R v BWP [2024] EWCA Crim 113

WARNING: reporting restrictions may apply to the contents transcribed in this document, particularly if the case concerned a sexual offence or involved a child. Reporting restrictions prohibit the publication of the applicable information to the public or any section of the public, in writing, in a broadcast or by means of the internet, including social media. Anyone who receives a copy of this transcript is responsible in law for making are that applicable restrictions are not breached. A person who breaches a reporting restriction is liable to a fine and/or imprisonment. For guidance on whether reporting restrictions apply, and to what information, ask at the court office or take legal advice.This Transcript is Crown Copyright.  It may not be reproduced in whole or in part other than in accordance with relevant licence or with the express consent of the Authority.  All rights are reserved.[2024] EWCA Crim 113Venue Royal Courts of JusticeThe StrandLondonWC2A 2LLDate Friday 9 th February 2024B e f o r e:LADY JUSTICE MACUR DBEMRS JUSTICE STACEY DBEHIS HONOUR JUDGE PICTON(Sitting as a Judge of the Court of Appeal Criminal Division)____________________
Between
R E XCrown- v -
Between
B W P____________________Computer Aided Transcription of Epiq Europe Ltd,Lower Ground Floor, 46 Chancery Lane, London WC2A 1JETel No: 020 7404 1400; Email: rcj@epiqglobal.co.uk (Official Shorthand Writers to the Court)_____________________Mr J Price KC and Mr S Wilshire appeared for Applicant CrownMiss J Dempster KC and Miss N Roberson appeared for Respondent Defendant____________________J U D G M E N T
[1]This case is listed as R v BWP. The provisions of section 71 of the Criminal Justice Act 2003 apply to these proceedings. By virtue of those provisions, no publication may include a report of these proceedings until the conclusion of the trial.[2]The respondent (BWP) faces an indictment alleging one count of murder, contrary to common law. His trial started on 4th January 2024. At the close of the prosecution case, on 26th January 2024, the judge acceded to the submission of no case to answer.[3]Following the ruling, the prosecution sought, and was granted, an adjournment until the next business day to consider an appeal against the judge's decision, pursuant to section 58 of the Criminal Justice Act 2003.[4]On 29th January 2024, the prosecution gave notice of its intention to appeal and gave the undertaking, pursuant to section 58(8) of the Criminal Justice Act 2003, that the respondent would be acquitted if either of the conditions mentioned in section 58(9) of the Criminal Justice Act 2003 were met: that is either that leave to appeal to the Court of Appeal was not obtained or the appeal was abandoned before determination by the Court of Appeal.[5]Deputy Circuit Judge Cahill KC ruled that the application should be expedited. The jury have been retained and the case adjourned pending the outcome of this application.[6]The Registrar of Criminal Appeals has referred the application for leave to appeal, with the appeal to follow if granted, to the full court. We grant leave.

The Facts

[7]The facts of this tragic case can be stated shortly. The respondent’s wife was, at the time of her death, 78 years old. She suffered from dementia. The respondent (then aged 81) was her carer. He had previously reported that it was becoming increasingly difficult to care for the deceased.[8]Shortly before ten o'clock in the morning on 4th January 2022, the respondent phoned his GP's surgery in a distressed state. He told them that he had hit his wife and that he might have killed her. The police were called and attended, where they found the deceased lying unconscious on the floor. She had an injury to her head and two separate pieces of cloth had been stuffed into her mouth. She was taken to hospital but did not regain consciousness and the decision was subsequently made to terminate her life support . She died on 7th January 2022.[9]A post-mortem examination revealed that the deceased had a laceration to the top of her head and a bruise around her left ear, which could be explained by being hit by a saucepan. There was bruising and abrasions to her upper lip. The pathologist concluded that the most plausible explanation for the cause of death was cardiac arrest caused by smothering.[10]The respondent was arrested and interviewed. He accepted that he was responsible for what had happened to his wife. He said that an argument had developed. He had walked from the kitchen into the bedroom with a saucepan and drying cloth in his hands. The next thing that he remembered was being astride his wife on the floor, that she had the drying cloth stuffed into her mouth, and that his hands were pressed over her face and mouth. She appeared to be unresponsive and had stopped breathing, as a result of which he had called the GP’s surgery.[11]In the first Defence Statement served on his behalf in accordance with section 5 Criminal Proceedings and Investigation Act 1996, it was indicated that the respondent would enter a plea of guilty to manslaughter by reason of diminished responsibility. The prosecution indicated that it did not accept the plea, albeit that we are told that a review of the evidence given at trial has led to a reappraisal of that stance. In any event, it was withdrawn prior to the amended Defence Statement in which non-insane automatism was raised on the basis of the conclusions of Dr B.[12]Therefore, the respondent's primary defence at trial was one of non-insane automatism; alternatively, whether the partial defence of diminished responsibility had been made out in the case.[13]The submission of no case to answer “focused exclusively” on the evidence of four "experienced consultant psychiatrists" who had given evidence: three for the defence, Dr B, Dr A and Dr F; and one for the prosecution, Dr S.[14]The psychiatric evidence was summarised by Miss Dempster KC and Miss Roberson, who appeared on behalf of the respondent at trial, to be that Dr B's evidence favoured the defence of automatism. The prosecution conceded that an evidential basis for the defence of automatism had been made out.The other three doctors conceded that it was possible that the respondent had acted in a state of automatism at the relevant time, although they did not think it likely. Consequently, no jury could convict the respondent of murder,[15]Miss Dempster cited R v Brennan [2014] EWCA Crim 2387, and R v Golds [2016] UKSC 61 in support of the principle that, absent any evidence that would entitle the jury to disregard the expert opinion, a verdict contrary to that opinion would not be a true verdict in accordance with the evidence.[16]Mr Wilshire, who appeared on behalf of the prosecution at trial, resisted the submission, arguing that the state of the psychiatric evidence given by the four doctors was not in accord. When taken individually and as a whole, all three doctors other than Dr B, were in disagreement with her opinion. There was no other evidence other than Dr B’s evidence to suggest that the respondent was in a state of automatism at the relevant time. Whilst each doctor had accepted the theory or argument advanced by Dr B as possible, each maintained their opinion that there were no grounds for automatism to apply in this case. Therefore, there was sufficient evidence upon which the jury could be satisfied that the Crown had made them sure that automatism did not apply.[17]Mr Wilshire cited R v Gian and Modh Yusofff [2009] EWCA Crim 2553, for the principle that the mere fact that a matter of scientific certainty is not possible to rule out a proposition consistent with innocence does not justify withdrawing the case from a jury since juries are required to consider expert evidence in the context of all other evidence and make judgments based upon realistic, and not fanciful, possibilities. Further, he cited R v Hookway and Noakes [2011] EWCA Crim 1989, for the principle that the fact that each of the experts held a different opinion and therefore actually disagreed with each other, without stating specifically that the other was wrong, nevertheless meant that it was open to the jury to consider the expert evidence and to place what they considered to be appropriate weight on the opinions of the experts.[18]The prosecution also sought to rely upon previous situations related by the respondent when he had been provoked by his wife but had not reacted as he had done on this occasion and his acknowledgement in interviews that he had been angered by his wife's behaviour on the morning of the fatal assault.[19]Miss Dempster KC responded that the evidential burden placed upon the defence had unequivocally been discharged and it was therefore a case whereby the jury must be made sure by the prosecution that automatism was not at play. She submitted that the jury should not be put into a position that they "attempt to make themselves into amateur psychiatrists".[20]In her ruling delivered the following day, the judge noted that the prosecution had accepted that the defence of automatism was raised on a proper evidential basis – that is by the report and evidence of Dr B – and had accepted that, as a result, they bore the burden of proof to make the jury sure to the criminal standard that the defence of automatism did not apply. The judge regarded the prosecution's submissions, to which we have referred, to be the “wrong way” of looking at the case. Dr B's evidence provided an evidential basis of automatism. She agreed with the defence that the first question for the jury in any route to verdict would be:
"Are you satisfied so that you are sure that at the relevant time, on 4th January 2022, the prosecution has proved that the [respondent] was not acting in a state of automatism?"
[21]The judge noted the background facts of the case, including the respondent's physical care for his wife in difficult circumstances and in the absence of appropriate support, despite his continuing best endeavours. He had been diagnosed by his GP with stress in September 2020. The judge said:
"[The] caring burden on him as a man of 81 was enormous. None of that evidence is challenged by the Crown, and indeed much of it is called as part of their case. All of it is relevant to the level of stress he was under, leading to the diagnosis of depression by the psychiatrists."
She determined that the respondent's depression was also relevant to the defence of being in a state of automatism, although only one factor of that. She also noted that all the psychiatrists agreed that the respondent suffered a genuine amnesia of "the material time".[22]The judge then quoted from parts of the psychiatric evidence from Dr A, Dr B, Dr F and Dr S that had been called before her. Having done so, she referred to the authorities that had been cited to her and concluded:
"The prosecution evidence in this case took only two and a half days. … There was no evidence called by the Crown as to the actual circumstances of the death … … … The evidence is that automatism is something which cannot be predicted; it can be temporary and it may never recur. Whether [the respondent] was or was not in a state of automatism on another occasion is not relevant to whether he was in a state of automatism on the 4th January. … Mr Wilshire relies on accounts given by [the respondent] on the day of his wife's death. … There is a grave danger in relying on anything said by [the respondent], not for the conventional reason that he might be lying or fabricating, but because the experts agree he has no memory of events. … … … in normal circumstances, [if] there are changing accounts and the weight to attach to it would be a matter for the jury, as would the person's demeanour, but in this case, analysis of both statements and demeanour has once again been part of the psychiatric evidence, and even Dr [S], who relied on an interpretation of [them adverse] to [the respondent] says that the state of automatism is possible. … For the defence of automatism to be rejected by a jury, it must be the case that a reasonable jury properly directed is sure that it does not apply. … the burden of proof lies on the Crown to prove that the [respondent] was not acting in a state of automatism to the higher standard. Automatism is very rare and juries should not, themselves, become amateur psychiatrists. In the absence of other cogent evidence, a jury could only reach a conclusion that they are sure the [respondent] was not in a state of automatism if there is psychiatric evidence upon which they can rely to be sure that it does not apply. Such evidence could only be from a psychiatrist who was himself or herself sure that it does not apply, because in a case such as this, which depends entirely on expert opinion, if the psychiatrist cannot be sure, then neither could a jury. In this case, none of the psychiatrists have said that they are sure the [respondent] was not acting in a state of automatism. They have all said at the very least it is possible. This is not a case where the psychiatric evidence could be buttressed by other evidence which would allow a jury to move from a psychiatrist unlikely or not probable to their own state of being sure, because there is simply no other such evidence adduced by the prosecution."
Consequently and conscientiously, according to her reasoning, the judge acceded to the submission of no case to answer. The Grounds of Appeal[23]In their original written submissions to this court, the prosecution contended:(1) The judge erred in her ruling. She had failed to consider the totality of the evidence from the experts. She was wrong in her analysis of the evidence and therefore reached a conclusion that in all the circumstances of the case was unreasonable. There was clear evidence from Dr F from which the jury could conclude that the respondent had not lost control on the morning of 4th January 2022. Further, Dr S's evidence had not been given sufficient weight.(2) The judge erred in finding that there was no other cogent evidence which would have supported the prosecution's case. She incorrectly dismissed two previous incidents of aggression by the deceased towards the respondent as being irrelevant. These incidents had not triggered an episode of automatism and it was a relevant consideration for the jury when assessing whether the incident on 4th January – an arguably less serious incident – had been the trigger for an episode of automatism.(3) She incorrectly dismissed matters raised in the respondent's police interview. Consideration as to its truthfulness should have been a matter for the jury to consider.(4) She incorrectly conflated the issue of the respondent having no memory of the act, but the issue is memory in relation to the build up to the incident. There was evidence to suggest that the respondent did have memory of the matter and it was the prosecution case that he had given conflicting accounts in relation to this.(5) The judge incorrectly rejected the prosecution's submission that the respondent's changing accounts of demeanour after the incident should be considered as other evidence. This was contrary to the approach taken by the court in R v Hussain [2019] EWCA 666.[24]The respondent entered a Notice of Opposition. On his behalf it was indicated:(1) The judge had approached the submission of no case to answer correctly, both in terms of the facts of the case and the relevant case law. Her ruling was reasoned and detailed. As the respondent had discharged the evidential burden, she correctly approached the issue for the jury, namely, whether there was evidence which showed that the respondent was not in a state of automatism. She took considerable time to consider all of the psychiatric evidence and did not unfairly focus on certain aspects of the evidence. She properly considered their initial conclusions as well as their later evidence, and noted that it was the totality of the evidence that needed to be considered.(2) All of the experts agreed that automatism was possible. The judge was perfectly entitled to reach the conclusion that she did, namely that there was no other evidence upon which the jury could conclude that the respondent was not in a state of automatism. It could not be said, therefore, that the judge's ruling was unreasonable or wrong in principle and the application should be refused.[25]Subsequently, Mr Price KC has been instructed on behalf of the prosecution to pursue this appeal together with Mr Wilshire. He has made further written submissions, to the effect that the judge's ruling was wrong in law. We have refused to hear him on those submissions. This was not the basis of the appeal and runs counter to the concession made by Mr Wilshire, whether it subsequently transpires to be mistaken or not. We agree with Miss Dempster KC that if it was in issue then the judge should have been asked to rule upon it. She was not. Accordingly, we are unequivocally of the view that Mr Price should not be allowed to "piggyback" quite a different basis of submissions into this appeal.[26]Additionally , however, Mr Price KC makes two points in support of the original grounds of appeal. He submits that there was other evidence besides the psychiatric evidence before the jury which could go to the issue of the defence raised, First, the evidence relating to the orderly scene of the fatal assault shortly after it had occurred. He submits that it would be possible for the jury to conclude that the lack of any disruption was at odds with the suggestion made by Dr B that the respondent had acted ‘completely out of control’. Secondly, he cites R v Coley [2013] EWCA Crim 223 for the proposition that the jury were entitled to look at the primary evidence and the quality of the acts of the accused in the context of the automatism defence .[27]Miss Dempster KC and Miss Roberson responded to the further written submissions made by the Crown, taking issue with the attempt to broaden the basis of the appeal. As indicated above, we accept that Mr Price KC is not entitled to make submissions in this appeal contrary to the prosecution case at trial. We therefore do not address the specific procedural points upon which Ms Dempster KC and Ms Robeson relied.[28]However, the written response they prepared gave useful insight into the trial process . That is, written legal directions were given to the jury prior to the psychiatrists giving evidence. Those legal directions were discussed with counsel. Mr Wilshire suggested amendments to the draft legal directions to insert reference to the necessity for an ‘external’ factor, but made no submission that an external factor could not comprise an acute stress reaction which then brought about a dissociative state.[29]Dr B's evidence was that, following a prolonged period of extreme stress which culminated in the events of 4th January, an acute stress reaction had manifested itself into a dissociative response in which the respondent had lost voluntary control over his thinking and his behaviour. It is said that the literature describes this phenomenon as a "psychological blow", so that a brief period of dissociation in response to an extreme trauma, which was unlikely to recur, was capable of distinguishing those cases to which Hennesy (1989) 89 Cr. App. R 10 referred. The legal directions were amended as suggested and no further opposition raised by the prosecution.

Discussion

[30]This is undoubtedly a most sensitive case on its facts and a most difficult case in terms of the primary defence of non-insane automatism which is advanced on behalf of the respondent. We have read the reports of the psychiatrist and the transcripts of their evidence. We have also read with care the transcript of the submissions of no case to answer.[31]The comparative brevity of the judgment that follows is not to suggest that the judge gave anything other than anxious consideration to the submission. It is clear that she did so. That is demonstrated in aspects of her written ruling in which she sets out the reasons why she acceded to the submissions. Nevertheless, for the reasons that we indicate hereafter, and with respect, we disagree with the judge’s reasoning. We reverse the ruling that there is no case to answer. We intend to direct the discharge of the jury and order a fresh trial to take place in the Crown Court for the offence of murder.[32]The ruling under appeal is that there is no case to answer. We agree with Miss Dempster that, whatever the merits of Mr Price's additional submissions regarding the defence of sane automatism, no prosecution submission was made to that effect and there is no such ruling under appeal. That does not mean that the prosecution may not advance any additional, pertinent arguments to those previously canvassed in support of the extant appeal. In so far as the written submissions of Miss Dempster and Miss Roberson suggested a procedural bar to entertaining such further arguments, we reject those arguments.[33]We note that the time limits imposed for the making of an application for leave to appeal a terminating ruling to this court and all statutory and procedural requirements have been observed: see CPR 38. CPR 36.3 provides that we may allow a party to vary any notice that the party has served. We have refused that variation in relation to the Notice of Appeal, but the court regularly gives parties permission to rely upon amended skeleton arguments. No prejudice has been caused to the respondent. Miss Dempster and Miss Roberson have had the opportunity to respond to the issues raised and we are satisfied that they have responded qualitatively. We have not been provided with, and do not have access to, a transcript of the discussions held between the judge and counsel prior to the psychiatrists giving evidence. But we are grateful to Miss Dempster for the further information with which we have been provided today as regards the progress of the trial as indicated below.[34]The trial commenced on 4th January 2024. The prosecution evidence endured over two and a half days. There was then a break in the proceedings before the court resumed. Discussions as to direction on the law, and therefore the law itself, took place on Tuesday 9th January and Wednesday 10th January. Miss Dempster indicates that, at the outset of the trial, she had provided a document, which was further modified throughout the trial, setting out her submissions as to the relevant law on the defence of automatism. She had made clear that the defence relied upon sane automatism and everyone was aware of that fact. While we take note of that explanation , we are somewhat surprised that the discussion as to the law was not lengthier. This is a very complex issue. As has been confirmed to us today, there was no discussion as to the distinction to be drawn between "sane" and "insane" automatism. We are surprised at this news.[35]What we do derive from the latest written submissions of Miss Dempster KC and Miss Roberson (at paragraph 24) is that the initial draft directions sent by email from the judge to counsel obviously did not refer to the necessity for an ‘external factor’ in sane automatism. It is agreed that Mr Wilshire therefore sought modification of the draft legal directions. His submission was acceded to. In their latest written response to Mr Price KC’s further submuissions, Miss Dempster KC and Miss Roberson make clear that they provided the judge "with a number of authorities in which automatism had featured (including all of the authorities now relied upon by [Mr Price], as well as the relevant extracts (on automatism) from each of the four psychiatric reports, which were consolidated into a single document. … This document was intended to offer assistance to the court with what is undoubtedly a complex area of the law and, in particular, to assist the drafting of the preliminary legal directions."[36]Paragraph 25 of the same document reads:
"Crown counsel suggested amendments to the Legal Directions (including paragraphs 9 & 10) which refer to an 'external factor' but no submission was ever made by the Crown that the 'external factor' could not properly arise from the evidence that the Respondent had most likely suffered an Acute Stress Reaction which had brought about a dissociative state. …"
[37]These paragraphs do not reassure us that the important distinction between "sane" and "non-insane" automatism was sufficiently well rehearsed, whether in the documents or indeed in oral submissions before the judge. In this respect we suggest that paragraphs 5.38 to 5.53 of the 2013 Law Commission Discussion Paper: Insanity and Automatism, underline the particular difficulties in the case law and the categorisation, particularly in circumstances such as these.[38]The resultant directions, which now appear in the Digital Case System, are said to be a faithful copy of those supplied to the jury. It has been necessary for the purposes of this appeal for Miss Roberson to amend the draft, in accordance with the directions that were provided. These directions can be found on the DCS at CACD Y1592. They read:
"The Defence of Automatism 8. The defence have raised this and it is the prosecution who must make you sure that it does not apply. What that means is that unless you are sure [the respondent] was not acting in a state of automatism you must acquit him of murder. 9. A person acts in a state of automatism if, at the time they commit an act, the act is not voluntary. An act is not voluntary if at the time it is committed the person committing it has suffered a complete loss of voluntary control. The factor(s) leading to the complete loss of voluntary control must be external, in other words it must not be something that is or could be under the person's control. 10. In this case, the defence say the external factor which led to the [respondent] acting in a state of automatism was the extreme, full-time caring role he was undertaking in respect of [the deceased], who was suffering from Alzheimer's disease. This care included aggressive and ungrateful reactions to his care, to which was later added the additional stress of the positive test for Covid, meaning [the deceased] could not go to a Care Home for at least another three weeks. As a result of the extreme stress brought about by caring for his wife, [the respondent] developed depression. You will hear evidence from a number of psychiatrists about [the respondent's] mental health at the time (including his depression). Their evidence is central to the key issue of whether the condition of his mental health led, as the defence say, to a complete loss of voluntary control. (Lest there be any possible misunderstanding in saying what I just have, no blame attaches to [the deceased]. Her reactions were outside her control and as a result of her Alzheimer's disease.) 11. As I have said, it is for the prosecution to prove to you, so that you are sure, that [the respondent] was not acting in a state of automatism. If you conclude that he was or may have been acting in a state of automatism, you will acquit him of murder."
[39]The common law offence of automatism has been described as a "quagmire". We agree with that view. We also agree with the view expressed in the 2013 Law Commission Discussion Paper, which we have referred to above, that it is "difficult to identify a consistent definition of automatism from the case law" and that this difficulty can result in classifications and verdicts which, when under review, appear to be incoherent and arbitrary. Further, we recognise the problems which arise from the important distinction that must be drawn between "sane automatism" and "insane automatism" in cases such as this.[40]We express our reservation concerning the judge's implicit classification of the factors evidenced in this case as amounting to sane automatism. , but we proceed to examine the merits of the appeal on the basis that it was, or may be, correct. We make clear that our reservations are not based upon hearing full argument on the point.[41]We also make clear our concerns that paragraph 10 of the legal directions given in relation to "sane automatism", inadequately reflects the distinction that must be drawn between sane and insane automatism. As per Dr F's evidence, ( see DCS Y345A):
“.. the first thing I look at is there an external factor. That is what you need for automatism. Now, that can be looked at from two perspectives because what I do understand is that stress can be considered to be an external factor and stress is an external factor for the definition of automatism so that is one part. However, again, I don’t want to comment on Dr Brown’s, you know, report or the evidence she gave but my understanding from her evidence was that it wasn’t the stress but it was what led from the stress to the acute stress reaction and the dissociative disorder that led to the total lack of control”
. The framing of direction number 10 raises the question for us as to whether the judge misdirected herself in her analysis of the psychiatric evidence and opinion when she considered the submission of no case to answer. It also forms the basis of our order that the jury will be discharged and that a fresh trial may take place, although we make clear that it will be for the trial judge to determine, on the basis of the evidence then led, as to the relevant scope and content of any directions of law on this point to be given to the jury.[42]We have already referred to that part of the judge's ruling which states that in this case none of the psychiatrists have said that they are sure that the respondent did not act in a state of automatism and have said that at the very least it is possible. We find this to be factually correct. But, with respect to the judge, it is an oversimplistic analysis upon consideration of the whole of the psychiatric evidence.[43]Accordingly, we have called upon Miss Dempster KC to address those parts of the evidence which Mr Wilshire identified to the judge during the submission, as indicating a real difference of professional opinion as to the availability of the defence of sane automatism. Miss Dempster, in characteristic fashion, took up the challenge and attempted to respond accordingly but conceded the difficulty in doing so. We are not persuaded by her attempts to dissociate relevant parts of the evidence of Dr A, Dr F and Dr S by reference to other answers that the psychiatrists made and when taken out of context, or indeed to return to the mantra that “all psychiatrists agreed a defence of automatism was a possibility.”[44]We are satisfied, upon a review of the psychiatric evidence that there is and was a genuine professional disagreement which ultimately would be for the jury to resolve. There were significant issues raised during the psychiatric evidence: first, as to whether the respondent had suffered a complete loss of control: see the evidence of Dr F and Dr A; and secondly, whether, if the state of depression had led to a dissociative state, which if it led to a complete loss of control, could be described as an external event. This, in short, was not a Brennan case.[45]In paragraph 65 of her ruling, the judge went on to say that this was not a case where the psychiatric evidence could be buttressed by other evidence which would allow the jury to move from the psychiatrists' "unlikely" or "not probable" to their own state of being sure, because there was simply no other evidence adduced. We have difficult in accepting that conclusion. As Miss Dempster KC and Miss Roberson indicated in their second written response – and Miss Dempster confirms today – the jury was entitled to analyse the events in the minutes, days and weeks before the incident on 4th January 2022, to see if they collectively amounted to an external factor in accordance with the evidence of Dr B. (This accords with Mr Price KC’s additional submissions.) However, if this is right, then it is surely right that they were able to do so in order to consider the context of the differing psychiatric opinion. As to this, we note in paragraph 2.7 of her addendum report, dated 30th May 2023, that Dr B indicates:
"Therefore, it is possible that a dissociative state could lead to a total absence of voluntary control. Whether there was a total destruction of voluntary control on the [respondent's] part is a matter for the court [ we assume her to mean the jury] to decide."
[46]Section 67 of the Criminal

Justice Act 2003 provides:

"The Court of Appeal may not reverse a ruling on an appeal under this Part unless it is satisfied — (a) that the ruling was wrong in law, (b) that the ruling involved an error of law or principle, or (c) that the ruling was a ruling that it was not reasonable for the judge to have made."
[47]For the reasons above, and assuming for the purpose of this appeal that the evidence did support the defence of sane automatism, we are satisfied that the ruling at least involved an error of principle and was one that it was not reasonable for the judge to have made. She subsumed her views on the evidence for those of the jury. We reverse the ruling.[48]There has been significant delay in the time between the jury hearing the detailed and complex psychiatric evidence which was on the basis of written directions about which we have expressed concern above.[49]Therefore, we direct the discharge of the jury in variation of the judge's decision not to do so and, as we have previously indicated, order that a fresh trial may take place in the Crown Court on the count of murder.[50]We intend to direct that this case should be referred to the Lead Presiding Judge for the Circuit in order for them to give directions as to the venue of trial, whether or not the case should be released, and indeed as to the identity of the judge who shall hear any trial that becomes necessary.[51]We conclude by thanking Miss Dempster and Miss Roberson, and indeed Mr Price and Mr Wilshire – and also Mr Burgess – for their assistance in this most sensitive matter. We thank them for their written submissions, their oral submissions, and their obvious commitment to this case. ____________________________ Epiq Europe Ltd hereby certify that the above is an accurate and complete record of the proceedings or part thereof. Lower Ground, 18-22 Furnival Street, London EC4A 1JS Tel No: 020 7404 1400 Email: rcj@epiqglobal.co.uk ______________________________