England and Wales Court of Appeal (Criminal Division) Decisions

Peter Alan Fell, R v. [2001] EWCA Crim 696 (22nd March, 2001)

Case No: 99/5908/S1

Neutral Citation Number: [2001] EWCA Crim 696

IN THE SUPREME COURT OF JUDICATURE

COURT OF APPEAL (CRIMINAL DIVISION)

ON REFERRAL FROM CRIMINAL CASES

REVIEW COMMISSION

Royal Courts of Justice

Strand, London, WC2A 2LL

Thursday 22^nd March 2001

LORD JUSTICE WALLER:

Introduction

1. On 10 May 1982, Ann Lee aged 44, and Margaret Johnson aged 66, when walking their dogs together on a small piece of Army Common land on the outskirts of Aldershot, were stabbed and suffered fatal injuries. Almost 18 months later, on 18 October 1983, the appellant was charged with two offences of murder arising from the incident. On 9 August 1984 in the Crown Court at Winchester before Nolan J and a jury, following a trial which had lasted some 24 days, the appellant was convicted by a majority (10-2) of both offences.

2. The appellant was refused leave to appeal against his conviction by the Court of Appeal (Criminal Division) in November 1985.

3. The case against the appellant was based to a very large extent on admissions made by him in the course of a number of police interviews which had taken place in July 1983. The admissibility of those interviews had been the subject of lengthy argument in a voire dire conducted at the trial. The judge ruled that the admissions were voluntary and the interviews were thus admitted in evidence.

4. In a report dated 29 September 1999 the Criminal Cases Review Commission referred the case to the Court of Appeal (Criminal Division). By their report the Commission reviewed many different lines of argument relied upon to demonstrate that the verdicts of the jury were arguably unsafe. After a detailed consideration of those grounds the Commission formed the view that in three areas there was a real possibility that the conviction would not be upheld if a reference were made to the Court of Appeal.

5. Their conclusion was in the following terms:-

"In this case the Commission concludes that there is such a real possibility. This conclusion is based upon new evidence concerning material non-disclosure (pages 14-18); psychological evidence concerning Mr Fell and the effect of the police interviews in the light of Mr Fell's compliant nature (pages 24-26) and the reliability of the evidence of Mr Harper (pages 18-21). "

6. By their grounds of appeal dated 3 August 2000 those acting for the appellant (a) did not seek to pursue the reliability of the evidence of Mr Harper but (b) sought to add further grounds of appeal. They also applied to call fresh evidence, being the psychological evidence concerning Mr Fell, and to put in fresh evidence relating to the non-disclosure aspect.

7. So far as the psychological evidence was concerned statements had been provided by Dr Gudjonsson and Professor Kopelman. Those statements were before the Commission. The Crown obtained evidence from a Dr Joseph. His conclusions were to the same effect as those of Dr Gudjonsson and Professor Kopelman that the admissions were unreliable. In the result, the appellant applied for bail and the matter came before the Court of Appeal on 1 December 2000. That court, consisting of Potter LJ, Poole J and Sir Brian Smedley, gave the appellant bail and also gave leave, unopposed by the Crown, under section 23 of the Act, for the calling of fresh expert evidence before this court.

8. The question for this court has thus been whether, in the light of the fresh evidence, the conviction of the appellant was safe.

The Facts

9. On the afternoon of 10 May 1982, Ann Lee and Margaret Johnson were walking their dogs on Army Common land on the outskirts of Aldershot. The common is about 400-500 yards long by a shorter distance wide, in other words it is quite a small area.

10. As the two women walked along a track bordered by a fence which marked the grounds of Anglesey House, a large house used by the military as a Courts Martial Centre, they were attacked by a man wielding a knife. That attack took place at approximately 3-0pm.

11. At about 3-15pm Molly Hogg, who was herself walking with a dog, found the body of Mrs Lee lying face down and she also saw the body of Mrs Johnson lying on her side, some distance away, at the foot of a gate. The dogs of the two victims were still in the vicinity.

12. The pathologist report showed that the death of the two victims was due to multiple stab wounds. Mrs Lee had suffered 11 wounds to the back of her upper body and Mrs Johnson had suffered 5 wounds to the front of her chest. The wounds in relation to both victims ranged from 1 to 1½ inches in width and penetrated up to 5 inches.

13. Mrs Johnson was measured and found to be 4'10'' tall. That was said to have some significance, in view of admissions made by the appellant in interview ultimately, that she resembled his mother.

Persons on the common

14. The jury heard evidence from six witnesses who said that they saw a young dark-haired man on or near the common between 2-10pm and 3-15pm. The appellant it should be said was aged 20 and about 5'11'' tall. His hair would be shortish. He had had a crew-cut when in the army, from which he had been discharged in March of 1982, and there is a photograph amongst the exhibits taken some few weeks before May 1982 which seems to show the appellant with short hair. It is right to add that we were also shown a photograph taken in September 1982 which showed the appellant with much longer hair. In any event his hair was dark brown.

15. Sylvia Swainson and Joy Whatley saw a young man wearing a camouflage jacket at about 2-10pm. Mrs Swainson described the man as aged 18 to 25, between 5'8" and 5'10", of slim build with short brown hair.

16. Brian Hackney, a Ministry of Defence policeman, was jogging across the common when (as he assessed) at about 2-58pm he overtook Mrs Lee and Mrs Johnson about 250 yards from the murder scene. About a minute later he passed a man who appeared to be unhappy. Mr Hackney described this man as being in his early twenties, 5'8" tall, of medium build with long, dark, unkempt hair and wearing a green anorak or an army-type jacket and possibly jeans.

17. Also on the common was a group of soldiers including Richard Nikolic, Ian Loisy and Corporal Peter Waterman. Between 2-15pm and 2-30pm Nikolic, who was about 250 yards from the scene of the attack, saw a young man wearing clothes with the colours camouflage and blue. Nikolic was lost and asked the man if he had seen any Landrovers and the man replied "Yeah, there are some jeeps down there", pointing out a track. Loisy and Waterman described the man as wearing either blue jeans and an army combat-jacket or a blue jacket and green combat trousers.

18. The relevance of the camouflage or green jacket was that the appellant denied at all times owning either a green or camouflage jacket, although a dozen witnesses gave evidence that the appellant did own such a jacket.

Telephone calls to the police

19. On 11 May 1982, the day following the attack, an anonymous telephone call was made at 11-43pm to Aldershot police station. The call was not recorded but a note of the call has been accepted as accurate. That note reads:-

"I met a man in a pub tonight who got very drunk and he started rambling on about the murders. He seemed to know a lot about them and kept saying how sorry he was that he'd done it. I only know he lives at 10 York Road Aldershot. He asked me to go back there with him tonight but I didn't."

20. At 11-47pm on 12 May 1982 PC Mills at Aldershot police station received a further anonymous telephone call from a male which was recorded in a note as follows:-

"I know where the man who did the killings lives. My brother rang you yesterday and told you. His name is Pete and he lives at 10 York Road."

It is undisputed that the appellant made those two calls.

21. Over a year later, on 4 June 1983, by which time the appellant had moved to Bournemouth, a series of anonymous telephone calls were received at Bournemouth police station. Once again it was, and is, undisputed that the appellant made these telephone calls.

22. The calls were made between 9-0pm and midnight on the `999' system and were tape-recorded. The caller was plainly drunk. The calls referred to the double murder at Aldershot on 10 May 1982 and mentioned the two women and the two dogs. The calls identified the attacker as the appellant living at an address in Bournemouth of 13 Wellington Road.

The first police investigation of the appellant

23. The appellant was employed as a sales representative for Olan Mills Portrait Studios, Grosvenor Road, Aldershot. His bed-sit at 10 York Road was on a direct route between the scene of the attack and his place of work. On 19 May 1982 police officers making house-to-house enquiries about the attack spoke to the appellant and obtained answers from him to a proforma questionnaire. The appellant stated in that questionnaire that he had not visited the common between 1-30 pm and 3-30pm on the day in question or on any other occasion. His account of his movements was that he had been in the Unicorn or Trafalgar public house until closing time and then from 2-30 pm to 3-30pm on 10 May he had possibly been at home. He then said that from 3-30pm until 8-0pm he had been at work. He stated that he had possibly been wearing a blue jacket and grey trousers. He did not mention visiting his bank.

24. On 23 May 1982 the appellant was again seen by police officers. He gave the same account of his movements. His room at 10 York Road was searched and there was no trace of a green jacket. The appellant said that he had never owned a knife other than one which he then produced and which was a pocket knife with a 1" blade.

25. On 4 July 1982 the appellant was again seen by police officers and he made a written statement about his movements. He said that he was in the Trafalgar public house until about 2-30pm and then went either to the Cosy Café or the Job Centre until about 3-30pm when he went to work.

Other evidence of the appellant's movements on 10 May 1982

26. According to two colleagues of the appellant, on 10 May 1982 Mark Keable, the appellant's employer, had asked his staff to start work half an hour earlier than usual, that is at 3-0pm. Those two colleagues, Jean Stone and Susan Bennett, remembered 10 May because they had objected to the above request as they both had to collect their children from school before then going to work. Having collected their children they went to Aldershot and then arrived at work at about 4-0pm. Both stated that when they arrived at work the appellant was not present and that when he did arrive, about 5 or 10 minutes later, he was unusually smartly dressed.

The alibi

27. At his trial the appellant relied on alibi. It could be demonstrated that between 2-0pm and 4-0pm on 10 May 1982 someone had cashed a cheque drawn on the appellant's bank account at Barclays Bank, Victoria Road, Aldershot. The appellant had written two cheques drawn to cash on that day and the second had only been entered on the bank's computer at 3-47pm. Mrs Hewer, an employee from the bank, gave evidence that in accordance with her standard practice that cheque must have been presented at the counter after 2-30pm. If it was the appellant who presented that cheque personally then it was common ground that he could not also have been on the common at the time that the murders took place.

Conversations on 11 May 1982

28. A conversation between the appellant and his girlfriend Lorraine Sturrock was relied on at the trial. It appears that their relationship was brought to an end on 11 May 1982 and Lorraine Sturrock said that she remembered the appellant using words to the effect "if it wasn't for you you bitch this never would have happened."

29. One evening after the murder the appellant had a conversation with Jean Stone . They were discussing the murders and Jean Stone expressed surprise that two women with dogs could both be attacked. The appellant had responded that it was possible to immobilise one woman, run after the second and kill her, then return to kill the first. The Crown attached significance to this conversation suggesting that the appellant whilst in the army may well have received training on such matters. The procedure described by the appellant would be consistent with the way in which the two murders had taken place.

Possession of a knife

30. The appellant's neighbour at York Road was a man named John Harper. Initially Mr Harper made a statement saying that the appellant had told him that he (the appellant) had a knife which he had retained from the army, but that when on a later occasion Mr Harper had asked to borrow the knife the appellant had told him that he had thrown it away.

31. Mr Harper made a later statement dated 10 August 1983 in which he claimed to have seen the knife. He described it as in total about 8½ inches long with a handle of about 3 inches and a blade of about 5 inches in length narrowing to a point, the blade being about 1 inch wide. Significantly, Mr Harper described the blade as being "maybe two-sided". Mr Harper explained that he initially failed to mention that he had seen the knife because he had not wanted to get involved in the investigation.

32. Mr Harper ultimately chose a knife from an array of knives in a shop in Aldershot as similar to the knife he had seen in the appellant's possession, and the pathologist, Dr Pullar, gave evidence that the injuries caused to Mrs Johnson and to Mrs Lee could have been caused by such a weapon.

33. A number of the appellant's fellow soldiers gave evidence to the effect that while stationed in Germany the appellant had had a knife of widely varying descriptions. At least two of the witnesses described the appellant throwing a knife at a door. The knife or knives described by these witnesses were not however double-edged or sharp at both edges.

34. The appellant, throughout the interviews and with whatever other admissions he may have made, maintained that he had never owned a knife of the type described by Mr Harper, and maintained resolutely that he did not remember using any knife on the day in question.

The photofits

35. Mr Hackney, who had been jogging on the common on 10 May 1982, provided details to the police which led to the first photofit picture being compiled on 11 May 1982. Further details were supplied by Mr Hackney which led to a second photofit being compiled on about 21 May 1982. Corporal Waterman also supplied a description used in compiling the photofit. These photofits were published in the press on 10 and 11 June 1982 and were regarded by many people as bearing a striking resemblance to the appellant. Undisclosed at the trial was the fact that many people had contacted the police saying that the photofit bore a striking resemblance to other people. Out of 157 people who had contacted the police only 5 suggested that the resemblance was of the appellant. Also undisclosed at the trial was the fact that Mr Hackney had identified other persons at the labour exchange as bearing a similarity to the person that Mr Hackney had seen on the common.

36. On 4 July 1982 the appellant himself spoke to police officers and stated that people were saying that the photofit pictures looked like him. He went on to say words to the effect of:-

"Are you accusing me of being the murderer because if you are I shall get a solicitor in order to sue the police for harassment."

The appellant's arrest and interviews

37. It was following the anonymous telephone calls to Bournemouth police in June 1983 that the appellant was arrested on 5 July 1983, and when arrested and cautioned he said in response "I know I don't have to say anything but I'll answer what you want to know".

38. While the appellant was being taken to Farnborough police station in a police car there was an incident involving another vehicle, and in the course of discussing the incident the appellant referred to the vehicle in question as a "jeep". That was relied on by the Crown as having some significance in that the man seen on the common by Nikolic had used the word "jeep" in order to describe a British Army Landrover.

39. The appellant was interviewed between 5-7 July 1983. The interviews were tape-recorded although that was not a requirement at that time. Before the first interview took place the appellant had asked about "legal aid" and that was understood to be a request for a solicitor. At various stages during the interviews the appellant made a similar request but at no stage was he provided with the services of a solicitor, that being a deliberate decision taken by DCI Long. The refusal to allow the appellant the services of a solicitor was the main ground on which the defence argued in the voire dire at the trial that the admissions made in interviews should not be admitted in evidence.

40. The appellant furthermore did not eat at all during the whole period of the interviews. He was offered food but declined it, not apparently by way of protest, but simply because he did not want to eat.

41. In the first interview between 1-25pm and 4-0pm the appellant was cautioned in clear terms and he clearly appreciated the effect of the caution. In that interview he denied that he had ever owned a green jacket. He admitted to making the telephone calls on 11 and 12 May 1982 and 4 June 1983. He said that he was bored and that he wanted to be involved and that he wanted to find out about the murders and draw attention to himself. He denied ever owning a knife of the type described by Mr Harper. He denied that he had ever visited the track on the common. He maintained that he had not visited the common and did not know where the murders had taken place, and he denied committing the murders.

42. The second interview took place on the same day between 7-30pm and 8-45pm. Again the appellant was cautioned. Again he admitted the telephone calls and stated that he had made them because he had wanted a bit of fun and because he was bored. He persisted with his denial that he did not own a green jacket and said that witnesses who had described him as wearing such a jacket were lying.

43. At 8-0am on 6 July 1983 the appellant was visited in his cell by Inspector Ransier who was in charge of the police station. The appellant made no complaints about the conditions of his detention and stated that he wanted "legal aid".

44. At 9-45am the appellant was seen by DCI Long and PS Searle who enquired after his welfare. He said that he was all right. He stated that he had been rethinking the questions and answers from the previous interviews but he was told to reserve anything he wished to say until a formal interview was conducted.

45. The third interview then took place between 10-40am and 12-50pm. Again he was cautioned. He was told that somebody had gone to see his wife to ensure that she was all right. The appellant continued to deny that he had ever been on the common. He stated that he did not know the whereabouts of Anglesey House. He stated that he did not know where he had been at the material time or during the day of 10 May 1982 at all. At the end of the interview the appellant again asked about legal aid.

46. The fourth interview took place on the same day, 6 July 1983, between 2-30pm and 4-0pm. Again the appellant was cautioned and he said that he did not know what he had been doing on the day of the murders. He said that he could not have been on the common. It was during the fourth interview that the appellant asked to be taken to the common.

47. The appellant was taken to the area of the attack by DCI Long, DS Vincent and PS Searle. Before embarking on the journey to the common the appellant was cautioned. It was during this visit to the common that the appellant stated that the fence surrounding Anglesey House appeared familiar. He said however that he had not visited that common before and that the only location on army land that he had visited was at Tweesledown racecourse, some two miles away, on a map-reading course.

48. Following his return to Farnborough police station at about 6-40pm the appellant asked to see DCI Long. The appellant said that he had been thinking about what had happened on the common and he said "it's that fence, I've been along there before I'm sure I have. I'm not sure I think it may have been the day." DCI Long interrupted the appellant and told him that it was in his interest for anything that he had to say to be tape-recorded. This led to the fifth interview which took place between 6-45pm and 7-45pm. The appellant again was cautioned and in summary stated (i) that he remembered walking on the common along the track past (what was in fact) the murder scene; that he was depressed and that he had stayed on the common for about ten minutes; (ii) he had seen a man catching rabbits, a man running, two landrovers and a boy and a girl with white hair; and (iii) had not seen nor killed the two women. In the course of this interview the appellant again asked about legal aid.

49. At about 11-20am the following day, 7 July 1983, DS Vincent and PS Searle took the appellant to the common to retrace his steps. Before the journey, in a conversation with PS Searle initiated by the appellant, the appellant said that he could not remember and then asked the difference between murder and manslaughter. The officer explained the difference between the two offences and said that if there was something wrong with him (that is the appellant) he could look at the fact that if he was convicted of manslaughter rather than murder that would have a bearing and relevance to the future. The appellant said that he had not killed anyone.

50. Following this second visit to the common, and on his return to the police station, DCI Long visited the appellant in his cell. During this conversation the following exchange took place. Appellant: "My wife, I'm worried about her. Will I have to go to court". DCI Long: "I don't know at this stage but a decision will have to be made sometime soon". Appellant: "What could I go to court for?" DCI Long: "If at the end of the day it is considered you are the person who killed the women, as you well know, we must look at it as murder". Appellant: "That's life isn't it?" DCI Long: "The maximum sentence but that is if you went out with the intention to kill those women. At least it would be manslaughter". Appellant: "What's the difference, that's life as well". DCI Long: "With murder you have the intention to kill before the act where manslaughter is where an unlawful killing takes place but there is no prior intention to kill". Appellant: "It doesn't make any difference. I've lost my wife and everything". The appellant said "I keep getting these flashes, I remember sitting by that concrete thing, (a reference to a tank-trap). I think I saw the two women, I think they spoke to me they were laughing, I attacked them." When asked "are you positive" he replied "I think I did there was two dogs one was running about. I went crazy and I don't remember a knife." The appellant was told not to say anything further until it could be recorded.

51. The appellant was then taken to an interview room and interviewed. That interview took place between 1-33pm and 2-45pm. After being cautioned the appellant admitted that he had seen the two women as he came down a hill and that they were by a derelict shed. The women were laughing at him, the trees and everyone were laughing at him. He attacked the dark woman (Mrs Johnson) with his fists causing her to fall down to the ground. He could not remember attacking the other woman. He was feeling "down" and his mind was confused. He thought he was still in the army. He gave as his motive for the attack the fact that he hated his mother and his treatment at her hands. In answer to the question "can you explain to me why both of them were killed?" he answered "no I just remember can't care she reminded me of my own mother, I think she is the same height." He went on to say that he had picked up a long white stick which broke. He denied owning a green jacket or any knife other than a 2½ inch sheath-knife which he no longer had.

52. At the end of the interview the appellant was returned to his cell. Within an hour he again asked to see DCI Long and he retracted his confession saying "I didn't kill anyone, I didn't see those women, I don't even think I was on the common."

53. The appellant was again taken to be interviewed and between 5-45pm and 6-03pm he retracted his admission but maintained that he hated his mother. He said that he had gleaned the information used in the admissions from press reports of the incident. In this interview he explained a number of times that he had made his previous admission because he thought that time was running out and the inference is clear that he had decided to make an admission which would be likely to lead to a charge of manslaughter rather than murder.

54. Nothing that the appellant said in the admission interview, nor indeed in any of the interviews, could be regarded as some special feature which could only have been known to the person committing the murders. Indeed, Mr Perry did not suggest that this was a case in which any reliance could be placed on special features of the admissions as demonstrating that only the murderer could have made the admissions that the appellant did. Indeed, the appellant's description of what he did was inconsistent with the way in which the murders were likely to have been carried out.

55. On 8 July 1983 the appellant was admitted to bail. Prior to being released from custody an identification parade was held. Brian Hackney, the only witness to attend the parade, failed to identify the appellant, although he said that the appellant was the only one on the parade who resembled the man he had seen. Because of the lapse of time however, he said that he could not be sure.

56. During the period July 1983 until October 1983 two areas of enquiries were pursued.

57. The appellant's mother was seen. Facially she was not dissimilar to the victim Mrs Johnson as portrayed in the photograph in the press in 1982. She was also of short stature, she being 5'1" whereas Mrs Johnson measured 4'10". It seems that Mrs Johnson's photograph, as published in the press at the time, was in fact a passport photograph taken many years previously, and would thus give the impression of a very much younger lady than Mrs Johnson in fact was. The two were otherwise dissimilar, in particular Mrs Johnson was slim whereas the appellant's mother was not.

58. The other area of enquiry related to the arrival of the appellant at his place of work on 10 May 1982. Up until the time of the telephone conversations from Bournemouth, the position was that the appellant himself had at all times maintained that he was at work at 3-30pm and there was no evidence from his colleagues or his manager to contradict that. However, between the time of those conversations and October 1983 the police made further enquiries. That led to Mrs Stone remembering that on the particular day the appellant had arrived late at work after 4 o'clock. She remembered that fact due to Mr Keable having given instructions that the employees were to come in to work earlier and to work a period without extra pay. In protest Mrs Stone and Susan Bennett had decided to arrive late. They both said that they arrived at work at 4-0pm and the appellant arrived after that, smartly dressed. It is right to say that Mr Keable did not fully support this version, albeit he too gave a further statement saying that the appellant arrived at work after 4 o'clock.

59. On 18 October 1983 the appellant was charged with murder.

The trial

60. The trial took place at Winchester Crown Court between 9 July and 9 August 1984. At the commencement of the trial an application was made to exclude the evidence of the interviews. The judge heard evidence from the police officers who conducted the interviews over a period of 5 days, and ultimately ruled, on 13 July 1984, in the following terms:-

"At the end of five days of evidence and argument upon the issue at present before me I have reached a clear conclusion, and I think it right to give expression to it. The issue goes to the admissibility of statements made by the defendant in interviews with the police and in particular one statement amounts to a partial confession. The question I have to ask myself may be taken from the speech of Lord Hailsham in the case of DPP v Ping Ling 62 Cr.Ap.R. and the passage I have in mind comes at p.21. There Lord Hailsham poses the question words which I adapt slightly for the purpose of the present case like this: "have the prosecution proved that the contested statement was voluntary - and I interject - have they proved beyond reasonable doubt that it was voluntary in the sense that it was not obtained by fear or prejudice or hope of advantage excited or held out by a person in authority or by oppression?""

61. The judge came to the clear conclusion that the Crown had established that the admissions were voluntary. He did so in effect under four heads. Firstly he was satisfied that the defendant was repeatedly cautioned. Secondly the partial confession was, in the judge's view, initiated by the appellant without prompting from the police in the course of the general conversation. Thirdly the acts and motives to which the appellant according to his evidence confessed differed in a number of important respects from any suggestions by the police. Fourthly the judge was satisfied that the interviews were not conducted in an oppressive or unfair manner.

62. The judge finally returned to the refusal of access to a solicitor which he found was a deliberate refusal. On that aspect the judge found that he for his part accepted that if a solicitor had been present the appellant would almost certainly have been advised not to make any confession and probably not to make any statement at all. But the judge found that it did not follow that the statements made were involuntary. The judge then went on as follows:-

"For the reasons given I am satisfied that they were wholly voluntary and are therefore admissible. I have carefully considered the argument that a breach of the principles set out in paragraph C, or for that matter in paragraph D, (of the then Judges' Rules) automatically and as a necessary consequence in law should lead to the exclusion of a statement obtained in disregard of those principles. The authorities which have been quoted to me do not establish that proposition.

It is not in my judgment part of the law of this country. Those facts are of the greatest importance and need to be carefully borne in mind, as I have carefully borne them in mind arriving at the decision which I have reached; but in the exercise of my discretion and in what I consider to be in the interests of justice I propose to admit the evidence of the conversations."

63. On 24 July, that is to say after the above ruling, PS Searle brought to the attention of counsel the exchange that had taken place with the appellant on 7 July 1983 (concerning the difference between murder and manslaughter). There was a further voire dire in which PS Searle gave evidence when the judge again ruled that he was satisfied that the trial should go on, he being satisfied beyond a reasonable doubt "that the partial confession made by the defendant has not been obtained from him by hope being held out by a person in authority."

64. The Crown's case closed on 1 August 1984. The defence submitted that the case should be withdrawn from the jury but that submission was rejected. The appellant did not give evidence, but evidence was called in support of the alibi. Mrs Hewer, an employee of Barclays Bank, thought that the cheque entered on the bank's computer at 3-47pm on 10 May 1982 was probably cashed between 2-30pm and 3-30pm that day. She did not think that it was cashed on the morning of that day and entered on the computer later.

65. At the conclusion of the evidence and at the request of the defence, the learned judge reminded the jury of their right to acquit the appellant and stated:-

"I mention this without in any way wishing to guide you. This matter is in your hands. You have heard a great deal of evidence. It is however proper that I should tell you the position in law."

66. The case proceeded and the judge summed up the case on 7 August 1984 concluding it on 8 August 1984. One or two minor criticisms were at one time sought to be made of the summing-up, albeit the Commission in its report did not think that the summing-up could be criticised in any way. We say straight away that we take the view that the summing-up cannot be criticised, and Mr O'Connor QC did not seek to make any criticisms.

67. Having given very full directions on the burden and standard of proof the judge at the forefront of his summing-up reminded the jury of the alibi evidence. Thus it was that he said:-"if by reason of the bank evidence you feel that there must be a real possibility not displaced by the prosecution that the ten pound cheque that was cashed during the afternoon by the defendant was cashed by him, then your deliberations over this case may be fairly short." (See p.6G). The judge then went through the evidence with great care putting both the prosecution case and the defence case. In particular, he made clear to the jury that the interviews had been conducted in breach of the rule that a person under investigation is entitled to consult with a solicitor, and warned the jury as to the reason for such a rule i.e. the danger of someone making a false confession without that assistance.

68. The jury retired at 10-08am on 8 August 1984. They were given a majority direction at 3-19pm. They retired overnight and resumed their deliberations on 9 August 1984. At 11-31am on 9 August the jury returned and by a majority verdict of 10-2 convicted the appellant of murder on both counts.

The appeal

69. The appellant applied for leave to appeal against conviction. The matter came before the full Court of Appeal, Lord Lane CJ, Boreham and McCowan JJ, on 14 November 1985. The grounds of the application were as follows:-

1. The judge erred in admitting the interviews and the confession because there had been inducements, oppression and a breach of the Judges' Rules in failing to allow the applicant access to a solicitor when he requested one.

2. The judge should have removed the case from the consideration of the jury at the close of the prosecution case, following the defence submission that there was no case fit for a jury.

3. Having heard the alibi evidence the judge should have directed the jury to acquit rather than remind them that they had a right to stop the case if they so wished to do.

4. The judge failed to put the defence case properly to the jury in his summing- up.

5. The verdict of the jury was unsafe and unsatisfactory.

70. The Court of Appeal delivered a reasoned judgment dismissing the application for leave. In their judgment they identified two potential inducements. The first occurred early in the interviews, when it was suggested to the applicant that it was to his advantage to indicate his movements on the day of the incident so that he could be eliminated or otherwise. This inducement was said to have led to the appellant admitting making hoax calls. The Court of Appeal said that it agreed with the trial judge who had heard all the evidence in a voire dire that these admissions had been voluntary.

71. The second inducement related to the evidence of an unrecorded conversation with the appellant in which there was a discussion about the difference between murder and manslaughter. In this regard the Court of Appeal were concerned in particular with the conversation held by PS Searle which only came to light after the first voire dire had been completed. In the event the suggestion being made on behalf of the appellant was that following such conversations the appellant had made his limited confessions regarding the attacks. The Court of Appeal again agreed with the trial judge that these admissions were voluntary as the police had not held out any hope which might have induced the appellant to confess.

72. When considering the question of admissibility the Court of Appeal also considered the appellant's submission that he had a fragile and childish personality which had been worn down during the interviews, and a tendency to boast and fantasise albeit no medical evidence was called in relation to those matters.

73. The Court of Appeal considered the question of whether the interviews had been oppressive in the light of the submissions made about the appellant's personality. After listening to the interview tapes the court agreed with the opinion of the trial judge that the interviews were not oppressive.

74. The Court of Appeal further considered the denial of access to a solicitor. The court found that there was no principle of law that stated that this meant that any subsequent confession should be inadmissible and found that the judge had a discretion as to whether or not to exclude the evidence. The trial judge had pointed out to the jury that they should consider how the absence of a solicitor affected the reliability of the confession. The Court of Appeal felt therefore that this aspect had been dealt with properly by the trial judge.

75. The Court of Appeal did not consider that the case should have been withdrawn from the jury after the alibi evidence had been heard. The court found that there were no grounds to criticise the handling of the alibi in the summing-up. The Court of Appeal indeed found that the trial judge had dealt properly with the defence case in the summing-up and considered that the summing-up was clear and had been carefully prepared.

The present appeal

76. As already indicated, this matter has been referred back to the Court of Appeal by the Criminal Cases Review Commission under the Criminal Appeal Act 1995. Section 13 of that Act provides as follows:-

"(1) A reference of a conviction . . . shall not be made under any of sections 9 to 12 unless-

(a) the Commission consider that there is a real possibility that the conviction . . . would not be upheld were the reference to be made,

(b) the Commission so consider -

(i) in the case of a conviction . . . because of an argument, or evidence, not raised in the proceedings which led to it or on any appeal or application for leave to appeal against it, or . . ., and

(c) an appeal against the conviction . . . has been determined or leave to appeal against it has been refused.

(2) Nothing in subsection (1)(b)(i) or (c) shall prevent the making of a reference if it appears to the Commission that there are exceptional circumstances which justify making it."

77. This was not a case where the Commission thought there were exceptional circumstances, but a case where they thought that there was a real possibility that the conviction would not be upheld, because of an argument or evidence not raised in the proceedings or in the application for leave to appeal. In forming that view, it also considered there was a real prospect that the Court of Appeal would receive evidence not produced at the original trial. Thus, by their report, the Commission drew a distinction between those matters already dealt with at the trial and by the previous Court of Appeal decision, and matters which were not before the previous courts or in relation to which fresh evidence was not material. Thus it was that they recommended that it was in three areas that they felt that another Court of Appeal might interfere.

78. By section 14(5) of the same Act, it is provided as follows:-

"(5) Where a reference under any of sections 9 to 12 is treated as an appeal against any conviction, verdict, finding or sentence, the appeal may be on any ground relating to the conviction, verdict, finding or sentence (whether or not the ground is related to any reason given by the Commission for making the reference)."

79. It follows that, once a reference to this court has been made, it is open to this court to examine all grounds whether or not those are grounds considered by a Court of Appeal on some previous occasion. As it is the safety of the conviction with which the court is concerned, this perhaps is not surprising. Logically however, it seems to us the starting point should be the new points, in the context of which should then be viewed points previously argued.

80. The first and critical new point identified by the Commission, related to the reliability of the admissions or partial admissions. The Commission had before them evidence of a psychiatric nature which had not been available at the trial or on the application for leave to appeal, which cast doubt on the reliability of the admissions made by the appellant. It is only in recent years that psychiatric evidence of this nature has come to be recognised as of assistance to the judge and to the jury. The history of cases which show the development is this area have been supplied by both sides in this appeal, and we will turn to those cases below.

81. The second area relied on by the Commission related to evidence not disclosed by the prosecution prior to the trial and undisclosed prior to the first appeal. It is unnecessary to dwell on the third area for the present as it does not figure in the appellant's grounds of appeal.

Fresh evidence

82. The question of whether fresh evidence should be admitted in the Court of Appeal and the safety of the conviction are usually inter-related questions. Section 23 of the Criminal Appeal Act 1968 provides as follows:-

"23 Evidence

(1) For [the purposes of an appeal under] this Part of this Act the Court of Appeal may, if they think it necessary or expedient in the interests of justice -

(a) order the production of any document, exhibit or other thing connected with the proceedings, the production of which appears to them necessary for the determination of the case;

(b) order any witness who would have been a compellable witness in the proceedings from which the appeal lies to attend for examination and be examined before the Court, whether or not he was called in those proceedings; and

[(c) receive any evidence which was not adduced in the proceedings from which the appeal lies.]

[(2) The Court of Appeal shall, in considering whether to receive any evidence, have regard in particular to -

(a) whether the evidence appears to the Court to be capable of belief;

(b) whether it appears to the Court that the evidence may afford any ground for allowing the appeal;

(c) whether the evidence would have been admissible in the proceedings from which the appeal lies on an issue which is the subject of the appeal; and

(d) whether there is a reasonable explanation for the failure to adduce the evidence in these proceedings."]

83. In this instance the defence wished to call the evidence of Dr Gudjonsson and Professor Kopelman to give evidence in accordance with the various reports prepared by them. Their reports, by different routes, concluded that the admissions made by the appellant were, in all probability, unreliable. The Crown obtained a report from Dr Joseph to deal with that evidence in the eventuality that the evidence was admitted. The conclusions reached by Dr Joseph, again not necessarily by the same route, were that the admissions made by the appellant might well be unreliable. The defence also obtained a statement from Dr Ilbert who was the prison doctor who examined the appellant at the time of the trial and could thus give evidence of his condition at that time. That statement supported the view that the admissions were unreliable.

84. On 1 December 2000, the Court of Appeal (Potter LJ, Poole J and Sir Brian Smedley) dealt with an application by the appellant for bail. Bail was granted, and without opposition from the Crown, that court gave leave to call the fresh evidence, including now of course the evidence obtained by the Crown from Dr Joseph and the evidence of Dr Ilbert. In the result before us were called Dr Ilbert, Dr Gudjonsson and Dr Joseph. They gave evidence in accordance with their reports and were cross-examined. Professor Kopelman, whose report was accepted, did not give evidence.

Approach of the court

85. Once the court is conducting a review as to whether a conviction is safe, it is impossible to conduct that review other than through present day eyes. In Regina v O'Brien , Hall & Sherwood CA Transcript 25/1/2000 Roch LJ giving the judgment of the court, put the matter succinctly as follows:-

"This court has to decide whether these convictions are safe or unsafe. To do that we must apply the substantive criminal law that was in force at the time of the trial in 1988. However, we judge the conduct of the investigation of the case, the conduct of the trial, the directions to the jury and the reliability of the evidence on which the jury acted in accordance with the standards that this court now applies, c.f. R v Mills [1998] AC 382 per Lord Hutton at page 397 C-G and R v Bentley not reported but decided on the 30th July 1998 by this court presided over by the Lord Chief Justice."

86. In R v O'Brien the court also made an important point in relation to expert evidence of the kind which this court has heard and it is worth putting that into its context.

False confession cases

87. In R v David Stuart Mackenzie (1993) 96 Cr.App.R. 98, the Court of Appeal had to consider the reliability of certain confessions. In that case (and it is unnecessary to go into the details) evidence was given on the voire dire by Dr Gudjonsson and Dr Eastman in relation to the reliability of the confessions. The judge having ruled that the confessions were admissible, that evidence was again given before the jury. It was in that case where Lord Taylor giving the judgment of the court said:-

"Nevertheless, applying the guidance given by this Court in Galbraith (1981) 73 Cr.App.R 124, we consider that where (1) the prosecution case depends wholly upon confessions; (2) the defendant suffers from a significant degree of mental handicap; and (3) the confessions are unconvincing to a point where a jury properly directed could not properly convict upon them, then the judge, assuming he has not excluded the confessions earlier, should withdraw the case from the jury. The confessions may be unconvincing, for example, because they lack the incriminating details to be expected of a guilty and willing confessor, or because they are inconsistent with other evidence, or because they are otherwise inherently improbable. Cases depending solely or mainly on confessions, like cases depending upon identification evidence, have given rise to miscarriages of justice. We are therefore of opinion that when the three conditions tabulated above apply at any stage of the case, the judge should, in the interests of justice, take the initiative and withdraw the case from the jury."

88. The Crown in that case submitted that the confessions were rightly admitted and that the convictions were safe and satisfactory. The argument was that the confessions showed special knowledge which would only have been available to the killer and that thus there was internal support for the reliability of the confessions. The court was unpersuaded of the special features but it is unnecessary to go into that aspect.

89. In R v Long (Transcript 13 July 1995) the Court of Appeal considered the case of George Long referred to them by the Home Secretary under the provisions of section 17(1)(a) of the Criminal Appeal Act 1968. Long had been convicted of murder in 1979. His application for leave to appeal had been refused by the single judge and in 1980 the full court refused an extension of time in which to renew the application. It was as a result of a petition based upon fresh medical evidence that the Home Secretary referred the case back to the Court of Appeal.

90. It was in 1992 that a consultant psychiatrist Dr McKeith had been instructed by solicitors for the appellant. Dr McKeith had made a special study of the psychiatric aspects of cases in which defendants had made confessions which they subsequently retracted. Dr McKeith gave evidence before the Court of Appeal. Dealing generally with the state of medical knowledge he said that at the time of the appellant Long's trial the possibility that a defendant charged with a grave criminal offence may confess and then retract his confession because he suffered from a mental disorder, stopping short of a frank mental illness, had not been recognised. He went on to give as his opinion that at the time when Long made his admissions he was suffering from a mental disorder such as rendered those admissions unreliable.

91. Another psychiatrist called on behalf of the appellant expressed the same opinion. Dr Joseph was called for the Crown and, although he did not come to precisely the same assessment of the appellant's state of mind as the defence experts, he was satisfied that due to his condition the appellant's admissions may have been unreliable.

92. In allowing the appeal Lord Taylor CJ giving the judgment of the Court of Appeal, expressed himself in this way:-

"We are firmly of the view that his confessions cannot now be regarded as reliable. If the jury had known of the appellant's mental history and if they had had the benefit of the medical evidence of (the experts called on the appellant's behalf) they may well, in our view, have been unsure of guilt." (14D).

Lord Taylor also said later in the judgment:-

"Unless guilt is proved by reliable evidence so that the verdict can be regarded as safe and satisfactory, a conviction cannot stand." (15B).

93. In R v Evans (Transcript) 3/12/97 the Court of Appeal (Lord Bingham CJ, Jowitt and Douglas Brown JJ) allowed the appeal of Andrew Evans. This was another case in which the conviction of the appellant for a murder (in 1973) resulted from his admissions to the police. In that case the prosecution case rested entirely on the confessions of the appellant. In that case Dr Joseph again gave evidence and Lord Bingham concluded that the court had to accept, having heard fresh evidence, that the "appellant's confessions were, as confessions, entirely unreliable." (P.11).

94. In R v Ashley King (Transcript 10/12/99) the Court of Appeal (Lord Bingham CJ, Morison and Nelson JJ) considered the reliability of the confessions of Ashley King who had been tried at the Crown Court in Newcastle in July 1986. King had made written confessions signed after a number of interviews with the police conducted in the absence of a solicitor. The confession was retracted 3 days after it was made when the appellant was provided with the services of a solicitor. The appellant had been assessed by two psychiatrists prior to his trial. It was again a case in which the case against the appellant rested entirely on his own confessions.

95. On the appeal a Mrs Tunstall, a professional psychologist, gave evidence. She concluded that the appellant was abnormally suggestible and abnormally compliant. This evidence the court accepted and found persuasive. In giving the judgment of the court, Lord Bingham made reference to the development of this kind of evidence:-

"There is, however, the additional finding that the appellant was suggestible and compliant to an abnormally high degree. That was not a matter which could, practically speaking, have been tested, assessed or quantified in 1985 to 1986. Although there had been some published work on the subject, this was a new and embryonic science."

96. Lord Bingham went on to set the appellant's admissions in the context of the significance of the development of this kind of psychological understanding and referred to "the real problem" as:-

"whether, in a case which depended solely on the reliability of the appellant's confessions, improperly obtained, the appellant is now scientifically shown, as he could not have been shown at the time, to be highly abnormal in respects directly related to the reliability of the confessions and in a way which throws doubt on their reliability.

Had the defence had the benefit of this new expert evidence, there would have been strong grounds for seeking the exclusion of the confession evidence under section 76 and 78 of the Police and Criminal Evidence Act 1984, and possibly even section 77. Had the jury heard the confession evidence but also heard the evidence of Mrs Tunstall, it would in our judgment have been very hesitant indeed to convict the appellant on the strength of his uncorroborated and retracted confessions, and rightly so. In the light of this new evidence we feel bound to regard the appellant's conviction as unsafe and we accordingly quash it." (Paragraphs 68, 69).

97. In R v O'Brien, Hall and Sherwood (Court of Appeal Transcript 25/1/2000) Roch LJ, Keene and Astill JJ heard another appeal calling into question the admissibility of a potentially unreliable confession. In that context they gave guidance as to the basis of the admissibility of fresh evidence in relation to such appeals. They said this at page 19:-

"At one time the law was thought to be that expert evidence of the kind we have heard could only be admitted if that evidence showed a recognised mental illness, this being the interpretation placed upon R v Turner [1975] QB 834. It has now been accepted that expert evidence is admissible if it demonstrates some form of abnormality relevant to the reliability of a defendant's confession or evidence, see for example R v Ward [1993] 96 Crim.App.R. 1. In the case of Ward at page 66 this court said:

"But we conclude on the authorities as they now stand that the expert evidence of a psychiatrist or a psychologist may properly be admitted if it is to the effect that a defendant is suffering from a condition not properly described as mental illness, but from a personality disorder so severe as properly to be categorised as mental disorder."

Despite what was said there in the case of Ward, the test cannot, in our judgment, be whether the abnormality fits into some recognised category, such as anti-social personality disorder. That is neither necessary nor sufficient. It is not necessary, because as R v Roberts showed, the real criterion must simply be whether the abnormal disorder might render the confession or evidence unreliable. It is not sufficient because an anti-social personality disorder does not necessarily mean that the defendant is a compulsive liar or fantasist or that his confession or evidence might be unreliable.

The members of this Court, as were all counsel who addressed us, are conscious of the need to have defined limits for the case in which expert evidence of the kind we have heard may be used. First the abnormal disorder must not only be of the type which might render a confession or evidence unreliable, there must also be a very significant deviation from the norm shown. In this case the abnormalities identified by the experts were of a very high level, Hall's test results falling within the top few percentiles of the population. Second, there should be a history pre-dating the making of the admissions or the giving of evidence which is not based solely on a history given by the subject, which points to or explains the abnormality or abnormalities.

If such evidence is admitted, the jury must be directed that they are not obliged to accept such evidence. They should consider it if they think it right to do so, as throwing light on the personality of the defendant and bringing to their attention aspects of that personality of which they might otherwise have been unaware.

The evidence, both factual and expert which has been placed before us has satisfied us that this is a case in which such evidence would now be admissible, and that a jury having heard such evidence may well have reached different verdicts."

The Importance of the admissions

98. The admissions were central to the Crown's case. That can be demonstrated most forcefully by the way in which the alibi defence was dealt with both in evidence and in the summing-up. In the summing-up the judge put the matter this way:-

"Was he the killer on the footpath? If within at the most about 30 minutes of the murders, that is to say, between 2.30 and 3.30, he was cashing a cheque at Barclays Bank in Victoria Road, Aldershot, the best part of two miles away, then you may feel for that reason alone that he is very unlikely to have been the killer. I shall return to that matter when I come to the defendant's case. If, by reason of the bank evidence, you feel that there must be a real possibility, not displaced by the prosecution, that the £10 cheque which was cashed during the afternoon by the defendant was cashed by him, then your deliberations over this case may be fairly short.

The other side of that coin, say the prosecution, is that if you are completely satisfied by the prosecution evidence that the defendant was the murderer, then there must be some other explanation of the second cheque cashed that day at Barclays Bank, for example, that he had simply given the cheque to somebody else to cash for him."

99. There was no effective challenge to the evidence of Mrs Hewer, and no evidence was called to establish that someone cashed the relevant cheque on behalf of the appellant. Thus the Crown's case had to be that the jury could be so sure that the admissions contained in the interview were reliable, that they could discount the alibi without knowing precisely on what basis.

100. There was some other evidence which, if the jury accepted it, could be said to support the admissions, i.e. the evidence that he had a knife of the kind likely to have been used (and his denial at all times of having such a knife); the evidence that he had a camouflage jacket (and his denial that he ever had such a jacket); the evidence that he arrived late for work and in smart clothing (and his assertion that he was in fact at work by 3.30 pm); the evidence that the appellant's mother was of small stature, as was Mrs Johnson, and the fact that the appellant maintained that he hated his mother even when retracting his admissions. But, as Mr Perry for the Crown very properly conceded, there could have been no case to go to the jury without the admissions. They were thus the central plank of the prosecution case.

Reliability

101. Following the approach of Roch LJ in O'Brien to the admissibility of this type of evidence, the questions are (a) whether the psychiatric evidence now available demonstrates something well outside the norm, and (b) whether there is something in the history of the appellant which supports the psychiatric evidence. To these questions, if they are answered in favour of the appellant, we would add, so far as the Court of Appeal is concerned, (c) whether an examination of the admissions in the light of the psychiatric evidence leads to the view that they may have been false or indeed that they were false so as to render the verdict of the jury unsafe.

The psychiatric evidence

102. Dr Gisli Gudjonsson's first report is dated 24^th November 1993. Dr Gudjonsson has a considerable reputation in the field of Forensic Psychology, and he has a particular expertise in the area of confessions, having developed certain techniques for measuring "Suggestibility" "Compliance" and the like. He interviewed the appellant in February 1988 and then in January 1993, and his first report is based on those two interviews. His conclusions are conveniently summarised at the end of his report, and include the following:

"2. Mr Fell's compliance score fell outside the normal range, indicating an exceptionally high level of compliance. His "shift" suggestibility scores were somewhat elevated, but nevertheless fell within normal limits. However, it is noteworthy that Mr Fell appears to have some difficulties in detecting discrepancies between what he observes and what is suggested to him through leading questions.

3. The results from the Gough Socialisation Scale and the Eysenck Personality Questionnaire are consistent with a diagnosis of personality disorder. In addition, Mr Fell is a rather introverted individual with marked neurotic traits. He appears to experience considerable emotional turmoil, which includes anxiety, depression and schizoid life-style.

6. The reasons for Mr Fell drawing attention to himself as a murder suspect are not entirely clear. He denies that it was due to feelings of guilt about the murders and his explanations are that he wanted the police to investigate his background as well as having a bit of "fun". His explanations for telephoning the police are consistent with his disturbed family background and low self-esteem. It is probable that his behaviour was a way of seeking excitement and enhancing his self-esteem (i.e. making him feel important).

7. It is my view that at the time of his interrogation in 1983 Mr Fell was a psychologically vulnerable individual, who would have benefitted greatly from the advice and support of a solicitor. Having studied this case in some detail, I have serious doubts about the reliability of Mr Fell's self-incriminating admissions to the police in 1983."

103. The Commission when reviewing this case, and in the light of the first report, requested Dr Gudjonsson to answer three questions. They were as follows:-

"1. Is there evidence to support the contention that Mr Fell had a false memory which prompted his admissions to the police?

2. In the light of the additional material now available and advances in the relevant areas of knowledge, is there anything to be added to your report dated 24th November 1993?

3. Is there any evidence that would tend to support the argument that the admissions made by Mr Fell were made as a result of an inducement which may have been made to him?"

104. Dr Gudjonsson interviewed the appellant again, and carried out a further detailed analysis of the interviews, setting out substantial portions in his report dated 21^st December 1998. His conclusions were (1) that the appellant was a psychologically vulnerable individual. He said that the present assessment confirmed his high level of compliance, anxiety proneness, introversion, and low self esteem; (2) he said there was no evidence of psychogenic amnesia nor any evidence of false memory; (3) As to inducement his conclusion was as follows:-

"5. Inducement

I believe there is evidence that would tend to support the argument that the admissions made by Mr Fell were the result of an inducement, which may have been made to him. I think it is quite possible, if not likely, that his confession to the murders of Margaret Johnson and Ann Lee was made because of the officers' persistent and determined implication throughout the interviews that Mr Fell was the murderer, Mr Fell's failed efforts at effectively challenging the officers' assertions, the failure of the police to take any notice of his repeated and determined efforts to seek legal advice (which is likely to have exacerbated his low self-esteem and a sense of helplessness), and seeing the confession as a desperate compromise to reduce the charge to manslaughter. The content of the confession interview supports the view that Mr Fell was attempting to present a case which supported a manslaughter charge as opposed to murder."

Then on reliability he said this:-

"6. Reliability

Having had the opportunity of seeing Mr Fell again and studying the case in more depth, my previous reservations about the reliability of his confession to the police concerning the murder of Margaret Johnson and Ann Lee are strongly reinforced. Indeed, I consider his confession unreliable. His making telephone calls to the police can be best understood in terms of his very poor self-esteem at the time and his need to feel a sense of self-importance."

105. Michael Kopelman, Professor of Neuropsychiatry at St Thomas' Hospital, eminent and highly experienced in this field, made a report on 9^th November 1998. He did not interview the appellant, but from certain documents relating to the period prior to the murders, and from an analysis of the interviews, concluded first that the possibility of the appellant experiencing a partial amnesia for the events as a "psychological defence mechanism" was extremely unlikely, and then summarised his views in the following way:-

"In summary, it seems likely that Mr Fell could not remember the events of the day in question. Following his hoax telephone calls (to which he readily admitted), he was interviewed by the police in May 1982, and again over three days in July 1983. He was man of low self-esteem, who had been dismissed from the army, and who was known to be something of a story teller. The transcripts of the police interviews suggest that Mr Fell came under considerable pressure from the police to make admissions regarding the offence. Despite this, he made consistent denials until he had been taken to the site of the murders on two occasions, after which he had an apparent `recovery' of his memories, which I do not regard as convincing, considering that it had occurred 15 months after the alleged offence and that the `confession' was retracted a couple of hours later. Mr Fell was not given the legal support that he requested, nor was he given medical advice when he sought it. In the light of these various factors, I would regard his conviction as unsafe to the extent that it was largely based on this `confession'."

106. Dr Joseph, as the cases cited show, is an eminent Forensic Psychiatrist again highly experienced in this field. He reported on 16^th November 2000. He interviewed the appellant, and made a study of the relevant papers. His conclusions were, so far as the appellant's mental state was concerned, that "There were no psychotic features evident at interview and he appeared of average intelligence" and that "Mr Fell was not suggestible during my interview with him and he was able to disagree when he felt it appropriate. I did not form the view that he was excessively compliant. I was aware however that 17 years had elapsed since he was interviewed by the police and there is likely to have been personality maturation during this period.". His conclusions then included the following:-

4. "Taking into account the evidence of childhood conduct disorder and his subsequent psychological difficulties during his adolescence and early adulthood, it is likely in my opinion that Mr Fell was suffering from a personality disorder in the period leading up to the killings, characterised by low self-esteem, psychological dependence on others, feelings of rejection and abandonment and displays of attention-seeking behaviour to compensate for these feelings. Although he was also suffering from depressed mood and was abusing alcohol, both these features are likely to have been secondary to his personality difficulties rather than constituting a mental disorder themselves.

5. On the basis of his personality disorder, I conclude that Mr Fell was psychologically an extremely vulnerable man in the period leading up to and including his arrest on the charges. It is notable that he made hoax telephone calls initially after he separated from his first girlfriend and subsequently after his wife left him to go to Liverpool in June 1983. On both occasions the hoax telephone calls served the function of drawing attention to himself at a time when Mr Fell was feeling abandoned, rejected and lost. It is recognised that he was prone to telling lies about himself in order to make himself seem more impressive to others and there is a previous example of his lying about a conviction for violence to the police.

6. Bearing these facts in mind, it is highly probable that the defendant was psychologically vulnerable at the time of his interviews with police so that any admissions he made would be unreliable. This is especially so when he was denied the assistance of a solicitor. If he had been interviewed today then he would have also required the services of an appropriate adult to protect him during an interview with the police.

7. Although I consider Mr Fell's admissions during police interviews to be unreliable, I did not find him to be a suggestive person, nor did I find him to be particularly compliant during interview with him. It is not those aspects of his personality which render him psychologically vulnerable, but rather it is the aspects of his personality disorder which caused him to suffer such low self-esteem and to believe that he could impress others by lying and drawing attention to himself. It is highly likely that Mr Fell's personality disorder and self-esteem have improved over the years and this is borne out by the psychological testing carried out by Dr Gisli Gudjonsson in 1988 and 1993."

He agrees there is no evidence of a "false memory".

107. Both Dr Gudjonsson and Dr Joseph were cross-examined on the differences between them, particularly so far as compliance was concerned. For example Dr Gudjonsson had to deal with the fact that the telephone calls could not have been made simply because the appellant was compliant; they were made without any intervention by the police or anybody else. Dr Gudjonsson accepted that and confirmed what he had said in his report that in his view these were made due to the appellant's low self esteem, and at times when he was particularly vulnerable after being left by, in relation to May 1982, his girlfriend, and in June 1983, by his wife. He also had to deal with how, if the appellant was unusually compliant, he should in the interviews still have resolutely refused to admit having the knife and/or a green jacket. Dr Gudjonsson was in difficulty in explaining how that came about save that the matter must have been "very important, and stuck in his mind.".

108. On the other hand Dr Joseph had to accept that despite his disagreement with Dr Gudjonsson on suggestibility and compliance, some different mechanism was likely to have produced the admissions in interview from that personality disorder which produced, or was likely to have produced, the telephone calls. He accepted that the pressure on the appellant having made the telephone calls, and admitted making them, would be very great, and that pressure would be increased by the following factors - no communication with anybody other than his interviewers over about 54 hours; the sheer length of the interviewing process; any misunderstanding as to the strength of the evidence against him; the anxiety over his wife who was pregnant and an epileptic; the lack of any food; indeed the fact that his appetite had been lost would demonstrate the psychological distress the appellant was in. He further accepted that under these sort of pressures the appellant might have been inclined to look for compromise.

Contemporaneous evidence

109. The material before us has been taken into account by the psychiatrists in forming the views they did. In summary that evidence included the appellant's regimental conduct sheet for the period up until March 1982, which demonstrated how he had made up stories about being attacked or having things stolen from him. It further showed there was no truth in his assertion volunteered to the police on 19^th May 1982 (i.e. during the inquiry into the murders) that he had been dismissed from the Army because of committing an offence of GBH. There is the photograph with our papers showing the appellant with a boxing trophy. He even convinced the police during the inquiry "that he could handle himself" by reference to this trophy. He was writing to his brother in December 1983 as if he had been a boxer. He never had won any such trophy and had bought it and engraved it himself. He had boasted of fighting in the Falklands, and there was simply no substance in that.

110. In Dr Pitcher's Report dated 9^th July 1984, made for the purposes of considering the appellant's mental health at the time of the trial, it said amongst other things "it is also evident from his account to me and from his statements to the police that almost nothing he says can be relied upon unless it is corroborated".

111. Dr Ilbert, who as already indicated was the prison doctor when the appellant was remanded in custody at Winchester Prison in 1983/84, was also the doctor allotted for the purpose of submission of court reports. He made a report in answer to certain questions from the solicitor acting for the appellant on 20^th November 2000. In his evidence before us, he said that he saw the appellant every day during the period of remand and saw him for the purpose of reporting on him 11 times. He also kept (as one would expect) clinical notes. He was not a trained psychiatrist, but in one note for example, made on 19th October 1983, Dr Ilbert recorded his view of the appellant as a "pathological confessor". In another made on 20^th February 1984 he recorded "It does seem to me that his hoax telephone calls are consistent with his boxing-Falklands ego trip and they also incorporate the element of inner relief from tension gained by confessing".

112. Dr Ilbert clearly had considerable anxieties about the reliability of the appellant's confessions during 1984, but felt uneasy (as he says) "about the probity of such doubts....for such concepts fell well outside the normal terms of reference governing the conclusions qua Homicide Act....Mental Health Act, which one was required to reach in murder charge cases." One of the sadnesses of this case is that in some way Dr Ilbert's anxieties were not made known at the time.

113. Thus the contemporary evidence supports the psychiatric evidence.

114. That evidence thus casts serious doubt on whether the telephone calls could be regarded as in any way reliable as admissions, and casts serious doubt on whether admissions made in interview were reliable.

The interviews and telephone calls themselves

115. In the context of the above evidence the features of the telephone calls and the interviews which were no doubt stressed both at the trial on behalf of the appellant and on the application for leave to appeal, take on a different complexion. The starting point seems to us to be that if the above evidence had been before the jury their approach to the telephone calls would have been very different. Furthermore, their approach to the interviews would have been very different. With the benefit of that evidence we would say that all the indications are that the telephone calls were the appellant seeking to draw attention to himself and not calls made by the murderer. Our reasons depend to some extent on the view we take of the interviews, in relation to which we again would say, with the benefit of the experience of other cases and of the evidence called, bear the hallmarks of being false. They bear out paragraph 5 of Dr Gudjonsson's second report quoted above.

116. One can see from the interviews how once the police had an admission that the calls were made by the appellant, the pressure on the appellant gradually built up until it became overwhelming. The sheer length of the interviews, the non-communication for 54 hours other than with the police, the lack of food, and the method of interrogation all contributed. We do not think it necessary to go through the detail but one aspect that particularly struck us was the obtaining of the admission that he was on the common. That was obtained as if that would not lead to the conclusion that he was the murderer (see pages 827A, 829B-G, 833-4, and 839-840), whereas once he admitted being on the common the reverse was true (1193B and 1198F). Constantly the appellant asked for a solicitor and constantly that request was ignored; on one occasion he asked for a doctor and that too was ignored (see 1241). Finally under the impression he may have gained that time was running out and that he was bound to be charged with something, it seems that he sought the compromise of seeking to admit to manslaughter, not murder. When however he did so, he not only did not talk of any special features that only the murderer would have known but was inaccurate about many of the features of the murders e.g. he suggested he hit and punched in the face but no sign of such injury was present on the victims; he described seeing two children, and DCI Long accepted in evidence that "no-one there of that description at the relevant time"; he said he spoke to a man going rabbit shooting, but there would not seem to have been any such person; he said he attacked the dark-haired woman first but DCI Long accepted that in all probability the fair-haired lady was attacked first; the dark-haired lady was in her 60s and slim, and bore very little resemblance to the appellant's mother who was 42 and certainly not slim. They were both short but even then there were 3 inches difference in height.

117. If the evidence we have heard had been before the jury, would the only reasonable and proper verdict of the jury have been one of guilty? We are clear that the answer to that question must be in the negative, and indeed, the longer we listened to the medical evidence, and the longer we reviewed the interviews, the clearer we became that the appellant was entitled to more than a conclusion simply that this verdict is unsafe. There are strange features of the case, not least his failure to support his own alibi, but the alibi exists from an independent source. But more important, since our reading of the interviews and the evidence we have heard leads us to the conclusion that the confession was a false one, that can only mean that we believe that he was innocent of these terrible murders, and he should be entitled to have us say so.

Other grounds

118. Mr Perry for the Crown took us through the other grounds relied on by Mr O'Connor one by one submitting in relation to each that they would not have provided a basis for suggesting that the jury's verdict in this case was unsafe. On grounds 1 and 2 (the decision by the trial judge to admit the interviews and the admissions contained therein), he pointed out how the trial judge, after a voire dire lasting 5 days including hearing the evidence of the police officers concerned, considered with care the question of oppression, tricks, inducements, and denial of access to a solicitor. The judge had furthermore considered whether in his discretion he should exclude the admissions, and had declined to do so. The judge's ruling was upheld by the Court of Appeal in refusing leave to appeal.

119. Those points may be powerful in the absence of the fresh evidence. But, as Mr Perry recognised, once the psychiatric evidence is admitted these other points must be looked at in a different light. The first question which was considered by both the trial judge and the Court of Appeal was whether the admissions were voluntary. This they did before considering the question whether, in the court's discretion, the admissions should be excluded.

120. The Court of Appeal in its judgment cited passages from DPP v Ping Lin [1976] 62 Cr.App.R. 14, and R v Rennie [1982] 74 Cr.App.R. 207. The essential question, according to those authorities, was whether the prosecution had established beyond a reasonable doubt that the admission was voluntary, in the sense that it was not obtained "either by fear or prejudice or hope of advantage excited or held out by a person in authority .... or by oppression".

121. In the modern era section 76 of PACE makes clear the question of reliability is the important consideration. But it is fair to say that in exercising his discretion in this case reliability was an important consideration so far as the judge was concerned, he recognised that it was to that question that the absence of a solicitor went (see the passage of the summing-up quoted on page 15 of the previous Court of Appeal's judgment and their conclusion on this aspect).

122. The evidence we admitted showed that experts with an experience which the judge and the jury, and indeed the previous Court of Appeal would not have, were of the view that the admissions were unreliable. There would have been a danger in allowing the admissions to go before the jury if the judge was clearly of the view that it would be unsafe to act on them. It seems to us that the evidence we have heard would have added significantly to a submission that there was a need for someone such as the appellant to have a solicitor present before reliance could be placed on any admission he was making. A simple listening to the tapes might well indicate the absence of bullying or oppression in that sense, but the evidence would also add force to the submission that the sheer length of the interviews, without food, and the method of interrogation without the protection of a solicitor, would be likely to lead to an unreliable and indeed a false confession.

123. It would seem plain now to us, with the benefit of this evidence, that the case for excluding the admissions would have been even more powerful than it was, and indeed we would have expected the judge to have excluded them.

124. Without the admissions there would simply have been no case fit to go to the jury.

Police conduct

125. So that there should be no misunderstanding we think it right to record that we think the decision of the police not to allow this appellant a solicitor was reprehensible. That is consistent with the view of the trial judge, and the previous Court of Appeal.

126. As regards other allegations made against the police, the trial judge listened to the tapes, as did the previous Court of Appeal. We have not. We have of course read all the interviews. We do not think it right to make findings of "oppression" in the sense of misconduct by the police, or of "tricks", insofar as that imputes bad faith on the police who conducted these interviews. We furthermore do not think that there is any evidence that the police in some way misconducted the inquiries during the period between the appellant being interviewed and the appellant being charged. It is true that during that period various witnesses provided changes to their statements; Mr Harper that he had seen the appellant with a knife of the type used in the murder: Mrs Stone and Mrs Bennet that they now remembered the appellant arriving at work after 4.0pm. Those changes by themselves establish nothing. The police had a particularly brutal and unpleasant murder to investigate, and it must come as no surprise that they made all efforts to find the killer and bring that killer to justice.

127. They were, as we have said, quite wrong not to allow this appellant a solicitor. They allowed their quest for a conviction to override their responsibilities to an accused, and particularly to a vulnerable accused. If that fundamental right had not been denied this appellant a false confession would not have been made. We do not however make any other finding of misconduct against the police in this case.

Ground 5 Alibi

128. No criticism is made of the way the judge dealt with this matter in his summing-up, and the argument that the judge should have withdrawn the case from the jury once the alibi evidence had been given was not pursued. As a self sufficient defence we can see force in the points Mr Perry made for the Crown. The appellant never himself gave evidence to support the alibi. He never mentioned visiting the bank when questioned on 19^th May 1982 or on 23^rd May 1982 or indeed on 4^th July 1982, this despite having made the telephone call on 11^th May 1982 at which time one might (argued Mr Perry) have expected him to have recalled where he was on the afternoon of the 10^th May. Indeed, pointed out Mr Perry, the consistency of the appellant was simply that he never could account for his whereabouts during the afternoon of the 10^th between 2.30 and 3.30 pm. The cashing of a cheque was only discovered by the hard work of the solicitor acting for the appellant at the trial during 1984, and thus it was that the statement of Mrs Hewer was only obtained in July 1984.

129. But the dismantling of the alibi by the Crown was not attempted on any basis other than that if the admissions were true, then the alibi must be false and there must be some explanation other than that the appellant cashed a cheque personally on that afternoon.

130. Once the admissions are shown to be suspect, the defeat of the alibi itself becomes suspect.

Grounds 6, 7 and 8 Non-disclosure

131. Mr Perry accepts that there was non-disclosure (1) that there were various people that Mr Hackney had identified as persons he recognised as being the person on the common; (2) that of the 157 calls to the police identifying persons that resembled the photofit pictures published, only 5 related to the appellant, and 152 related to a variety of different other persons; and (3) of the recognition by Peter Carter and Christine Dye of the photofit resembling a person that had attacked them on the common a year before the murders, at a time when the appellant was in Germany in the Army.

132. Mr Perry argued that the disclosure would not have made any difference. He pointed to the fact that a photograph of the appellant taken in September 1982 bore a marked resemblance to the photofit pictures assembled as a result of Mr Hackney's description. Thus that the appellant did resemble the photofit was a fact which the jury would always have taken in. They would also have taken in from the interviews that the police made clear that there were others whom other people had identified from the photofit (Interviews 949 and 997). They would also have appreciated that Mr Hackney could not be sure of the appellant's identity on an identity parade.

133. If these had been free-standing grounds of appeal, there would be force in Mr Perry's points. He would have been able to rely on the admissions and the telephone calls, which, if reliable, would have provided a very strong case. To that would have been added the telling of lies about the knife and the camouflage jacket, all of which might well have followed if the reliability of the admissions could have been relied on.

134. The fact is however that they are not free standing. We doubt still whether they add much to a conclusion which we had already reached that these convictions are unsafe.

135. We announced at the conclusion of the hearing on Friday 2 March 2001 that the convictions would be quashed for reasons to be given later. These are those reasons.

MR O'CONNOR: My Lord, Mr Fell would personally wish to thank my Lord for the care with which my Lords have approached this difficult case. He feels obviously very moved.

I only have one brief application and it is for an order for defence costs out of central funds. That need not trouble the Court under the regulations about amount because there is a procedure whereby it is assessed within the Criminal Appeal Office by a sort of taxing master. My Lord, this is more aimed at Mr Roberts, the defence solicitor at trial, who assiduously carried on working, obviously with no legal aid existing, for many years after this conviction and he will of course have to justify his claim and it will be assessed in due course. I do ask my Lord to make that order. My Lord it is made in very many successful conviction appeals.

LORD JUSTICE WALLER: You may have your costs out of central funds.