The last pieces of legislation

One of the last pieces of legislation to slip through its parliamentary stages before the start of the new parliamentary year was the Coroners and Justice Act. This omnibus Act traverses a wide range of issues with which criminal practitioners and others will need to familiarise themselves: from coroners' work, to measures for the protection of vulnerable witnesses, further measures against child pornography and the creation of a Sentencing Council, data protection measures and new laws to prevent convicted criminals from profiting from their crimes through the publication of their memoirs. The full list of topics touched on by this gargantuan Act is somewhat longer!

Amongst the sections in Pt 2 of the Act amending the substantive criminal law are those reforming the partial defences to murder. The gestation of these provisions has been lengthy - but the offspring to which the Act gives birth bears only passing resemblance to what its original progenitors, the Law Commission, wished it to be.

The reform project began when the partial defences to murder were referred by the then Home Secretary to the Law Commission in June 2003. The Commission was asked to examine and make recommendations relating to the law and practice of provocation and diminished responsibility, with particular reference to their impact in cases of domestic violence; and to consider whether there should be a partial defence to murder for cases of excessive self-defence. The Commission pointed out at that stage that considering the partial defences without also addressing the offence of murder itself was less than satisfactory. And the Government responded to the Commission's report (Partial Defences to Murder, Law Com No.290) by referring a wider review of homicide in its entirety to the Commission. But the Law Commission's hands were tied in one crucial respect: the mandatory life sentence was sacrosanct - any recommendations would have to proceed on the basis that it would remain the penalty for murder. And so there followed in November 2006 the Law Commission Report Murder, Manslaughter and Infanticide (Law Com No.304) which recommended wholesale reform of the law of homicide, reshaping the offences into a more coherent, graduated “ ladder” of offences which better reflected different degrees of culpability than do the current offences of murder and the various forms of involuntary manslaughter. Amongst those recommendations was the key reform: that murder should be divided into first and second degree murder, the former - to which the mandatory life sentence would uniquely apply - being reserved for those who intended to kill and those who intended to cause serious injury provided they were also aware of a serious risk of causing death. Those who intended only to cause serious injury unaware of any such risk would only be guilty of second degree murder, to which full sentencing discretion would apply. Meanwhile, reformed partial defences of provocation and diminished responsibility would mitigate the liability of those prima facie guilty of first degree murder by way of a conviction instead for second degree murder and its discretionary sentencing. Further recommendations were also made in relation to duress as a defence to murder, complicity in homicide cases, infanticide; and a government-led public consultation on mercy killing was urged.

The Government's response to these recommendations came in the new Ministry of Justice's July 2008 consultation paper Murder, manslaughter and infanticide: proposals for reform of the law. The Government adopted the title of the Law Commission's report, but its paper notoriously “ cherry-picked” (and in crucial respects redrafted) certain aspects of the Commission's work, notably the partial defences, declining to move forward with the wider structural changes to the underlying homicide offences themselves. The consultation paper promised - and the Government during debates on the present Act continued to promise - that these proposals - and this Act - were “ step one” of what, following a promised evaluation of the first set of reforms, would be a staged reform process. It remains to be seen, not least with an election pending, whether and when “ step two” will take place. But it is worth noting Opposition criticism of the Government's technique of squeezing these reforms of murder into a larger Bill, thereby depriving “ huge and important topics” of the “ full parliamentary scrutiny” that they would have acquired had they been presented in a Bill devoted to reform of murder (Lord Hunt of Wirral, October 26, 2009, HL Deb col.1036).

The Government's reasons for this staged approach are, with respect, difficult to understand, and scarcely rebut the widely held view that one cannot sensibly reform the partial defences without at the same time considering the wider framework in which they sit. During the Public Bill Committee debates earlier this year (March 3, 2009, PBC Deb col.413) Maria Eagle, Parliamentary Under-Secretary of State, remarked that the Law Commission's proposals “ did not command total agreement between all stakeholders and users of the system, whom we must get on board to ensure that the system works and is credible” . If the Government's measure of acceptable law reform is that it must command “ total agreement” from all “ stakeholders” (whomever they might be), then the partial defences provisions in the Bill should have been withdrawn by the Government some time ago. During the passage of the Bill through the Lords in the summer, the partial defence sections were described by former Law Lord, Lord Lloyd - in refreshingly scathing language - as being “ all over the place” , “ beyond redemption” and “ a dog's breakfast” , a description in turn regarded by Baroness Mallalieu as “ a kindness” .

Lord Lloyd moved amendments designed to do away with the partial defence clauses entirely and to replace them with Professor Spencer's proposal for an “ extenuating cir cumstances” rider that could be added by the jury to a murder conviction in order to allow the judge full sentencing discretion (see Archbold News, issue 7, “ Lifting the Life Sentence” ; June 30, 2009, HL Deb col.150 onwards and October 26, 2009, HL Deb col.1008 onwards). The “ extenuating circumstances” amendment was met with very wide approval in the Lords amongst the distinguished array of expert participants in the debates, including criminal barristers, a past Lord Chief Justice, a former police commissioner, a former Home Officer minister, a psychiatrist, a bishop and many others. But the amendment failed when pressed to a vote at Third Reading (by which time the Opposition had decided not to support it). However, the widespread support expressed for the amendment, together with public opinion evidence collected for the Law Commission indicating that the public is not as wedded to the monolithic application of the mandatory life sentence for murder as the Government apparently assumes it to be, should give pause to politicians fearful (as Lord Lloyd suggests) of headlines in the popular press about going “ soft on crime” . We should credit the public (if not parts of the media) with a rather more nuanced view of the law and sentencing of murder than that. This article now examines the homicide-related sections of the Act, focusing in particular on the new partial defence of loss of control (the replacement for provocation) and the reformed diminished responsibility defence. One particular concern of the Government has been to facilitate battered spouses' use of the defences, and that issue will be examined throughout. Practitioners should note that these sections will not come into force until a commencement date yet to be appointed, and so the pre-Act law prevails until then.

Section 56 abolishes the common law defence of provocation, s.3 of the Homicide Act will cease to have effect, and both are replaced with the new partial defence of loss of control, drafted in rather labyrinthine fashion across ss.54 and 55. The new defence consists of three components, broadly identified in s.54(1): D's conduct; the trigger to which D reacted in losing control; and an objective test. The burden of proof rests with the prosecution, the jury being required to assume that the defence is satisfied wherever the judge finds sufficient evidence adduced to raise the issue, i.e. evidence on which a properly directed jury could reasonably conclude that the defence might apply (s.54(5),(6)).

The law of provocation required that D suffered a “ sudden and temporary loss of self-control” (Duffy [1949] 1 All E.R. 932 n.). While delay might indicate a recovery of control leading to a revenge killing, it no longer precluded reliance on the defence as a matter of law (Ahluwalia (1993) 96 Cr.App.R. 133). This development in practice went some way to accommodate defendants who might respond in a “ slow burn” fashion rather than explode in the face of the provocation. However, some commentators remained concerned that the continued requirement of a sudden loss of control posed an obstacle to some deserving defendants; and juries may have been stretching the rubric of the defence to let them through. The Act responds to that concern by removing the suddenness requirement (s.54(2)) while making clear that the defence cannot be relied upon where the defendant “ acted in a considered desire for revenge” (s.54(4)); the question of delay will therefore remain evidentially important - the longer the delay between trigger and killing, the more likely (the jury may think) that D was acting in revenge. On the face of it, this is a welcome development. But difficult questions remain. Is there such a thing as a loss of control which is other than sudden? Is there such a thing as an “ unconsidered” desire for revenge? And, more fundamentally, are the very cases which the Government is most concerned to cover in the new law - the battered spouses - aptly described in terms of loss of self-control at all? The Law Commission would have dispensed with that requirement entirely, leaving the defence arguably amply restricted by the express preclusion of revenge, the qualifying triggers (to which we turn next) and the objective test.

The “ qualifying trigger” required by s.54(1)(b) is explained in s.55. Two triggers of D's loss of self-control may be relied upon either alone or in combination (s.55(5)): first, “ D's fear of serious violence from V against D or another identified person” (s.55(3)); second, “ a thing or things done or said (or both) which constituted circumstances of an extremely grave character and caused D to have a justifiable sense of being seriously wronged” (s.55(4)). These are further elaborated upon in s.55(6)(a) and (b) which, in short, prevent D from relying upon a qualifying trigger which he has incited “ for the purpose of providing an excuse to use violence” . This resolves the issue of self-induced provocation addressed under the previous law by Edwards ([1973] A.C. 648) and Johnson ([1989] 1 W.L.R. 740). Finally, and most controversially, “ the fact that a thing done or said constituted sexual infidelity is to be disregarded” (s.55(6)(c)). Of the qualifying triggers, the s.55(4)(6) trigger - which might be dubbed the “ seriously wronged” trigger - is the closer equivalent of the provocation defence. But by contrast with the extremely broad notion of what can count as provocation - anything said or done, whether wrongful or not (Doughty (1986) 83 Cr.App.R. 319), trivial or serious - this section dramatically limits access to the new defence. Adjectives and intensifiers of the sort found in these sections are perhaps not generally to be found in the criminal law, but their role here is critical. It is not enough that D might have felt wronged - only if he might have had a justifiable sense of being seriously wronged will the trigger apply. Likewise, to experience “ grave” circumstances might have been thought to be bad enough - but no, only extremely grave circumstances will suffice. The intention here is clearly to push the defence as far up the scale as it can go. And in the process, many people who currently enjoy access to a defence will be deprived of it at this point.

The sexual infidelity section made its way back into the Bill during the final “ ping pong” stages of the parliamentary process. This section - described in the Lords as an exercise in “ gesture politics” - is of doubtful practical utility given the Government's own evidence that jealous husbands are not currently getting away with murder by relying on provocation (see Impact Assessment accompanying the Government's consultation paper, pp.11-12). And the drafting has been heavily criticised: how, for example, can “ a thing said” be said to “ constitute” sexual infidelity? And are we not really concerned with sexual jealousy or envy ? (See Professor Horder's Memorandum to Public Bill Committee). Rather than leave the matter - as the Law Commission would have done - to the judgement of the judge (under s.54(6)) and then to the jury and the limits already imposed by the express exclusion of revenge killings and the objective test, the Government determinedly restored the clause in a Commons vote, concerned to send a clear message about what conduct is and is not acceptable in the 21st century. But the arguments for its inclusion are somewhat weak - as Lord Henley put it, the Government won the Commons vote only after losing the argument (November 11, 2009, HL Deb col.841). There are a number of circumstances which many people might feel should not provide an excuse for murder in this day and age, so why single out this one rather than leave the matter to juries? An obvious example of another candidate for express exclusion would be so-called honour killings (cf. Mohammed [2005] EWCA Crim 1880). And how, in reality, is the jury to “ disregard” sexual infidelity when it may be intimately bound up with a number of other (permissible) triggers arising in a domestic context? (See Lord Thomas of Gresford, November 11, 2009, HL Deb col.840). This sort of “ micro-management” of the defence may prove to do more harm than good.

The other qualifying trigger - “ fear of serious violence from V” - is intended to accommodate cases such as the battered spouse's. It might be described as offering a defence based on the notion of excessive defence of self or another (cf. Clegg [1995] 1 A.C. 482). But whether the defence, as a whole, will meet that objective is questionable. One criticism of the previous law's attempts to accommodate these cases is that while the provocation paradigm is based on anger, defendants in these cases are motivated by fear and so struggle to fit that model. Section 55(3) expressly recognises that motivation. But, unlike under the Law Commission's scheme, the loss of self-control requirement applies. The better characterisation of D's conduct in these cases may be that she is - through desperate measures - seeking to regain some control through killing her long-term abuser, seeing no alternative way out of her predicament. On one view, so long as D's purpose was genuinely defensive, it ought not to matter whether she suffered a loss of self-control. But the Act shoehorns the battered spouse's case into a model that it does not readily fit.

The final stage of the defence is the objective test, under which D will only enjoy the defence if a “ person of D's sex and age, with a normal degree of tolerance and self-restraint and in the circumstances of D, might have reacted in the same or in a similar way to D” (s.54(1)(c)). This is elaborated upon in s.54(3), which states that the reference to D's circumstances covers “ all of D's circumstances other than those whose only relevance to D's conduct is that they bear on D's general capacity for tolerance or self-restraint” . The Act in this regard essentially codifies the outcome of the long-running saga relating to this limb of provocation, concluded by the Privy Council decision in Attorney General for Jersey v Holley ([2005] UKPC 23).

The application of this objective test, together with the requirement of a loss of self-control (albeit not sudden), to defendants relying on the fear trigger might deprive deserving defendants of a defence: if the degree of force used by the fearful D is objectively excessive, will she be deprived of the defence on the basis that an ordinary person with normal degree of tolerance and self-restraint would, by definition, not have used excessive force? However, the reference to D's “ circumstances” rather than mere “ characteristics” may be significant and helpful to D in these cases. If we can contrast the conduct of D with the conduct of another ordinary person in D's circumstances broadly construed - that is to say, having enduring the ten gruelling years of emotional and physical abuse which have generated the present fear of further serious violence - D's conduct may be more readily understandable, and an ordinary person similarly situated might well have reacted similarly, even if the force used is (viewed more abstractly) excessive. This approach, suggested by Lord Millett in his dissenting judgment in Smith ([2001] 1 A.C. 146), is not incompatible with the Holley test codified in the Act.

To sum up, the loss of control defence is not straightforward. Lord Lloyd was concerned that judges would find it difficult to direct juries satisfactorily and that juries may, despite the best efforts of the trial judge, flounder with it. The overall effect of these sections is undoubtedly to close off defences in lots of situations in which the argument could at least be raised under the previous law. Those who have qualms about anger-based reactions providing a partial defence to murder at all will doubtless be glad that the defence has been tightened up so significantly. But some cases now excluded from the defence might give us pause. Whether Mr Doughty ((1986) 83 Cr.App.R. 319) should have access to this particular sort of defence (rather than, as various commentators have suggested, a more widely drawn defence based on abnormal emotional states such as those induced by things such as extreme stress or fatigue, or the “ extenuating circumstances” amendment) is a moot point. But the effect of the Act - and its piecemeal approach to homicide reform - is that future Mr Doughtys, absent a plea of diminished responsibility, will now be guilty of murder and not, as the Law Commission's scheme might have had him, guilty only of second degree murder on the basis of what a jury might have found to be his less serious mens rea.

What then of the prospects for defendants wishing to plead diminished responsibility? The intention behind the Act here was merely to clarify and modernise rather than to change the scope of the defence. But s.52, which substitutes an entirely new s.2 of the Homicide Act 1957 for the original text, may have the effect of narrowing the defence, in the process depriving it of much of its considerable utility as a safety valve on the mandatory life sentence.

Section 52 replaces the concept of “ abnormality of mind” with “ abnormality of mental functioning” and requires that it arise from a “ recognised medical condition” . This, in many ways welcome, development will tie the defence to medical science in a way that the original defence - leaving most key matters to the judgement of the jury - did not. That condition in turn must have substantially impaired D's ability to do one of the effects listed in new subs.(1A): to understand the nature of his conduct, to form rational judgment, or to exercise self-control (cf. Byrne [1960] 2 Q.B. 396). And finally there must be some link between D's mental condition and his conduct in killing, specifically that the mental impairment caused or was a significant contributory factor in causing D to act as he did. As in Dietschmann ([2003] UKHL 10) this leaves room for other operating causes of D's conduct. While this test may sound perfectly reasonable, from a psychiatric perspective proving even a contributory causal link can be extremely difficult, if not impossible, to do in practice (see for example Baroness Murphy, June 30, 2009, HL Deb col.177 and 180). This may therefore be too restrictive a limitation on the defence.

Overall, the changes effected to diminished responsibility by the Act may in practice reduce the potential usefulness of the defence. Currently, what the Law Commission called a “ benign conspiracy” may operate between psychiatrists, defence, prosecution and trial judges, the CPS accepting pleas of guilty to manslaughter on grounds of diminished responsibility where it is not entirely clear whether the facts strictly meet the requirements of the defence, but where a sympathetic jury might acquit of murder anyway (see Law Com No.290, para.2.34). The more tightly drafted defence in the Act might inhibit prosecutors from adopting that approach, resulting in more murder trials and murder convictions. Lord Lloyd's “ extenuating circumstances” amendment would have provided a mechanism for enabling current beneficiaries of that approach - such as mercy killers - a route to mitigation that the more tightly drafted diminished responsibility defence might no longer afford. The battered spouse who is suffering from post-traumatic stress disorder or severe depression as a result long-term abuse may be able to bring herself within the defence. But commentators have long objected to a law which requires such defendants to plead their own mental illness to obtain mitigation, rather than offering them a defence which narrates a story of self-defence against unjustified violence.

Another controversial omission from the Act relates to children. The Law Commission had recommended including developmental immaturity in a child as a form of diminished responsibility. Not least given the recent abolition of the doli incapax defence for children aged 10-14 (affirmed in T [2009] ULHL 20), child defendants to murder face a peculiarly harsh sanction if there is no scope for considering whether immaturity mitigates their offence. (An amendment was moved during the Lords stages, but withdrawn following debate, to repeal s.34 of the Crime and Disorder Act 1998: June 30, 2009, HL Deb col.206). Many commentators therefore support the inclusion of a specific defence to murder based on developmental immaturity, which - psychiatrists argue (see again Baroness Murphy, from col.178, supported by former President of the Family Division, Baroness Butler-Sloss, from col.184) - can be exhibited by normal children. The Opposition front bench sought to amend the Bill to include developmental immaturity. But the Government opposed that move as unnecessary and creating the potential for abuse by its being routinely pleaded. The Government view is that if the child has no recognised medical condition (and so does not fall within diminished responsibility), no defence should be afforded. But without any recognition of developmental immaturity, the paradox highlighted by Professor Spencer will remain: a man of 40 with a mental age of 10 will be treated as having diminished responsibility - but a child of 10 will not.

The remaining sections in this chapter of the Act deal with the anomaly created by Gore ([2007] EWCA Crim 2789) in relation to infanticide, whereby D might be convicted of that offence where she could not have been convicted for either murder or manslaughter; and replace s.2 of the Suicide Act 1961 with a modernised offence of encouraging or assisting suicide, or attempting to do so. The latter reform does nothing to deal with the lack of guidance provided by the law which formed the basis of the successful appeal in Purdy v DPP ([2009] UKHL 45). The DPP's current consultation on his Interim Policy for Prosecutors in respect of Cases of Assisted Suicide, and associated concerns about the scope of the criminal law being effectively determined in policy guidance that may change at any time, rather than in the language of the statute, therefore remain as pressing as ever.

So, we await a commencement order. Sadly, the worst fears of the distinguished and vocal critics of the Bill may come to be realised. In relation to the core reform - the replacement of provocation with the new loss of control defence - concerns about the intellectual incoherence created by squeezing together two rather differently motivated defences may at best baffle juries and at worst deprive of any defence those whom the Government most wished to assist. Only time will tell. As to whether “ step two” of the homicide reform process ever sees the light of day…

The author is grateful to Professor Graham Virgo for his comments on an earlier draft.

Arch. News 2009, 10, 6-9