The Merits of Ambiguity: Provocation from the Irish Perspective

Abstract

The old common law defence of provocation has been abolished in many jurisdictions, including England and Wales and Northern Ireland, but it is still part of the law south of the Irish border. Up to now the Irish law of provocation has been built up solely by the courts, and has developed a number of peculiarities, not least the apparent rejection of the objective test by the Court of Criminal Appeal in the 1978 case of MacEoin. However, it is not until recently that the defence has come to the attention to the Supreme Court, in the 2020 case of McNamara. In this case the court set out to clarify the law, but the case has been criticised by commentators for its failure to clear up some of the existing ambiguities; indeed, it has been said that it has even introduced further ambiguities into the law that did not exist before. Be that as it may, this paper argues that McNamara is to be welcomed, in so far as it promotes a flexible approach and one which recognises the essentially normative nature of the provocation defence.

John E Stannard ^♦

Some twenty years ago the Law Commission for England and Wales in its Consultation Paper suggested that the then current law of provocation was ‘profoundly unsatisfactory’ and that its defects were ‘beyond cure by judicial development of the law’. ¹ In response, the overwhelming majority of those consulted agreed that this was the case. ² Certainly it could not be denied that the law was riddled with anomalies and ambiguities, despite a previous attempt at legislative reform ³ and a torrent of decided cases. ⁴ What was the rationale of the defence? ⁵ What sort of conduct could constitute provocation? ⁶ To what extent did the capacity of D (the defendant) to control his or her conduct have to be impaired? ⁷ What was meant by the objective requirement that a reasonable person would have done as D did, ⁸ and to what extent in deciding that question was the court required to take into account his or her personal idiosyncracies? ⁹ As is well known, these consultations resulted in a wholesale legislative reform of the defence in the Coroners and Justice Act 2009, and yet similar questions remain. ¹⁰ In particular: (1) the theoretical basis of the new defence is no clearer than that of the old; ¹¹ (2) no guidance is provided in the Act as to what is meant by a ‘loss of self-control’, ¹² except to say that it no longer has to be ‘sudden’; ¹³ (3) though the requirement of a ‘qualifying trigger’ has narrowed the law considerably, there are still unresolved issues, most notably with regard to sexual infidelity ¹⁴ and (4) there is still disagreement as to the function of the objective requirement. ¹⁵ Indeed, there are even some who argue that the defence should be abolished in its entirety. ¹⁶ One might very well conclude that, at least as far as England and Wales is concerned, the defects of the defence are ‘beyond cure’ not only by ‘judicial development of the law’ but by legislative reform as well.

With this in mind, it is worth looking, if only by way of comparison, at the approach in Ireland, ¹⁷ where there has been no legislative reform at all despite recommendations to that effect. Instead, the common law defence is still in force, albeit modified to a radical degree by the courts over the last 50 years. In 2020 the defence was considered for the first time by the Supreme Court in the case of People (DPP) v McNamara, ¹⁸ and though the approach in that case has been subjected to adverse criticism, not least for its lack of clear guidance to juries, the present paper will argue that perhaps it has something to be said for it.

Development of the Law in Ireland Prior to McNamara

Prior to 1978, the defence of provocation was not considered at all either by the Irish courts or by the legislature. ¹⁹ As a result, the law remained as inherited from that in force prior to partition, ²⁰ the basic requirements being stated as follows ²¹ : (1) there had to be a provocative act; (2) the act had to be such as to send a reasonable person out of control; (3) the mode of retaliation had to be proportional to the insult; (4) the plea of provocation had to be genuine; and (5) the homicidal act had to be directed at the provoker and not against any other person. However, in 1978 the Irish Court of Criminal Appeal introduced a radical reinterpretation of the law in People (DPP) v MacEoin. ²² The facts of the case were relatively unexceptional: D, who had killed the victim with a hammer, pleaded provocation on the basis of an attack by the victim with that same hammer. At his trial the judge, Butler J, directed the jury that D could not avail of the defence unless it could be shown that the provocation was such as to render him incapable of forming the necessary malice aforethought required for murder. ²³ Though this approach was not without authority, ²⁴ it was decisively rejected by the Court of Criminal Appeal; the defence of provocation, so far from negativing the proof of malice aforethought, presupposed its existence. In the words of Kenny J, who delivered the judgement of the court: ²⁵

… [T]he provocation relied on usually is one, if not the sole, cause of the formation of the intention to kill or cause serious injury to another. To speak of provocation negativing or depriving a man of the intention to kill or cause serious injury is to confuse cause and result.

That alone was sufficient to dispose of the case. However, the court then went on to undertake a general review of the defence as it stood, particular criticism being levelled against the objective requirement that the provocation should have been sufficient to cause a reasonable man to lose his self-control and do as D did. Despite this requirement having been approved at the highest level in England and Wales ²⁶ and elsewhere, ²⁷ there was no Irish authority on the point, which left the court free to consider the objective test on its merits. On this basis the court refused to follow the test any longer. ²⁸ In the words of Kenny J: ²⁹

The objective test is profoundly illogical: we assume that the reasonable man whom it propounds as the criterion is not the accused. If he were, the question would not be whether the reasonable man would be provoked but whether the accused was provoked. But what are the characteristics of this reasonable man? Is he to be endowed with the knowledge and temperament of the accused? Words which would have no effect on the abstract reasonable man may be profoundly provocative to one having knowledge of what people say about him. A hot-tempered man may react violently to an insult which a phlegmatic one would ignore. These are difficulties which those who support the objective test have never attempted to answer.

In this connection Kenny J went on to cite with approval the following passage from the dissenting judgement of Murphy J in Moffa v R: ³⁰

The test cannot withstand critical examination. It is not clear whether the reasonable or ordinary man, if he was subjected to the same provocation, would (or might) have lost control, or would have lost control to the extent of killing the deceased, or would have lost control to the extent of killing in the manner he did. Is he a complete stranger subjected to the provocative conduct or a person in the same circumstances as the accused? To be in the same circumstances, he should be taken to be in the same relationship with the deceased (in this case, a marital relationship) and must have experienced the relationship. In a case such as this, he should have lived the life of the accused, or it would be impractical to speak of what a reasonable or ordinary man would do in the circumstances. For example, it might have been an unbearable insult to a person of the accused's origin to be called ‘a black bastard.’ Once the full circumstances are taken into account, the objective test disappears because it adds nothing to the subjective test. For this reason, those who adhere to the objective test have rigidly excluded individual peculiarities of the accused (for example, low intelligence, impotence, pugnacity).

From now on, said Kenny J, juries should no longer be directed in objective terms. Instead, the procedure should be as follows: ³¹

When the defence of provocation is raised, we think that the trial judge at the close of the evidence should rule on whether there is any evidence of provocation which, having regard to the accused's temperament, character and circumstances, might have caused him to lose control of himself at the time of the wrongful act …

So far so good: the court should apply a purely subjective test. However, Kenny J then muddied the waters to some degree by adding the words: ³²

… and whether the provocation bears a reasonable relation to the amount of force used by the accused.

All of this left the law in a very uncertain state. ³³ In particular, if the test was purely subjective, how did the proportionality requirement fit in to the picture? In a number of cases following on from MacEoin the Irish courts suggested that the only relevance of this was as a matter of credibility: in cases where such a reasonable relation could not be demonstrated, the jury would be entitled to conclude that there was no loss of self-control in the first place. ³⁴ Others conceded that the test, though correct in principle, ³⁵ was confusing to juries ³⁶ and that the law was perhaps ripe for further restatement. ³⁷ Concern was also expressed at the application of a purely subjective test in cases where allowing it would promote moral outrage, as in cases of ‘road rage’ and other comparable types of ‘socially repugnant violent reaction’. ³⁸ In their Consultation Paper published in 2003, the Law Reform Commission observed that though the traditional objective test clearly no longer formed part of Irish law, some normative elements seemed to have survived, most notably the requirement of a sudden and complete loss of self-control. ³⁹ However, as things stood, there was some justification to the charge that as things stood it was virtually impossible for the defence of provocation to be rebutted once it had been raised. They added that Ireland appeared to be the only one of the common law jurisdictions that had ‘saddled itself with this dispensation’. ⁴⁰

McNamara

The recommendations of the Law Reform Commission were not taken up, and the problems with the common law defence accordingly persisted for a good number of years. However, in 2020 the Supreme Court had a chance to consider the law afresh in People (DPP) v McNamara. ⁴¹ In this case D was a member of a motorcycle gang, the ‘Caballeros’, which was currently engaged in a territorial dispute with the ‘Road Tramps’, a rival gang. One evening in June 2015 D, who was wearing his uniform gang jacket with the ‘Caballeros’ insignia, was drinking in a bar in the village of Doon, near Limerick, which happened to be within the territory claimed by the ‘Road Tramps’. Word of this presumably got out, and on leaving the bar, he was duly accosted by members of that gang, who forcibly ripped off his jacket and took it away. And when his wife tried to intervene, she was badly assaulted. Later on that evening, members of the ‘Road Tramps’ drove to D's home, where they made murderous threats.

Following all of this, a meeting was held between D and other members of the ‘Caballeros’ gang at which an appropriate response was considered. ⁴² The next day, following a further encounter between one of the assailants and D's stepson, D armed himself with a sawn off shotgun and drove to the heavily fortified headquarters of the ‘Road Tramps’ in the nearby village of Murroe. There he found two members of the rival gang waiting, including the victim (V), whom he shot in the head with fatal results. At the time V was holding a metal bar, though D claimed that he believed this to be a firearm. ⁴³

D was subsequently tried for murder, and raised two defences, one being self-defence and the other provocation. At the trial the judge allowed the first of these to go to the jury, but withdrew the second on the basis that there was no evidence of provocation fit to be considered. ⁴⁴ Having been convicted of murder, D appealed to the Court of Appeal on a number of grounds, one of them being that on the basis of MacEoin the defence of provocation should have been allowed to go forward. ⁴⁵ The Court of Appeal dismissed the appeal, but referred four questions of general public importance to the Supreme Court, these being: (1) whether the defence of provocation required that the provoking action or words come from the ultimate victim; (2) the extent to which background circumstances could found or inform that defence; (3) whether the defence of provocation contained any objective element either as to reaction or as to mode of response or as to time of response; and (4) the proper role of the trial judge in deciding whether the defence should go to the jury. ⁴⁶

At first sight it appears that if there is any case that merited the withdrawal of the defence of provocation from the jury, this is it. Even on the basis of a purely subjective test, it is hard to see any evidence whatever of the sort of loss of control required by the cases. The forcible removal of D's uniform jacket followed by the assault on his wife may very well have been a gross insult to his honour, and on that basis he might have been able to defend a lethal response then and there or even later that day. However, this was not done; instead, following a discussion with other members of the gang, D waits until the next day, deliberately arms himself with a lethal weapon, drives to the rival headquarters and then shoots not the provoker himself but another member of the gang! A weaker case could hardly be imagined. Nevertheless, it gave the Supreme Court the chance its first opportunity to review the defence of provocation, and for that we must be grateful.

Not surprisingly, the appeal was dismissed, but what is of interest are the observations made by Charleton J, who delivered the judgement of the court. His judgement is both lengthy and complex, and the reasoning is not always easy to follow, but the approach we shall adopt here is to summarise it as best we can before going on to consider how Irish law stands as a result.

What Does McNamara Say?

Having set out the facts and the issues for decision, the judge began by analysing the early development and rationale of the defence of provocation. ⁴⁷ In this connection he said that the law on the point had never been static; rather, it was ‘a defence made by judges and had been adapted over the centuries to new social mores or societal change at several junctures’. ⁴⁸ He also stressed that provocation was ‘never about excusing rage or bad temper, but rather lessened culpability as a concession to human frailty in the face of severely provocative conduct by the deceased’. ⁴⁹ In the words of Coleridge J: ⁵⁰

…though the law condescends to human frailty, it will not indulge human ferocity. It considers man to be a rational being and requires that he should have reasonable control over his passions.

It was against this background that Charleton J turned to the case of MacEoin. The defendant had argued that the effect of this case had been to remove the objective element altogether from the requirements of provocation. ⁵¹ However, Charleton J stressed that though MacEoin had undoubtedly changed the law radically away from its common law boundaries, ⁵² the provocation offered in that case had been ‘physical, repeated and severe’, and it had reaffirmed that provocation involved a sudden and complete loss of control. ⁵³ It had moreover given a key role to the trial judge, a role which could hardly be said to exist at all if the elements of the defence were entirely at large and to be adjudicated by no objective standard whatever. ⁵⁴ In the words of Charleton J: ⁵⁵

This appeal has centred on the actions of McDermott J in withdrawing provocation from the consideration of the jury. Yet, that is what … The People v MacEoin … mandates and does so in order that judicial control be maintained over circumstances which could not fairly be regarded as raising the defence of provocation. Here, the defendant's argument on this appeal is that no objective element remains within the definitional elements of the defence of provocation. Further, it is asserted that provocation is now a defence which cannot be defined in any way to include any objective element since, the contention goes, all objective elements were removed in The People v MacEoin and that this was confirmed in subsequent decisions. Yet, it is clear that there are limitations. Otherwise, no judge would ever be entitled to withdraw a case from a jury once provocation was mentioned and the element of genuineness must be grounded in socially understandable circumstances of provocation and resentment because otherwise criminal gang members would be foremost in producing the defence where slights or betrayals generated murders. The time element would, on the defendant's argument, also be gone since a claim of loss of self-control might be made days after the provocative conduct. Who would there be to say that this was not acceptable if everything was to be judged from inside the accused's own perceptions and reactions?

Having disposed of an argument by D that any reinterpretation or restatement of the law would infringe the doctrine of the separation of powers as set out in Article 15.2.1 of the Irish Constitution, ⁵⁶ Charleton J now moved on to the key issue in the case, namely the extent to which the defence of provocation in Irish law preserved an objective element. In this connection D had contended that all aspects of provocation as a defence in criminal law were entirely to be judged from the perspective of the accused at the time he or she killed the victim, ⁵⁷ but this was certainly not a stance shared by other common law jurisdictions. ⁵⁸ As Charleton J said: ⁵⁹

Yet, even as a word, provocation requires to be defined and any such definition is, since it is a legal definition, required to be of universal application. If, as the defendant has asserted, provocation is entirely in the realm of what the accused thought or felt, what has always buttressed the law in the form of objective requirements for conduct would entirely be removed. That would neither equate with the common law tradition nor would it be just.

That said, Charleton J had equal reservations in relation to the submissions of the prosecution. Their argument here was for a three stage test involving: (1) a complete loss of self-control in consequence of what V did such that D could not prevent himself or herself from killing the deceased intentionally; (2) this loss of self-control was to be adjudicated entirely from the subjective viewpoint of D; but (3) the mode of resentment, that is to say how D reacted to the provocation, should render the defence inoperable if it exceeds what is reasonable. ⁶⁰ As Charleton J pointed out: ⁶¹

Immediately a difficulty arises both in logic and in the realm of how that might work in experience. If there is no objective element in the second part, in accepting that the accused must first completely lose self-control as a matter of genuine fact, how could it be possible to fairly say that if the second part of the proposed test is entirely subjective, that somehow it is unreasonable for the accused, judged objectively, to have picked up a knife as opposed to have punched the victim or merely used bad language? Hence, the prosecution submission must be rejected.

Summing up on this issue, Charleton J declared: ⁶²

There, nonetheless, has to be a common and sensible standard which takes into account the variability of people as to age or sex or pregnancy and state of health or capability in physical or mental terms. Equally, social norms must now exclude violent responses to ordinary stresses such as a lover moving on or to phobic reaction to the right of people to choose their own lifestyle or path. Provocation is a defence which has always been limited by objective elements and by the need for the account of loss of self-control to be genuine and not contrived or bogus or set up to enable murder. There must be a sudden, and not a considered or planned, loss of self-control. That deprivation of self-control must be total to the degree that it is not merely a loss of temper but such a complete overwhelming of constraint, in consequence of what was done or said, that the accused cannot help intending to inflict death or serious injury, and cannot at all prevent himself or herself inflicting such deadly violence. That rage must not be fuelled by intoxication on drink or drugs. Loss of self-control must be in response to a serious provocation, not a mere insult, by the victim. The provocative act, by action or gross insult, is required to be outside the bounds of any ordinary interaction acceptable in our society. The defence does not apply to warped notions of honour and the proper sexual conduct of males or females, or mere hurt to male pride, or to gang vengeance, or to situations where sober people sharing the same fixed characteristics as the accused, where relevant, as to age, or mental infirmity, or sex, or pregnancy, or ethnic origin, would be able to exercise self-restraint in the same background circumstances as apply to that accused. The loss of self-control must genuinely cause the lethal violence. If any of those features are absent, the defence is not applicable.

Charleton J then went on to address a number of other points which were germane to the issues in hand. The first of these was the mode of retaliation. Traditionally, as Charleton J pointed out, this was part of the objective test, and might remain part of a modified objective test, since the question was whether an ordinary person could have been so provoked to be driven through transport of passion and loss of self-control to the degree and method and continuance of violence which produced the death. ⁶³ However, as we have seen this analysis was at odds with the purely subjective test applied in MacEoin. In McNamara, both defence and prosecution had sought to resolve this conundrum, albeit in different ways, but both proposed approaches were rejected by Charleton J, as can be seen from the following passage: ⁶⁴

Under the subjective test, according to the submissions made on behalf of the defendant, this is to be looked at from entirely the accused's mind. Yet as a person losing self-control is not thinking or reacting rationally, this makes little sense. The Director of Public Prosecutions, in submissions on this appeal argues, nonetheless, that the objective element to the defence of provocation occurs when assessing how much force was used. In all other jurisdictions, it is used simply in assessing credibility. This also fails to account for the fact that in picking up a knife, or as in The People v MacEoin … when replying with a hammer to an attack with a hammer, the accused is not acting in accordance with rational calculation, but in consequence of such loss of self-control as the law requires to be total.

According to Charleton J, the key to resolving the conundrum was to be found in the fact that the provocation defence was only relevant in cases where the necessary malice aforethought for murder was present. In the words of Charleton J: ⁶⁵

It should be remembered that in every murder case, the prosecution is required to prove the key subjective element: in doing what he or she did to kill the victim, did the accused intend to kill or to cause serious injury? That subjective element is not changed nor compromised by objective elements in the provocation defence since provocation only applies where the accused is proven by the prosecution to have killed the victim while intending to cause serious injury to him or her. What is at issue in the provocation defence is what the accused did in response to what the deceased person is alleged to have done to him or her in the context of any relevant background. All assertions of a state of mind are to be considered by a jury against what is claimed to justify having such a state of mind…It is from circumstances that an inference may, not must be made, and it as against the background of real events that any claimed loss of control must be considered by a jury with shrewdness and common sense. People can be provoked but there are degrees of provocation and degrees of reaction. In ordinary people who are not intoxicated or drugged, people assess what was done against what provocation was allegedly offered by the deceased. What the accused did, and his claimed mental state must be judged against that background. Such total loss of self-control to the degree of an intentional use of fatal violence must be genuine.

One of the problems for the defence in McNamara was the fact that the provocation did not emanate from V. The defence had argued that V was more than a mere bystander as he was a member of the rival gang, but as Charleton J pointed out, no just system of law could encompass retribution on a group basis. Rather, the law required any retaliation to be against the provoker, save perhaps in exceptional circumstances such as where a number of people were acting in concert, or where the doctrine of transferred malice came into play. In the words of Charleton J: ⁶⁶

In general, misdirected provocation does not accord with the law. If the accused kills someone who did not provoke him or her, the defence does not lie in any context where the accused says that he or she was provoked by another individual.

Another problem for the defence was the fourteen hours that had elapsed between the provocative act on the Friday evening and the killing the following afternoon. The defence had argued that on the basis of the subjective test the question of delay was purely one for the jury, but as Charleton pointed out, historically retaliation had to be sudden, and there had to be some limits even if a purely subjective test was applied. While some delay might be tolerated, as in cases of ‘slow burn’ provocation, there had to be limits. Though no entirely prescriptive rule could be laid down: ⁶⁷

…there is also a sense in which delay may bring about a situation where it would be contrary to the duty of a jury to judge facts fairly should they find for an accused on provocation. A case where there has been such a delay that the defence could not be fairly found must result in the defence being withdrawn from the jury by the trial judge.

This brings us on to Charleton J's final point, the role of the trial judge. Once again the defence had argued that once there was any evidence of provocation the defence had to be left to the jury. However, this was once again rejected by Charleton J, in the following words: ⁶⁸

In this context, the burden of proof is on the accused to produce evidence, or to point out evidence on the prosecution case, whereby as a matter of reality a jury would continue to act judicially be finding that the prosecution had failed to negative whatever evidence might be so adduced. But the evidence must be such as to be capable in law of amounting to provocation. That is a judicial decision. If a jury would be acting perversely in finding provocation, the judge cannot leave the defence for their consideration.

In conclusion, Charleton J set out six points that should be borne in mind by judges when instructing the jury in cases of provocation: ⁶⁹

To consider the defence of provocation, the jury must first be satisfied that the accused killed the victim and that in doing so he or she intended to kill the victim or to cause to the victim serious injury.

For the defence of provocation to apply, the jury must be satisfied that the prosecution has not rebutted such evidence as the jury considers raises a reasonable doubt in their minds that the accused may have been acting under provocation in killing the victim. All elements of the defence of provocation must be present to reduce an intentional killing of the victim from murder to manslaughter.

For provocation, there must be a sudden, and not a considered or planned, loss of self-control. There must be a total loss of all control to the degree that it is not merely losing your temper but, instead, is such a complete overwhelming of ordinary self-restraint, in the face of what was done and said, that the accused cannot help intending to inflict death or serious injury, and could not stop himself or herself inflicting this deadly violence.

That total loss of self-control in consequence of provocation cannot be because of intoxication on drink or drugs. The accused's actions are to be considered as if he or she was not acting under the influence of drink or drugs when the accused killed the victim.

Loss of self-control must be in response to a genuinely serious provocation, not a mere insult, by the victim. The provocative act, by action or gross insult, is required to be outside the ordinary interaction acceptable in our society. The defence does not apply to warped notions of honour or to any unacceptable ideas as to the proper romantic or sexual conduct of males or females; nor hurt to male pride; nor to gang vengeance.

The defence of provocation does not apply in situations where ordinary people, sharing, if relevant, the same fixed characteristics as the accused, as to age, or sex, or pregnancy, or mental infirmity, or ethnic origin, or state of health, would be able to exercise self-restraint in the same background circumstances as apply to that accused. People can be provoked, but juries should always remember that there are degrees of provocation and that there are also degrees of reaction to being provoked. What the accused did, and that accused's claimed mental state, must be judged against that background. Such total loss of self-control to the degree of an intentional use of fatal violence must be genuine. Thus, a jury will reject the defence if it is regarded as fabricated.

What Does McNamara Mean?

The rather curious upshot of all this is that, without any legislation on the topic, Ireland has arrived at a position not dissimilar to that seen in England and Wales under the Coroners and Justice Act 2009. In particular:

In both jurisdictions it is presupposed that D has killed with the necessary intent for murder; if this is not the case, then the defence does not come into play at all. ⁷⁰

Both jurisdictions require evidence of a loss of self-control on the part of D. ⁷¹ In Ireland this is defined by McNamara in terms of ‘such a complete overwhelming of ordinary self-restraint’ that D ‘cannot help intending to inflict death or serious injury, and could not stop himself or herself inflicting this deadly violence’. ⁷² In England and Wales the legislation gives no guidance as to what is meant by a loss of control, though in Jewell it was said to involve ‘loss of the ability to act in accordance with considered judgement or a loss of normal powers of reasoning’. ⁷³ One difference is that the law of Ireland, unlike that of England and Wales’ continues to insist that the loss of self-control be ‘sudden’. ⁷⁴

The law of Ireland, unlike that of England and Wales, ⁷⁵ has no specific requirement of a qualifying trigger, but does require a ‘provocative act (either by action or gross insult)’ which is ‘outside the ordinary interaction acceptable in our society’. ⁷⁶ There is no specific exclusion of sexual infidelity ⁷⁷ ; rather, such cases are left to the jury to decide to what extent such infidelity meets the general requirement of failure to conform to acceptable standards.

For many years the law of Ireland, as we have seen, purported to eschew any objective requirement, ⁷⁸ but now we are told quite clearly that ‘the defence of provocation does not apply in situations where ordinary people, sharing, if relevant, the same fixed characteristics as the accused, as to age, or sex, or pregnancy, or mental infirmity, or ethnic origin, or state of health, would be able to exercise self-restraint in the same background circumstances as apply to that accused’. ⁷⁹ Similarly the law of England and Wales requires that ‘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 acted in the same or in a similar way to D’. ⁸⁰

In Ireland the jury must be told to disregard cases where D's loss of self-control was due to intoxication by drink or drugs. ⁸¹ The current law of England and Wales contains no such explicit limitation, ⁸² but it is reflected in the requirement that a person of D's age and sex with normal powers of self-control would have responded in the same or a similar way.

In both jurisdictions, D bears an evidential burden with regard to the defence. In Ireland, the jury ‘must be satisfied that the prosecution has not rebutted such evidence as the jury considers raises a reasonable doubt in their minds that the accused may have been acting under provocation in killing the victim’. ⁸³ In the same way the Coroners and Justice Act 2009 in England and Wales provides that there must be sufficient evidence to raise an issue with respect of the defence, but once that is done the jury must assume that the defence is made out unless the prosecution proves beyond reasonable doubt that it is not. ⁸⁴

In Ireland, the judge has always been entitled to withdraw the defence from the jury in cases where there is no evidence to back it up, or where a finding of provocation would be perverse. ⁸⁵ This was not the case in England and Wales under the Homicide Act 1957, where the defence had to be left to the jury in any case where there was evidence that D might have been provoked to lose his or her self-control. ⁸⁶ However, this is no longer the case under the Coroners and Justice Act 2009, which requires evidence to be adduced on which, in the opinion of the trial judge, a jury properly directed could reasonably conclude that the defence might apply. ⁸⁷ This brings the two jurisdictions back again into line.

Problems with McNamara

The defence of provocation in Ireland has been described on more than one occasion as a ‘graveyard for judges’ when directing juries. ⁸⁸ The case of McNamara, and the guidance given there by Charleton J, may have cleared up some of the uncertainties surrounding the defence, but questions still remain in relation to all three elements of the defence. In particular:

As far as loss of control is concerned, McNamara does nothing to address the question of the relationship between provocation and automatism. ⁸⁹ We are told that a mere loss of temper is not enough, and that there must be such a complete overwhelming of ordinary self-restraint that D cannot help intending to inflict death or serious injury, and could not stop himself or herself inflicting this deadly violence. ⁹⁰ On the face of it this looks very much like automatism, but clearly this cannot be so. ⁹¹ In this connection it has been suggested that the traditional formulation used by judges is basically no more than a matter of rhetoric, ⁹² being designed to convey to the jury and to the public at large that the defence is very restrictive in nature. ⁹³ As well as this, there is the ongoing requirement that the loss of self-control be ‘sudden’, ⁹⁴ a factor which has given rise to problems in relation to so-called ‘slow burn’ cases in England and Ireland alike.

In relation to the relevant trigger, McNamara as we have seen avoids the problems of a closed list such as that seen in the Coroners and Justice Act 2009; ⁹⁵ instead, it simply requires evidence of a ‘provocative act (either by action or gross insult)’ that is ‘outside the ordinary interaction acceptable in our society’. ⁹⁶ What counts as ‘acceptable’ in this context is clearly is a normative question that can only be decided by the jury, ⁹⁷ with the attendant risks of uncertainty and inconsistency between one jury and another. ⁹⁸

As far as the so-called ‘objective test’ is concerned, the jury must now be told that ‘the defence of provocation does not apply in situations where ordinary people, sharing, if relevant, the same fixed characteristics as the accused, as to age, or sex, or pregnancy, or mental infirmity, or ethnic origin, or state of health, would be able to exercise self-restraint in the same background circumstances as apply to that accused’. ⁹⁹ This test, though very similar to that seen in the Coroners and Justice Act, ¹⁰⁰ gives rise to a number of problems. One of these is the nature of the question the jury has to decide. As has been pointed out, the old ‘reasonable man’ test could be read in two ways, that is to say either as a heuristic for whether D's actions were reasonable in the circumstances, or as a construct for the ordinary or empirically average person. ¹⁰¹ The substitution of ‘ordinary’ for ‘reasonable’ in the Coroners and Justice Act 2009 and its adoption in McNamara suggest that the jury should take the latter approach, but both approaches suffer from the same basic flaw, which is that neither a reasonable nor an average person would normally perform a killing even under provocation. ¹⁰² Another problem is the reference of the test to ‘fixed’ characteristics; ¹⁰³ the rationale for this is unclear, and as has been pointed out, none of the characteristics listed other than ethnic origin are in fact ‘fixed’! ¹⁰⁴ As well as that, the same problem arises in relation to the role of the judge; if the judge is to exclude from the jury cases where there is no evidence to back up the defence, on what material can such a decision be based?

All in all, it is hard to disagree with Andrea Shieber's conclusion that while Charleton J in McNamara offers a compelling argument for some practical limits on what should qualify as provocation, he is less successful in charting a principled basis for the parameters of the combined subjective and objective test. ¹⁰⁵ Rather, at the end of the day a great deal is left to the judgement of the jury. The judge may have a role to play in weeding out the hopeless cases, but on the basis of the analysis given above this is unlikely to happen very often. In particular, the judge may still have a significant role to play where the issue is whether D was suffering from a loss of control at the relevant time, as in McNamara itself, but where the issue concerns the other two limbs of the defence (the nature of the provocation and whether an ordinary person would have reacted in the same way), it is hard to see what material the judge could use to justify withdrawing the defence from consideration. Most cases will be left to the jury to decide, albeit on the basis of an appropriate direction by the judge. This gives rise to the question of whether this is an acceptable approach for the criminal law to take.

The Merits of Ambiguity

The proposition that law, especially criminal law, should be formulated with precision seems to be so obvious as not to be worth arguing. ¹⁰⁶ After all, if legal norms are supposed to regulate human conduct, how can a norm that lacks precision fulfil this function? ¹⁰⁷ Indeed, it could be argued that such a norm is not really a norm at all. ¹⁰⁸

Be that as it may, a moment's reflection will show that even the criminal law is replete with imprecise terms. This applies not just to the two examples often cited in this connection – dishonesty in theft and gross negligence manslaughter ¹⁰⁹ – but to other crimes too. Thus one of the most commonly charged crimes, that of driving without due care and attention, ¹¹⁰ would seem to fail the test of precision, as would the crime of neglecting a child in a manner likely to cause unnecessary suffering or injury to the child's health or seriously to affect his or her wellbeing. ¹¹¹ Indeed, an element of vagueness ¹¹² can be seen even in such basic concepts as subjective recklessness, which consists not just of the taking of a known risk but doing so in circumstances where it would be unreasonable to take that risk. ¹¹³

The fact is that the arguments with regard to precision in the criminal law are not so stark as might at first sight appear. The issue is one which has in recent years generated a great deal of academic debate both at the legal and the philosophical level, ¹¹⁴ and whilst there is not the space here to cover the relevant arguments in any great detail, a few key points can be highlighted.

The first is that precision in law as a desideratum is not simply self-evident, but has to be seen in the context of its purpose. According to Timothy Endicott, these can be described in terms of ‘guidance’ and of process’. ¹¹⁵ Thus a fixed speed limit provides more useful guidance to drivers than would a law saying that one should not drive too fast; a fixed voting age provides more useful guidance to officials than would a law making the right to vote dependent on their maturity, and so on. The same applies in relation to process; thus a law prohibiting a person from driving with more than a certain amount of alcohol in the blood has two advantages over one providing that one should not drive while drunk; not only can a police officer with a breathalyser use the reading as a guide for deciding whether to arrest the driver in question, it also reduces the need to decide in every case whether D was ‘drunk’ or not.

However, this precision comes at a price. In particular, as Endicott points out, it can lead to a certain degree of arbitrariness, in so far as it detaches the norm in question from the rationale for imposing it. Let us once more take the example of the law forbidding a person from driving with more than a certain amount of alcohol in the blood. Clearly the rationale for this is to deter potential drivers from imperilling other road users when their abilities are impaired by intoxication. In a particular case driver A who is just over the limit may be less impaired than driver B who is just under it. Yet it is driver A who is guilty of the crime, not driver B.

As well as this, a vague norm can have advantages of its own. In this connection Jeremy Waldron draws a distinction between ‘rules’ and ‘standards’; the latter do not function in the same way as rules, but nevertheless they can be effective at guiding action in the sense of practical deliberation involved in the exercise of agency. ¹¹⁶ As well as that, as Endicott argues, a vague standard may serve the function of delegating the power of resolving debates over the application of the standard to some other person or body who may be in a better position to do so. ¹¹⁷ Thus in relation to careless driving, it makes more sense for the law to let the courts decide in the light of the circumstances whether D has driven without due care and attention rather than to try and draw up in advance a list of things that D should or should not do. ¹¹⁸ Last but not least, Endicott says, a standard couched in broad terms can encourage desirable forms of private ordering that achieve the law's purposes. ¹¹⁹ Thus the uncertainty of negligence law may give private parties the incentive to avoid the creation of risks, or to contract out of liabilities in a way which will allocate the cost of the risk to the least cost avoider of the risk. In the same way, the imposition of broad standards in criminal law may provide a more effective incentive for the regulation of conduct than would a requirement couched in more precise terms. ¹²⁰

So how does all of this relate to the defence of provocation? There are four reasons why a more flexible and ‘open textured’ ¹²¹ approach may be particularly suitable here.

First of all the demands of legality are relatively weak in the context of provocation. A driver is entitled to know how fast he or she can go without breaking the law, and a drinker is entitled to know how old he or she should be before being allowed into the bar, but it makes no sense to say that a killer is entitled to know in advance if and on what terms the provocation defence is going to be available to them. ¹²² For one thing, the defence does not provide a complete justification for the killing; it merely mitigates the crime to manslaughter. ¹²³ Moreover, as Prendergast says, there is no action-guiding function in the provocation defence; the relevant killing by definition is not planned, and simply should not be done. ¹²⁴

As well as this, the risks of ‘arbitrary precision’ highlighted by Endicott are particularly high in this context. Many of the problems associated with the defence of provocation in the past have arisen in an attempt to set out in advance excessively rigid rules for the application of the defence, whether in relation to the loss of self-control requirement, or the triggering event, or the requirement that the reaction of the accused be a reasonable response to the provocation offered. Sometimes this has led to the exclusion of cases which are arguably clearly within the spirit of the defence, such as Bedder v DPP, ¹²⁵ R v Thornton, ¹²⁶ R v Ahluwahlia ¹²⁷ and possibly R v Acott. ¹²⁸ In other cases it has led to the inclusion of those which are clearly not, as in R v Doughty ¹²⁹ and R v Dryden. ¹³⁰ In yet others it has led to unnecessarily fine distinctions having to be drawn, as in R v Morhall, ¹³¹ Luc Thiet Than v R, ¹³² R v Morgan Smith, ¹³³ A-G for Jersey v Holley ¹³⁴ and R v Clinton. ¹³⁵ It could be argued that none of these cases would have caused such problems if the courts had been allowed to apply a more flexible approach based on the basic rationale of the defence of provocation.

The third factor worth mentioning is the role of the jury. It is the jury which will make the final decision in the vast majority of provocation cases, ¹³⁶ and though the judge may do his or her best to give directions in accordance with the law, the extent to which these are understood, let alone followed, is much less clear. ¹³⁷ Indeed, the more precise and complex the instructions are, the less likely it is that a jury will be able to follow them, still less apply them. ¹³⁸

Last but not least, we have the essentially normative nature of much of the provocation test. Some twenty years ago the present writer, in the pages of this journal, drew attention to this, ¹³⁹ and referred to the remarks of Lord Hoffmann in the Morgan Smith case, in which Lord Hoffmann highlighted this aspect of the defence. Referring to the discretion allowed to the jury under section 3 of the Homicide Act 1957, he commented: ¹⁴⁰

I do not think it possible to attribute to Parliament, in making this change, any intention other than to legitimate the relaxation of the old law in those cases in which justice appeared to require it and to allow the jury in good conscience to arrive at a verdict which previously would have been perverse. In other words, the jury was given a normative as well as a fact-finding function. They were to determine not merely whether the behaviour of the accused complied with some legal standard but could determine for themselves what the standard in the particular case should be.

It was argued by the present writer that this approach was in many ways the key to reconciling the inconsistencies with regard to the defence of provocation both in English and in Irish law, and that it offered the best way forward on both sides of the Irish Sea. Since then the relevant law has moved on considerably in both jurisdictions, but there is nothing that has led the present writer to resile from that opinion.

Conclusion

In assessing McNamara a distinction needs to be drawn between the general approach seen in that case and the detailed dicta of Charleton J. With regard to the latter, there are still infelicities and consistencies (for instance in relation to the ongoing requirement of ‘suddenness’ ¹⁴¹ and the caveat that allowance can be made for the attributes of the individual defendant only in so far as they are ‘fixed’ ¹⁴² ). But these are minor matters that can be ironed out in the context of future judicial guidance. What is more important by far is the extent to which the approach in that case, whether or not by design, ¹⁴³ allows for the judgement of the jury, especially in relation to such issues as the gravity of the provocation and the defendant's reaction to that provocation. The detailed guidance suggested by Charleton J will certainly have a part to play in helping the jury reach their decision, ¹⁴⁴ but at the end of the day, save in exceptional cases, they will have the final word. Certainty and precision in the law is undoubtedly an important value, but sometimes it may make it difficult to see the wood for the trees. In so far as a more flexible approach to provocation enables Irish juries to keep in mind the general principles underlying that defence, the case of McNamara is to be welcomed.