Feature Archives
Offsite Links



<< return to front page


Although the Ministry of Justice has spent the last two years reviewing homicide law, there still remains a serious problem around “one-punch killers” which means punishment for bad luck, according to

Nearly three years ago the government set up a review of the homicide law following a recommendation by the Law Commission in its report on Partial Defences to Murder (Law Com No. 290, 2004, London: TSO, at para. 1.12). The vast majority of the debate since then has focussed on the more serious cases – how murder should be defined and what partial defences which would reduce the offence to manslaughter should be recognised by the law, though the Law Commission did acknowledge that there should be “a limited review of the law on involuntary manslaughter”. The Coroners and Justice Bill currently being debated in the House of Lords would amend the law on provocation and (to a lesser extent) diminished responsibility, but consideration of offences at the bottom end of the spectrum is deafening by its silence. The government has said nothing on the subject since its brief paper in 2000, (Reforming the Law on Involuntary Manslaughter: The Government’s Proposals, London: TSO) and it is understood there are no plans to review any other parts of the homicide law.

The problem of the one-punch killer

One of the criticisms the writer and others have of the current law is that it makes it too easy for a person to be convicted of manslaughter, essentially because the law pitches the minimum level of culpability too low. A typical example of this is what's often called a one-punch manslaughter. A punches B once in the face. B stumbles over a crease in the carpet on which he was standing, hits his head against the wall, fracturing his skull and he dies. A is guilty of manslaughter by an unlawful and dangerous act. Such cases are not unduly rare – the writer has just completed a survey of 127 manslaughter cases since 1995 and there were at least two cases where the only violence consisted of a single punch, (as well as a small group of other cases where the fatal assault was a punch to the head or face). Whilst there's no doubt that A should not have punched B in the face, B's death was extremely unlikely - no-one would expect B to suffer anything more than a bruise or slight cut. No serious injury is foreseeable, let alone death. The element of luck is so great that many criminal lawyers are unhappy about holding A criminally liable and punishing him for killing B. A does not deserve to be treated as a manslayer.

In favour of the present position, it is said that we cannot entirely control the outcome of our actions and thus must recognise the possibility that something unexpected might occur. The fact that A threw a punch means he effectively crossed a moral threshold and cannot complain when he is held criminally liable for the consequences. Moreover, whilst we would not normally expect a single punch to result in the victim’s death, we are subconsciously aware that the unexpected does occasionally happen. On the other hand, the main argument against the existing law is that although a moral threshold was crossed, there is still the problem of the significant gap between what was foreseen or foreseeable (namely some relatively minor injury), and what the puncher is held liable for (namely causing death). Furthermore, there is the quite separate argument against the current law, that the offence of involuntary manslaughter is being stretched too far. At the upper end of the spectrum there are those who only just avoid a murder conviction – cases such as Hyam, where the defendant set fire to a house knowing that death or serious harm to any occupants was highly probable, would now be treated as examples of reckless manslaughter. Is it really justifiable to treat one-punch killers as guilty of precisely the same offence? The reckless manslayer displays a far higher level of moral blame for killing his victim. Of course, the sentence for the one-punch killer will be more lenient, but that misses the point: he should not be formally and publically labelled in the same way.

Obviously, the precise circumstances of a homicide will vary from one case to another, and some activities are inherently much more dangerous than others. Compare, for example, driving – which is invariably a potentially fatal activity – with punching someone in the face. Even one-punch cases will differ from one another in this respect. The precise degree of risk – the degree of dangerousness – will fluctuate from one instance to the next. It may be possible for experts to provide an arithmetical calculation of the likelihood that in a given set of circumstances someone would die, but it would be profoundly unsatisfactory to identify a particular statistical probability as representing the boundary between what was and was not legally acceptable. Any such figure would be wholly arbitrary and thus indefensible. In deciding whether a defendant should be criminally liable for killing another, we should look at the nature of the risk created by what the defendant did and the victim’s death. Only if the gap between the two is sufficiently close might liability be merited.

Being What should be the bottom line for manslaughter?

In deciding how close it should be, one of our most eminent criminal lawyers, Professor Andrew Ashworth (Oxford University), recently argued that the defendant should, as a minimum, be aware of the risk of causing serious harm, albeit not necessarily fatal harm, (see “’Manslaughter’: Generic or Nominate Offences?”, in C.M.V.Clarkson and Sally Cunningham (eds) Criminal Liability for Non-aggressive Death (2008) Aldershot: Ashgate, chapter 11). That would create a symmetry between murder and manslaughter – murder only requires an intent to cause (non-fatal) serious injury, and it seems that will still be true after the government has completed its review of the homicide law.



Professor Ashworth’s suggestion would also recognise that we do not have complete control over the outcomes of our actions, that there is almost inevitably an element of luck as to the precise consequences of what we do, as well as bringing a sense of proportion between the risk and the ultimate outcome for which the defendant is held liable. But the gap between what was foreseen or foreseeable and the victim’s death would obviously be reduced. Clearly, many one-punch killers would not foresee a risk of serious harm, some might. Consider a very sober defendant and very inebriated victim. The defendant throws a single forceful punch at the victim who is overtly unsteady on his feet and off balance. The victim falls heavily backwards, cracking his head against a brick wall, fracturing his skull, and he dies shortly afterwards. In those circumstances it would not be reasonable for a jury to infer that the defendant knew that the victim might be seriously injured. The chances of causing serious harm might not be that high, but the defendant might be aware, at least subconsciously, of the risk nonetheless.

The Irish Solution

In a relatively recent report, the Irish Law Reform Commission shared the writer’s reluctance to treat all one-punch killers as manslayers and suggested a new crime be created for these cases called "assault causing death"; (see Irish Law Reform Commission, Homicide: Murder and Involuntary Manslaughter, [ LRC 87 – 2008] at para 5.39). Whilst accepting that the puncher should not have thrown the punch in the first place, the Commission was concerned that ordinary people would not want to put him in the same category as other, more deserving, manslayers. In this respect the Commission were supported by a small-scale study the writer conducted in this country about ten years ago; (see “Further Evidence of the Relationship Between Legal and Public Opinion on the Law of Homicide” [2000] Criminal Law Review 814 at 819, 820). At the same time the Commission thought that people would want the fact that someone had been killed to be reflected in the offence. Substituting a conviction for assault causing death would obviously separate out the one-punch killers from more serious cases, and structuring the new offence in this way would show much more accurately and precisely what the defendant had done – he assaulted his victim and as a result the victim died. The label “manslaughter”, on the other hand, wraps up the criminality in a single word, merging the defendant’s act with the ultimate consequence and that fails to reflect the true limitations of the defendant’s moral culpability.
A final suggestion

In the writer’s view this proposal is a step in the right direction, but it is unnecessarily and unjustifiably narrow. Some academics have argued that it is right to convict one-punch killers of manslaughter because the punch is of the same family of wrongs as causing death, namely they are both crimes against the person. But whilst that is true, its significance is purely descriptive: there is no significance in the similarity of a punch and causing death to the defendant’s moral culpability. The justification for holding the defendant liable for killing the victim is that he should not have thrown the punch because of the risks it created. A defensible argument for holding the one-punch killer criminally liable for killing must surely be founded on moral blameworthiness, and not simply on an (albeit understandably) emotional reaction to the fact that someone has been unlawfully killed. Throwing a punch creates a risk – a risk which is likely to cause relatively minor injury but one which, as we know, might in exceptional cases, result in much more serious harm. But the point is that such a risk might arise from other forms of seemingly minor wrongdoing; it is not confined to minor assaults against the person. Thus, an act of criminal damage such as throwing a stone at a window might create the same kind and level of risk – recall the two teenage boys who pushed a paving stone off the parapet of a bridge which struck a train as it passed beneath and killed the guard (DPP –v- Newbury and Jones [1977] AC 500). There is no material distinction between throwing stones at windows (or trains) which result in death and throwing a punch at a person, with the same outcome.

If therefore, the prevailing wisdom is that it is right to convict the one-punch killer of some sort of homicide offence (rather than just the common assault which the single punch would normally constitute), then it should be of committing “an unlawful and dangerous act causing death”. The adjective “dangerous” here would mean that the defendant’s act created a risk of causing serious, though not necessarily fatal, injury. Conviction should also depend on the defendant being aware of the risk when he throws the punch. The gap between what was foreseen and the harm for which the defendant is held criminally liable would be much less than the law currently allows, but the moral basis and thus the rationale for this offence would be






Search WWW Search The Barrister