. In the end I decide to cut the rope that attaches you to me: it’s better that one of us (that is, you) should die than that we should both die. I cut the rope and you fall to your death. Am I guilty of murder? My cutting the rope undoubtedly caused your death. But when I cut the rope, did I intend to kill you? If we define ‘intention’ in the way it is normally defined, I did not intend to kill you. When I cut the rope it was not my aim or purpose to kill you. Had you by some miracle survived your fall, I would have been delighted. I would not have snapped my fingers and thought, ‘Confound it! I thought I’d finished you off! Now I’m going have to find some other way of getting rid of you!’ However, according to the courts’ definition of intention, I did have an intent to kill in this situation: while my aim or purpose in cutting the rope may not have been to kill you, I did know at the time I cut the rope that it was virtually certain that you would die as a result.

Would it be a good result to find me guilty of murder? Of course not. I don’t exhibit anything like the callous disregard for people’s rights to life that is shown by either the assassin or by the plane bomber in the case we were considering above. But if my actions caused your death, and I had an intent to kill you when I performed those actions, then I will be guilty of murder unless the courts recognise that I had a lawful justification or excuse for my actions. So – what sort of justification or excuse can I rely on in this situation? The most obvious one is a defence of necessity: ‘The circumstances meant that I did the right thing when I cut the rope. If I had not cut the rope, both of us would have died. By cutting the rope, I ensured that at least one of us (that is, me) survived.’

So – problem solved. We can recognise a defence of necessity in this case, and thereby ensure that I am not convicted of murder. But if we do that, what does that imply for the following case? Bob is wheeled into Dr Strange’s hospital with a rare blood condition that threatens to kill him in the next two hours. The only cure is to give Bob a complete blood transfusion, replacing all of his blood with new blood. Unfortunately, there are no blood bags available of the same blood type as Bob’s. Another patient, Eric, is in Dr Strange’s hospital. He suffers from cancer. His cancer is incurable and it is predicted he probably only have six more weeks to live. Eric’s blood type is the same as Bob’s. Dr Strange orders that Eric be sedated, and all his blood drained from his body and given to Bob. Eric dies as a result of this treatment. Is Dr Strange guilty of murder? He has the actus reus and mens rea of murder – but can he take advantage of a defence of necessity here to establish that he had a lawful justification or excuse for doing what he did? If we allow a defence of necessity in the mountaineering case set out above, it is hard to see why we would not allow one here. Why could Dr Strange not argue, ‘The circumstances meant I did the right thing when I drained Eric’s blood from his body. If I had not done that, both Bob (immediately) and Eric (eventually) would have died. By draining Eric’s blood from his body, I ensured that at least one of them (that is, Bob) survived.’

I think most people would be unhappy if Dr Strange were allowed to escape a murder charge in this situation by relying on a defence of necessity. But it would seem that we have to allow him such a defence if we allow such a defence to be raised in the mountaineering case. Of course, we only have to allow such a defence to be raised in the mountaineering case because the courts have ruled that a defendant will be held to have had an intent to kill if he acted as he did knowing that death was a virtually certain consequence of his actions. If we got rid of that rule and returned to the idea that you can only be said to intend the consequences of your actions if it was your aim or purpose to produce those consequences, then we could find that there was no murder in the mountaineering case because there was no intent to kill. However, if we find there was no intent to kill in the mountaineering case, then we would have to find there was no intent to kill and therefore no murder in the plane bomber case and in Dr Strange’s case as well. (Dr Strange could argue that it was not his aim or purpose to kill Eric when he drained Eric of his blood – had Eric by some miracle survived without his blood, he would have been delighted.)

We seem to be stuck. If we adopt a strict line on when someone can be said to have had an intent to kill, some defendants who deserve to be convicted of murder will be acquitted. If we relax the definition of when someone can be said to have had an intent to kill – and say that you will have an intent to kill if you act knowing that death is a virtually certain consequence of your actions – then you will bring within the scope of the law of murder people who do not deserve to be convicted of murder. The only way of saving them from being convicted of murder will be to provide them with defences to being charged with murder; which defences may then be taken advantage of by people who do deserve to be convicted of murder.

Is there any way out of this mess? One way would be to redefine when someone is guilty of murder and say that someone will commit the crime of murder if he acts in a way that causes another’s death and at the time he acted he exhibited a callous disregard for other people’s rights not to be unjustly deprived of their lives. Under this rule, the plane bomber and Dr Strange would be convicted of murder. Both exhibited a callous disregard for other people’s rights to life in acting as they did. In contrast, I would be acquitted of murder for cutting the rope in the mountaineering case set out above. In that case, I did not exhibit a callous disregard for your right to life in acting as I did.

Morality and the criminal law

The above discussion shows what an important role is played by considerations of morality in developing the criminal law. The perception that a particular person is just as bad as someone who is clearly guilty of murder leads the courts to reshape the rules on when someone can be found to have an intent to kill. And when those rules operate in such a way as to threaten that a defendant who does not deserve to be convicted of murder will be convicted, the courts respond to that threat by reshaping the law on when someone can be said to have had a lawful justification or excuse for acting as he did, and provide that someone with a defence.

Should considerations of morality play an even greater role in the development of the criminal law? Should the criminal law take on a third function – that of encouraging people to act morally, by punishing them for acting immorally? The Victorian philosopher John Stuart Mill thought that it should not. In his book On Liberty (published in 1859), he argued:

‘[T]he only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinion of others, to do so would be wise or even right…’

The modern consensus is that the criminal law should observe Mill’s ‘harm principle’, so that conduct should only be criminalised if it violates other people’s rights or if it harms the public interest. The fact that acting in a particular way is immoral does not give us sufficient reason to make it a criminal offence to act in that way. Having said that, two fairly recent cases may be interpreted as breaking with this consensus.

R v Brown

In R v Brown [1994] 1 AC 212, a group of sado-masochists were charged with committing offences under ss 20 and 47 of the Offences Against the Person Act 1861. (Someone will commit an offence under s 20 if he maliciously wounds or inflicts grievous bodily harm on another. Someone will commit an offence under s 47 if he commits an assault on another person that occasions actual bodily harm.) The defendants deliberately inflicted pain on each other by doing such things as driving nails into each other’s bodies; burning each other with candles; and cutting each other with knives. Each of the defendants fully consented to having these things done to them.

The House of Lords had to decide whether the defendants had committed offences under the 1861 Act in acting as they had. By a majority of three to two, the House of Lords decided that they had. Some of the judgments of the majority give the impression that they thought the defendants should be punished because they had acted immorally. Lord Templeman remarked, ‘Pleasure derived from the infliction of pain is an evil thing. Cruelty is uncivilised.’ Lord Lowry argued that no exemption from the 1861 Act should be made for people who ‘wish to satisfy a perverted and depraved sexual desire.’ In contrast, Lord Mustill argued in dissent that the defendants should not be convicted. He based himself on Mill’s harm principle: ‘[Questions as to whether the activities of the members of the group were right or wrong] are questions of private morality;…the standards by which they fall to be judged are not those of the criminal law;…if these standards are to be upheld the individual must enforce them upon himself…or have them enforced against him by moral pressures exerted by whatever…community to whose ethical ideals he responds.’

A couple of arguments might be made in favour of the view that convicting the defendants in Brown did not violate Mill’s harm principle. Neither of them are convincing.

(1) Violation of rights. At first sight, it seems that the activities in Brown did not violate the rights of those who were the object of those activities. Rights can almost always be waived. For example, I have a right that you not smash up my car. But if I invite you to smash my car and you do so, you will not have violated my rights. I will have waived my right that you not smash up my car.

In a similar way, it can be argued that each of the sado-masochists in Brown waived their rights not to be physically harmed by the other sado-masochists; so whenever a particular defendant physically harmed a fellow sado-masochist, he did not violate that sado-masochist’s rights. However, this argument overlooks the fact that some rights are inalienable. They cannot be waived. It is arguable that one’s right not to be treated in a cruel and inhumane manner is one such right. If this is correct, then it can be argued that every time one of the sado-masochists in Brown was injured by a fellow sado-masochist, his rights were violated. So punishing the defendants in Brown fulfilled the criminal law’s core function – that of protecting our rights by punishing those who deliberately violate them.

The weakness in this argument is that it is not clear why one’s right not to be treated in a cruel and inhumane manner should be inalienable. Nor is it clear why the criminal law should go out of its way to protect rights that the right-holders are plainly not interested in having.

(2) The public interest. The majority in Brown placed a good deal of stress on the harm that might be done to the public interest if activities such as those that went on in Brown were allowed to continue. Lord Jauncey of Tullichettle pointed out, ‘it would appear to be good luck rather than good judgment that…prevented serious injury occurring [in this case]. Wounds can easily become septic if not properly treated, the free flow of blood from a person who is HIV positive or who has Aids can infect another…’ Underlying this statement is a concern that if someone was seriously injured as a result of engaging in sado-masochistic activities, the public would have to pick up the tab for treating his injuries on the National Health Service (NHS). It follows that engaging in sado-masochistic activities is contrary to the public interest, and in punishing those who engage in such activities, the criminal law is fulfilling its secondary function – that of deterring people from acting in ways that are contrary to the public interest.

If this argument in favour of the decision in Brown were correct, then many other activities could justifiably be criminalised. As Lord Mustill pointed out in Brown, the same argument could be used to justify criminalising homosexual relations among men, as this is the principal cause of the spread of HIV and Aids in the UK. It could also be used to justify making criminals of people who spend their spare time skateboarding or mountain-climbing or smoking cigarettes in private. In effect, if this argument in favour of criminalising the activities in Brown were accepted, the existence of the NHS in the UK would make it justifiable to deprive people of the freedom to decide what to do with their own bodies. The socialisation of heath care in the UK would have the effect of justifying the socialisation of people’s bodies. This cannot be right. The fact that the State has created a national health care system cannot be invoked by the State as a reason for taking away from people liberties that they would have enjoyed in the absence of such a health care system.

So it seems that the House of Lords’ decision in Brown did violate Mill’s harm principle. In Brown, the criminal law was used to force people to act morally – no other explanation of the case makes sense.

R v Hinks

In R v Hinks [2001] 2 AC 241, a 38-year-old called Karen Hinks befriended a 53-year-old man called John Dolphin. Dolphin was a naïve, trusting man of limited intelligence. Hinks became his principal carer. Dolphin had substantial savings and money inherited from his father in a building society account. From April to November 1996, Hinks and Dolphin went to Dolphin’s building society virtually every day. Dolphin would withdraw £300 each time, and then give the money to Hinks. By the end of the six months, Dolphin had given Hinks approximately £60,000.

Hinks was convicted of theft and the House of Lords upheld her conviction, by a three to two majority. The majority found that all the elements of the offence of theft were present in Hinks’ case. A defendant will commit the offence of theft if he dishonestly appropriates property belonging to another with an intention permanently to deprive that other of that property. It was found that when Hinks accepted Dolphin’s money, she had appropriated property belonging to Dolphin. It was further found that Hinks had acted dishonestly in accepting Dolphin’s money and that when she accepted that money, she had an intent permanently to deprive Dolphin of that money.

On the facts of Hinks, the decision to convict Hinks of theft was unobjectionable. The law grants people of limited intelligence a right that other people not take advantage of their weakness, and will set aside any gifts or transactions that are the product of a violation of that right on the ground that the gifts or transactions are ‘unconscionable’. So Dolphin had a right that Hinks not exploit his limited intelligence, and Hinks violated that right when she prevailed upon him to give her £60,000. So punishing Hinks for theft did not breach Mill’s harm principle: Hinks was straightforwardly punished for deliberately violating Dolphin’s rights.

However, the real danger of the decision in Hinks is that it opens the door to people being convicted for theft whenever they dishonestly accept a gift or act dishonestly in buying something. For example, consider the following two cases:

(1) Freddie is in love with Linda. Linda does not care for Freddie, but she does not make her true feelings clear to Freddie because he lavishes so much attention and, more importantly, so many gifts on her. On Linda’s 21st birthday, Freddie gives Linda a diamond necklace, worth £20,000.

(2) Howard walks into Bill’s antique shop where a painting is on sale for £50. Howard recognises that the painting is by Rembrandt, and is in fact worth £2m. Howard gives Bill £50 for the painting on the spot and walks out of the shop with it.

If we find that Linda’s and Howard’s conduct in these situations was ‘dishonest’ then the decision in Hinks suggests that each of them will be guilty of theft. However, neither of them has violated anyone’s rights in acting as they did. The law does not confer on Freddie a right that Linda not take advantage of his infatuation for her. The law says that in affairs of the heart, Freddie has to look after himself. If Linda makes a fool of him, then that is his look-out. Similarly, the law does not confer a right on Bill that Howard not take advantage of Bill’s ignorance of the true value of his goods to snap up a bargain for himself. If Bill ends up selling a painting to Howard for far less than it is actually worth, then that is his look-out.

So if Linda and Howard are convicted of theft in the above situations, even though they did not violate anyone’s rights, the law will in effect punish them simply for acting ‘dishonestly’. In so doing, the law will violate Mill’s harm principle, under which the mere fact that someone’s conduct is immoral (or, here, ‘dishonest’) is not enough to justify criminalising it.

Evaluating Mill’s harm principle

Of course, the fact that John Stuart Mill thought that the criminal law should not be used to force people to act morally does not establish that the criminal law should not be used in this way. Why shouldn’t the criminal law be used to force people to act morally? Three reasons can be given.

First, it is not clear what the point is of forcing someone to act morally. Let’s assume that it is immoral for Wendy to cheat on her husband by sleeping with Ethan. Threatening to punish Wendy if she sleeps with Ethan may deter her from cheating on her husband, but forcing Wendy not to cheat on her husband will not make her into a better or wiser person. Instead, it will simply make her feel cowed and resentful.

Secondly, in today’s society there is no consensus on what sort of behaviour is moral and what sort of behaviour is immoral, particularly in the areas of sexual activity and drug taking. There is not even a consensus on how to resolve disagreements over what sort of conduct is moral or immoral. Given this, the idea of using the criminal law to force people to act morally is particularly problematic. Whose view of what is moral and immoral should govern? The Church’s? The State’s?

Thirdly, criminalising behaviour that is commonly believed to be immoral may have the effect of entrenching mistaken moral beliefs. For example, in parts of the United States it was at one time thought that it was immoral for a black man and a white woman to go out with each other. Had such behaviour been criminalised in the United States, it would have been very hard to challenge this thoroughly evil belief. Indeed, the lack of examples of black men and white women going out with each other that would have been produced by criminalising such behaviour would have encouraged people to think that there was indeed something wrong, or unnatural, in black men and white women going out with each other.

In light of these arguments, it seems that Mill was right. The criminal law should not play a role in forcing people to act morally, but should instead confine itself to performing the two functions identified at the start of this chapter – that of punishing people who deliberately violate other people’s rights, and that of deterring people from acting in ways that are contrary to the public interest.

The mens rea for murder

What is it?

A defendant who has caused another’s death will have had the mens rea for murder if, at the time he acted (or at the time he wrongfully failed to prevent that other’s death), he had an intent to kill or an intent to cause someone to suffer grievous bodily harm.

Is this area of the law in need of reform?

Many commentators criticise the rule that a defendant can be found to have had the mens rea of murder if he merely had an intent to cause someone to suffer grievous bodily harm. I want to make a more fundamental criticism: that even if we got rid of what is called the ‘grievous bodily harm rule’, and said that someone will have the mens rea of murder if, and only if, he had an intent to kill, the law of murder will still be in an unsatisfactory state. Simply asking whether the defendant had an intent to kill will invariably result in our classifying as murderers some people who should not be classified as murderers, and our classifying as non-murderers some people should be classified as murderers – and this is so whatever meaning we give to the phrase ‘intent to kill’. In other words, a definition of the mens rea of murder that is centred round the concept that you are a murderer if you acted with an intent to kill will result in our drawing the line between murderers and non-murderers in the wrong place. To draw it in the right place, we need a different definition of the mens rea of murder.

The multiple meanings of ‘intent’

I’ve claimed that whatever meaning we give to the phrase ‘intent to kill’, focussing on whether the defendant had an intent to kill will result in our drawing the dividing line between who is a murderer and who is not a murderer in the wrong place. I’ll now make out this claim by considering some different definitions of when we can say that a defendant had an ‘intent to kill’.

(1) Aim or purpose

This is the most straightforward definition of what it means to have an intent to kill – the defendant acted with an intent to kill if killing was one of his aims or purposes in acting as he did. We can test whether someone had the aim or purpose to produce effect x in acting as they did by asking what their reaction would have been had their actions not produced effect x. If their reaction would have been one of indifference, then they did not act with the aim or purpose of producing effect x. If their reaction would have been one of frustration, then they acted with the aim or purpose of producing effect x.

So if D puts poison in V’s glass of water, we can determine whether or not D is acting with the aim or purpose of killing V by considering what D’s reaction would be if V did not die after drinking from the glass. If he wouldn’t mind, then killing V is not his aim or purpose – maybe he is simply putting the poison in V’s glass in order to make her sick. (And we can test for that by considering what D’s reaction would be if V’s was completely unharmed by drinking from the glass.) If D would be very frustrated if V did not die after drinking from the glass, then it’s clear he is acting with the aim or purpose of killing her in putting the poison in her glass.

Now – asking, in a case where a defendant has caused another’s death, ‘Did the defendant act with an intent to kill, in the sense of acting with the aim or purpose of killing?’ will usually produce the right result. The Assassin who shoots a politician whom he has been hired to kill will be rightly found guilty of murder – he acted with the aim or purpose of killing. So will the Incompetent Assassin who attempts to shoot the politician whom he has been hired to kill, and who misses and hits and kills a bystander instead. The Drunk Driver who unthinkingly runs down and kills someone crossing the road in front of his car will rightly not be found guilty of murder because he did not act (either in getting into the car drunk, or in running down the person crossing the road in front of his car) with the aim or purpose of killing.

However, in some cases defining intent in terms of ‘aim or purpose’, will produce some false negatives – cases where we will find that a defendant is not guilty of murder when we would really want to find that the defendant is guilty of murder. For example, the Plane Bomber who blows up a plane that he owns in mid-flight, killing everyone on board, in order to collect on the insurance that he has on the plane does not act with the aim or purpose of killing: if everyone aboard the plane survived the explosion, Plane Bomber would have been happy or indifferent, not frustrated. His aims or purposes in acting as he did were: (1) to destroy the plane; and (2) collecting the insurance on the plane. Destroying lives was not part of the plan; it was just a side-effect of the plan. So if we define ‘intent to kill’ in terms of acting with the aim or purpose of killing, we will end up saying that Plane Bomber did not have an intent to kill, and acquit him of murder. But I think most people would really want to find him guilty of murder.

Consider also the case of Transplant Surgeon, who takes the heart and lungs out of Patient One who is dying of kidney failure and transplants them into Patient Two whose heart and lungs have failed. One dies and Transplant Surgeon is charged with murder. If we define having an ‘intent to kill’ in terms of acting with the aim or purpose of killing, then Transplant Surgeon should be acquitted. He could legitimately say that he did not act as he did with the aim or purpose of killing One – had One survived having his heart and lungs taken out, then Transplant Surgeon would have been quite happy (and amazed). But I think most people would want to see Transplant Surgeon convicted of murder.

(2) Aim or purpose OR acting believing that death is the virtually certain consequence of your actions

In order to cover cases like Plane Bomber and Transplant Surgeon it might be suggested that the definition of when a defendant should be held to have acted with an ‘intent to kill’ should be adjusted so that a defendant will be held to have acted with an ‘intent to kill’ if: (i) he acted with the aim or purpose of killing; or (ii) when he acted as he did he believed that it was virtually certain that someone would die as a result of his actions. On one reading of the House of Lords’ decision in R v Woollin (1998) (which reading was adopted by the Court of Appeal in the conjoined twins case, In re A (2001)) this is the position now taken by the courts on when a defendant will be held to have had an intent to kill.

(Note two common errors that students tend to slip into as a result of the decision of the House of Lords in Woollin. First, they think that it always has to be shown that a defendant believed that death was virtually certain to result from his actions for it to be shown that the defendant had an intent to kill. This is not true. If the defendant acted with the aim or purpose of killing then he had an intent to kill, no matter how improbable he thought it was at the time that he would succeed in killing. So, for example, if I aim a gun at you when you are half a mile away, and pull the trigger with the aim or purpose of killing you, the fact that I thought it was very unlikely I would actually hit you should not stop us finding that I had an intent to kill when I pulled the trigger. Secondly, Lord Steyn’s endorsement in Woollin of Lord Lane CJ’s direction in Nedrick (1986) that a ‘jury should be directed that they are not entitled to infer the necessary intention, unless they feel sure that death or serious bodily harm was a virtual certainty (barring some unforeseen intervention) as a result of the defendant’s actions and the defendant appreciated that such was the case’ (emphasis added) leads some people to think that if, objectively, death was not virtually certain to result from the defendant’s actions, the defendant cannot be found to have had an intent to kill (unless he acted with the aim or purpose of killing). This cannot be true. When we are asking whether someone had an intent to kill, we are focussing on their mental state. The objective facts on the ground as to whether or not death was virtually certain to result from the defendant’s actions cannot be relevant to the issue of what the defendant’s mental state was in acting as he did. So suppose a Plane Bomber – who wants to collect the insurance on a plane he owns – places a bomb on the plane that due to its faulty construction is only 50% likely to go off. The Plane Bomber does not know that his bomb is faulty and thinks that the bomb is virtually certain to explode when the plane is in mid-air, with the result that everyone aboard will be killed. We should not allow the fact that the bomb was not virtually certain to go off to stop us finding – in the event that the bomb does go off and everyone is killed and the Plane Bomber is charged with murder – that the Plane Bomber had an intent to kill when he put the bomb on the plane.)

This re-definition of when someone will be held to have an ‘intent to kill’ seems to solve the problems created by our first definition of when someone will be held to have an ‘intent to kill’. We can find that Plane Bomber and Transplant Surgeon had an intent to kill because when they acted as they did, they believed that death was virtually certain to result from their actions. However, while our re-definition of the meaning of intent enables us to convict Plane Bomber and Transplant Surgeon of murder, it does so at the cost of bringing within the scope of the law of murder other people who most people would not want to see convicted of murder.

For example, consider the example of the Mountaineer who finds himself trapped on the side of the mountain because his climbing companion, to whom Mountaineer is attached by a rope, has slid into a seemingly bottomless crevasse. The deadweight of his companion prevents Mountaineer from moving, and Mountaineer’s companion is unable to climb up and out of the crevasse. Night is falling and Mountaineer and his companion will freeze to death if they stay where they are. Mountaineer decides to cut the rope and his companion falls to his death. Most people would not want to find Mountaineer guilty of murder here, but our redefinition of the concept of having an ‘intent to kill’ indicates that we must find that Mountaineer had an intent to kill when he cut the rope – when he cut it, he believed that death was virtually certain to result from his actions.

And consider the case of the Doctor who gives a cancer patient who is in great pain a dose of morphine that is precisely calibrated to relieve the patient’s pain. Doctor is aware that giving the patient a dose of that size will inevitably hasten the patient’s death, but killing the patient is not part of his aim or purpose. If the patient survived the dose, he would be happy, not frustrated. The patient dies soon after being given the dose of morphine. Most people would not want to find Doctor guilty of murder here, but on our redefinition of when someone can be said to have acted with an intent to kill, we must find that Doctor acted with an intent to kill when he gave his patient the dose of morphine as he believed when he injected the morphine that doing so was virtually to result in the patient’s death being accelerated.

It might be argued that we can deal with the problems created by Mountaineer and Doctor being brought within the law of murder by our redefinition of when a defendant will be held to have had an intent to kill by giving defendants like Mountaineer and Doctor defences to being prosecuted for murder. However, this proposal is fraught with difficulties.

First of all, there are serious civil liberties objections to defining murder in such a way that people who we all know should not be found guilty of murder will be held to have committed the crime of murder unless they can raise a defence. Secondly, it is not clear that any defences we constructed or invented to cover cases like Mountaineer or Doctor can be kept under control and prevented from providing other, more sinister, defendants with defences to being charged with murder. For example, if we say that Mountaineer has a defence because if he had not acted as he did, then both he and his companion would inevitably have died, then Transplant Surgeon might try to take advantage of the same defence: ‘Had I not acted as I did, then both my patients would have died – as it is, I enabled one to live by sacrificing the other.’ And if we say that Doctor has a defence because he acted with the object of relieving pain, then Euthanaser – who gives his patient a very high dose of morphine in order to kill him, thereby bringing his patient’s suffering to an end – might be able to take advantage of the same defence when he is charged with murder.

Considerations such as these indicate that our redefinition of what it means to have an intent to kill does not really work to distinguish satisfactorily between cases where we want to find someone guilty of murder and cases where we do not.

(3) An intent to kill is established if you acted with the aim or purpose of killing, and may be inferred if when you acted as you did, you believed that death was virtually certain to result from your actions

This is the alternative reading of the House of Lords’ ruling in Woollin, on when a defendant should be held to have acted with an intent to kill. On this definition, the fact that a defendant believed that death was virtually certain to result from his actions does not mean that he did have an intent to kill – it is only evidence from which it may be inferred that he did have an intent to kill.

The fact that someone believed that death was virtually certain to result from their actions can often be useful evidence that they had an intent to kill, in the sense that they acted with the aim or purpose of killing. For example, in Hyam v DPP (1975) the defendant had an affair with a Mr Jones for a number of years. He then broke off their relationship because he was getting engaged to a Mrs Booth. Booth had three children. One night, the defendant went to Booth’s house (having first made sure that Jones was at home and not with Booth), poured half a gallon of petrol through the letter box, and then set the petrol alight by stuffing a lighted newspaper through the letter box. If we freeze the story right there and ask, Was the defendant’s aim or purpose in doing all this to kill someone? I think most people’s reaction would be to say ‘I don’t know – she could have been doing it to kill Booth, but she could also have been doing it to scare Booth away, or to punish Booth – by frightening her – for stealing her man.’ But if we add the fact (which was not actually established in the case) that when the defendant acted as she did, she believed that death was virtually certain to result from her actions, then I think most people would conclude from that that the defendant acted as she did in order to kill. The alternative explanation of why the defendant did what she did – that she was trying to frighten Booth, either to scare her away or to punish her – would become completely implausible. If the defendant foresaw that death was virtually certain to result from what she did, she would have chosen some other means of frightening Booth, if that’s all she was trying to do.

However, in other cases it would be wrong to conclude from the fact that the defendant foresaw that death was virtually certain to result from his actions that the defendant acted with the aim or purpose of killing. Other evidence will make it clear that that was not what the defendant was trying to do, even though he foresaw that death was virtually certain to result from his actions. For example, if we saw Mountaineer hacking away at the rope that is the only thing that is keeping Mountaineer’s companion from falling to his death, we might initially think that Mountaineer is trying to kill his companion. The fact that Mountaineer must foresee that death is the inevitable result of his actions would count towards our thinking that he is trying to kill his companion in acting as he does. But everything else in the situation counts against us reaching such a conclusion. The fact that the two climbers are trapped on the mountain, that Mountaineer has hung on for as long as he has, hoping for some help to arrive, the fact that Mountaineer and his companion are tearfully saying goodbye to each other – all these things count towards our concluding that, actually, Mountaineer is not trying to kill his companion but is hoping against hope that something will happen to save his companion’s life (while realising that that is a desperate hope).

So on this third definition of when we will find that someone had an intent to kill, we will not find that defendants like Mountaineer or Doctor had an intent to kill. While the fact that they foresaw that death was virtually certain to result from their actions may allow us to infer that they had an intent to kill, all the other facts of their case will lead us in the end to conclude that they were not acting with an intent to kill. But the trouble is that on this third definition, we will not find that Plane Bomber or Transplant Surgeon had an intent to kill, either. If we are honest, we will say that while those defendants did foresee that death was virtually certain to result from their actions, all the other facts of their cases indicate that they were not trying to kill anyone when they acted as they did – so we will not infer that they acted with an intent to kill, despite the fact that they believed that death was virtually certain to result from their actions. But that is only if we are honest. This third definition of when someone will be held to have an intent to kill does create the space or ‘wiggle room’ for a judge or a jury to be dishonest – to decline (honestly) to infer that there was an intent to kill in cases like Mountaineer or Doctor but to infer (dishonestly) that there was an intent to kill in cases like Plane Bomber or Transplant Surgeon in order to find them guilty of murder. So justice can be done under this third definition – we can find those who should be guilty of murder, guilty of murder; and those who should not be found guilty of murder, not guilty – but only under the cover of darkness created by giving judges and juries the scope to choose whether or not they will infer, or not infer, that a defendant had an intent to kill because the defendant foresaw that death was virtually certain to result from their actions. This is not satisfactory – not least because we cannot guarantee that judges and juries will make the right choices, particularly when they are given no guidance as to when they should infer from the defendant’s foreseeing that death was virtually certain to result from his actions that the defendant acted with an intent to kill.

(4) Aim or purpose OR acting with the aim or purpose of producing an effect, the inescapable consequence of which is that someone will die

Andrew Simester’s article ‘Moral certainty and the boundaries of intention’ (1996) 16 Oxford Journal of Legal Studies 445 might be taken as suggesting a fourth definition of when someone will be held to have acted with an intent to kill, which is set out above.

In the case of OBG Ltd v Allan (2008) – a case that has absolutely nothing to do with the law of murder – the House of Lords might be taken to have agreed with this definition. Lord Nicholls said (at [167]):

‘Take a case where a defendant seeks to advance his own business by pursuing a course of conduct which he knows will, in the very nature of things, necessarily be injurious to the claimant. In other words, a case where loss to the claimant is the obverse side of the coin from gain to the defendant. The defendant’s gain and the claimant’s loss are, to the defendant’s knowledge, inseparably linked. The defendant cannot obtain the one without bringing about the other. If the defendant goes ahead in such a case in order to obtain the gain he seeks [the defendant will be held to have intended to harm the claimant].’

Lord Hoffmann used the same imagery (at [134]): if the claimant’s loss was ‘the other side of the same coin’ as the defendant’s gain, then an intention by the defendant to make a gain for himself would also amount to an intention to cause the claimant to suffer loss.

Does this idea do a better job than our first three definitions of drawing a satisfactory dividing line between those who will be found guilty of murder and those who will not be found guilty of murder? The idea has most promise when applied to Transplant Surgeon: assisting Two by giving her One’s heart and lungs simply cannot be done without killing One. The very reason why Two will be assisted by being given One’s heart and lungs is the reason why One will die once his heart and lungs are taken away – the reason is that having a functioning heart and lungs is essential for life. So we can say in Transplant Surgeon that One’s death was an inescapable consequence of Two’s being given One’s heart and lungs, and that Transplant Surgeon not only intended to keep Two alive by giving her One’s heart and lungs, but also intended to kill One in the process. One’s death was the ‘other side of the same coin’ as Two’s being kept alive.

However, once we extended the idea of an ‘inescapable consequence’ to Plane Bomber and say that Plane Bomber intended to kill the passengers aboard the plane when it blew up because killing those passengers was an ‘inescapable consequence’ of the plane being blown up, then it is very hard to distinguish the idea of something being an ‘inescapable consequence’ of an event occurring from it being ‘virtually certain’ to result from an event occurring. In other words, if we say that there was an intent to kill in Plane Bomber because death was an ‘inescapable consequence’ of what Plane Bomber was trying to do, it is hard to see why we should not also say that Mountaineer and Doctor also had an intent to kill – thus reviving all the problems that afflicted our second definition of when someone can be said to have had an intent to kill.

But if we don’t extend the idea of an ‘inescapable consequence’ to Plane Bomber, then on our fourth definition of what it means to have an intent to kill, we will find that Plane Bomber had no intent to kill and acquit him of murder – which is not a consequence that most people would find palatable.

Alternatives

It seems, then, that if we say that someone will have the mens rea of murder if and only if they had an intent to kill, there is no way of making sense of the concept of having an ‘intent to kill’ that will not result in unacceptable consequences for the scope of the law of murder. Given this, we should start thinking about whether the mens rea of murder should be defined exclusively in terms of whether or not someone had an intent to kill.

In his article ‘The mental element in the crime of murder’ (1988) 104 Law Quarterly Review 30, Lord Goff suggested that a defendant should be held to have had the mens rea of murder if: (1) he acted with an intent to kill, in the sense of acting with the aim or purpose of killing; or (2) he acted with wicked recklessness, in the sense that he did not care whether or not the deceased lived or died. New York criminal law adopts a similar rule. Under section 125.25 of their Penal Code, a defendant will be held to have committed murder in the second degree if: (1) the defendant acted with an intent to kill without any extenuating circumstances; or (2) under ‘circumstances evincing a depraved indifference to human life, he recklessly engage[d] in conduct which create[d] a grave risk of death to another person’. (Murder ‘in the first degree’ under section 125.27 is reserved for cop killings, witness killings, killings for hire, serial killing, and so on.)

Either of these definitions would do a better job of drawing a dividing line between who is a murderer and who is not than a definition of the mens rea of murder which was focussed on whether or not the defendant acted with an intent to kill. Assassin and Incompetent Assassin and Euthanaser would all be found guilty of murder because they each acted with an intent to kill in the sense that they acted with the aim or purpose of killing. Plane Bomber and Transplant Surgeon would not be held to have acted with an intent to kill but would still be held to have had the mens rea of murder because they acted with wicked recklessness/depraved indifference to human life. Mountaineer and Doctor would not be held to have had the mens rea of murder because they did not have an intent to kill – they did not act with the aim or purpose of killing anyone – and they did not act with wicked recklessness/depraved indifference to human life.

Moreover, adopting a definition of the mens rea of murder along the lines proposed above would allow us to drop the ‘grievous bodily harm rule’. Instead, where a defendant acted with the aim or purpose of causing another grievous bodily harm, we would ask whether they acted with wicked recklessness/depraved indifference to human life in acting as they did. If they did (as would be the case where D intended to beat V to within an inch of his life with a baseball bat to punish him for not paying a gambling debt), they would be found guilty of murder. If they did not (as would be the case where D intended to break one of V’s legs to punish him for not paying a gambling debt, and complications set in and V died as a result), they would not be. This might be thought to be more satisfactory than the current situation, where all those who cause death while acting with an intent to cause grievous bodily harm are held to have had the mens rea for murder.

Partial defences to murder

What are they?

A defendant who committed the actus reus of murder with the mens rea will only be found guilty of voluntary manslaughter if he can raise a partial defence of loss of control or diminished responsibility. (There are other partial defences such as infanticide, or acting pursuant to a suicide pact, but I am not concerned with those here.) The purpose of this note is to suggest that there is no need for these partial defences to exist.

The mandatory life sentence

Before 1965, it made a huge difference whether or not you were convicted of murder or voluntary manslaughter: if you were convicted of murder, you could be executed. Nowadays, the difference it makes is that someone who is convicted of murder is punished with a ‘mandatory’ sentence of ‘life imprisonment’; someone who is merely convicted of voluntary manslaughter is given a sentence that fits the circumstances of this case.

But what does it actually mean for a murderer to be given a ‘mandatory’ sentence of ‘life imprisonment’? What actually happens to him or her? To answer these questions, we need to look at the Criminal Justice Act 2003, which sets out the statutory framework for sentencing someone who has been found guilty of murder.

Under the 2003 Act, the judge sentences the prisoner to life imprisonment but also specifies a minimum term that the prisoner must serve before the prisoner will be eligible for parole. In setting this minimum term, the judge must first take into account the prisoner’s age. If the prisoner is an adult who is 21 years old or over, the judge must first select a starting point for determining the minimum term he or she will spend in prison. There are four possible starting points:

(1) The prisoner will never be released on parole (a ‘whole life order’) – this is only appropriate if the murder was extremely serious, which will be the case if: (i) the prisoner murdered two or more people and the murders involved a high degree of premeditation, abduction of the victims, or sexual or sadistic conduct; (ii) the prisoner murdered a child, and the murder involved the abduction of the child, or was motivated by sexual or sadistic reasons; (iii) the murder was carried out for political, religious or ideological reasons; (iv) the prisoner had already been convicted of murder previously.

(2) The prisoner will serve a minimum of 30 years in prison – this is the appropriate starting point in cases such as the murder of a police officer, murders for gain, murders intended to obstruct the course of justice, murders involving sexual or sadistic conduct, ‘hate’ murders where the race or religion or sexual orientation of the victim played a part in their being killed.

(3) The prisoner will serve a minimum of 25 years in prison – this will be the appropriate starting point where the prisoner committed the murder with a knife or other weapon that he took to the scene of the crime intending to commit an offence, or to have it available as a weapon.

(4) The prisoner will serve a minimum of 15 years in prison – this will be the appropriate starting point where the prisoner committed a murder that does not fall under (1), (2) or (3), above.

Having decided what his or her starting point will be, the judge must consider any aggravating or mitigating factors that might make it inappropriate to make the prisoner serve the ‘starting point’ minimum sentence. Mitigating factors include: (i) the prisoner only intended to cause gbh and not to kill; (ii) the murder was not premeditated; (iii) the prisoner was acting under the influence of a mental condition; (iv) the prisoner was provoked to act as he or she did.

Under s 28 of the Crime (Sentences) Act 1997, once a murderer has served his or her minimum sentence, his or her case must be referred to the Parole Board and – if the Parole Board declines to direct his or her release – every two years subsequently. If the Parole Board is satisfied that ‘it is no longer necessary for the protection of the public that the prisoner should be confined’ (s 28(6)) then they will direct that the prisoner be released, and he or she must then be released. But the prisoner is only released ‘on licence’ and his or her status as being released ‘on licence’ continues until he or she dies (s 31(1)). The prisoner’s licence to be released may be revoked at any time on the recommendation of the Parole Board (s 32(1)). And one of the conditions of the prisoner’s being released on licence is that the prisoner place him or herself under the supervision of a probation officer.

What if the partial defences didn’t exist?

If the partial defences didn’t exist, a defendant who intentionally killed another in circumstances currently covered by the defences of loss of control (Coroners and Justice Act 2009, ss 54-55) or diminished responsibility (Homicide Act 1957, s 2, as amended) would be found guilty of murder. This would make the following practical differences to the defendant:

(1) The defendant would incur the stigma of being convicted of murder rather than manslaughter.

(2) The defendant would be given a mandatory sentence of life imprisonment, as opposed to being given a sentence appropriate to the circumstances of his or her case. However, the defendant would probably do no more time in prison as a result of being given a mandatory sentence of life imprisonment than he or she would have if he or she were convicted of manslaughter. This is because the judge, in fixing the minimum term the defendant would have to spend in prison, would almost certainly have a 15 years minimum sentence as his or her starting point, and then could reduce that minimum sentence down to take account of the element of mitigation in the defendant’s case. Having served his or her minimum sentence, in a loss of control case, the defendant would almost certainly be released as his or her further detention would no longer be ‘necessary for the protection of the public’. In a diminished responsibility case, the position might be more complicated depending on the defendant’s mental condition.

(3) Once released, the defendant would come under a lifelong obligation to report to a probation officer, and would always live in fear of being recalled to prison on the recommendation of the Parole Board.

These are the differences that the existence of the partial defences currently makes to the position of a defendant who is accused of murder but can take advantage of a partial defence. Given this, the question is – Should it make any difference to us if the partial defences were abolished?

In the case of someone who has killed in circumstances covered by the defence of loss of control, the abolition of the defence will almost certainly not make any difference to how long the defendant spends in prison. So the defence exists solely to save the defendant from the effects (1) and (3) that being convicted of murder would have on him or her. It is not clear that we should want to save someone who has killed someone because a ‘qualifying trigger’ caused them to lose control of themselves from effects (1) and (3). Having chosen to kill someone else, it might be argued that they deserve the stigma of being labelled a ‘murderer’; and having shown themselves capable of ‘losing it’ so badly that they end up intentionally killing someone else, it might be argued that they should be monitored once they leave prison.

For the same reasons, it could also be argued that we should not want to save someone whose responsibility for intentionally killing someone else was ‘diminished’ by their ‘suffering from an abnormality of mental functioning’ from effects (1) and (3) that being convicted of murder would have on them. Moreover, it could be argued that giving such a defendant a mandatory sentence of life imprisonment would have the positive advantage of allowing the Parole Board to evaluate the defendant’s mental condition once his or her minimum term is up, and allow the Parole Board to decline to order the defendant’s release if his or her mental condition is such that there is a real danger that the defendant might kill again if released.

At any rate, the above discussion should make it clear that justifying the existence of the partial defences simply on the basis that they are needed to mitigate the rigours of the mandatory life sentence is overly simplistic. In fact, it is hard to avoid the impression that the fuss that the criminal establishment makes over the partial defences may be attributable to the fact that they have fooled themselves into thinking that a mandatory sentence of life imprisonment means more than it does.