Goodwin, R v

LORD JUSTICE DAVIS:

In order to give the legal context it is convenient to set out the provisions of the relevant sections of the Coroners and Justice Act 2009 at the outset:

The background position is this. The appellant is now aged 63. On 7 November 2007, after a trial in the Crown Court at Maidstone before His Honour Judge Carey (the then Recorder of Maidstone) and a jury, the appellant was convicted by unanimous verdict of the jury of murder. He was subsequently sentenced, as required by law, to life imprisonment. The minimum term specified was 12 years and 191 days. No verdict was taken on count 2, which was an alternative count of manslaughter.

The facts are that on 19 May 2017 the appellant killed the deceased, Harry Messenger. The deceased had been his 75-year-old neighbour and friend. They had been friendly for about 5 years and spent much of their time together in their local pub. It seems that drink formed a significant part of the life-style of each of them. Furthermore, they had carried out a small and seemingly unlawful business of selling alcohol and cigarettes which they had purchased in France.

In the early hours of the morning neighbours heard banging and shouting from the deceased's address. There was evidence that the deceased had in the previous period been drinking alcohol. At 04.11 am the appellant sent a text to his brother in Ireland in which he said that he had "murdered someone". At 16.50 that day he made a 999 call in which he said that he had had a dispute with someone and had been trying to contact him by telephone without success. He was to say that the deceased had attacked him with a hammer (a hammer which the deceased had in the house; we have seen a photograph of it, it was a large claw hammer) and it was further said by the appellant that he had punched the deceased before using the hammer in retaliation to an attack on him. He was to say that the deceased had been in a bad way when the appellant had left and had not been responding. He had checked his pulse, had been unable to find it and had left. He was to say that the dispute that had arisen had been about something silly. Police officers found the deceased's body at 17.17 on the 19 May 2017. He had sustained significant head injuries. There was a hammer lying beside his head.

Put shortly, the prosecution case was that the appellant had murdered the deceased, doing so by striking him repeatedly with the hammer. Amongst other things, reliance was placed upon the evidence of a neighbour, a Ms Gibson, who claimed that she had overheard the appellant making threats against the victim a few weeks before the killing, as well as hearing arguing and banging on the night in question. There was considerable challenge to the reliability and accuracy of her evidence.

There was also evidence from another neighbour, who said that she had heard loud voices on the night of the killing.

A very prominent part of the evidence adduced at trial on the part of the prosecution was scientific evidence. In particular, (not rebutted by any expert evidence adduced by the defence) there was scientific evidence as to blood distribution which showed that the deceased had been attacked whilst lying on the ground. There was also evidence extrapolated from the evidence of blood that the attack may have extended over a period of at least 5 minutes, if not more, and that the assaults on the deceased, whilst he was lying on the ground, had occurred when he was lying in two different positions.

Post mortem evidence indicated that the cause of death was blunt force head injury. Alcohol and drug intoxication were also noted. It was estimated that the deceased had sustained at least 18 blows to his face, head and neck and these were consistent with his having been struck with a hammer. Further, there were a number of superficial injuries to his chest which had been inflicted at the time of or near to or after death. The use of a second weapon could not be excluded. It was also noted that there were injuries to the deceased's hands and arms which might have been caused by defensive action.

A toxicology test indicated a level of alcohol of almost three times the legal drink drive limit. In addition, there was noted to be a significant concentration of prescribed analgesics which exceeded the therapeutic dose.

In the course of interview, the appellant was also to say that he had washed the clothes which he had been wearing at the time of the attack. In the course of the interview the appellant had also spoken of the deceased having initially punched him to the head.

The defence case advanced at trial was one of self-defence. Put shortly, the appellant gave evidence that the deceased had attacked him with a hammer without warning, and that, in fear of his life, the appellant had grabbed the hammer and then struck the deceased. He could recall only striking three blows. He could not explain the post mortem findings as to the extent of the deceased's injuries. There was, it should be added, evidence of bruising to the appellant's face.

The principal issue for the jury was that of self-defence in circumstances where the trial judge had declined the invitation to leave the issue of loss of self-control to the jury.

So far as the defence evidence was concerned, the appellant was to say that he had first met the deceased in a pub. He explained the background of how they would frequently meet up and how they would buy alcohol and tobacco and sell it on. On the day in question he had gone to the local pub and had drunk 3 or 4 pints of beer and a vodka. He left at about 8 o'clock and went to see the deceased at about 9 o'clock in the evening. They watched television and chatted in what was a normal conversation. They were both drinking. During a conversation about a particular incident involving the deceased's son the deceased, he said, seemed to take offence at a remark which he, the appellant, had made; and the deceased then became foul mouthed, which was out of character for him.

The appellant was to say that the deceased then walked past him and when he returned he struck the appellant on the head with a hammer. The appellant had to accept that he had not said this in interview, rather he had referred to being punched: although he did say that he had at one stage referred to a hammer during the course of the interview.

At all events, in his evidence at trial, he was to say that there was then a tussle in which the appellant had tried to grab the hammer. The appellant had felt heavy blows coming from the deceased and told him that enough was enough. He was to say that he feared for his life. Eventually he managed to get up, having fallen to the ground, and took the hammer from the deceased. He was to say that the deceased had behaved "like a mad dog". He was to say that he then struck the deceased on the head with the hammer but had not hit him hard. They had struggled and the deceased had inflicted injuries to himself. He, the appellant, had wanted to leave but the deceased had charged at him. He said that a further struggle had ensued during which he hit the deceased. He said that he recalled hitting him three times altogether before the deceased went to the floor. He claimed no recollection of striking the deceased beyond the three blows which he had identified.

At one stage in his evidence, having said he did not remember what had happened after those blows he referred to "lights out". In dealing with this part of the evidence the judge was, amongst other things, to say this in his recount in the summing-up of what the appellant said:

It may also be noted that in the course of his summing-up and in dealing with the issue of self-defence, the judge, towards the end of the summing-up, gave what might be described as something of a steer to the jury, albeit quite properly leaving it to them to decide whether they accepted it, to the effect that the jury might well conclude that it was indeed possible that it was the deceased who had first attacked the appellant and invited them to focus in particular on the second limb of the self-defence direction, that is to say, whether his response was, objectively speaking, reasonable. In the result, as we have said, the jury convicted of murder.

At the conclusion of the evidence there was, as entirely conventional practice, a discussion between the trial judge and counsel in the absence of the jury.

Mr Higgs QC, then as now appearing for the appellant, argued (and we gather his argument had also been foreshadowed at the outset of the trial) that loss of control should be left to the jury. Whilst self-defence had been advanced as the principal defence (and of course if not disproved that would have resulted in an acquittal) it was alternatively said that on the evidence adduced loss of control was also in issue (which, if not disproved, would of course have resulted in a conviction for manslaughter). The prosecution, appearing then as now, through Mr Bennetts QC, opposed that suggestion.

One particular feature of the matter was that the appellant had at no time in his evidence said that he had lost control. The submission, however, was and is that it was a fair inference that he had lost control, given in particular the sheer number of blows administered to the deceased whilst on the ground over at least 5 minutes in circumstances where, as the appellant had said, the deceased had first attacked the appellant with the hammer and whereby it was said the appellant was in fear of serious violence: moreover, in circumstances when, after the third blow to which the appellant had admitted the appellant had said that it was "lights out" for him, connoting, so it was said, at the least that he had no recollection or consciousness of what happened thereafter. It was said that here there had been loss of control and that here there was a qualifying trigger, for the purposes of the statutory provisions, in the form of fear of serious violence on the part of the deceased towards the appellant, by reference to section 55(3). As to the third statutory component, that is to say that set out in section 54(1)(c), it was said that a person in the shoes of the appellant, with a normal degree of tolerance and self-restrain, might have reacted in the same or a similar way.

The judge rejected that submission. He did so having gone to the trouble of preparing a written ruling and, as is to be gathered, having considered the matter over an intervening weekend.

In that ruling the judge referred to certain relevant authorities. He then recited in some detail the account which the appellant had given to the jury in his evidence in-chief. He recorded, amongst other things, that in his evidence the appellant had said that he was fearful for his life. He noted that, amongst other things, that the appellant had said that, after he had been getting up from the initial attack by the deceased "Harry was coming at me like a mad dog and it got very frightening - I was petrified for my life". At a further stage, he had said that he was concerned that the deceased might try to reach for a sword in the bedroom. He is also recorded by the judge as having said this in his evidence in-chief (and as the judge was, as we have recorded, also to recount in the summing up):

Turning then to deal with the first component arising under the statutory provisions, that is to say loss of control, the judge in paragraph 5 of his ruling said this:

The judge then referred to the submissions of Mr Higgs, to the effect that the prosecution case was that there were at least 18 blows to the deceased's head and:

The judge then went on to deal with the second component arising under the statutory provisions, that is to say, relating to the qualifying trigger. As to that, the judge said this:

The judge having reached that conclusion that there was no qualifying trigger, it necessarily followed that the defence of loss of control should not be left to the jury. However, the judge went on to indicate his conclusion on the third component, that is to say, by reference to section 54(1)(c). As to that, he said this:

“10.

The third component: In these circumstances it is not necessary for me to make a finding at to the third component, which is concerned with the way in which the defendant has reacted in the circumstances and lost control but had it been necessary to do so, I would have reached the conclusion, having heard and considered all the evidence in this case, that there was insufficient evidence that this 64 year old male defendant, whom I have seen and heard giving evidence, who was very well acquainted with Mr Messenger a man some 10 years his senior, overweight and in poor health, if superimposed upon a person with a normal degree of tolerance and self restraint, might have reacted in the same or in a similar way as the defendant.”

The overall conclusion thus was that there was insufficient evidence, even on an interpretation most favourable to the defendant, to raise an issue of loss of self control. Accordingly, the jury would not be directed to consider it.

The appellant now says that that ruling was wrong. It is further said that since loss of control was not left to the jury when, as it is argued, it should have been, it follows, as it is argued, that the conviction for murder itself cannot be considered safe.

So far as the submissions before this court are concerned, Mr Higgs, on behalf of the appellant, in effect, has repeated the submissions which he made to the trial judge. Mr Higgs noted that the judge in fact had ruled in his favour on the first component of the defence arising under section 54(1)(a), that is to say finding that there was sufficient evidence of loss of control. But as to the second component, it is said that the judge was simply wrong to conclude that there was no sufficient evidence of a qualifying trigger and indeed it is said that the judge's reasoning was both garbled and mistaken in this regard. In particular, Mr Higgs draws emphasis to the specific evidence of the appellant, which it is said the jury could have accepted about the deceased coming at the appellant like a "mad dog" and about the appellant's stated fear of loss of his life in consequence and so on.

As to the third statutory component, Mr Higgs shortly submitted that this was clearly a matter for the jury in circumstances where both limbs of self-defence were in any event to be left to the jury for their consideration and having regard to the evidence.

For the Crown, Mr Bennetts has submitted to us that this was a matter for the trial judge, his task being to assess whether there was sufficient evidence justifying leaving loss of control to the jury. Mr Bennetts said that the judge had applied the correct sequential approach for the purposes of section 54; and overall, there was no proper basis for this court to interfere with the trial judge's ultimate conclusion based on the trial judge's appraisal of the evidence. However, in saying that, Mr Bennetts did in fact dispute the correctness of the judge's decision on the first component: that is to say, whether in fact, as a matter of sufficiency of evidence, the appellant had lost control.

We think that in a case of this kind there are a number of general considerations which need to be borne in mind which we should list. In doing so, we do not proffer this list as being necessarily an exhaustive list of the kinds of points that a trial judge, where such an issue arises, will need to bear in mind.

Those propositions are borne out by a number of authorities, including amongst others R v Gurpinar [2015] EWCA Crim 178; [2015] 1 Cr App R 463 and R v Jovan Martin [2017] EWCA Crim 1359; cases indeed to which the judge himself referred. Thus in Gurpinar, Lord Judge LCJ, after citing from the previous decisions in R v Clinton [2012] EWCA Crim; [2012] 1 Cr App R 26 and R v Dawes [2013] EWCA Crim 322; [2013] 2 Cr App R 3 said this at paragraph 12:

It also follows from all this that, putting it bluntly, there is no room for what may be called a "defensive" summing-up on such an issue. A trial judge cannot - tempting though it may sometimes seem - simply leave loss of control to the jury in order to seek to avoid generating a potential ground of appeal, if a conviction for murder results, on the basis of a failure to have done so. That would go against the scheme and the wording of the statute. As stated in the court by R v Skilton [2014] EWCA Crim 154 (itself a case where self-defence had been the principal defence advanced) a trial judge should not "clutter up" a jury's deliberations by inviting them to consider issues which do not arise on the evidence.

As to the relationship between self-defence and loss of control, that was considered in cases such as Skilton and Jovan Martin (supra). It is at all events to be emphasised that the two defences are, of course, legally distinct. One obvious distinction, amongst others, is that self-defence can arise where a defendant genuinely believes that there will or may be violence directed at himself or another. Loss of control however only arises for the purposes of section 55(3), where there is a fear of serious violence (emphasis added). There are, of course, other distinctions as well which we need not set out. Of course, there may well be cases where a murder trial is conducted on the footing of self-defence where, on an appropriately rigorous evaluation of the evidence, it can properly be assessed that there is also a sufficient evidential basis for the loss of control defence to be left to the jury. But it is most emphatically by no means the case that a defence of self-defence in a homicide case necessarily of itself will carry with it a sufficient evidential basis in the alternative for a defence of loss of control. That most certainly is not the law - see, for example, paragraphs 47 and 49 of the judgment in Jovan Martin and the decision in the case of R v Charles [2013] EWCA Crim 1205.

Against those general observations we turn to this particular case. As we have said, the judge had had cited to him and had referred to a number of cases including Jovan Martin. As to that case of Jovan Martin, Mr Higgs submitted to us, as he had to the judge below, that that was distinguishable on the facts. In that case, all trial counsel had positively agreed that the defence of loss of control did not run. That was a significant point in that case, albeit not decisive; it certainly was not the position in the present case where the point most certainly was advanced before the judge. But quite apart from that, in Jovan Martin the defendant, who had initially, been the subject of an attack by the victim in a house with a hammer, then picked up a knife and chased the victim out into street and there stabbed him many times as the victim lay on the ground. In the present case, says Mr Higgs, the attack by the appellant on the deceased was altogether more proximate, both in time and in space, to the alleged prior attack by the deceased on the appellant. At all events we, of course, accept that each case is fact specific and that Jovan Martin is indeed different on its facts.

Turning to the first component, loss of control, we have to say that we have considerable difficulty with the judge's conclusion, which in the event was favourable to the defence on this first component of the statutory defence: that is to say, whether there was sufficient evidence of loss of control.

As the judge himself had noted, at no stage had the appellant claimed to have lost control. That is particularly striking when his evidence was that he had remembered and had admitted to inflicting three blows on the deceased albeit, as he said, he thereafter remembered nothing and "lights out". Thus, for the blows that the appellant remembered administering, he did not claim to have lost control.

Of course, it will not necessarily be decisive of the issue that a defendant does not in his evidence positively assert loss of control, at all events if there is other evidence which is capable of indicating loss of control. But on any view it is a powerful point against the issue arising in such a case - see also the observations made in the cases of Charles and Jovan Martin (cited above) in this context.

Here, from his shortly stated ruling, as set out in paragraph 7, the judge appears to have accepted the submission that loss of control could be inferred "based on inferences which the jury are entitled to draw from the pathology and forensic science evidence." But how that could be inferred from such evidence is not explained. Indeed, such evidence in effect goes to the number of blows that were administered to the deceased whilst he was lying on the ground and over a period of at least 5 minutes if not more. Mr Higgs sought to say that this evidence suggests that the attack was "frenzied". With all respect, that is simply his own proffered gloss on what happened. The evidence simply does not go that far. Indeed, the fact that the administering of the blows seems (on the forensic evidence) to have lasted over a period of at least 5 minutes, and when for the most part if not the entire part of which the deceased was lying on the ground, would tell strongly against that.

In such circumstances, we do have great difficulty, as we say, with the judge's conclusion which he does not in any way amplify. We certainly do not think that the fact that there were at least 18 blows of itself, in the circumstances of this case, gives rise to an inference of loss of control. There is no other evidence identified which might support such a conclusion.

We do, of course, understand that respect, some kind of margin of appreciation if you like, is to be accorded to the evaluation of the trial judge, not least so experienced a trial judge as this. But as we have said, this is not a matter of discretion but a matter of evaluation, which is to be assessed as either right or wrong. However, we ultimately do not think it necessary to decide this appeal simply by reference to the first component in circumstances where the trial judge had found in favour of the appellant on that first component. We therefore, having expressed our reservations on that aspect of his ruling, go on to consider the judge's conclusions on the second and third components, which were adverse to the appellant and which Mr Higgs attacks.

We have to say that with regard to the second component, that is to say the qualifying trigger, we here see considerable force in Mr Higgs' criticisms of the judge's reasoning and conclusion. True it is that the jury would only be considering a loss of control once self-defence had been rejected. But it simply does not follow, as some aspects of the judge's reasoning might suggest, that if the defence of self-defence was rejected then there could be no qualifying trigger. That cannot be right: as Mr Higgs pointed out, the jury might well for example have accepted the appellant's version of events as to how he was initially attacked but nevertheless have concluded that, objectively speaking, he had used unreasonable and disproportionate force in response. Certainly the appellant had given clear evidence, which the jury were free to accept, that he had been in fear of his life and that the deceased had first attacked him with a hammer and so on.

Mr Bennetts suggested that the judge's reasoning was, or may have been, predicated on the fact that so many blows were administered (on the uncontradicted expert evidence) whilst the deceased was lying on the ground and not in a position to defend himself. That would connote in effect dividing the incident up between the first three blows (when the appellant did not claim to have lost self-control) and the remaining blows whilst the deceased was on the ground (and so, it was said, not in a position to give rise to a fear of serious violence). But not only did the judge not put it in terms like that, we do think, in the circumstances of this particular case, that it is very difficult indeed to divide the whole incident up in the way in which Mr Bennetts suggested. Moreover, to do so would run against the whole tenor of the way in which the issue of self-defence was left to the jury in the summing-up. So on this particular component of the statutory defence, here too we have significant reservations as to the correctness of the judge's rulings.

But, all that said, in our judgment the matter is settled, and adversely to the appellant, by the judge's assessment on the third statutory component. This was an assessment which we consider was entirely justified. Although the judge had expressed himself strictly on an obiter basis, he made clear what his conclusion was and explained why. Here, on the evidence, after the first three blows struck by the appellant himself, the appellant then himself had repeatedly struck the deceased, a much older man, whilst he was lying on the ground and in no real position to resist. The assault, moreover, lasted over a period of over 5 minutes. We do appreciate that under section 54 the emphasis is on how a person with a normal degree of tolerance and self-restraint, in the circumstances of a defendant, "might" have reacted: not "would" have reacted. Even so, the circumstances of this case were clearly such as to justify the judge's conclusion, in our judgment.

Mr Higgs submitted that on this third component, and in circumstances where self-defence was in any event going to be left to the jury, this was pre-eminently a jury matter. With all respect, that is in effect an argument by reference to the former law. As Gurpinar itself emphasises, no less a "rigorous evaluation" by the trial judge on the third component is called for than on the first and second components. Here, such an evaluation entirely endorses the judge's conclusion on this third component: a conclusion which we confirm.

Consequently, and whilst, as we have indicated, we do not necessarily accept the judge's conclusions on the first two components, we do think that he reached the right overall conclusion. In such circumstances, he was correct not to leave the issue of loss of control to the jury.

We are satisfied that this conviction is safe and we dismiss this appeal.