Mukeh Joseph Kawa & Anor v R

In the early hours of 20 January 2017, Djojo Nsaka, then aged 19, was killed by a stab wound to the chest. In August 2017, after a trial at the CCC before HHJ Leonard KC and a jury, these applicants, Mukeh Kawa and Donald Davies, were convicted of his murder. They were later sentenced to life imprisonment. Their co-defendant Ali Tas was acquitted of murder but convicted of manslaughter.

For convenience, and meaning no disrespect, we shall generally refer to persons by their surnames alone.

Neither of the applicants commenced any appeal at that time. Tas appealed against his conviction, but his appeal was dismissed by this court in 2018.

In November 2021 Kawa applied for an extension of time of 1,513 days to apply for leave to appeal against his conviction. His applications were refused by the single judge. He renewed them to the full court. On 14 October 2022 the court, differently constituted, indicated a provisional view that the grounds of appeal were arguable, but adjourned the hearing so that the respondent could be represented. Davies then applied for an extension of time of 1,911 days to apply for leave to appeal on like grounds against his conviction. His applications have been referred to the full court by the Registrar.

In considering Kawa’s renewed applications, and Davies’ referred applications, we have been assisted by the submissions of Mr Emanuel KC on behalf of Kawa and Mr Butler on behalf of Davies, neither of whom appeared below, and of Mr Boyce KC and Ms Przybylska, who represent the respondent in this court as they did at trial. We are grateful to them all.

The fatal stabbing occurred in an incident involving two groups of young men, one comprising the three defendants and the other comprising Nsaka and his friends Chris Bonda and Daniel Tamfuri. It appears that a trivial incident on the previous evening had given rise to ill will between these groups. They met near a student hall of residence, to which the three defendants had travelled by car, apparently in search for the other group. The three defendants spotted Nsaka and his friends, left their car and approached Nsaka’s group.

The prosecution case at trial was that the three defendants were acting together to carry out a pre-planned attack, and that both applicants were armed with knives. Tamfuri gave evidence that he saw Kawa produce a large knife from his waistband. It was alleged that Kawa and Davies chased and caught Nsaka, and that Kawa stabbed him with a knife which Kawa had brought to the scene. The three defendants then left in their car, driven by Tas.

Neither of the applicants answered any questions when interviewed under caution, but both gave evidence at trial. Kawa’s case was that Nsaka initially ran away but then stopped, produced a knife and attacked Kawa. Kawa said that the knife had been knocked to the ground, he and Nsaka had wrestled for it, and in the course of the struggle the knife had accidentally entered Nsaka’s chest, though Kawa did not realise that at the time. Kawa denied any intention to cause serious harm. Davies’ case was that he had not been armed, and had played no part in that violent episode.

At trial, each of the three defendants was represented by leading and junior counsel. An important issue in the trial was whether the jury could be sure that it was Kawa, and not Nsaka, who brought to the scene the knife with which the fatal wound was inflicted. In his summing up, the judge directed the jury that the prosecution had to prove that the man who stabbed Nsaka was not acting in lawful self defence. He went on to direct the jury:

“If the knife was or may have been brought to the scene by [Nsaka], then the prosecution would have failed to prove its case and that would result in you returning verdicts of not guilty in respect of all the defendants.”

It should also be noted that the judge further directed the jury that they had to be sure that, either as the stabber or as a participant in the joint enterprise during the course of which Nsaka was stabbed, the defendant whose case they were considering intended to kill Nsaka or at least cause him really serious bodily harm if the need arose. The judge continued:

“If you are sure that the defendant whose case you are considering had a knife, whether it was used to stab [Nsaka] or not, or that he knew that one or more of the other defendants had a knife, then that is evidence you can take into account when answering [that] question.”

It is unnecessary, for present purposes, to say more about the facts. Further details can be found in the judgment of the court dismissing Tas’s appeal: see [2018] EWCA Crim 2603, [2019] 1 Cr App R 26.

We turn to the issues relating to bad character evidence which are the subject of the applicants’ grounds of appeal. It is convenient to begin by setting out the provisions of ss101 and 103 of the Criminal Justice Act 2003, so far as is material for present purposes:

“101 Defendant’s bad character

(1)

In criminal proceedings evidence of the defendant’s bad character is admissible if, but only if – …

(d)

it is relevant to an important matter in issue between the defendant and the prosecution,

…

(g)

the defendant has made an attack on another person’s character.

…

(3)

The court must not admit evidence under subsection (1)(d) or (g) if, on an application by the defendant to exclude it, it appears to the court that the admission of the evidence would have such an adverse effect on the fairness of the proceedings that the court ought not to admit it.

(4)

On an application to exclude evidence under subsection (3) the court must have regard, in particular, to the length of time between he matters to which that evidence relates and the matters which form the subject of the offence charged.

…

103 “Matter in issue between the defendant and the prosecution”

(1)

For the purposes of section 101(1)(d) the matters in issue between the defendant and the prosecution include –

(a)

the question whether the defendant has a propensity to commit offences of the kind with which he is charged, except where his having such a propensity makes it no more likely that he is guilty of the offence;

…

(2)

Where subsection (1)(a) applies, a defendant’s propensity to commit offences of the kind with which he is charged may (without prejudice to any other way of doing so) be established by evidence that he has been convicted of –

(a)

an offence of the same description as the one with which he is charged, or

(b)

an offence of the same category as the one with which he is charged. …”

The prosecution applied pursuant to s101(1)(d) to adduce against Kawa, as evidence of bad character, evidence of three matters: the convictions in April 2016 of both applicants for offences committed in December 2015 of possessing an offensive weapon, namely a knife, in a public place; an occasion in December 2016 when police went to search an address linked to Davies, and found Kawa wearing a coat and gloves and carrying 2 knives; and an occasion only 10 days before the stabbing when the applicants purchased two knife blocks containing a number of knives, and also other household items, at a supermarket. A similar application was made to adduce against Davies the 2016 convictions of the applicants and the recent purchase by the applicants of knife blocks.

The basis of the application was stated as follows on the notice of application relating to Kawa:

“The defendant denies being in possession of any weapon. He takes issue with the eyewitness evidence that he was carrying a knife. Whether he was in possession of a knife is an important matter in issue between him and the prosecution. It is submitted that all three matters set out above are relevant to this issue and therefore admissible pursuant to section 101(1)(d).

The defendant maintains that he was not aware that any other defendant had a knife. His knowledge of the presence of knives at the scene is therefore an important matter in issue between him and the prosecution. It is submitted that his 2016 joint conviction with Davies for possession of knives is relevant to this issue and therefore admissible pursuant to section 101(1)(d).”

It will be noted that the application did not make any reference to any propensity on the part of either applicant. In a skeleton argument, the prosecution confirmed that the issue to which the material was relevant was whether a defendant had a knife at the time of the stabbing, not whether he had a propensity to behave in a particular way.

The application was resisted by the applicants, on the grounds that the prosecution were trying to establish a propensity but the evidence was not capable of proving one. In the alternative, it was submitted that the evidence should be excluded because it was overwhelmingly prejudicial.

The judge ruled that the evidence was admissible, and declined to exclude it pursuant to s101(3). In order not to delay the trial, the judge initially gave only a short oral ruling. He gave his reasons in writing at a later stage in the trial. By that time, further relevant evidence had emerged, which the judge regarded as supporting the decision he had made. In particular, Kawa’s counsel had attacked the character of Tamfuri by adducing evidence of his previous convictions for violence; both applicants in their evidence had attacked the character of Nsaka by saying that it was he who had brought the knife to the scene; and it was accepted that the evidence of the 2016 convictions and of the character of the applicants was admissible pursuant to s101(1)(g) of the 2003 Act.

The judge ruled that the evidence relied on by the prosecution was relevant and admissible pursuant to s101(1)(d) to assist the jury on the central issue of who had the knife, and whether Kawa or Davies would have known about the other carrying a knife. The evidence of the April 2016 convictions was relevant to both applicants because it involved the same two defendants who were alleged to have been carrying knives on 20 January 2017; the event in December 2016 was relevant to Kawa because it could assist the jury as to whether Kawa did (as he had claimed) stop carrying knives after his April 2016 conviction; and the purchase of the knife blocks was relevant to show that both applicants possessed knives and would have had them readily available on 20 January 2017.

After the judge’s oral ruling, the evidence which the judge had admitted was dealt with by means of formal admissions. The judge intervened in the reading to the jury of a series of admissions to say the following:

“Before you go on, there is something I need to say to the jury about the last two paragraphs that have been read to them and indeed in relation to the knife blocks and the buying or the receipt in relation to the purchase of knife blocks with knives. You will get full directions on this in writing before you retire to consider your verdicts, but because that evidence has come for you at this stage, some time [before] I will be able to give you those directions, I am just going to give you a warning about it. The reason why you have heard of that evidence through admissions contained in paragraphs 34 and 35, you have also heard about the knife block, is because, say the prosecution, it is relevant as to whether the defendants had access to them and were armed with knives that night, or indeed what they knew about others and whether others who they were with had a knife or knives with them. What you must guard against is concluding that just because they had knives on them in the past that they used knives on that night, or had a knife in their possession on that night or knew someone else had a knife in their possession on that night. It is part of the evidence that you can use in order to assist you in coming to a conclusion as to what they did on the night in question, but it is not conclusive of it, and you will get further directions at a later stage which puts this evidence into context.”

When the judge later came to give his directions of law, he did so in terms which had been discussed with all counsel, and which were provided to the jury in writing. In summarising the bad character evidence he said:

The judge directed the jury that they had heard that evidence for two reasons: because they were entitled, when deciding the truth of the allegations made against Tamfuri and Nsaka, to know the character of the men who had made those allegations; and because the evidence was capable of being relevant to an important matter in issue between prosecution and defence as to whether either or both of the applicants were carrying knives on 20 January 2017. His directions continued:

“If you are sure that the evidence demonstrates that the defendant whose case you are considering armed himself with knives in the past, or has bought knives in company with his co-defendant for their own use, then that is capable of supporting the prosecution’s case that he carried a knife on this occasion. Alternatively, if you are sure that Kawa knew that Davies had carried knives in the past or that Davies knew that Kawa carried knives in the past, then that is capable of supporting the prosecution’s case that he would have known that the other was armed with a knife on this occasion.

You must not convict the defendant whose case you are considering wholly or mainly on the evidence of what, if anything, you find he has done in the past. The fact that a defendant has carried knives in the past or bought knives in the past does not prove he did so on this occasion but you may use it as some support for the prosecution case.

Both defendants have told you that they were involved in supplying cannabis, because they could not explain why they had knives on them when they were stopped on 22 April 2015, and it has become relevant to what they were doing in January 2017 – so, pausing there, that is why you have been told about it, I’m sure. You may take the evidence into account when considering the case generally. But the fact that they were supplying cannabis cannot of itself affect whether they are guilty or not guilty of the murder.”

It does not appear that either of the applicants’ counsel argued against those directions at the time. Nor, as we have noted, was any ground of appeal relating to them put forward by trial counsel.

Both applicants have given their explanations for the delay in filing their respective notices of appeal. Before considering that aspect of the case, we will address the merits of their grounds of appeal.

The grounds of appeal put forward on behalf of Kawa are, first, that the judge erred in admitting non-conviction bad character evidence of Kawa’s previous possession of knives; and secondly, that the judge erred in failing to direct the jury that the bad character evidence could only assist the prosecution’s case if they were sure it meant that Kawa had a propensity to carry knives in public.

Davies’ ground of appeal is that the judge misdirected the jury by failing to direct them properly about the need to be sure that the bad character evidence showed a propensity on the part of either Davies or Kawa to carry knives before it could be taken into account.

In brief summary, the core of Mr Emanuel KC’s argument on behalf of Kawa is that, although the prosecution disavowed any reliance on propensity, they were in reality seeking to invite the jury to find that Kawa had a propensity to be armed with a knife or knives in public, and was for that reason more likely to have been carrying a knife at the time of the fatal incident. He submits that the terms of the judge’s written directions show that the judge also understood the prosecution to be relying on the bad character evidence as demonstrating habitual behaviour. He places particular reliance on the words of the judge’s direction which we have quoted at paragraph 20(i) above: “the prosecution say that it was their habit to go out armed and subsequent events show that they had not learned their lesson”.

Mr Emanuel KC accepts that for the purposes of s101(1)(d), the matter in issue need not necessarily be an issue as to whether a defendant has a propensity to behave in a particular way; but in this case, he argues, the prosecution were in reality seeking to rely on such a propensity. He submits that the judge should therefore have directed the jury that before they could treat the bad character evidence as supporting the prosecution case, they must first be sure that the evidence did establish that Kawa had a propensity habitually to carry knives in public. He submits that the directions given by the judge were insufficient in the circumstances of this case, and that Kawa’s conviction is unsafe.

Mr Emanuel KC further submits that if, as the judge indicated, the evidence was not admitted on the basis that it was capable of establishing a relevant propensity, then it should not have been admitted at all. He argues that, unless the jury were sure that they established a relevant propensity, the earlier incidents could not provide any support for a finding that it was Kawa, not Nsaka, who took a knife to the scene.

Mr Butler adopts those submissions on behalf of Davies. He too accepts that the “important matters in issue” are not necessarily limited to issues of propensity, but submits that the reality of this case is that the jury were being invited by the prosecution to rely on a relevant propensity; he too relies, in particular, on the direction which we have quoted at paragraph 20(i) above. He submits that the jury should therefore have been directed that they must be sure the evidence did establish the relevant propensity before it could support the prosecution case. In the alternative, he submits, a direction should have been given which clearly prohibited the jury from relying on any issue of propensity, and which explained precisely what use could properly be made of the bad character evidence against Davies. As it was, he contends, the judge’s directions allowed the jury to use that evidence as propensity evidence, but did not afford the necessary protection of the jury being sure of the relevant propensity, with the result that Davies’ conviction is unsafe.

On behalf of the respondent, Mr Boyce KC submits that the judge’s directions were correct. He argues that the bad character evidence was plainly admissible on the issue of whether either applicant himself had a knife and knew that his co-accused had a knife. At the core of his argument are his submissions that evidence of the applicants’ previous possession of knives was plainly capable of assisting the jury with whether or not either of them was in possession of a knife at the material time; and that the evidence of their joint conviction, and later joint purchase of knife blocks, was plainly capable of assisting the jury with whether either knew that the other was carrying a knife at the material time. In Kawa’s case, the evidence was also relevant to rebut his evidence that he had stopped carrying knives after the 2016 conviction and that it was Nsaka who brought the knife to the scene of the killing. Mr Boyce also points out that the evidence of the 2016 convictions was in any event admissible under s101(1)(g) of the 2003 Act. He goes on to submit that the prosecution at trial were not relying on showing a propensity and that the judge’s directions were correct in law and sufficient in the circumstances of this case.

Although we have summarised the submissions very briefly, we have in mind all the points made by all counsel. Having reflected on their submissions, and on the case law to which they have invited our attention, we have reached the following conclusions.

By s101(1)(d), evidence of a defendant’s bad character is admissible if it is relevant to an important matter in issue between the defendant and the prosecution. Such relevance is the sole criterion of admissibility through this gateway, though the court of course has power under s101(3) to exclude the evidence on grounds of unfairness.

By s103(1)(a), the matters in issue between the prosecution and the defendant include, but are not limited to, the question whether the defendant has a propensity to commit offences of the kind with which he is charged. Section 103(2) goes on to identify different ways in which such a propensity may be established. It is important to note the precise nature of the propensity to which those provisions relates, and to note that it is not the only possible matter in issue for the purpose of s101(1)(d). Past conduct may therefore be relevant to an important matter in issue between the prosecution and the defence, and admissible under s101(1)(d), even though it is not capable of showing a relevant propensity. As was said by this court in R v Bowman and Lennon [2014] EWCA Crim 716 at [64] –

“Section 101(1)(d) is not directed at evidential sufficiency but instead it principally concerns the relevance of the evidence that it is proposed should be introduced, and particularly it focuses attention on the issue of whether the bad character evidence will throw light on the real issue or issues in the case.”

There is a distinction between evidence of a propensity and evidence of past behaviour which makes it more likely that a defendant acted in a particular way on the occasion of the alleged crime, albeit that the distinction may sometimes be a fine one: compare R v Richardson [2014] EWCA Crim 1785 at [20] (a case in which the bad character evidence was relevant to an important issue of identification). On the facts, it is a fine distinction in this case; but the prosecution drew the distinction, as they were entitled to do, and did not seek to prove a relevant propensity. The citation of case law (eg R v CN [2020] EWCA Crim 1028) in which the prosecution had sought to prove a relevant propensity therefore does not assist the applicants.

The question of who brought a knife to the scene was an important matter in issue between the prosecution and both applicants. So, too, was the question of whether either applicant knew that the other was carrying a knife.

Each of the applicants denied having a knife. The only evidence that Nsaka was armed was the evidence of the defendants that it was Nsaka who produced the knife which shortly afterwards killed him: it was no part of the applicants’ cases to say that there was more than one knife at the scene.

The facts of the three matters relied on by the prosecution were not in dispute, and the jury could be sure about them. In our judgement, each of the three matters was relevant to one or both of the important issues. The prosecution had made clear in their application that they were not seeking to prove that either applicant habitually carried knives, still less that either had a propensity to commit offences of the kind with which he was charged. They were seeking to prove that it was one or both applicants, and not Nsaka, who came to the scene carrying a knife. The evidence of the applicants’ previous possession of knives, including criminal offences committed together of possessing knives in public; Kawa’s possession of knives in Davies’ home; and, albeit to a lesser extent, their joint activity in purchasing multiple kitchen knives in blocks, was relevant because the jury could find that it made it more likely that one or both of them brought a knife to the scene, and less likely that it was Nsaka who brought to the scene the knife with which he was stabbed. To put it another way in relation to Kawa, the jury could find that the evidence made it more likely that Kawa brought with him the knife with which he caused the fatal injury, and less likely that he was the victim of an attack by Nsaka who, alone, was armed with a knife. The applicants’ shared activities were relevant because it made it more likely that one applicant would know if the other was carrying a knife.

It does not appear that any evidence was adduced by the defence to show any bad character on the part of Nsaka. If the applicants’ submissions to this court were correct, the jury deciding the important issue of who had a knife would have been denied any knowledge of previous matters relating to the applicants which could shed light on that issue.

We recognise, of course, that even a single previous incident may sometimes be capable of being relied upon as evidence of a relevant propensity: see R v Hanson and others [2005] 1 WLR 3169. In this case, however, the prosecution did not seek to establish a propensity.

In arguing to the contrary, the applicants understandably rely on the judge’s words which we have quoted at paragraph 20(i) above. We accept that the reference to “their habit” suggests a propensity to behave in a particular way. That, however, is contrary to the basis on which the application under s101(1)(d) had been made and granted; is not echoed in any other part of the judge’s directions; and it is not echoed in the words of his intervention, much earlier in the trial, which we have quoted at paragraph 19 above. With respect to the judge, it was an incorrect statement of the prosecution case, though not one to which prosecution counsel objected at the time. In our view, however, that error – affecting a single sentence in the judge’s summing up – did not alter the prosecution case, did not undermine the directions which the judge correctly gave, and did not render the convictions even arguably unsafe.

We note that similar issues arose in R v Okokono and Wilson-Moonie [2014] EWCA Crim 2521. The appellant Wilson-Moonie was accused of being involved in a revenge attack, with one or more knives and bats, on one Ailton, said to be a member of a rival gang. The trial judge admitted evidence of Wilson-Moonie’s previous conviction for possession of a lock knife. That ruling was challenged on appeal, and one of the submissions made on behalf of Wilson-Moonie was that the jury may have thought (contrary to the judge’s ruling) that the evidence was relied on as showing a relevant propensity. In dismissing the appeal against conviction, this court said at [69]:

“It was never the Crown’s case that the appellant had a propensity himself or that he was one of those who stabbed Ailton. The conviction was introduced to demonstrate two things: that he was a party to the attack seeking to revenge the death of his friend in a gang-related feud which he may have triggered, and his realisation that a knife or knives might be used with lethal intent if gang violence broke out.”

For those reasons, we can see no basis on which it could be argued that the judge was wrong to find this evidence admissible under s101(1)(d) as well as under s101(1)(g). Having admitted it, there was no basis for excluding it on grounds of unfairness; and indeed, the applicants do not challenge that aspect of the judge’s ruling.

What, then, of the judge’s directions? We have explained why we do not accept that the safety of the convictions is undermined by the single reference to “their habit”. The principal remaining criticism is that he did not direct the jury that they must first be satisfied that the evidence established that an applicant had a propensity to carry knives. A direction of that kind would certainly have been necessary if a relevant propensity had been alleged: see R v Mitchell [2016] UKSC 55 (a decision of the Supreme Court based on Northern Ireland legislation which was materially the same as the relevant provisions of the Criminal Justice Act 2003), in particular at [44]. But as we have said, that was not the purpose for which the evidence was adduced and not the basis on which it was admitted. It follows that the judge was not obliged to give such a direction. To have done so could only have confused the jury.

We have considered whether, in the circumstances of this case, it was incumbent upon the judge to give an express direction to the jury that they should not treat the evidence relating to all or any of the three matters as being capable of establishing any propensity. We accept that in some cases, such a direction might be necessary. The need for it depends, however, on the circumstances of the individual case; and here, in our view, it would not have assisted the jury for the judge to identify a possible propensity, solely in order to say that it was not alleged by the prosecution and should not be considered.

It is in our view relevant to note that no objection was taken at the time by the experienced leading counsel who represented the applicants at trial. Of course, if the judge’s directions were wrong in law, failure to object at the time could not make them correct; but it is relevant that those who were fully tuned in to the feel of the trial were satisfied with the directions which the judge, equally fully tuned in, had drafted.

We conclude that the directions given were correct in law and sufficient in the circumstances of this case, and that the contrary is not arguable.

We return, finally, to the applications for very long extensions of time. We recognise the practical problems which can face legal representatives who come into a case for the first time after conviction and sentence, and we note that delays were caused by events such as the elevation to the Bench of one counsel who was initially instructed, and by the problems caused by the Covid-19 pandemic. We need not explore these matters in further detail because, for the reasons we have given, the grounds of appeal are not arguable. It follows that no useful purpose would be served by our extending time, even if we could be persuaded to do so.

The applications for extensions of time and for leave to appeal against conviction accordingly fail and are refused.