|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
Court of Appeal in Northern Ireland Decisions
You are here: BAILII >> Databases >> Court of Appeal in Northern Ireland Decisions >> Latimer, R v  NICA 3 (09 February 2004)
Cite as:  3 BNIL 51,  NIJB 142,  NICA 3
[New search] [View without highlighting] [Printable RTF version] [Help]
Latimer, R v  NICA 3 (09 February 2004)
Introduction The appellant Neil Fraser Latimer was on 1 July 1986 found guilty, along with three other defendants, by Kelly LJ sitting without a jury of the murder of Adrian Carroll on 8 November 1983 and sentenced to imprisonment for life. The defendants all appealed against their conviction, but their appeals were dismissed by this court on 4 May 1988. The case was on 25 July 1991 referred to the court by the Secretary of State for Northern Ireland under the provisions of section 14(1)(a) of the Criminal Appeal (Northern Ireland) Act 1980. After consideration of fresh evidence adduced the court allowed the appeals of the other three defendants, but dismissed the appeal of this appellant.  The appellant was released from prison in 1998, but by a reference dated 9 May 2001 and made under the powers contained in the Criminal Appeal Act 1995 the Criminal Cases Review Commission referred the appellant's case once again to this court for a review of the conviction. Under section 9(2) of that Act the matter then was treated as an appeal from the appellant's conviction. It was heard by a court differently constituted from that which considered the two earlier appeals. We gave the appellant leave to call fresh evidence, consisting of the testimony of Dr FWA Browne, Lord Alderdice and Professor Gisli Gudjonsson. Dr Philip Joseph and Dr Michael Heap were, with the leave of the court, called on behalf of the Crown. We heard the appeal between 17 and 26 November 2003 and reserved our decision. This judgment, which is the judgment of the court, contains our decision and our reasons for reaching it.  The material facts have been set out in meticulous detail by the trial judge and in the judgments of this court on the earlier appeals, and we refer to them for the full account of the incident out of which the prosecution arose, the police investigation and interviews of the defendants and the course of the evidence given at trial. We have considered all the evidence given at trial, together with the fresh evidence given on the second appeal and the reference before us. It is not necessary for us to recount all of this in the present judgment and we shall confine ourselves to a summary of the facts material to our present consideration.
The Factual Background On 8 November 1983 shortly after 4.30 pm Adrian Carroll was shot in an alleyway off Abbey Street, Armagh, close to the house in which he lived. The gunman fired three shots with a handgun, inflicting wounds from which Carroll died that evening. The gunman was seen by an eyewitness Elaine Faulkner (later Mrs Dunne), then a girl of 17 years, going up Abbey Street and into the alleyway, from which she said she heard two shots a few moments later. She saw a man lying in the alleyway and took fright, running to her place of work nearby and then reporting the matter to the police. In her statement made that day to the police she described the gunman as being "20-25 years, about 5' 1" or 5" 2", small build, wearing a light blue duffle coat and a tartan or check cap, small sort of face and he had a light moustache and wore ordinary type glasses with a gold rim." She said that she would know the man again and that she thought he had dirty fair hair, tidy cut, and had on light casual shoes. She repeated and adhered to this description in her evidence, but omitted any reference to shoes and said that she was not sure about the moustache. In cross-examination she stated that the man whom she saw was definitely not Latimer, whom she knew to speak to, as he was a neighbour of hers at the time of the incident.  The news media carried reports of the shooting, some at least of which referred to the gunman as being of medium build and wearing a tartan or checked cap and gold-rimmed glasses. On 22 November 1983 a woman who lived in the Armagh area and who was known for the purposes of the trial as Mrs A, contacted an Armagh priest Father Murray and told him that she had seen a man dressed in civilian clothes, wearing a tartan cap and gold-rimmed glasses, accompanying two UDR soldiers from the technical college in Lonsdale Street, Armagh into a UDR Land Rover shortly before 4.30 pm on 8 November. She said that she recognised him immediately as the appellant, whom she knew well from having worked with him in her previous employment. She was surprised to see him so dressed, for he did not ordinarily wear glasses. She put the incident down to a joke, until she heard of the shooting of Carroll and the description of the gunman given on the news. The police were contacted and interviewed Mrs A on 2 December. She made a formal written statement to them on that date.  In consequence of the information which she had given them they arrested the appellant on 29 November and interviewed him at length over a period of seven days. On the evening of 29 November the appellant made oral and written admissions in which he confessed to having shot Carroll. On the morning of 30 November he retracted the admissions, which he termed all lies, and maintained this stance for some time. He continued to deny his involvement over a number of subsequent interviews, then commenced to qualify his denials, until on 2 December he again admitted shooting Carroll, though this time he gave a different version of his part in the incident and the preceding and subsequent events. We shall return later in this judgment to examine in greater detail the evidence of Mrs A given at trial and the course of the interviews of the appellant.
The Trial and Appeals The case presented by the Crown at the trial of the appellant and his co-defendants was that they had concocted a clever plan to provide an alibi for the appellant, who was purportedly on a UDR patrol all day long in a Land Rover, whereas he in fact changed into civilian clothes and tailed and shot Carroll before rejoining the patrol group and changing back into uniform. He arrived shortly afterwards back at the police station, along with the other occupants of his Land Rover, as if he had been with them on duty all day. The incident seen by Mrs A at Lonsdale Street was a charade designed to make it appear that the appellant was a civilian who was being arrested by soldiers and placed in the Land Rover. It was only the fact that he was recognised by Mrs A in spite of his disguise that proved their undoing and led to the apprehension of the appellant and the other soldiers involved. Latimer's counsel advanced the case on his behalf that the incident recounted by Mrs A was a complete fiction, designed to discredit the UDR, and that neither Latimer nor any other soldiers had had anything to do with the shooting of Adrian Carroll.  After a lengthy trial the learned trial judge found the appellant and three of the other defendants guilty of murder. He admitted the appellant's confessions in evidence and accepted that their contents and the evidence of Mrs A were a true and reliable account of the Lonsdale Street incident and the subsequent shooting. He held that Mrs Dunne's evidence that the gunman was not the appellant was an incorrect statement, but did not find it necessary to decide whether that evidence was falsely given or a gross error, or if falsely given why it should be. He therefore was satisfied beyond reasonable doubt that the appellant was guilty of the murder of Adrian Carroll.  When the case went to appeal it was thoroughly considered by this court, with Lord Lowry LCJ presiding. The court expressed the view that –
"The evidence of Witness A and Mrs Dunne, while important in pointing the finger at Latimer, was not sufficient in itself to convict him of the murder of Carroll."
It went on to examine in detail the admissions made by Latimer and the other defendants, accepting that the case against Latimer rested almost entirely on those statements. Counsel for Latimer did not contend on that appeal that the trial judge was wrong to decline to rule out those statements in the exercise of his discretion, and confined himself to arguing that the weight to be attached to them was insufficient to justify conviction. The court emphatically agreed with the judge's finding that the confessions challenged were freely and voluntarily made by Latimer and their content came from himself. It also agreed with the judge's conclusion that the essential part of Mrs A's evidence was true and reliable and that Mrs Dunne's evidence must, for whatever reason, be wrong in her assertion that the gunman was not Latimer. The court accordingly dismissed his appeal and affirmed his conviction. It also, dismissed the appeals of the other defendants, holding that their confession statements had been properly admitted and were reliable accounts of their complicity. The case of the four defendants was referred to this court by the Secretary of State in 1991 because of the discovery by use of the ESDA process that some of the police interview notes bore signs of having been rewritten. It is unnecessary for present purposes for us to deal with the detail of the operation of the electro-static detection apparatus, and it is sufficient to say that it reveals traces of indentations left on a document when another page has been written on top of it. It became apparent from the fresh evidence of the results of ESDA testing presented to the court that some of the interview notes had been rewritten and that they had not all been signed by senior officers at the times at which they had purported to sign them for the purpose of authenticating them. This meant that where detectives had stated at trial that the notes had all been written in the course of the interviews and that no notes had been rewritten at a later stage that part of their evidence was untrue. It also meant that the authentication of notes by senior officers was in some cases false.  The court made it clear in its judgment that there was no indication from the ESDA examinations that the police falsely concocted any confessions and attributed them to the appellants. Hutton LCJ said at page 18 of his judgment:
"It appears to be clear from the ESDA examinations that in rewriting the notes police did not insert any confessions which the appellants had not made or any suggestions made by the police which the appellants had not accepted, although in rewriting the notes the police did delete requests which Hegan, and possibly Bell, had made to see a solicitor and which were not granted." The court examined the case of each of the four appellants and assessed the impact of the ESDA evidence upon the safety of the conviction of each. Noel Bell had made the case in his defence at trial that Detective Superintendent Mitchell had suggested to him matters, obtained by him from statements made by Latimer, and that he Bell had simply agreed to each such suggestion, because he was frightened and "could not take any more." This was denied by the interviewing officers, and the trial judge had accepted that the essential content of the oral and written confessions made by Bell came voluntarily from him and were a reliable account of his complicity in the murder. The Court of Appeal found, however, that it had come to light that certain interview notes had been rewritten and that Detective Superintendent Mitchell and Detective Inspector Milligan appeared to have appended false certifications to the interview notes. Since there was a serious conflict of evidence between the police officers and Bell about the way in which his oral confessions were made and his written statement was taken, it was unsafe to rely upon them in support of his conviction. As this was the only evidence implicating Bell in the murder, his appeal was allowed.  James Irwin Hegan had made the case that he had been misled by police officers into making a statement, which was compiled by the interviewing officers and was not the product of admissions made by him. The judge did not accept this and preferred the evidence of the interviewing officers that Hegan was slow, thoughtful and deliberate in the dictation of his confession statement. He held that Hegan was master of himself at the time and not submissive or broken, nor was he yielding to a statement being concocted by others. He rejected the contention that the statement was induced by threats or promises or impropriety. He held that the conclusion was irresistible from a sensible interpretation of the statement and from the other evidence in the case that Hegan knew of the plan to murder Carroll and was a willing accomplice to it. The Court of Appeal in 1988 approached Hegan's case in the knowledge that certain of the interviewing officers had given untruthful evidence to the judge when they said that his notes had not been rewritten. Hutton LCJ said at pages 119-20 of his judgment:
"This court further knows and the trial judge did not know, that the interview notes of interview no. 18 had very probably been rewritten to conceal the fact that Hegan asked to see a solicitor and that this request was not complied with. With this knowledge and having regard to the serious conflict of evidence between the police and Hegan as to how his written statement was made, we have a reasonable doubt as to the reliability of his written statement which constitutes the only evidence admissible against him and we consider that it would be unsafe and unsatisfactory to rely on that written statement to ground his convictions." Alfred Winston Allen made the case at trial that his statements of admission were the product of ill-treatment at the hands of the interviewing detectives. The trial judge rejected this, accepting the officers' evidence that no ill-treatment occurred. The Court of Appeal in 1988 upheld this conclusion. He also claimed that the statements were not made by him but were made for him by the detectives. The judge rejected this contention and his decision was affirmed by the Court of Appeal in the 1988 hearing. The latter issue, the only one argued on behalf of Allen, was re-examined in depth by the court in the 1992 hearing. It looked closely at the conflict of evidence between Allen and the interviewing officers in the light of its knowledge that untruthful evidence had been given by certain of the officers about the writing of the notes. It came to the conclusion that there was a reasonable doubt about the reliability of the confessions made by Allen and that it would be unsafe to rely on them in support of his conviction. His appeal also was accordingly allowed.  The court came to a different conclusion when it came to consider Latimer's appeal. On this occasion counsel advanced the contention that the appellant's oral and written confessions should have been rejected in the exercise of the judicial discretion, since the effect of the evidence relating to the ESDA tests was to show that the interviewing officers had been untruthful in material respects. It followed that the court could not in that knowledge be satisfied beyond reasonable doubt that the allegations made by the appellant about the conduct of the officers in the interviews were unfounded, and accordingly the court should in its discretion have ruled out the admissions. He also submitted that the statements were unreliable and should not be accepted as of sufficient weight to found a conviction.  The court rejected both submissions. It held that the statements were admissible by virtue of section 8 of the Northern Ireland (Emergency Provisions) Act 1978, and that it was not a case for the exercise of the judicial discretion to exclude them as being the product of unfair pressure. The false statements of the police officers in evidence did not affect the issue, in that the court was satisfied from evidence other than that of the officers that the appellant did not make the admissions because of threats, promises or improper pressure on the part of the police. The court considered in detail the course of the interviews and the evidence given by Mrs A and Mrs Dunne. It concluded that it remained satisfied beyond reasonable doubt that Mrs A's evidence was true, that Mrs Dunne was wrong in her assertion that the gunman whom she saw was not Latimer and that Latimer's confession made on the night of 2/3 December 1983 was "a truthful confession made by a man who realised the game was up". It accordingly held that his conviction was safe and satisfactory and dismissed his appeal.
The Reference to this Court The Criminal Cases Review Commission referred the appellant's case back to the court because of the emergence of medical evidence, not previously considered by the trial judge or the Court of Appeal, which might be regarded as throwing a fresh light on the safety of the conviction. We gave the appellant leave to adduce this evidence under the terms of section 25 of the Criminal Appeal (Northern Ireland) Act 1980, although some of it could possibly have been adduced at the time of the 1992 appeal. The burden of this evidence, on which the submissions presented on behalf of the appellant were based, was that the conviction is unsafe in two respects:
(a) Mrs A's evidence is less reliable than the judge supposed, for her medical history and psychological make-up show that it should be accepted with caution;
(b) doubt is cast on the truth of the appellant's confessions by the psychological evidence showing that he is unusually compliant and so more willing to agree under pressure with the version put to him by the interviewing officers than the judge had supposed.
In order to consider these submissions we shall examine in more detail the evidence given by Mrs A at trial and the psychological evidence called before us which related to her, then the course of the interviews of the appellant and the psychological evidence which related to him.
Mrs A's Evidence The evidence given by Mrs A at trial was fully and accurately summarised by the judge, who described her in his judgment as a simple woman, direct in manner and also sharp. He set out her evidence at pages 28 to 33:
"On the afternoon of the murder, a Tuesday, she drove her car into the city with her young son. She left home around 4 pm. She drove to the head Post Office in Upper English Street arriving there about 4.05 pm and parked opposite. Her son left the car and went into the Post Office. He was away 'just a few minutes'. He came back to the car and having been given money to buy a musical record he went off to McKeever's in Scotch Street to buy it while she waited in the car. He was away roughly 10 or 15 minutes. When he returned they drove off down College Street, turned left past Joshua White's car showroom into Lonsdale Street where Armagh Technical College is. As she drove up to the college buildings which were on her right-hand side, she saw two Land Rovers parked in Lonsdale Street at the college. The first one she encountered was parked at an angle with its front out on the road and its rear towards the college buildings. The second was parked further along against the footpath also on her right. She probably could have driven past the Land Rovers, but she said she thought there was a bomb or somebody had been shot and she expected to be told to reverse back. Therefore she stopped. And she stopped close to the first Land Rover. She noticed a UDR soldier at each wheel of the Land Rover in a crouched position.
Then she looked up to her right towards the college and saw 'two UDR soldiers with a civilian running in the middle of them'. They came down the entrance between the college and a portakabin. The three were running with their heads down and the uniformed soldiers were carrying rifles. They jumped the fence surrounding the college. When the civilian was close to the fence about to jump he looked up and she recognised him as the accused Neil Latimer. She said he was dressed then in a brown anorak and a tartan cap and wearing small gold-rimmed glasses, square type.
She knew Neil Latimer. She had worked with him in the factory for just over a year and had met him every day there and would have spoken to him every day. After the factory had closed down, he was a member of a UDR patrol that stopped her about two weeks before and she had spoken to him then.
When she first saw the civilian that afternoon in Lonsdale Street she thought he was being arrested, but then when she saw his face and recognised him as Neil Latimer and realised also he was a member of the UDR, she thought his fellow soldiers were playing a prank on him, that he was getting married and that, to use her phrase, they were 'doing him up'.
The two soldiers and Latimer she said disappeared into the back of the Land Rover. Then a third Land Rover came into Lonsdale Street from the Albert Place direction (ie. the opposite end of the street from her approach). The Land Rovers flashed their lights and they all drove off. This time Mrs A estimated to be about 4.25 pm. She continued her journey along Lonsdale Street up Albert Place up to the roundabout at the Shambles and stopped at Gillespie's mill there. Her son got out to go to a corner shop. Meantime she turned the car and waited for him in the Shambles yard. She saw from the Cathedral clock it was then 4.30 pm. She waited around eight to ten minutes and while waiting heard the sound of the siren of a police car or ambulance. Then she [sic] came back and they drove home.
That night she heard on the BBC1 television news at 9 pm an announcement about the murder and a description of a man running away from the scene of the shooting and that the police were appealing for information about a man who had been seen with a dark brown anorak and tartan cap and glasses. It was also said that it was thought he was a middle-aged man going grey. She connected the description of the way he was dressed with Neil Latimer as she saw him that afternoon. She felt 'very peculiar'. She couldn't sleep that night. Thereafter it played on her mind, but she didn't want to get involved. Eventually, as she described it 'to clear her mind', 'to take a load off my mind', she contacted Father Murray, a priest in Armagh, and some days later she saw senior police officers. The records show she made a statement about the events in Lonsdale Street to Father Murray on 22 November 1983 and a statement to the police on 2 December 1983.
Although the evidence of Mrs A did not directly implicate any of the accused other than Latimer, its significance in the whole round of the case was recognised and counsel for each of the six accused then cross-examined her. They challenged her evidence on many fronts: bias, truthfulness, recollection, accuracy and reliability. Generally they suggested her evidence was a 'bundle of lies made up by you', a 'pack of lies', a 'bundle of untruths', a 'fantasy', an 'invented story'. Their suggestion in sum was that she had invented what she'd seen in Lonsdale Street, although during the course of one or two of the closing speeches it was suggested, rather mildly, that she may have seen Latimer in uniform in Lonsdale Street at the time and that this led to her mistake or was used as a basis for her invention.
Their cross-examinations of Mrs A did reveal inconsistencies in her evidence, mistakes, some faulty recollections, some contradictions, some inconsistencies between her evidence in court and what she had said in her statements to Father Murray and to the police. Many of these were labelled 'lies' by defence counsel. I need not set them out exhaustively. Here are some: She undoubtedly hedged about admitting a conviction for larceny of a pound of butter in 1967. She was wrong as to when she came forward for the first time to Father Murray and to the police, not 6 or 7 days or 10 days, but 14 days and 24 days respectively. She did not make her statement to Father Murray in the parochial house, but in her own home. She did try more than once to telephone him. She was quite wrong in denying that she sought directory inquires as to Father Faul's telephone number. The BBC 9 pm news contained no mention of the Carroll murder that evening and certainly not a description of the gunman or anyone running from the shooting. She was confused as to how many soldiers in Lonsdale Street were about, or got into the Land Rovers. She only assumed but did not see Latimer and the others getting into the back of a Land Rover. She showed contradiction as to whether the Land Rovers at the time had headlights on. She had said they had back doors, but admitted she couldn't see the back. She described the garment worn by Latimer as a dark brown anorak.
Then she was asked why she hadn't come forward earlier to the priest and the police. It was suggested she wished to discredit the UDR and this is one reason why she sought out Father Murray and Father Faul rather than her own parish priest. The comment was made that it was strange that no other traffic or pedestrians appeared to be in Lonsdale Street that afternoon, although it was a busy thoroughfare and it was put to her that she was 'the only person in the world' who had seen what she said she'd seen." Mrs A was recalled to give further evidence at a much later stage of the trial. The judge summarised it at page 33 of his judgment:
"Sometime after she'd given evidence she had been confronted by Bell's mother and sister in Scotch Street, Armagh, on 10 May. She was confronted and spoken to in circumstances which do no credit to Bell's sister. Allowing for family distress what Bell's sister did and said then was quite improper. It apparently led to Mrs A telephoning Mrs Bell on 5 June. Mrs A in the witness box again on 9 June put forward her views on certain aspects of the case and that the accused were innocent. But she did not retract at all what she'd seen in Lonsdale Street in the afternoon of 8 November 1983. On the contrary, she confirmed it. Her appearance on the second occasion underlined her unhappy state at being caught up as an important witness in a murder trial." The judge also pointed out that it was never suggested that Mrs A did not know the appellant well or that she might have been mistaken in her identification of him as the civilian whom she saw in Lonsdale Street. The case made on his behalf was solely that her evidence was a complete concoction, invented for some malicious purpose of discrediting the UDR. The judge gave careful consideration to the possibility that her view of the civilian might not have been sufficient for a clear identification, and held that the circumstances were such that an identification of the best quality could be made. He also considered and roundly rejected the suggestion that her whole story was a fabrication, for the reasons which he set out at pages 34 and 48 to 50 of his judgment, with which we agree. He expressed his conclusions about Mrs A's evidence in a passage at pages 47 to 48:
"I have already set out and commented on at some length the evidence of Mrs A. In this I have recognised the frailties of some parts of her evidence. If I have not set them all out here I have considered them all in reaching a conclusion on the weight of her evidence because they have been helpfully canvassed by all counsel in their closing speeches.
I repeat, I find her discrepancies or inconsistencies or whatever term one may wish to use as peripheral and comparatively speaking unimportant. My conclusion is that the essential part of her evidence is true. I have not the slightest doubt about it. She was quite unshaken either in content or demeanour on that essential and central evidence.
I am satisfied that she is telling the truth when she says that she saw two Land Rovers stationary in Lonsdale Street about 4.25 pm on the 8th November 1983 and that she saw Neil Latimer dressed as a civilian in the way she described in evidence being run down the steps of the college into the back of a Land Rover by two uniformed members of the UDR. I reach this conclusion not only after observation of her as a witness and hearing her evidence over the course of days but also after attempting to analyse why she might or might not be a witness of truth and after considering the evidence of the accused, that of Mrs Dunne, the evidence of the UDR witnesses and the other relevant evidence in the case.
I am also quite satisfied that she saw the accused in Lonsdale Street at a time before Adrian Carroll was shot and I accept her evidence that at a time later than her sojourn in Lonsdale Street, in the Shambles yard, she saw the cathedral clock showing 4.30 pm." The Court of Appeal in the 1988 appeal also rejected the suggestion that Mrs A had invented a completely false story, saying that as a scenario they found it quite incredible. They expressed their own opinion of her evidence at pages 8 to 9 of their judgment:
"Even from the cold print of the transcript, Witness A appears to us to be someone who witnessed an interesting, perhaps amusing, incident of no especial significance at Lonsdale Street and thought little of it until she heard a description of Carroll's assailant on the media. The cap and the glasses were a link, but she knew Latimer and did not believe he would commit a murder. We are not surprised that she wrestled with these two apparently conflicting factors for two weeks before consulting a priest. Her continuing mental conflict emerged clearly during her evidence and nowhere more clearly than when she was re-called on the day of the trial and confirmed her evidence about what she had seen at Lonsdale Street. No doubt many parts of her evidence were quite wrong, but the learned trial judge was well aware of the discrepancies which had been emphasised in counsel's submissions and we are not surprised that, having taken note of the facts and the arguments, he was prepared to accept her account of having seen Latimer as a party in the mock arrest. With only the transcript and counsel's submissions to guide us, we would have reached the same conclusion."
At the 1992 appeal the court again firmly rejected the suggestion that Mrs A's evidence was a deliberate fabrication and expressed the view that the evidence which she gave when recalled, both in its manner and content, confirmed very strongly that her account of what she had seen happen in Lonsdale Street was true. Evidence was given about Mrs A's psychological make-up by three consultant psychiatrists, Lord Alderdice and Dr FWA Browne called on behalf of the appellant and Dr PLA Joseph called on behalf of the Crown. None of the three had at any stage examined Mrs A, who died some two years ago, and their evidence was based on examination of the medical notes and records relating to her, particularly hospital notes from St Luke's Hospital, Armagh covering a period in 1964-5. Copies of the latter notes had first been given to Lord Alderdice in 1993 by Mr Ken Maginnis MP, who requested him to examine them and advise as to the significance of the findings contained in them, but we were not informed how the latter had come to have them in his possession.  Mrs A, then aged 30 years, was admitted to the hospital on 7 September 1964, having previously been well until about two weeks before. The history showed increasing depression and irritability, apparently triggered by money worries, concerns over her husband's gambling habit, and a feeling of persecution and isolation. She was reported as thinking that she was being shot at by her husband's friends and having heard people around the house at night. She was discharged on 30 September, with no medication, having made considerable improvement following LSD treatment, then in regular use to improve insight. In his discharge letter Dr WAG MacCallum, the consultant psychiatrist in charge of her case, diagnosed the disorder as a "mixed psychoneurotic reaction". This term was explained by Lord Alderdice as meaning that there was more than one psychoneurotic symptom, being a mixture of hysterical, anxious, depressive and obsessional symptoms, and by Dr Browne as being a mixture of anxiety and depression.  She was readmitted to hospital on 5 December 1964 as an informal patient and "sectioned" on 11 December when she wished to leave. She had supposed that her neighbours were against her and persecuting her and she imagined people around the house all the time. She stated that she wanted a separation from her husband and wished to live in a flat in Armagh. On 11 December she was described as emotionally labile and unstable, varying from depression to aggression, especially towards her husband. Her husband described a pattern of somewhat extreme conduct when she had rows with him, and she was described in the notes as having "a most unstable background both from personality and environment". She was discharged on 2 January 1965 on some medication and attended once as an out-patient. She was not seen at the hospital again. On discharge Dr MacCallum diagnosed her as being of a "psychopathic personality". It was agreed by the medical witnesses that the concept of psychopathology had changed since 1964 and psychopathic disorders were more narrowly defined, so that what was then termed a psychopathic personality would now be classed as a personality disorder. Personality disorders generally manifest themselves in a person's teens and persist through life, though they are more marked in young adults and are tempered with time by the maturation process. There was, however, no evidence from Mrs A's general practitioner that she had suffered any personality disorder before 1964, of any hereditary mental illness or of any psychiatric illness subsequent to her discharge from hospital.  Lord Alderdice expressed the opinion that persons with hysterical disorders of the personality may well be unreliable historians, with their accounts being influenced by what they feel is expected, or by wish phantasies, or by a desire to excite or make dramatic. Such a patient's intention is not to deceive, but to hold interest or attention. He concluded his opinion by stating:
"The material in these [hospital] notes would therefore suggest a very unhappy and unstable woman with a disturbance of personality which would make her unreliable and difficult in relationships, and the disorder is likely to be lifelong, though it may with certain circumstances be modified as she grows older, and with long-term psychological treatment.
If the question is asked, 'would such a patient be a reliable witness?', I would have to give the opinion that, on the evidence of these notes, one would be advised to be extremely cautious in setting too much store by a witness with such a personality and history. Such a patient could be given to fantasies and stories that are more to do with her own wishes than with objective reality, as is pointed out a number of times in these notes."
He said in his oral evidence that Mrs A might think that she saw a specified person, then go away and think about it, and become positive about it. It was possible that she made it up completely, but even more likely that she stitched it together from threads which she embroidered and built upon. Lord Alderdice did not have the opportunity of seeing Mrs A's general practitioner's notes and records or of reading a transcript of the evidence which she gave at trial. The latter would have allowed him to appreciate that she was not seeking attention in giving evidence, but was on the contrary a reluctant witness. Dr Browne was somewhat less positive in his conclusions, but emphasised the need for caution in accepting Mrs A's account. He stated at page 14 of his report:
"In my opinion Mrs [A's] psychiatric records do not show that Mrs [A] is incapable of providing accurate and reliable evidence, but they urge for caution in accepting her evidence as true."
He said in evidence that the disorder would not prevent her giving an accurate account. He could only say that she was more likely to fantasise than normal, but he could not say if she would go so far as to invent the episode. Dr Joseph strongly rejected a diagnosis of psychopathic personality disorder. He expressed the opinion that there was no medical evidence to show that in 1983 or 1986 Mrs A was suffering from any form of mental disorder, either mental illness or any longstanding personality disorder, which would affect her reliability as a witness. He considered that she suffered from a brief period of stress-related mental illness in 1964 due to marital difficulties, from which she made a full recovery within a year. This was of no significance when considering her reliability as a witness 22 years later. He vigorously rejected Lord Alderdice's diagnosis and disagreed with his conclusions. Her medical history would not raise a question mark in his mind about her reliability compared with that of the average person. He considered that if she had suffered from a personality disorder she would have had problems in school, in the workplace or in relationships with her children and others, but her general practitioner's notes and records showed no evidence of this.  We do not find it possible to accept any one of these opposing views in its entirety, rejecting any conflicting opinion. We think that the judge, if faced with the evidence of the three psychiatrists, would have recognised the need for rather more caution in accepting Mrs A's evidence as reliable than in the case of most witnesses, and this is the approach which we have adopted. If one took the opinion most adverse to accepting that evidence, that of Lord Alderdice, it appears clear that he regarded Mrs A as a person who would embroider an account and build an incorrect story out of threads in order to hold interest or attention. When one examines the sequence of events relating to the production of Mrs A's evidence, it appears to us very difficult to suppose that it was the product of any desire to seek attention or hold interest. She delayed for some time before she went to Father Murray and the tenor of her evidence does not appear to be that of a person desiring to thrust herself forward in order to gain attention. She seems rather to have seriously regretted that she had ever become involved. She went so far as to say that she did not want to get involved, she wished someone else had been there and seen the incident and that she had been led to believe that she would not be giving evidence.  If we approach Mrs A's evidence with caution, as adjured by the medical witnesses, we still find it impossible to suppose that she would have invented the whole incident in Lonsdale Street. The most that one might suppose is that she saw a person who she thought might have been Latimer, then persuaded herself positively over the next couple of weeks that it was Latimer. If the pattern of her behaviour fitted a possible embroidering of her account in this way, then we in this court, knowing her medical history, might have hesitated to accept the reliability of her account. As we have stated, however, the pattern does not fit such a conclusion and we would not regard her evidence as discredited or undermined. As recommended by Lord Bingham of Cornhill at paragraph 19 of his opinion in R v Pendleton  1 All ER 524, we have tested our provisional view by asking whether the evidence adduced on appeal before us, if given at the trial, might have affected the decision of the trial judge to convict the appellant. We are assisted in this exercise by the fact that, in contrast to an appeal from a jury verdict, we have a full judgment containing his reasons as well as his conclusions. The judge might well have exercised some caution before accepting Mrs A's evidence, if he had known her history, but we are satisfied that he would have reached the same conclusion regarding her identification of the appellant as the man whom she saw in the tartan cap and gold-rimmed glasses in Lonsdale Street as reliable. We therefore do not regard the conviction of the appellant as having been rendered unsafe on this ground.
Interviews of the Appellant We turn now to consider the course of the police interviews of the appellant during the period of seven days following his arrest on the morning of 29 November 1983. A total of 29 interviews took place in Castlereagh Police Office, in the course of which he made two written statements to which we shall refer.  Interview 1, 29 January, 10.35 am to 1 pm.
The appellant gave an account of his movements on 8 November, stating that he commenced duty at about 1 pm and was engaged in a foot patrol in the centre of Armagh. The members of the patrol later returned to the police station and were ordered to take two Land Rovers to the Portadown Road and set up a vehicle checkpoint. It was put to him that he was identified in Abbey Street prior to the shooting (strictly this was not correct, as it was an inference drawn by the police when they combined the information given by Mrs A and Mrs Dunne). The appellant denied this and said that he was with his patrol in the town. Interview 2, 29 January, 2.05 to 4.05 pm.
This time it was put to the appellant that he had been in Lonsdale Street in a Land Rover on 8 November, which the appellant denied. It was further put that he had been in civilian clothes and wearing a cap and glasses and that a witness had recognised him. The following exchange took place, which the Crown submitted was significant:
"Q. Why would this person state that you were in Lonsdale Street in civilian dress with members of the UDR in uniform.
A. I don't know.
Q. Why not.
A. It's bad I know but I can't explain.
Q. What do you mean that it's bad. But you can't explain.
A. Things look bad for me.
Q. Neill explain your involvement in the shooting of Adrian Carroll.
A. I can't, I wasn't involved.
Q. That's not the truth.
A. It is." Interview 3, 29 January, 4.05 to 7.05 and 8.05 to 11 pm.
In the first part of the interview the appellant denied that he had been involved in the shooting, but the detectives put it to him that there was strong information that he was. The note records that he remained silent and had his head down between his hands. It was put to him that he had shot Carroll and he said "Alright, I shot him but I don't want to say any names." An extended series of questions and answers is then recorded, in the course of which the interviewers asked the appellant questions to bring out his account of the events of 8 November and he pieced together that account, apparently answering the questions freely and willingly. At 9.05 pm he agreed to make a written statement, which was recorded between 9.05 and 9.37. This statement, Exhibit 7, read as follows:
"I want to tell you above what happened on the day of Adrian Carrolls shooting. The Cortina a greenish/bluish colour arrived at my house from Belfast a couple of days before the shooting. I don't want to say who brought it down. I put it in my father's garage down Bennetts Lane and opened the boot and seen the gun. I looked at the gun and she was full of rounds. I'm not sure now if the number plates were in the boot or I changed them. I put the gun back in the boot and closed it down and locked the garage and went up home. On the day of the shooting I phoned into work and told them that I wouldn't be in to late that I had to go to the Doctor's. I went down to Lonsdale Street that morning in my own car and put the duffle coat and cap and the glasses that were in the duffle coat in a blue bag under one of the huts. Then I went back up home. I messed about up at home and then I put my uniform in a white plastic bag and headed off down the town to the market, near Pinkerton's where I showed you on the map and I got changed into my uniform. I waited for the rest of the boys on the patrol to come in for a smoke and whenever they came and I had a smoke I headed off with them. I walked with them up to the Police Barracks and we got into two landrovers. I got my rifle there from one of the landrovers and said I had to go down to Lonsdale Street. We headed down to Lonsdale Street when it was just getting dark. I went to the back of the hut and got changed into the jeans, duffle coat cap and shoes. I left my uniform in the blue bag that they were in under the hut with the white bag I had the other clothes in. I told the rest of the patrol that I had to go somewhere to do something and they said that I just couldn't head off like that. After a wee bit of an argument I headed off and got the car. I went down the top of the Mall and saw that Carroll was still working there painting railings. I drove up round the shambles to Market Street and parked the car there. I left the keys in it on the floor. I headed down to Linenhall Street and waited for a while and then I went into McCrums Court which I showed you on the map and waited there until I seen him. I followed him out of McCrums Court and down English Street. I put the glasses I had on when I started following him. I followed him up passed the Post Office to where he lived in Abbey Street. He turned round just as he was coming up towards his house and that's where I shot him twice. Then I run off down College Street and into Lonsdale Street. I collected the bags and took the duffle coat off and changed into the denim jacket and then headed off up home. I stayed at home until around about 11 o'clock when I took my dog for a walk. I went down around the river where I marked on the map and hid the gun by the side of the river. I went back again the next day with the dog but the gun was gone. I realise that I was wrong in what I done and I am very sorry." Interview 4, 30 November, 9.10 to 10.30 am
The significance of this interview and the light which its content throws on the motivation of the appellant are such that we shall quote the body of the interview note in full:
"Latimer was brought from his cell to interview room escorted by a uniformed officer. He was already aware of our identities and the nature of our enquiries I cautioned him. He nodded. He stated he had thought about last night and hadn't changed. He was asked if he meant he still didn't want to tell us who else was involved he replied 'yes'. He was asked if he was willing to provide his fingerprints. He agreed. He was asked if he did not think that he should tell us everything he knew about this as he was obviously not doing this. He sat with his head in his hands but did not reply. It was put to him that other UDR men were involved and he was trying to protect them and take it all on his shoulders. He stated that the rest of the patrol knew nothing about it. He was asked about the Cortina car which he said he had used. He said he didn't remember changing the number plates and didn't want to mention any names. It was put to him that we knew he was telling lies about what he did that day. He was shown a blue cash book marked issue book B vehicles (RL1) and it was pointed out that he had apparently signed out a vehicle at 1 pm that day the 8th November. He was asked to explain this. He sat for a while thinking and then he looked up and said 'None of the patrol were involved and I didn't shoot the man'. He was asked what he meant by this. He said 'I was on duty that day and I didn't shoot anybody'. It was put to him that he had already made a statement admitting the shooting and we knew he was trying to protect the others involved. He was asked about the notes we had recorded before the statement. He said that both the notes and statement were lies. It was put to him again that we were in no doubt that he was trying to protect others who were involved as we knew that things he had said in his statement were wrong. He appeared quite depressed and held his head in his hands. He stated 'it's all lies'. The interview ended. The notes were read over to him. He agreed they were correct but declined to sign them."
We shall comment later in this judgment on the conclusions which we consider should be drawn about the factors which caused the appellant to retract his earlier confession, with particular reference to the theory advanced by Professor Gudjonsson to explain it.
 Interviews 5 to 11, 30 November and 1 December
The appellant maintained an adamant denial of any involvement in the shooting, stating that his previous admissions were all lies and that he had made them up because he thought that it would be easier for him if he did. He gave an account of taking the Land Rovers to carry out a search on the Moy Road, but averred that they did not stop at Lonsdale Street. Interview 12, 1 December, 2.25 to 5.45 pm
When taxed again with the identification evidence the appellant stated that the Land Rovers may have stopped in Lonsdale Street. When pressed on this he said "All right we did stop and the boys got out to give cover." He was the driver of the second vehicle and stopped because the lead vehicle did. He did not get out of his Land Rover. He then said that a man wearing a blue duffle coat, cap and glasses came out from between the huts and got into the front Land Rover. He denied that he knew who the man was, but named the two soldiers who were with him. He only put two and two together later and realised what was going on. He said that he had made the statement earlier to "keep the other boys out of it". Interview 13, 1 December, 7.0 to 9.40 pm
The appellant maintained his account, but in this interview referred to two soldiers Bell and Worton coming out of the grounds of the technical college with a man between them. He said that he did not want to say who the man was, implying knowledge on his part of his identity. When they reached the police station they were sent out to do the VCP on the Portadown Road, because of the shooting of Carroll a short time before. As they were travelling there Lcpl Hegan told him not to hurry, as it was "our own side" that had done it and they did not want to catch him. Interview 14, 1 December, 9.10 to 11.30 pm
When the appellant was asked why he had made a statement that he had shot Carroll, if he now denied that he had, he sat with his head in his hands and said "Jesus Christ, his wife's father has just died. Do you not think he has enough troubles?" He was asked whether it was his brother's wife to whom he referred, and he replied that it was. He agreed when asked further about this that he had made up his original story to protect his brother and his other colleagues in the UDR. Later in the interview he was asked whether the civilian in Lonsdale Street was his brother, to which he replied "I hope to God it wasn't". Asked again he said "I don't know". He repeated an account of picking up the civilian at Lonsdale Street and seeing him heading towards McCrum's Court after being dropped off at the Mall. He was asked if that civilian was his brother, to which he replied "I'd rather not say". Interviews 15 to 18, 2 December
The appellant continued to develop the version which he had put forward in Interview 14, that the party had rushed the search at the Moy Road and that he was told that there was a "wee job" on and was instructed to stop and pick up a person at Lonsdale Street. He stated that the civilian and two accompanying soldiers had got into his own Land Rover, which was the lead vehicle, and that he had slowed down in the Mall at McCrum's Court to drop the civilian off. In Interview 16 he said that he knew him but declined to say who he was, then in Interview 18 he said that he recognised him as his brother David. In Interview 17 he agreed that he had been willing to go to jail for murder to protect the others, saying that he did not think he would do all that long and that it was easier on a single man than a married man. Interview 19, 2 December, 7.40 pm to 12.25 am.
The interviewers told the appellant that a witness had been interviewed and said that she definitely saw the appellant getting into a Land Rover in Lonsdale Street and not his brother David. They put it in detail to him what she had seen and he appeared to get very depressed and put his head down in his hands. They told him that all he had said before was a lot of lies and they believed that he did shoot Adrian Carroll and they wanted him to tell the truth now once and for all. The appellant was quiet for a while and the detectives asked him now to tell them the truth. DC Orr asked him "You shot him, didn't you?", to which the appellant replied "Aye". He said that the first statement had a "pile of lies" in it, then he gave a detailed account of his part in the incident in response to a series of questions put by the interviewers. Following this he agreed to make a written statement, which was commenced at 11.40 and completed at 12.15 am. The statement, Exhibit 9, read as follows:
"I want to tell you the truth about shooting of Adrian Carroll. Me and Jim Hagan were walking out of the camp after finishing the day before the shooting and Hagan said to me we're going to shoot Carroll tomorrow and I was going to do the shooting. He said to bring in clothes tomorrow when you come in to go to work, a jacket and trousers. I just said `yes'. I came in the next morning and went and got a Rover and brought it over in front of the loading bay. I went up to my car and got my clothes. Then I put the clothes under the seat. I lifted the loose seat and then the lid of a sort of box over the wheel and put the clothes in there and then went into the brief. I came out again and went over to the loading bay and loaded my rifle with everybody else and got into the Rovers. We headed off towards the Moy and Hagan asked me did I bring my clothes and I said yes they're in the back of the Rover. Worton and Bell were in the back of the Rover and they knew what was happening because we discussed it earlier. Then on the way out Hagan told me that Sergeant Rolson would be leaving the search early and I would be going with them. He said I would be dropped off at Lonsdale Street where I was to change behind the Technical School Huts. We went on out to the search and during the search four of us or five got together at one stage and worked out what was happening. There was me and Hagan, Worton and Bell and I don't know if Sergeant Rolson was there or not. Then just before the search was finishing I headed off with Sergeant Rolson in his Rover. I was in the back and Sergeant Rolson was in the front I think Winston Allan was driving. I had moved my other clothes into Sergeant Rolson's during the search just before we headed off. I left my SLR Rifle in L/Cpl. Hagans Rover the one I had been driving. We headed off to Lonsdale Street and I got out just beside the huts in Lonsdale Street and went in behind them and took my clothes with me. Sergeant Rolson headed off in the Rover to the Police Barracks. I put my jeans on over my UDR trousers and I took my UDR jumper and jacket and beret off and put on the other jumper and the duffle coat. Then I waited for the Rovers to come. The Rovers came along I heard them coming and I started to walk out. I put the cap and glasses on and I met Worton and Bell. I got into the back of the Rover and Hagan was driving. We headed off towards the Mall and Hagan handed me the gun over from the front into the back. We got up as far as McCrums Court and we stopped and I got out. I headed up to McCrums Court through Pinkerton's yard where I waited for Carroll to come along to clock off. I waited for all the rest of the boys to move off and then I followed him down English Street and up round by the Post Office round by Abbey Lane. He was heading up towards his house and then he turned round and that's when I shot him twice. Then I run off down College Hill where I had arranged to meet the Rover. I got into the Rover at the bottom of College Hill and we headed off up the Mall again and I got changed back into uniform again and then we headed off up to the Police Station. I give the gun back to Hagan and he asked me Did I do it and I said Yes and one of the other boys in the back said did anybody see you and I said I don't know I don't think so and then we headed on up into the Police Station and just acted as if nothing had happened." Interview 20, 3 December, 10.20 am to 12.0
The appellant was asked about other incidents in the area and specifically the shooting of Peter Corrigan on 25 October 1982. He said that he remembered the shooting, but denied that he had been involved in it. Interview 21, 3 December, 2.45 to 3.15 pm
The appellant was again asked if he had been involved in anything else and again denied it. He agreed to a confrontation with his co-defendant Colin Worton and was taken to the room where the latter was being interviewed. Detective Sergeant O'Sullivan asked the appellant and Worton a series of questions, which they answered as follows:
"O'Sullivan: Do you know this man?
Latimer: Colin Worton.
O'Sullivan: Do you know this man, Colin?
Worton: Neil – Neil Latimer.
O'Sullivan: Neil, have you made a statement about the shooting of Adrian Carroll?
Latimer: I have."
Worton interrupted Latimer and shouted him down, saying "Look at me when you are telling lies." Latimer said "I wrote it in the statement about the shooting. That is the truth." Worton replied "Thanks, Neil, lying cunt." Interview 22, 3 December, 3.35 to 6 pm
The appellant agreed that he had told the complete truth and admitted that he was the gunman who shot Carroll. He was again pressed about other incidents, with particular reference to the shooting of Peter Corrigan, but continued to deny that he had been involved. Interviews 23 to 26, 4 December
The appellant was asked a number of times if he had been involved in any other terrorist incident, but continued to deny this. In Interview 25 he was asked if he was sure that he was correct in stating that he fired two shots at Carroll, and replied "I'm not sure. It could have been two or more." Interview 27, 5 December, 1045 am to 12.15 pm
The appellant identified a blue parka jacket as the one which he had worn when he carried out the shooting. He said that he had carried the gun in the right hand pocket. Interview 28, 5 December, 2.20 to 3.35 pm
The appellant agreed to go to another room where Lcpl Hegan was and to tell the truth in front of him. When he was taken to that room he identified the written statement which he had made and confirmed that the admissions which he had made in it were true and correct. When informed that the appellant had named him as being involved in the murder of Carroll, Hegan said to him "What in God's name are you involving me in this for?" Interview 29, 5 December, 7.35 to 8 pm
The appellant was again asked if there was anything more he ought to tell the interviewers, but replied that there was not.
Assessment of the Interview Evidence At trial the appellant challenged the reliability of all his oral and written confessions, although he did not then attempt to argue that they should be excluded from admission in evidence. In his testimony he said that the pressure and impropriety of the interviewing officers frightened and depressed him. He accepted that he had said at least the large majority of what was attributed to him, but claimed that that was all a bundle of lies brought about by his reduced state of mind and by his overpowering desire to get out of Castlereagh. The judge summarised at pages 20 to 22 of his judgment the reasons put forward by the appellant for his retraction of his first admissions and final confession in different terms:
"He explained the retraction of his confessions of that evening. When they showed him the UDR records that indicated he had signed out a Land Rover on the day of the murder at 1.00 pm he realised that he had been on duty that day that Carroll was shot and that did not fit in with the lies he had told the police in Exhibit No. 7.
Latimer explained that what led him to the later phase of involving the UDR in the murder but not himself as the gunman came from the detectives' suggestion that if they were willing to accept he was not the gunman he must still know something about it as the driver of a Land Rover that day. He went along with this although it was not of course true. Again it came from the detectives that the gunman might have been someone who resembled him in appearance such as his brother David and eventually he went along with that one too.
When he confessed again that he was the gunman and his comrades were involved in the murder, this was made up by him from what the detectives had told him, some of it was his own invention or came from talk about the murder afterwards around the camp. And it was given by him because the conduct of the detectives had put him in such a state that he wanted to get out of Castlereagh and he would have done anything to achieve that.
He said he would have named his own mother as the gunman to get himself out. He confronted Worton because he was told it would be better for him if he did. He said he'd put the gun in the right-hand pocket of the jacket they produced because he was going along with the police and because he was right-handed."
 The trial judge regarded the evidence given by the appellant in a poor light, saying that he "cut a sorry figure" in the witness box. He said that he lied constantly and that his evidence was full of absurdity and untruths. He pointed out that whereas the appellant attempted to explain his admissions by saying that he wanted to go along with the police and was telling them what they wanted to hear, on very many occasions he did not do so and, as he put it at page 24 of his judgment –
"the record of the evidence abounds with instances of when he did not agree with their suggestions or qualified or varied or added to them."
Nor did he attempt at any time, when he had made his first statement, to hold the detectives to the promise to release him from Castlereagh which he said they had made or complain of any ill-treatment or oppressive behaviour. The judge said that the appellant's efforts to explain why he made the second written statement were "pathetically weak". He rejected his explanations of why he went through with the confrontations with Worton and Hegan, which he regarded as admissions of very considerable weight. He expressed the view at page 23:
" … I find it hard to believe that a soldier after only two interviews at which the sum of the ill-treatment was nothing more than shouting, a threat and a promise of favour, would falsely confess to a murder and to the prime role in it. And this even when the only additive to his upset at the third interview was persistent questioning by two fresh detectives."
Again he said at pages 44-5:
"… I am considering the weight of the confessions of a young man, but an adult, of average intelligence and one who is not an idiot or a fool. But more than that. I am considering the weight of the confessions of a member of the Security Forces, a serving soldier in the Ulster Defence Regiment given to other members of the Security Forces, RUC detectives.
That such a person should confess to a murder of which he is completely innocent, and a sectarian murder at that, I find unbelievable. That he should confess to such a murder within hours, not days, of questioning at Castlereagh and because of impropriety no more formidable than shouting verbal abuse, threats and promises and persistent questioning I find quite unbelievable. And that he should agree in his confessions to the extraordinary, if not incredible, incident involving disguise and mock arrest in Lonsdale Street is hard to swallow." The judge concluded that the appellant had reasons which were responsible for the changes in his story and that he was in full control of his situation at Castlereagh. He set out at pages 23-4 his own assessment of the course which the appellant took in making admissions, retracting them and then finally confessing again:
"His verbal confession and the written statement that followed were carefully thought out to avoid bringing in his fellow soldiers to the plot. His change the next day to retraction of his admissions and an assertion of non-involvement yet an involvement of his fellow soldiers was made because he thought this might be acceptable to the detectives. Then when he realised the detectives were not content to let his story rest with an unknown civilian as the gunman and later an unnamed civilian, he was prepared to insinuate, but with feigned regret, his brother David as the gunman. And when he felt they did not dismiss this out of hand he was bold enough to say his brother was the gunman.
It was when he was faced finally on Friday, 2 December, with the statement that witness A, who had been re-interviewed that day, was quite definite that it was he and not brother David whom she saw get into the Land Rover in Lonsdale Street that brought his resigned 'Aye' to D/C Orr's assertion 'You shot him, didn't you?' and the long detailed verbal admission and written confession that followed." The Court of Appeal in the 1988 appeal accepted the judge's conclusions as correct, not merely because he saw and heard the witnesses and the court entertained the customary reluctance of an appellate tribunal to disturb a trial judge's findings of primary fact, but because the court itself considered that his findings were unimpeachable. In giving the judgment of the court Lord Lowry LCJ reviewed the course of the interviews of the appellant and concluded at page 14:
"It is difficult to imagine a stronger case for the Crown. It is impossible to accept that a reasonably sharp individual, as Latimer was found by the learned trial judge to be, a member of the UDR, who was in no way physically ill-treated, would wrongly confess to a murder after only a few hours of questioning. It is impossible to accept that such a person would tell a series of lies in the course of interrogation if he was, as he alleged, a totally innocent person. It is equally impossible to accept that a person who had made confessions which he alleged were false would confront a fellow member of the UDR, admit that he had made these confessions and state that they were true.
The defence, however, faced with what we consider to be an impossible task, proffered the novel defence that he was such an unreliable witness that it was impossible to accept even his confessions of guilt at face value. Reduced to its simplest terms, this contention means that the more lies a defendant tells while being interrogated by the police, the less likely it is, when he confesses, that his confession will be true. Experience would suggest otherwise. The more lies that are told to investigating officers before a confession is made, the more likely it is that the eventual confession is true. Any tribunal of fact, listening to a series of lies told to investigating officers and admitted to be such by a defendant, would pause to ask the question 'Why were these lies told?' The obvious answer is that the person telling the lies is endeavouring to put off the final moment of having to confess. In the absence of any indication that the defendant is or was mentally unstable, any suggestion that the mere telling of lies to investigating officers makes a confession subsequent to such lies unreliable is in our view a totally untenable proposition."
The court "unhesitatingly agreed" with the judge's finding and dismissed the appellant's appeal. The Court of Appeal in the 1992 appeal reached a similar conclusion. It gave extended consideration to the effect of the ESDA evidence, which showed clearly that on a number of occasions the interview notes had been amended and the authentication certificates of senior officers had been appended, not immediately after the interviews concluded, as they deposed, but at some later time. The court stated its conclusions on this matter at pages 14-15 of the judgment of the court given by Hutton LCJ:
"In the course of the trial the detective officers who had interviewed the four appellants all relied on the written notes which they produced in the witness box as containing an accurate account of what had been said in the course of the interviews, and their evidence as to what questions they had put to the appellants and what the appellants' answers had been was entirely dependent on those notes. The detective officers also gave evidence that the notes which they relied on in the witness box were the notes which had been written in the course of the interviews and that no notes had been rewritten at a later stage after the respective interviews were concluded.
The ESDA findings have clearly shown that detective officers gave evidence which was untrue when they told the trial judge that none of the interview notes had been rewritten.
It is also clear from the ESDA examination that the dated and timed signatures of senior officers on some of the presented interview notes authenticating the notes, which authentications were intended to demonstrate that the notes had been completed before the authenticating signatures had been appended and to show that there had been no rewriting of the notes, had in some instances been falsely appended at times and dates later than the authentications stated."
Hutton LCJ pointed out at page 18, however, that there was no evidence that the police concocted any of the confessions of any of the appellants or any suggestions made by the police which the appellants had not accepted. He observed that all of the notes recording Interviews 12 to 19 (except 26 and 28) were signed by Latimer, which appeared to indicate that any signed pages in those notes which bore signs of rewriting must have been rewritten during the course of the interviews themselves. One could hardly suppose that in such circumstances there was anything sinister in the rewriting of those notes. He concluded accordingly on this part of the appeal:
"Therefore, the ESDA findings cast no doubt on the fact that Latimer did make the confessions, both verbal and written, and did take part in the confrontations with his co-accused, which the police interviewers said he did make and take part in. It is clear beyond a doubt that Latimer did make these confessions and did take part in the confrontations with his co-accused, because in his own evidence at the trial he agreed that he made those confessions and took part in those confrontations." The court went on to review the course of the interviews and the appellant's confessions and retraction. It took a similar view to that adopted by the court in 1988. It found the remark of the appellant in Interview 2 that he could not explain but things looked bad for him very difficult to reconcile with his innocence. In relation to the appellant's claim that he made admissions to "get the police off his back" the court expressed the view at pages 72-3:
"Furthermore, it is clear that he did not make the confession on the night of 2nd/3rd December 1983 because he was improperly pressurised by the police and because the pressure from the police caused him to be in a condition in which he just told the police what they wanted to know. Such a suggestion is clearly invalid because after making the verbal confession and the written statement in the third interview that he had shot Carroll (although giving an account to the effect that he alone was involved in the shooting and that none of the other members of the patrol were involved) Latimer retracted that confession in the first interview the next day, 30th November, and he maintained his retraction and his denial of involvement in the shooting during 14 further interviews.
This undisputed fact, that Latimer retracted his first confession and then maintained that he was not involved in the murder throughout 14 further interviews undermines his claim that the pressure of the police was such that to get them 'off his back' he just told them what they wanted to hear. For 15 interviews, during interviews 4 to 19, Latimer did not tell the police what they wanted to hear. It is clear that Latimer made his full and detailed second written confession on the night of 2nd/3rd December, not because of improper police pressure, but because the police told him that they had interviewed Witness A and she confirmed that it was he who had got into the landrover in Lonsdale Street wearing civilian clothes. It is clear that at that stage Latimer realised that the game was up because he was clearly implicated by the evidence of Witness A.
A further consideration is that if Latimer was an innocent man pressed by the police into making a false confession, he would never have given the detailed and elaborate account of the shooting contained in his first written statement, which was clearly designed to protect other members of the UDR patrol."
Hutton LCJ expressed the final conclusion of the court on the appellant's confessions at page 81 of the judgment:
"We are satisfied that Latimer was not an innocent man pressurised by the police into making a false confession. Rather it is clear that, once Latimer was told on the first day in Castlereagh, that a witness had seen him getting into the landrover in civilian clothes in Lonsdale Street, he realised that the police had a case against him, and he twisted and turned to avoid telling the full story and to try to protect his comrades until, faced on 2nd December, with Witness's A confirmation of what she had seen in Lonsdale Street, he realised that the game was up and he made a full and truthful confession."
The Psychologists' Evidence We have set out in some detail the course of the interviews of the appellant and the findings of the several courts which have considered them, because it is necessary to take critical account of them when assessing the fresh evidence adduced on this appeal. The evidence on this issue came from Professor Gisli H Gudjonsson and Dr Michael Heap. Dr Heap was unfortunately taken ill shortly after commencing his examination-in-chief and had to return home. By consent his reports were received in evidence, together with such oral evidence as he had given. We shall have regard to his opinions, but bear in mind that there was not an opportunity to cross-examine him. To that extent his opinions are untested and the weight which we can place upon them is accordingly limited.  Professor Gudjonsson is a forensic psychologist, who has made a speciality of investigating the vulnerability of persons detained by police for questioning, in which he has a great deal of experience, has carried out very considerable research and has published books and papers on the subject. He is acknowledged as the leading authority on this topic and if his opinion had been available to the court of trial, the judge would clearly have been required to pay considerable attention to it.  Professor Gudjonsson carried out a battery of tests on the appellant on 7 April 1999 over the course of a whole day. He repeated one of the tests on 8 May 1999, for reasons to which we shall refer, and prepared several reports which were produced to the court. He also gave oral evidence by way of supplementing his reports and was cross-examined by counsel appearing for the Crown.  He expressed the view that the appellant was of average intellectual ability. In most of the tests he found that the appellant came within normal limits, and it is not necessary for us to refer to them in any further detail. The Eysenck Adult IVE test showed that he tended to act more impulsively than normal. Other tests showed a high degree of trait anxiety and state anxiety, mild clinical depression and a moderate degree of anxiety and of hopelessness. The Eysenck Personality Questionnaire showed moderately elevated scores of psychoticism and neuroticism and the addiction and criminality propensity scores fell well outside normal limits. The profile, according to Professor Gudjonsson, was that of an unstable (emotionally labile) person who has rather introverted tendencies and personality difficulties.  The appellant's suggestibility, measured by a test and on a scale devised by Professor Gudjonsson himself and bearing his name, was within normal limits. He found, however, that the appellant was abnormally compliant. There was a significant number of confabulations in memory recall, outside normal limits, which indicates that what the appellant recalled he recalled very inaccurately. Professor Gudjonsson explained that the suggestibility test is designed to elicit the extent to which the person tested is willing to accept something untrue which is put to him and to believe it to be correct. The compliance test, on the other hand, assesses the degree to which he tends openly to agree with information, suggestions and instructions from others, despite his private wishes or beliefs to the contrary, that is to say, even if he knows the material to be incorrect. As Dr Heap put it, abnormally compliant persons appear motivated to do so through an eagerness to please and to avoid conflict with other people, particularly those in authority.  Professor Gudjonsson also administered the Minnesota Multiphasic Personality Inventory (MMPI) test, on which Dr Heap commented, although he did not himself see or test the appellant, who did not give permission for that. The MMPI test is described as a comprehensive and widely-used personality assessment procedure, which gives scores on a very large number of personality dimensions by comparison with the general population and so may give an indication of the personality profile of the person tested. Dr Heap emphasised in his letter of 27 September 2001 to the DPP that one should not make assertions about a particular individual's personality and mental state from the scores registered on this test and that a psychological profile should be interpreted only with reference to all other information gathered about him. Professor Gudjonsson did say in the course of his evidence that the MMPI test was not in general terms of particular relevance to the present case.  Dr Heap pointed to the very high "F score" of 101 registered on the first occasion when the MMPI test was administered to the appellant. In his letter of 27 September 2001 he stated at paragraph 4:
"4. The MMPI has a number of validity checks in the form of scales on which very high or very low scores may indicate that caution should be exercised in how the remaining profile is to be interpreted. A respondent may, for example, lose concentration or be unmotivated to participate, and may therefore complete the questionnaire in a careless and inconsistent manner. Some respondents may be in a situation in which it is in their interests to appear 'normal' and to deny any personal difficulties, or this may be their usual defensive style. On the other hand, it is sometimes in a person's interest to present as having psychological problems or even a mental disorder. In these cases the person may, deliberately or unwittingly, bias his or her replies in the desired direction. The validity scales on the MMPI allow one to detect if the respondent is being inconsistent in his or her replies or is biasing them in a particular direction. This means that sometimes a profile has to be declared to have a strong possibility of being invalid."
In his report of 18 December 2002 at paragraph 9.34 he stated that he was not drawing a conclusion of deliberate malingering on the appellant's part. He repeated in several passages, however, that a high F score may be an indication of unconscious exaggeration or even tailoring of the answers to the result which the respondent might perceive was the subject of interest, for, like the compliance scale, it was a self-reporting test, with all the concomitant weaknesses of such tests. He expressed the opinion that a high F score was sufficient to invalidate the whole test and to throw doubt on the validity of the findings made as the result of other tests. Professor Gudjonsson said that when he considered the results of the first MMPI test he thought that the F score might be the result of fatigue, as the test had been taken after a long day's tests and assessment. He therefore arranged to administer the test again on 8 May 1999. On this occasion the F score was recorded as 98, again seriously elevated and indicative of the same tendencies. He was not satisfied that the result was correct when he examined the answers from the second test and discovered that there had been a computer error, the correct score being 70, which is a score of normal level.  Professor Gudjonsson set out a number of conclusions in his report of 16 October 1995. In paragraph 2 he stated that the appellant's verbal memory scores, indicating his capacity to process new information and remember it, fell well outside normal limits, suggesting that his verbal memory is much worse than that expected from a person of his intellectual abilities. We might comment that this may be against rather than in favour of the case made on behalf of the appellant. His written statements contain a high degree of detail, which on his case was fed to him by the police. He accepted in his evidence in chief, however, that he dictated the first statement and the police "just wrote it down" (transcript, 9 March, page 47). In cross-examination he said that he thought that he dictated the second statement (transcript, 12 March, page 8). If he was giving these accounts from material which the police had been putting to him, it would be a considerable feat of memory. Accordingly, if he did not have a good memory, as Professor Gudjonsson's test indicated, it may tend to prove that such material was not in fact the source of what he dictated to the interviewers.  Professor Gudjonsson went on to say in the same paragraph that on one of the tests the appellant produced abnormally high confabulation scores, which suggested that he had a high capacity for imaginative thinking. Such a person could fill in gaps in his account by inventing plausible details.  On the issue of compliance Professor Gudjonsson stated his opinion at paragraph 5 of the conclusions:
"5. On the Gudjonsson Compliance Scale Mr Latimer obtained a score of 17, whereas when assessed by Mrs Tunstall he obtained a score of 15. The difference in scores is not significant (ie it is a kind of discrepancy one would expect by chance when the test is re-administered). Both scores fall outside normal limits. The present score falls in the 98th percentile rank, which suggests that Mr Latimer is more compliant in his temperament than 98 per cent of the general population. This indicates that he is exceptionally eager to please people and tries hard to avoid conflict and confrontation with people in authority."
He then continued in the final paragraphs 9 to 14:
"9. The present assessment also indicates that Mr Latimer has a number of psychological problems, most of which are probably durable characteristics. These relate to personality disturbance, impulsivity, very poor self-esteem, a strong tendency towards compliance, poor verbal memory, and confabulation responses during memory recall. When assessed on some of the same test in 1995 by Mrs Olive Tunstall, the scores were consistent with the present scores. Taken together, these findings suggest that Mr Latimer is a psychologically vulnerable individual, whose ability to cope with police interrogation and custodial confinement would be greatly impaired.
10. Although it is not known how Mr Latimer would have performed on the various tests at the time of his arrest in 1983, it is highly probable that the problems highlighted by the test were present at the time of his arrest. The personality tests reveal reasonably stable characteristics. As an example, the very low score on the Gough Socialisation Scale largely relates to past behavioural problems which are evident from childhood onwards, and is consistent with poor self-evaluation. In addition, there is support for Mr Latimer's longstanding poor self-esteem from the Semantic Differential Scales. At least on the basis of his self-report on the test, in some respects he also had a poor self-evaluation in 1983.
11. It is evident from the trial judge's judgment and the appeal judgments, that the following factors weighted heavily against Mr Latimer:
- The trial judge considered Mr Latimer to be a liar, stating that the cross-examination soon exposed the `absurdity and untruths' of Mr Latimer's evidence.
- The judge clearly found it impossible to believe that Mr Latimer would have made a false confession without physical ill-treatment.
- Mr Latimer appeared to be of average intellectual abilities. This was apparently seen as a great strength which gave credibility to the confession.
- Mr Latimer was a member of the Security Forces (The trial judge seemed to assume that this made Mr Latimer's confession more believable, which is without a proper foundation).
- The appeal judges were concerned that at the time of his arrest Mr Latimer did not make a protest, which they had expected if he was innocent of the murder (This appears to be an unfounded assumption and should not be used as evidence of Mr Latimer's guilt, particular in view of what we now know about his personality).
- Mr Latimer's ability to maintain his retraction between interviews 4 and 19.
What the trial judge and the appeal judges appear to have failed to appreciate, which very probably relates to the lack of scientific knowledge about false confessions at the time, are the following:
(a) that persons of average intellectual abilities and without mental disorder do on occasions make false confessions to murder;
(b) psychological pressure and the belief that making a confession, even when false, may expedite their release from custody, are powerful forces which do sometimes result in detainees making a false confession to murder. Physical pressure is not a necessary condition for making a false confession.
12. With regard to Mr Latimer's confession in the third interview is concerned, he was still able to resist pressure to the extent that he refused to name others allegedly involved and refused to sign his statement. At this point in time, he had not broken down to the extent that he was agreeing with everything requested of him by the police. The following day he retracted his confession and maintained his innocence over a period of three days. Once he broke down during the 19th interview his behaviour became completely `reactive' (ie he was now signing his statement and implicating others; by this time his will appears to have been completely broken). The retraction in the fourth interview is best construed as an activation of `strategic coping', which sometimes occurs when suspects feel that they have been pressured too much by the police (ie the excessive pressure results in a `boomerang effect' and increased critical thinking and resistance which then becomes difficult to break down). I believe that the trial judge may have misconstrued the psychological nature of Mr Latimer's retraction and inappropriately used it as evidence of his guilt.
13. The present assessment indicates that Mr Latimer possesses unusual personality characteristics, which very probably made him psychologically vulnerable during the detention and interrogation in 1983. If these idiosyncratic features had been known to the trial judge then he may have interpreted the confession and Mr Latimer's demeanour during his testimony at trial differently. There remains, of course, the testimony of Witness A, whose reliability I have not directly assessed.
14. In view of the findings from the present psychological evaluation, I have serious reservations about the reliability of the self-incriminating admissions Mr Latimer made to the police in 1983." These conclusions appear on their face to be difficult to reconcile with the appellant's retraction and his sustained maintenance throughout a number of interviews of the denials he then advanced. Professor Gudjonsson explained this by putting forward a theory in paragraph 12 of his conclusions that the retraction is best construed as an activation of "strategic coping", when excessive pressure results in a "boomerang effect" and the person engages in resistance which is difficult to break down. At the request of the Criminal Cases Review Commission he expanded on this suggestion in a further report dated 3 March 2001. After referring to the sources of the concept he stated at page 2 of the report:
"I have also shown from my research of actual police interrogation how discontinuation of reactive behaviour and sudden activation of strategic coping can result in the lowering of normal suggestible and compliant behaviour during interrogation (Gudjonsson, 1995). The most likely explanation for such a change is a strong sense of injustice and/or feelings of anger, which can be sufficient to cause a sudden activation of critical thinking, focussed mental energy, and assertive behaviour."
We would observe, however, that the appellant did not aver at trial that he had any sense of injustice or feelings of anger. Professor Gudjonsson advanced the view that the additional pressure in Interview 4 to reveal the names of other people involved in the murder caused the appellant to retract his confession. He then returned to the point in a supplementary report dated 22 October 2003. He referred to the appellant's statement in evidence that when the police showed him the book which proved that he signed out a Land Rover on 8 November 1983, contrary to his previous assertion that he had been on a foot patrol, there was nothing else that he could do but tell them the truth, that he did not shoot Adrian Carroll that day. He said at page 21 of this report that those comments in the appellant's evidence provided an important explanation of his retraction of his earlier confession:
"The previous day he had given the police an account, which in part the police could prove was false. When confronted by the police with the evidence about his having signed out the Landrover, which undoubtedly came as a complete surprise to him, Mr Latimer reacted in a global, rigid and overgeneralized (all or nothing) fashion by retracting his confession and going back to his initial denial. At the time he seems to have believed that this was the only course of action available to him. In reality he of course had other options, but his cognitive rigidity, which was evident on a number of other occasions during his testimony, meant that he focused on small pieces of information and overgeneralized interpretations of his predicament." It is to be noted that at page 3 of his report of 3 March 2001 Professor Gudjonsson puts forward another theory, that the police had already suggested to the appellant the possible involvement of his brother in the murder rather than voluntarily implicating his brother. They then used it as additional pressure to obtain another confession from him, putting him in the position where he had to choose between implicating himself and implicating his brother. The passage which he quotes from the interview note "We put it to him 'Was that civilian your brother?' does not, however, give any support for that theory, for it relates to a later part of Interview 14 (page 138 of the book of interview notes). The sequence recorded in the notes and not challenged by the appellant is set out at page 134:
"He was asked why he had then made a statement saying he had actually shot Carroll. He sat with his head in his hands and stated 'Jesus Christ his wife's father has just died. Do you not think he has enough troubles?' He was asked was this his brother's wife he was talking about, he said it was."
It seems entirely clear from this passage that the suggested involvement of his brother came from the appellant himself, without any detectable prompting from the police. This discredits Professor Gudjonsson's theory on this aspect and may indicate undue readiness on his part to accept what the appellant told him. We note that he says at page 17 of his report of 30 August 2000 that the appellant claims that he repeatedly requested to see a solicitor, but this was refused until he was transferred from Castlereagh to prison. The custody record has no trace of any such request by the appellant, which one would expect to find if he had made requests during his detention there. He gave no evidence at trial that he had asked for a solicitor. It is not clear whether Professor Gudjonsson was again uncritically accepting the appellant's averment. In any event, it shows that the appellant's recollection in 1999 was badly at fault or he was embroidering his account.
 There were several other inconsistencies between the account which the appellant gave to Professor Gudjonsson and the evidence which he gave at trial. The matters which these involved included the following:
• the timing of the incident recounted by the appellant when he claimed that a police officer stuck a pen-knife into the table between his fingers;
• the implicit threat that police officers would beat him up if he did not confess, which he did not allege at trial;
• his claim to Professor Gudjonsson that he was intimidated by officers shouting at him, whereas he stated in evidence that the officers who took his first confession statement did not shout or make any threats or promises;
• the allegation that he heard shouting from another room during the night as if the police "were kicking the life out of somebody", which he made to Professor Gudjonsson but did not mention at trial;
• the appellant told Professor Gudjonsson that he thought that he would be released from custody if he gave the police some incriminating admissions, but he had eventually admitted in cross-examination that he did expect to go to prison after admitting shooting somebody and that he did not think he would get back home again.
Professor Gudjonsson does not appear to have noted these inconsistencies, which we would regard as significant.
 We would also draw attention to the fact that he told the professor that he was motivated to make the confession because he was shocked by what the police told him concerning the murder of somebody in retaliation for the murder of Adrian Carroll. He also said this at trial and stated that the police told him that the Darkley massacre was perpetrated in retaliation for Carroll's murder. He said that his mother's cousin was shot in that incident and that it was "playing" (sic) on his mind. We find it difficult to believe that this could have been a reason for his confession unless he was in fact involved in the murder of Adrian Carroll. Dr Heap rejected Professor Gudjonsson's interpretation of the appellant's behaviour at the police interviews, on the ground that the psychological assessments did not provide reliable information for such interpretation. As we have stated, in the absence of cross-examination we cannot attribute the same weight to Dr Heap's views as we might if they had been fully tested and accepted by us, and we bear in mind that he did not have the opportunity to examine and assess the appellant for himself. Nevertheless, some of the points which he made do appear to us to have substance. First, the tests were administered by Professor Gudjonsson some thirteen years after the trial, during most of which time the appellant had been detained in prison. One might reasonably question whether in consequence of that experience any of the psychological findings concerning the appellant might have varied in 1986 from those made in 1999. Secondly, a confession resulting from pressure may be true as well as false, though one must recognise that if it is not truly voluntary there must be a risk that it is false. It does seem to us nevertheless that psychological evidence may in general be of more relevance in determining the voluntary nature of a confession rather than the reliability of its content. Thirdly, there is very little background material, such as evidence from the appellant's early life, to demonstrate the traits determined only by psychological testing. Fourthly, the clearest evidence of what prompted the appellant to confess, retract and confess again is likely to be his own testimony given in court. If this is found to be at variance with the psychological theories, it tends to throw the latter into question.
Admissibility of the Confessions We come then to consider the admissibility of the appellant's confessions and the weight to be attached to them, in the light of the fresh evidence adduced. Although no application was made to the trial judge to exercise his discretion to reject the oral and written statements on the ground of oppression, it was argued before this court in the 1992 appeal and again at the hearing before us. In his judgment in the 1992 appeal Hutton LCJ considered the issue in detail. He stated first that the confessions were admissible under section 8 of the Northern Ireland (Emergency Provisions) Act 1978. He went on in a passage at pages 55-8 which we would quote in full:
"Latimer's confessions were admitted under section 8 because he made no allegations of torture or of inhuman or degrading treatment. However, at the trial and on this appeal Mr Cinnamond submitted that the confessions should be excluded in exercise of the court's discretion. In R v Howell  5 NIJB 10 at 12 in referring to the court's discretion to exclude confessions which were admissible under section 8 Hutton J (as he then was) stated:
'Before considering the issues which arise from the evidence given on the voire dire it is desirable to state the legal principles which are applicable:
1. Even though a statement by the accused is not inadmissible under section 8 of the Northern Ireland (Emergency Provisions) Act 1978, the Court has still a discretion to exclude a statement. This has recently been reaffirmed by the Court of Appeal in R v Cowan, Llewellyn and McAllister (not yet reported).
2. However the discretion should not be exercised so as to defeat the will of Parliament as expressed in section 8 of the Northern Ireland (Emergency Provisions) Act 1978 (R v McCormick  NI 105 at 114H).
3. The mere absence of voluntariness at common law is not by itself a reason for the discretionary exclusion of a statement in the trial of a scheduled offence (R v O'Halloran  NI 45 at 48A), and it was the intent of Parliament in enacting section 8 of the 1978 Act that, provided there had not been torture or inhuman or degrading treatment, statements made by a suspect after periods of searching questioning whilst in custody should be admitted in evidence, notwithstanding that at the outset the suspect did not wish to confess and that the interrogation caused him to speak when otherwise he would have stayed silent (R v Dillon and Gorman  11 NJB at 10).
at 456E).4. In considering whether to exclude a statement in exercise of its discretion the Court should have regard to the public interest as well as to the interests of the accused (R v Llewellyn  15 NIJB at 21) and to the consideration that the conviction of the guilty is a public interest, as is the acquittal of the innocent (R v Sang  AC 402
5. Where the evidence raises a conflict as to the factual background against which the discretion has to be exercised and where it is necessary for the conflict to be resolved, the onus rests on the Crown to prove the facts it alleges or to disprove the facts which the accused alleges beyond reasonable doubt, but once the facts have been established the decision whether or not to exercise the discretion to exclude the statement is not reached by applying the concept of the onus and standard of proof (R v O'Halloran  NI 45 at 48B, R v McAllister  10 NIJB, at 78, and R v Cowan, Llewellyn & McAllister at 14).
6. It appears from the judgment of the Court of Appeal in R v Cowan, Llewellyn & McAllister that, subject to the important qualification that the discretion should not be exercised so as to defeat the will of Parliament as expressed in section 8 of the 1978 Act, the judgments of the House of Lords in R v Sang  AC 402 afford guidance to the courts in Northern Ireland as to the exercise of the discretion in the trial of scheduled offences. However the qualification arising by reason of section 8 means that searching and persistent questioning over a period which causes the suspect to speak when otherwise he would have remained silent should not in itself be regarded as 'oppression' even though it might be regarded as 'oppression' (see per Sachs J in R v Priestly (1967) 51 CAR 1) under the common law rules relating to voluntariness.
7. Although the discretion is a judicial one and is to be exercised in a judicial manner, each case must be considered on its merits and the discretion is that of the individual trial judge (R v McCormick (1977) NI 105 at 114E).'
Mr Cinnamond further submitted that even if Latimer's confessions should not have been excluded in the exercise of the court's discretion, the confessions were unreliable because of the way in which, according to Latimer, he was pressurised into making them, and that therefore the conviction of Latimer in reliance upon those confessions was unsafe and unsatisfactory.
We observe that on the previous appeal to this court, Mr Cinnamond did not argue that the confessions should have been excluded in the exercise of the court's discretion and in delivering the judgment of the previous court of appeal in this case Lord Lowry LCJ stated at page 11:
'On appeal, however, Mr Cinnamond expressly declined to take issue with the failure on the part of the learned trial judge to exercise his discretion, and contented himself with the argument that the weight to be attached to these statements was insufficient to justify conviction.'
In support of his submissions before this court in relation to the discretion and the weight of the confessions Mr Cinnamond advanced the following argument. He submitted that, notwithstanding that Latimer accepted that he made the remarks and the confessions which the police described, there was a conflict of evidence between the interviewing police officers and Latimer as to the conduct of the officers in the course of the interviews. For example, Latimer alleged that the police threatened him with a long prison sentence if he did not confess, but the police officers denied the making of such a threat. Mr Cinnamond further submitted that as the ESDA examination proved that the police had lied to the trial judge about the rewriting of the interview notes and about the authentication of those notes and because senior detective officers were clearly involved in this lying and deceit, it followed that, applying the fifth principle stated in R v Howell, the Crown could not disprove beyond reasonable doubt the facts which Latimer alleged as to the conduct of the police in the interviews. Accordingly this court should hold that at the trial Latimer's confession should have been excluded in the exercise of the trial judge's discretion, or alternatively that the confessions cannot be relied upon for the purpose of convicting Latimer, with a consequence that his conviction should be quashed. In support of this submission Mr Cinnamond relied on a number of decisions of the Court of Appeal in England in well know cases in that jurisdiction. Mr Cinnamond cited the judgments of the Court of Appeal in R v Armstrong, Richardson, Hill and Conlon (The Guildford Four case), R v McIlkenny and others (The Birmingham Six case) and R v Silcott, Raghip and Braithwaite (The Broadwater Farm case). Mr Cinnamond submitted that those cases established the principle that if, in relation to the admissibility of a confession or the weight to be given to a confession, there was at the trial a conflict between the evidence of the police and the evidence of the accused as to what happened and what was said in the course of an interview or a series of interviews, and it was subsequently shown that the police lied in the witness box as to the manner in which the notes of the interviews were taken, then the Court of Appeal is bound to reject the confession of the accused, either as being inadmissible in evidence or as being of no weight."
Hutton LCJ then examined in depth the English cases to which he had referred and concluded at page 70:
"In relation to Latimer, having regard to his own acceptance at the trial of the confessions and other statements which the police said he made to them, and also having regard to his alleged reasons and explanations for making them, we are satisfied that he did not make those confessions because he was threatened or subjected to improper pressure by the police, or that he made the confessions, which he knew to be untrue, because he hoped that if he made a confession to the shooting of Carroll he would be given bail or would be able to leave Castlereagh."
We respectfully agree with and adopt the reasoning and conclusions set out in the passages which we have quoted. We therefore consider that the judge was correct to admit the statements under section 8 of the 1978 Act and the conviction is not rendered unsafe on that ground. We must deal shortly at this stage with another submission advanced by Mr Harvey QC on behalf of the appellant, based on the decision of this court in R v Magee  NI 217. The European Court of Human Rights had held in Magee v United Kingdom (2000) 8 BHRC 646 that in the circumstances of the detention of the appellant in Castlereagh there had been a violation of Article 6(1) of the European Convention on Human Rights, taken in conjunction with Article 6(3)(c) as regards the denial of access to a solicitor. It founded its decision, not on any direct evidence of the conditions in Castlereagh when the appellant was detained there, but on a report of the Committee for the Prevention of Torture made some four and a half years later. Moreover, the appellant had specifically requested on two occasions to have a solicitor's advice and gave evidence that he was unsure about the effect of the caution administered under Article 3 of the Criminal Evidence (Northern Ireland) Order 1988. He also showed symptoms of being materially more distressed and vulnerable than many other suspects in the same position. In those circumstances, notwithstanding our considerable reservations about the material upon which the ECtHR based its decision, we concluded that we would not be justified in holding that the conviction was safe in the light of a finding of the Court in the application brought by that appellant that there had been a breach of Article 6. We were careful to add, however, that if other cases concerning detention in Castlereagh came before us it would be a matter for consideration in each case how far the Court's findings in Magee's case were material in determining the safety of the conviction.  Our decision in R v Magee has, however, been overtaken in domestic law by the decisions of the House of Lords in R v Lambert  2 AC 545 and R v Kansal (No 2)  2 AC 69. The effect of these decisions is that the retrospective effect of the Human Rights Act 1998 and the direct enforcement of Convention rights do not apply where a defendant convicted before the Act came into operation on 2 October 2000 brings an appeal after that date. In that respect our decision in R v Magee was wrong in that we had held that the 1998 Act did apply retrospectively to the case. It also follows that the appellant in the present appeal cannot found a claim that his confessions should not have been admitted upon the ground that the conditions of detention at Castlereagh were in breach of his Convention rights.
Reliability of the Confessions We come finally to the issue of the reliability of the appellant's confessions and the weight which should have been attached to them, which is at the heart of the case. At the conclusion of his first report Professor Gudjonsson expressed "serious reservations" about the reliability of the confessions, and Mr Harvey QC on behalf of the appellant contended that his evidence is sufficient to give rise to a reasonable doubt whether they can be accepted as true. This opinion evidence must carry weight with the court, since the substance of much of it has not been fully controverted by expert testimony, but its weight is materially reduced by the evidence given by the appellant. We must examine all the evidence, including the expert evidence adduced on the hearing of the appeal, in order to see whether it creates a sufficient element of doubt about the truth of the confessions to render the conviction unsafe. In doing so we have to take into account the assessment which we have made of the reliability of Mrs A's evidence and the evidence given by the appellant himself about the course of the series of interviews and his own state of mind and motives. We must re-examine the judge's estimate of the appellant's character and his conclusion about the unlikelihood of his making an untrue confession in the light of Professor Gudjonsson's opinion evidence, to see whether they carry sufficient force or whether we would modify or reject them.  Mr Terence Mooney QC for the Crown mounted a strong case against accepting that Professor Gudjonsson's theory was a plausible explanation for the appellant's confession, retraction and subsequent confession in different terms. He submitted that a number of factors pointed strongly to a conclusion that the appellant was telling the truth when he confessed and that the mainspring of his desire to do so was not a psychological weakness on his part which made him agree to a story foisted on him by the police in order to escape the pressure of his situation:
(a) The first confession was made at a very early stage, when there can have been relatively little build-up of pressure on the appellant.
(b) The series of lies told by the appellant in his evidence at trial are a strong indication of readiness on his part to twist and turn in order to evade conviction.
(c) He was not sufficiently compliant to admit any of the other incidents about which the police questioned him, which tends to rebut the theory.
(d) The amount of detail which he included in his oral and written confessions was such that it is hard to suppose that he could have confabulated it, and it appears rather to represent true recall of the events which took place.
(e) The explanations of the appellant's retraction of his first confession accepted by the trial judge and this court on the first two appeals are overwhelmingly more likely than Professor Gudjonsson's "boomerang" theory of strategic coping.
(f) Professor Gudjonsson's opinion that the appellant's will was "completely broken" by the time he made his second confession is unsupported by the interview evidence.
(g) There is a notable absence of background material to corroborate the validity of Professor Gudjonsson's opinion that the appellant is a compliant person who would have accepted an untrue and incriminating version of the facts rather than dispute the authority figures in the person of the interviewing officers. On this point he cited the decision of the Court of Appeal in R v O'Brien  Crim LR 676, in which Professor Gudjonsson gave evidence in a case referred by the Criminal Cases Review Commission about the level of compliance of one of the appellants, which he considered made his confession unreliable. In giving the judgment of the court Roch LJ said at page 25 of the transcript:
"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."
In that case there was such a previous history, added to which the circumstances of the receipt of the appellant's confession in police interviews were, as Roch LJ said at page 24, unsatisfactory. The court accordingly allowed the appeal of the appellant Hall. We respectfully agree with Roch LJ's remarks.
Conclusions We have given extended and very careful consideration to all of the evidence in the case, in particular all the fresh evidence, and to the contentions fully and skilfully advanced by counsel. We have reached the conclusion that the course of the interviews points strongly to the appellant's confessions being reliable and true.  In Interview 2, only a few hours after his admission to Castlereagh, the appellant was faced with the allegation that a witness had recognised him coming out of technical school buildings in civilian clothes, wearing a cap and glasses, and getting into a Land Rover. The appellant made a response which appears at once to be incriminating. Interview 3 continued without a break, but with a change of interviewers. It was put to him that the police had "strong information" that he was involved. After some thought the appellant replied "Alright, I shot him, but I don't want to name any names." He thereupon gave an account which commenced with his receipt of a car and a gun, his being on foot patrol, joining it late after hiding his civilian clothes at the technical college, changing into those clothes there, leaving the patrol there and committing the murder on his own. When it came to Interview 4 he was being pressed with the proposition that other UDR men were involved and that he was trying to protect them. The vehicle issue records were then shown to him, which showed that he was with a crew in a vehicle on 8 November 1983, with the obvious implication that other UDR members were involved in the shooting incident. In our opinion it is clear from a careful reading of the interview note that this factor is what caused the appellant to retract. The only way that the appellant could think of to shield his comrades, which was uppermost in his mind, was to deny the whole of his previous version of the incident, including his own admission, which was partially correct but not the complete truth about the affair. We consider that this explanation of his retraction is very much more likely than Professor Gudjonsson's "boomerang" theory, which we do not find at all persuasive.  This view of the appellant's motivation in retracting his confession also explains his conduct over Interviews 5 to 11, when during the rest of 30 November and into 1 December the appellant maintained that his first confession had been all lies. In our view that is a much more convincing reason for his continued denials than the theory that he was fuelled by anger and indignation, which has no support from the evidence given and seems to us to savour of speculation. His maintenance of the denials then starts slowly to be worn away, and he begins to admit the involvement of the other soldiers. It may be that he was then moving towards throwing the blame on to his brother, which he did in Interview 14. We are satisfied that this was introduced by himself and did not, as Professor Gudjonsson propounded, originate in suggestions inserted by the police. By Interview 18 he was openly throwing the blame on to his brother David. That bluff was called in Interview 19, when the evidence which came from Mrs A was put to him and he eventually admitted once again that he had shot Carroll and made a detailed oral confession and written statement. There followed the confrontations with his fellow-soldiers, in which he twice confirmed that he had told the truth about the shooting.  We cannot see a sufficient foundation for Professor Gudjonsson's conclusion that by Interview 19 the appellant's will was completely broken and that he was ready to admit to anything, true or false, that the police put to him. It seems us vastly more convincing that he had tried in vain to protect his comrades by taking the blame in a partly true confession, retracting when he saw that that would not succeed, paving the way for throwing the blame elsewhere, then conceding when he realised, in the words of Hutton LCJ, that "the game was up". We are satisfied, notwithstanding the psychological evidence, that the content of the confessions made by the appellant was true and reliable. This conclusion is supported and reinforced by the identification evidence of Mrs A (discounting that of Mrs Dunne), which we regard as reliable, for the reasons which we have given. Applying the test contained in paragraph 19 of the speech of Lord Bingham of Cornhill in R v Pendleton to which we have referred, we consider that the trial judge, if he had had before him the psychological evidence adduced on appeal concerning the reliability of the appellant's confessions, would have reached the same conclusion as he did, that the content of his confessions was true and reliable.  Having considered the case against the appellant in the light of the evidence given on appeal, we are satisfied, for the reasons which we have given, that the conviction of the appellant for the murder of Adrian Carroll is safe. We therefore dismiss his appeal.