Kelly, R. v

J U D G M E N T (Approved)

1. LORD JUSTICE BURNETT: On 22 May 2013, in the Crown Court at Stafford before HHJ Tonking (the Recorder of Stafford), the appellant was convicted of murder. He was later sentenced to life imprisonment with a minimum term of 17 years.

2. This is his appeal against conviction brought by leave of the full court. There are three grounds of appeal. First, that evidence of bad character, which had been ruled inadmissible by the trial judge, was introduced by Miss Gould (prosecuting counsel) when she examined in chief Dr T Richardson (consultant psychiatrist on behalf of the prosecution). Secondly, that during her cross-examination of Dr Manny Bagary (consultant epileptologist and neuropsychiatrist) prosecuting counsel elicited from him his opinion that at least on one view of the facts the defence of loss of control was not made out. That was in the face of agreement between the parties, which the judge had endorsed, that the experts, of whom there were many, should not give their view on that matter. Thirdly, that the prosecution failed to give proper disclosure in three respects. The over-arching submission of Mr Joyce QC, on behalf of the appellant, is that these features demonstrate conduct on the part of the prosecution which was so unfair that it has resulted in the conviction being rendered unsafe.

3. The appellant, who is now aged 31, was arraigned for the murder of Brian Townley, which it was said had occurred on 16 or 17 July 2012. He had a co-accused, Thomas Heames. The three men had known each other for some time, perhaps 3 months. All were users of a drug known as Monkey Dust, which is a stimulant which acts in a way similar to amphetamine or cocaine. Mr Townley was 42 years old when he died. He was found dead in the flat belonging to Heames on 17 July. He had been savagely beaten. He had sustained 143 injuries, of which 121 were described by the pathologist as recent. There were substantial injuries to his head and neck, which were the cause of his death.

4. There had been an incident some days before involving the appellant and Mr Townley in which the police were involved, but no charges were brought. Although initially the appellant suggested that it was his co-accused who had inflicted the fatal injuries upon Mr Townley, shortly after the beginning of the trial he recanted and accepted responsibility for causing the death.

5. He also accepted that the violence inflicted upon Mr Townley was unlawful, but raised two issues before the jury. First, his case was that he did not intend to cause really serious injury; secondly, he ran the defence of loss of control found in sections 54 and 55 of the Coroners and Justice Act 2009. A second count of manslaughter was put on the indictment, to which the appellant pleaded guilty.

6. The circumstances in which the loss of control was said to arise were essentially these. As a young man of 19 the appellant had been the victim of a rape, or an attempted rape, late one night when he got into a car which he understood to be a taxi. His case was that as a result he suffered from post traumatic stress disorder (PTSD) and functional non-epileptic attack disorder (FNEAD). The latter condition caused him to suffer frequent epileptic like fits.

7. In the weeks leading up to the killing Mr Townley and the appellant had apparently become good friends. At some point the appellant told him, or gave him sufficient information for him to guess, the nature of the sexual attack some 10 years earlier. The appellant's case, as advanced at trial, was that Mr Townley had taunted him or wound him up about that incident by feigning a sexual interest in him. Mr Townley was a twice married man with nine children. Evidence adduced on behalf of the prosecution suggested that he was exclusively heterosexual.

8. In a very detailed account of the circumstances leading to the killing given by the appellant to Dr Richardson, the appellant described what on their face appeared to be sexual advances by Mr Townley, but himself explained that he thought they were feigned. He repeated evidence to the same effect before the jury. Be that as it may, the taunting by Mr Townley of the appellant was said to have led to his loss of self-control in circumstances where all the requirements for the partial defence of "loss of control" found in the 2009 Act were made out. There was a plethora of expert evidence concerning the question of PTSD and FNEAD. Some of that was directed towards the issue whether the appellant might be able to establish the partial defence of diminished responsibility. However, it was also contended on his behalf that it was relevant to the partial defence of loss of control. Diminished responsibility was not run at the trial.

9. As one would expect, each of the experts questioned the appellant about his background and history. In the course of that questioning the experts enquired about his previous history relating to violence. The appellant had three previous convictions for assaults. The details do not matter for the purposes of this judgment. One of those convictions predated the attack upon him. The prosecution eventually accepted that the sexual attack had occurred. The two other convictions were more recent.

10. The prosecution were also aware of many allegations of violence by the appellant from friends and associates and family members, none of which had been the subject of a prosecution. Indeed many of them had not been the subject of any investigation at all.

11. Following the appellant's acceptance of his factual responsibility for the death of Mr Townley, the murder charge was dropped against Heames. He pleaded guilty to a different offence. It was following those events that the prosecution made their application for the appellant’s bad character to be admitted. It was, as the judge pointed out in his careful ruling, a late application in any event. The judge ruled against the prosecution on that issue and excluded all evidence of bad character.

12. We note that the appellant, whilst of course accepting that he had been convicted of assault on the three occasions, did not accept any of the broad allegations made against him relating to the other incidents. One of the matters of understandable concern to the judge was that the trial would be deflected along a path of satellite litigation concerning a multitude of peripheral issues.

13. It was the contrast between the ruling that all evidence of bad character should be excluded and Dr Richardson's reliance, at least in part, upon that history of violence in forming one of her conclusions which gave rise to the complaint under ground 1. Prosecuting counsel sought to elicit the conclusion shorn of all its evidential foundation.

14. The prosecution argued that evidence relating to the appellant's PTSD and FNEAD was irrelevant to the partial defence of loss of control. As material, section 54 of the 2009 Act provides:

"(1)Where a person (“D”) kills or is a party to the killing of another (“V”), D is not to be convicted of murder if—

(a) D's acts and omissions in doing or being a party to the killing resulted from D's loss of self-control,

(b) the loss of self-control had a qualifying trigger, and

(c) a person of D's sex and age, with a normal degree of tolerance and self-restraint and in the circumstances of D, might have reacted in the same or in a similar way to D.

(2)For the purposes of subsection (1)(a), it does not matter whether or not the loss of control was sudden.

(3)In subsection (1)(c) the reference to “the circumstances of D” is a reference to all of D's circumstances other than those whose only relevance to D's conduct is that they bear on D's general capacity for tolerance or self-restraint."

15. The prosecution contended that the appellant's medical condition, of which the PTSD was disputed, fell within section 54(3) and thus was irrelevant for the purposes of the defence. The judge disagreed. By reference to the features of PTSD described in ICD-10, he concluded that the diagnosis of PTSD was relevant to more than the defendant's general capacity for tolerance and self-restraint. It was relevant to the defendant's circumstances as a whole because it was part of his character. He was also satisfied on the basis of the expert evidence which had been disclosed on behalf of the appellant that PTSD might provide an explanation for what appeared on its face to be a series of partial and contradictory accounts given by the appellant of the events, rather than a careful elaboration as suggested by the prosecution to meet their case.

16. The scope of section 54(3) has not been an issue in this appeal. We emphasise that the point has not been reargued before us.

17. In a shorthand written joint document produced by two of the appellant's experts, they expressed their conclusions about whether each of the factors identified in the statute necessary to found the partial defence of loss of control were established. Unsurprisingly there was concern that were evidence of that nature to be adduced it would trespass into the arena exclusively the domain of the jury and well beyond that of the expert. The issues for the jury are whether the defendant in question in fact lost his self-control, whether there was a qualifying trigger for the purposes of section 55, and whether a person of the same sex and age of the accused with a normal degree of tolerance might have acted in the same way.

18. There was an extensive discussion concerning the proper ambit of the expert evidence in the light of the judge's ruling that it was material to the defence being run by the appellant. There was agreement between the parties, which was endorsed by the judge, that the experts should not be asked whether in their opinion the defence of loss of control was or was not established. All agreed that that was not a matter of expert opinion because it involved an evaluation of non-expert factual matters entirely within the province of a jury.

Ground 1: Bad character

19. Dr Richardson found no sign of PTSD, but considered that the appellant had suffered from anxiety and depression. She thought his psychopathology was complicated by chronic dependence upon illicit drugs. She also thought he had traits of dissocial personality disorder. In paragraph 7.7 of her report she expressed this conclusion:

"I am concerned that Mr Kelly had a previous history of assaulting people when he is angry and this has been an established pattern of behaviour. He has previous convictions for violence. Mr Kelly, in my view, therefore demonstrates that he deals with conflict using physical force."

20. It was the last part of this paragraph that prosecuting counsel sought to get into evidence in the course of the evidence-in-chief. Both before and during the evidence of Dr Richardson there were many references in open court to what were described as "the parameters" of her evidence. That was a guarded allusion to the fact that the judge had excluded evidence of bad character and that all concerned should be careful to respect the ruling.

21. The evidence-in-chief followed very closely upon the content of the report. Having asked about drugs and dissocial personality disorder, to which we shall return, prosecuting counsel referred to “the parameters” and asked:

"Did you consider whether or not Mr Kelly had any established patterns of behaviour?"

Before Dr Richardson had an opportunity to answer the question the judge intervened. It was clear to him, as indeed no doubt it was to many others in court, that counsel was seeking to elicit the conclusion in paragraph 7.7. It was usual for a short break to be taken during the course of the evidence in the morning. The judge deftly contrived to do so at this point. The jury left court. The judge reminded counsel that he had ruled against the prosecution application on bad character and that she must respect that ruling.

22. It is then clear, from the exchanges that followed, that counsel thought that it would be permissible to ask Dr Richardson for her conclusion without establishing the factual premises upon which she had reached that conclusion. The judge sought to explain that such a course was not permissible.

23. There are circumstances in which an expert gives primary evidence of the factual matters upon which the expert opinion is then formed. That is often the case with a pathologist describing injuries found on a deceased and then expressing an opinion about their cause or possible causes. Psychiatrists may rely upon what they observe directly when interviewing a patient, but often depend to a great extent upon information provided to them by the person under examination or background information gleaned from other sources. In such cases the expert opinion is formed upon a series of factual premises which may or may not be correct. If the factual premises are uncontroversial it may well be appropriate to distil the conclusion for a jury and not complicate the evidence with unnecessary and sometimes complex underlying facts. That may be the case, for example, with some DNA evidence but otherwise the facts upon which the opinion is based must be described. If they are controversial they may be challenged.

24. Here the underlying facts were controversial in that the appellant disputed the general allegations of previous violence; but even insofar as the underlying facts were not disputed, that is in relation to the fact of previous convictions, that evidence had been ruled inadmissible by the judge.

25. In the lead up to the question we have identified Dr Richardson was also asked about her conclusion that the appellant showed "some traits of dissocial personality disorder", albeit that she thought the picture was confused by the use of illicit drugs.

26. Mr Joyce submits that the questioning, particularly when seen in the context of Dr Richardson's earlier description of the speciality of a forensic psychologist as being to deal with violent patients and prisoners, gave rise to an obvious risk. The risk was that the jury would infer that there were matters in the appellant's background about which they had not been told, but which the prosecution were hinting at. We note that he was also critical of an attempt by prosecution counsel during her cross-examination of Dr Bagary to revisit the issue. She raised the matter in the absence of the jury and received a firm answer from the judge whose patience, whilst by that point sorely tested, was commendable. He said, "I am sorry but I am not going to reopen bad character".

27. Mr Bennetts QC for the prosecution, who did not appear below, submits that nothing was admitted that should not have been.

28. It is regrettable, to put it mildly, that counsel prosecuting at the trial did not appear to appreciate that the ruling on bad character impacted upon the extent to which it was appropriate to explore the full amplitude of Dr Richardson's expert conclusions. Even more so that after a clear reminder from the judge during that evidence she came back to the topic in a different context during Dr Bagary's evidence. However in our judgment, Mr Bennetts is right. The quick action of the judge resulted in no harm being done with respect to the content of paragraph 7.7.

29. We do not consider that the evidence relating to dissocial personality disorder could give rise to the inference suggested by Mr Joyce. There was evidence other than the appellant's previous alleged or accepted violence, which could support such a conclusion, namely his description of the attack itself, the previous incident involving Mr Townley, and also much of his family history that was described in some detail by his own experts. The judge summed up this part of the evidence. No criticism was made of the summing-up and it clearly demonstrates that the judge did not consider that that part of the evidence was inadmissible.

30. Furthermore, we do not consider that a general description from a forensic psychiatrist of the specialist role of that discipline of medicine could give rise to, or provide any support for, an inference of the sort contended for on behalf of the appellant. It should not be forgotten that the basis upon which the expert evidence was eventually given was that the appellant accepted that he had launched a vicious fatal attack upon Mr Townley, in the context of a defence which suggested that, at the time, he was out of control.

31. In our judgment, the judge dealt with this problem admirably. We note that no submission was made to the judge that more should be done in respect of this matter, or that any harm had arisen in consequence of this undoubted lapse by prosecution counsel.

Ground 2: Loss of Control

32. As we have indicated, the appellant gave conflicting and evolving accounts of events in the period between his arrest and trial. He did so not only in providing a prepared statement at interview, much of which he later contradicted, but also in his several accounts to the various experts. In the course of her cross-examination of Dr Bagary, prosecution counsel wished to place before the jury the different accounts that the appellant had given to him at various times. In his reports Dr Bagary dealt with a number of those accounts. One at least was not supportive of the suggestion that there was a loss of control. In addition to eliciting the appellant's account, prosecuting counsel then asked:

"And your conclusion was he doesn't have a defence of loss of self-control? "

Dr Bagary answered:

“Not based on what he told me with that description."

The judge immediately intervened and said:

"Well, whether or not he does is not a matter for any expert witness, it is a matter for the jury."

The following exchange then took place with the jury still in court.

"MISS GOULD: I agree, but in your report, whether it be a matter for the jury or not, I accept of course your Honour it is, you have expressed the opinion --

JUDGE TONKING: No, I am sorry, Miss Gould. I am going to say this. We have discussed this in the absence of the jury and it has been decided and indeed agreed, the elements of loss of control which are susceptible to expert evidence and the elements which are not.

MISS GOULD: I agree.

JUDGE TONKING: And it is agreed that no expert can give an opinion as to whether they think there is loss of control or evidence of loss of control because that is a question for the jury.

MISS GOULD: I agree completely with that, your Honour. That's not the point I am seeking to make. If you will allow me, I hope I can show you the point.

JUDGE TONKlNG: I won't allow you. I am very sorry. The parameters are clear. You are drawing out from the witness his opinion of whether or not there was evidence of loss of control.

MISS GOULD: No, your Honour, I am not seeking to do that.

JUDGE TONKING: Will you give us a moment, ladies and gentlemen, please. I am very sorry."

The jury then went out. The discussion continued:

"JUDGE TONKING: I suspect that the witness doesn't need to leave for this discussion.

MISS GOULD: He does, please. Yes, he does.

JUDGE TONKING: Dr Bagary, I am sorry. Thank you.

...

JUDGE TONKING: Yes.

MISS GOULD: I am not seeking to draw out what your Honour thinks. What I am seeking to draw out is firstly having then acquired fresh facts by the process that he then goes through, which is the further questioning, there is a change of position, that is one aspect and, secondly, that this is

contained in his report when we have had evidence from Mr Kelly that he has had the various psychiatric reports and the fact that the doctor has indicated that in the first instance and then there has been an alteration in position for Mr Kelly.

JUDGE TONKING: Well, you can put that but you must not refer, as you have done, to what Dr Bagary's opinion was.

MISS GOULD: Forgive me. I hope your Honour understands the point I am seeking to make.

JUDGE TONKING: I do understand.

MISS GOULD: If I have done it clumsily then I am sorry.

JUDGE TONKING: I do understand, now that you make it clear. The difficulty was you just directly elicited what Mr Bagary's opinion was or at that time and that is not permissible. It's as simple as that.

MISS GOULD: Forgive me.

JUDGE TONKING: Right, can we have the witness back and can we have the jury back."

33. Mr Joyce then intervened and drew the judge's attention to a further problem. That problem was that the opinion elicited by prosecuting counsel was only part of the story because later in the report Dr Bagary had expressed his view that the nature of the attack admitted by the appellant, if it was accepted by the jury, indicated a loss of self-control when sexually charged comments were made to him. Prosecuting counsel then made clear that she had indeed wished to introduce evidence from this expert witness that on the early account given by the appellant the defence of loss of control would not run. We note that the early account preceded the appellant's acceptance of his responsibility for causing the death.

34. Prosecuting counsel appears to have been blind to the acceptance all round that no such course should be followed.

35. The judge reminded her that the doctor's opinion on whether the legal defence was made out was irrelevant. A discussion ensued about how to sort out the problem which concluded in these terms:

"JUDGE TONKING: I think what I had better do is when the jury come back I had better explain to them that the opinion of Dr Bagary or any expert as to whether or not Mr Kelly lost control is irrelevant and in fact is inadmissible.

MISS GOULD: Yes.

JUDGE TONKING: But the question was asked and his opinion was elicited because a point is being made by the prosecution about an alleged change of account and it only goes to that and whether or not Mr Kelly lost control is not for any expert, it is for the jury.

MISS GOULD: Absolutely.

M R JOYCE: No, it doesn't go far enough. The opinion has been elicited.

JUDGE TONKING: Yes.

MR JOYCE: It has been elicited in absolute contradiction of the agreement and your Honour's ruling.

JUDGE TONKING: I follow that.

MR JOYCE: And it must be corrected specifically. That was Dr Bagary's opinion at one stage but in fact it's not a total change of story that is being referred to here either in this report if we can look at it. What is being adduced here is Dr Bagary saying the account is not hugely changed. It changes later. This is being asked about in terms of four to five blows to Mr Townley. "It is my opinion that Mr Kelly did experience a loss of self-control when he punched Brian Townley four to five times". This is towards the end of that self-same report and it has to be put as plain as a pikestaff. It shouldn't have been asked. The concluding opinion was this: Ignore both of them otherwise it is unfair.

JUDGE TONKING: I rather gathered and implicit in what I said was that the concluding opinion would have to be given as well to address the balance.

MR JOYCE: Absolutely.

JUDGE TONKING: It must be given as well to address the balance."

The jury then returned to court and the judge immediately gave them a direction:

JUDGE TONKING: Members of the jury, there is a problem which I am going to try to put right. Questions of fact are for you and you alone. They are not for me, they are not for the lawyers, they are not for the expert witnesses. It is as simple as that.

One of the questions of fact which you will have to decide in this case is whether or not Mr Kelly lost control and it is for you to decide on all of the evidence. It follows from that that whether an expert or anybody else at any stage thinks that there is evidence which does or does not support loss of control is totally irrelevant and for that reason the experts were not to go there.

Miss Gould has elicited that at one stage, having heard an account from Mr Kelly, Dr Bagary was of the view that there was no evidence with supported loss of control -- inadmissible, irrelevant. Miss Gould's purpose, and I say this to be fair to her, was not to just get that out before you. The purpose is that Dr Bagary changed his opinion and he did so having had further discussion with Mr Townley (sic) Mr Townley having given him further and other information. That is the point of the question. It is right that you should know that Dr Bagary was of the opinion that has just been stated but at a later stage he changed his view about loss of control to a degree by saying that there was loss of control, though he opined that it was not complete.

I tell you all that so that you know the full picture and you know why the question was asked but I must stress the opinion of any expert as to whether or not there was loss of control is totally irrelevant. It is a question for you to answer and it's a factual question and so the opinion of somebody else is not to the point. It is not to the point at all and that applies whether the opinion is yes or no or maybe. It just is not relevant. That is why I intervened when I did and that is why we have had the discussion and that is why I am giving you this direction now so that it is absolutely crystal clear, and I can see from you nodding that you understand and have taken the point. Thank you.

Now may we have the witness back, please."

36. Mr Joyce takes no point on the judge mistakenly referring to Mr Townley on two occasions during the course of that direction. It was obvious that he was referring to various accounts given by the appellant. In our view, the judge explained to the jury in the clearest of terms that an expert's opinion on whether the defence of loss of control was made out was irrelevant, and also put right the partial picture that had emerged, as it was explained to him by Mr Joyce, by stating that Dr Bagary had come to a different conclusion elsewhere in his report. Both those conclusions were thus in one sense before the jury, then HHJ Tonking told the jury to ignore that evidence altogether.

37. No submission was made to the judge that his corrective action was inadequate, or that at that stage or later the jury should be discharged. Mr Joyce has explained that at the time that these events unfolded his focus, and the focus of his team, was on keeping the trial going in circumstances where the appellant was having repeated fits. That may well be so, but it is nonetheless a striking omission that there was no contemporary complaint of the nature now advanced.

38. Mr Joyce submits that because loss of control was the crucial issue in this trial, the introduction of the evidence by the prosecution "must make the conviction unsafe". We do not agree. We accept that what happened was more than unfortunate and led to Dr Bagary's inadmissible conclusion on the availability of the defence on one set of facts being placed before the jury. Mr Joyce has also submitted that even that was a partial account of the evidence found in the relevant part of his report. Nonetheless, it is clear that after careful consideration of submissions advanced on behalf of the appellant, which the judge accepted in full, the direction he gave neutralised any harm because the jury were told that Dr Bagary came to a different conclusion elsewhere in his report and, in any event, that the jury should ignore his opinion or opinions on this issue.

39. We would add that in his summing-up the judge dealt meticulously with the question of loss of control and produced a route to verdict for the benefit of the jury, which was agreed by the parties.

40. Whilst we understand fully the frustration felt both by the appellant and his team and indeed the judge at the turn of events, the issue surrounding Dr Bagary's evidence provides no basis for a conclusion that the conviction is unsafe.

Ground 3: Disclosure

41. Mr Joyce complains, on behalf of the appellant, in relation to disclosure and what are called "general matters". They are distilled in paragraph 48 of the advice and grounds upon appeal and are in these terms:

"In general terms, apart from the above, the prosecution of this case was characterised by unfairness in particular with reference to the following:

(a) Failure to produce a chronology to the defence relating to their failed application to extend custody time limits. This was despite repeated requests and was only produced during the course of the hearing when it became apparent to the defence that there was such a document in the prosecution's possession which they had failed to disclose.

(b) Informing the defence that the pocket notebook entry of PC Slater was not disclosable material. When these were finally disclosed after repeated requests it became plain that the document clearly assisted the defence and should have been disclosed earlier. The statement of PC Slater and the notebook require examination to see the point.

(c) Cross examination of Mr Kelly in relation to the use he had made of the deceased's phone (in the possession of the defendant)... This was based upon telephone examination reports in the unused material not disclosed to the defence let alone served as evidence upon which the prosecution relied. To understand the significance of this it is important to consider the defence submissions in relation to this line of cross-examination...

'the problem I have is this, I don't know and I don't know because since material on the face of it quite clearly has been used that has been marked clearly not disclosable, I don't trust the rest of it.'"

The last observation encapsulates an argument advanced by Mr Joyce to the judge and also before us that the problems with disclosure have led to a general lack of trust.

42. We note that the approach to questions of non-disclosure in criminal trials was settled by the Supreme Court in McInnes v Her Majesty's Advocate [2010] UKSC 7 relying upon earlier domestic authority, which in turn took full account of Strasbourg jurisprudence on Article 6 ECHR. In paragraphs 19 and 20 of his judgment Lord Hope identified two questions that fall to be considered in disclosure cases. The first is whether material under consideration should have been disclosed and the second is the consequence of any failure to disclose. As to the second he said:

"The test that should be applied is whether, taking all the circumstances of the trial into account, there is a real possibility that the jury would have arrived at a different verdict."

Both Lord Walker and Lord Kerr agreed with Lord Hope. Lord Rodger at paragraph 30 and Lord Brown at paragraph 35 expressed the test in slightly different terms, but to the same effect.

43. The first complaint identified under this ground does not in fact touch upon the evidence adduced at trial. It is nonetheless a fair criticism of the prosecution. The second relating to PC Slater's notebook gave rise to no observable difficulty at the trial, although unquestionably it was disclosed late. As to the third, it is correct that prosecuting counsel asked the appellant questions about whether he used the deceased's phone and indeed Heames' phone in the weeks before the killing. It is also right that phone records were not disclosed before the questions were asked.

44. Mr Joyce submits that the questions were asked in a pointed way suggesting that the appellant had something to hide. The questioning came as a surprise to Mr Joyce and his team, given the absence of any disclosure of phone records. After the cross-examination had concluded Mr Joyce's team set out a series of written questions for the prosecution directed towards establishing the evidential foundation for these questions. In particular, they sought to understand whether phone records, which had been marked as "Not for Disclosure", had been used.

45. There was no response to the questions from Mr Joyce. It appears the prosecution lost the piece of paper on which the questions had been carefully set out. Thus it was in argument the next day that the issue was ventilated orally. Prosecuting counsel accepted that the foundation, at least in part, was to be found in the undisclosed records relating to what was described as an "unattributable phone". Some of the questions appear to have been merely speculative. There can be no doubt, as it seems to us, that having determined to ask the questions and used the records to craft some of them, that there was a duty to disclose those records before the questions were asked. Indeed, it would be necessary to disclose the records in time to give the defendant's team an opportunity to consider them and take instructions upon them.

46. Mr Bennetts did not seek to suggest otherwise. In our view, this was a serious lapse, alas one of many in the conduct of this prosecution. However, it is necessary to stand back and look at the quality of this evidence and its relevance. In our judgment, the exchanges were harmless and the judge rightly considered that part of the cross-examination to be irrelevant. A reading of the transcript leaves no real clue as to why this line of questioning was pursued at all. In a long and difficult trial this was perhaps the most peripheral of peripheral evidence.

47. As we have noted, these discrete issues, collected together under ground 3, support a submission from Mr Joyce that this was not only a poorly but unfairly conducted prosecution, particularly when viewed in the light of the earlier complaints under grounds 1 and 2. He has drawn our attention to a number of observations made by the trial judge during the course of the trial, which were critical of the way in which the prosecution was conducted. We can well understand why the judge said what he did. But criticism of the prosecution, even if fully justified, does not necessarily lead to the conclusion either that a conviction is unsafe, or that a trial was unfair.

48. We have dealt thus far with the individual complaints seriatim, nonetheless we have stood back, as Mr Joyce invited us to do, and asked ourselves whether the conviction is unsafe or whether the appellant was denied a fair trial. In our judgment the answer to both questions is "no".

49. Mr Joyce advanced legitimate criticism about the way the prosecution conducted itself. The judge too was, in our view, rightly critical on a number of occasions. Both he and Mr Joyce displayed commendable patience during the course of this trial, but the lapses by the prosecution in this case were not such as, in our judgment, undermined the conviction. In those circumstances this appeal must be dismissed.