England and Wales Court of Appeal (Criminal Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CRIMINAL DIVISION)

		Royal Courts of Justice Strand, London, WC2A 2LL
		31st July 2003

Lord Justice Kennedy:

1. On 19^th March 1970 Patrick Colin Murphy, Michael Graham McMahon and David Cooper were convicted at the Central Criminal Court of the murder of Reginald Stevens at Luton on 10^th September 1969. On 26^th February 1971 their applications for leave to appeal were dismissed by this court, but in 1973 the case of Murphy was referred to this court by the Secretary of State for the Home Department pursuant to section 17 of the Criminal Appeal Act 1968, and on 11^th March 1973 his appeal was allowed. The Secretary of State then referred to this court the cases of Cooper and McMahon, but on 12^th February 1975 their appeals were dismissed. There was a second reference by the Secretary of State in 1976, but on 22^nd July, 1976 their appeals were again dismissed, and on 11^th April 1978 this court declined to receive fresh evidence tendered by McMahon following a reference by the Secretary of State under section 17(1)(b) of the 1968 Act. On 17^th July 1980 Cooper and McMahon were released from prison by order of the Secretary of State, and the remainder of their sentences was remitted, because of the "widely felt sense of unease" about the safety of their convictions. Cooper died in 1995 and McMahon in 1999. Their cases were referred to this court by the Criminal Cases Review Commission on 15^th March 2001.
2. This is not a case in which there is any criticism of the conduct of the trial by the trial judge. What is said by Mr Emmerson QC for the appellants is that over the years it has become increasingly apparent that there are serious grounds for concern about the way in which the police investigated the matter, and put the prosecution case together; about the failure of the prosecution to disclose certain matters to the defence, and about the impact of subsequent events upon the safety of the convictions. Some matters which we are asked to consider have already been considered by this court at one or more of the hearings to which we have just referred, but it is the cumulative effect of the criticisms we have to consider, and we are indebted to all three counsel for the work which they have done and the clarity with which they have presented the matter to us, both orally and in writing. Before we turn to the grounds of appeal which have been developed before us it is necessary to outline the case as presented at trial, and the main points considered at each of the subsequent hearings in this court.

The case at trial.

3. At about 6.10 p.m. on 10^th September 1969 Mr Stevens, the sub-postmaster of a post office in Luton was getting into his car, which was parked in the car park of a bank close to his post office, when he was shot with a sawn-off shotgun at point blank range. It was apparently a bungled robbery. Two women, Mrs Crawley and Mrs Calvert, were in their gardens nearby. They heard the shot and saw three men running from Mr Stevens' car. They then got into a green van which was driven off at speed. When it got to Luton railway station, not far away, Mr Herbert Andrews saw that it contained four men. They all got out, one clad in a boiler suit, and they got into two cars, a Vauxhall Cresta and a red sports car which may have been a Mercedes or an MG. Mr Andrews got the number of the Vauxhall Cresta, but his recollection varied about the make of the sports car, and he only got part of the number of that vehicle. He did however notice that the man in the boiler suit threw a bag over a fence on to the railway embankment before leaving. That bag contained the murder weapon in a pillow case, and it was recovered at 8.40 a.m. next day.
4. The Vauxhall car belonged to Alfred Mathews, who became the principal prosecution witness in the case. He was 53 years of age, and he was arrested on 22^nd October 1969. He was interviewed and, after initially denying any involvement, he made statements as a result of which, or partly as a result of which, the appellants and Murphy were arrested. Cooper had in fact already been arrested on 21^st October 1969 when he made a statement and was released. The two appellants and Murphy were all aged 25 to 27, and all three of them, like Mathews, had previous convictions for significant offences. Murphy was on bail awaiting trial for the armed robbery of a post office in London, for which he later received a sentence of 12 years imprisonment.
5. Within hours of the murder Scotland Yard was asked to conduct the investigation, and Detective Chief Superintendent Drury became the officer in charge. Once he began to admit any form of involvement Mathews told the police that he had gone to Luton with Cooper, McMahon and Murphy, because Cooper, having visited him at home several times, had persuaded him to go to pick up parcels. He had travelled alone in his own car, but in convoy with the sports car and the green van. The two cars were parked at Luton Station, and the others went off. When they returned in the van in a hurry he was picked up a short distance away and was driven back to the station. During that short journey McMahon said to Murphy (or vice versa) "you've shot him, you killed him", and McMahon also said "I didn't mean to kill him. He pulled the gun and it went off". Mathews also said that McMahon had blood on his clothes, and was in the back of the van putting on a boiler suit. When they got to the station car park Cooper and Murphy left in Cooper's sports car (Cooper did own a red Mercedes sports car) and McMahon left with Mathews in his Vauxhall. They parted on the way to London, and he arrived home at about 8 p.m. The murder was reported on the 10 p.m. television news, and at about 10.20 p.m. Cooper called. Mrs Mathews was there, and Cooper in effect confessed to the murder. Next day Mathews left his Vauxhall car unlocked in a side road in Hackney, and sent off a change of ownership form. In evidence he explained his initial denials to the police by saying that Cooper had made it clear to him that if he talked someone would be injured.
6. The sawn-off shotgun which was the murder weapon had belonged to a man named Good. He had been arrested on 12^th May 1969 for armed robbery, and claimed to have arranged for the disposal of the gun after his arrest. On 24^th October 1969 he was sentenced to 7 years imprisonment. We will look more closely at the position of Good when we turn to those grounds of appeal developed before us.
7. Superintendent Drury came to the conclusion that in order to convict Murphy, McMahon and Cooper it would be necessary to rely on the evidence of Mathews, and so in a report to his Commander which was dated 27^th November 1969, and which was submitted to the Director of Public Prosecutions, Drury said that Mathews was an essential witness and that "almost everything he has said is corroborated somewhere or another by a completely independent and impartial witness." The DPP agreed that proceedings against Mathews should be discontinued, and he became, as we have said the principal prosecution witness. When summing up the trial judge said that it would be idle to pretend that the case "did not really turn on Mathews". Of course the jury was warned of the danger that he might be an accomplice with his own interests to serve, and they were instructed to look for evidence which corroborated what Mathews said against each of the accused.
8. In the case of Cooper there were three pieces of evidence capable of amounting to corroboration, namely –

(1) The evidence of Mrs Mathews as to his visits to Mathews' home on the 9^th and 10^th September:

(2) His ownership of a red Mercedes sports car, and –

(3) The fact that the pillow case in which the gun was wrapped contained hair similar to the hair on Cooper's Labrador dog.

9. In the case of McMahon there were four pieces of evidence capable of amounting to corroboration, namely –

(1) He was identified by Mrs Crawley as one of the men in the bank car park when Mr Stevens was shot.

(2) Two serving prisoners, Jackson and Weyers, said that McMahon had given each of them a detailed account of the killing including, in the case of Jackson, that the gun had been obtained from the brother-in-law of a man now doing seven years for armed robbery (which seemed to be consistent with the evidence of Good).

(3) When identified by Mathews, McMahon was heard to respond aggressively, and –

(4) Mr Andrews said that one of the men in the station car park was wearing a boiler suit, and Mathews said that at that stage McMahon was wearing a boiler suit.

10. There was also some evidence capable of amounting to corroboration of what Mathews said about Murphy, but as his appeal was allowed in 1973 we need not itemise that evidence.
11. All three defendants gave evidence at the trial. Their case was that they did not go to Luton, and they were not in any way involved with the killing of Mr Stevens. Mathews was lying, and Cooper never visited him at home on either the 9^th or 10^th September. It was not suggested that Mathews could be mistaken. Cooper pointed out that the registration number of his soft top Mercedes did not contain 15 or 51, those being the numbers which Mr Andrews recalled seeing on the red car in the station car park.
12. As to what they were doing on 10^th September 1969 Cooper said that he had called on his tailor, Morris Lerman, for a suit. Mr Lerman recalled a visit at about 5 p.m. on a Wednesday, but his statements were inconsistent as to the date. In evidence he did say that it was 10^th September. McMahon said that he met Cooper at about 2.30 p.m. on 10^th September at Thames Magistrates'Court. They then went home, and watched the St Leger horse race on television, after which he went to his girl friend's. Murphy gave a broadly similar account and alibi witnesses were called.

Previous decisions of the Court of Appeal.

13. (1) February 1971: Application for leave to appeal (Fenton Atkinson LJ, Lyell and Mars-Jones JJ). When this application was made Murphy sought to call fresh evidence from three witnesses, the first of whom was Mrs Brooks, who described the way in which the expedition set off in terms which were not consistent with the evidence of Mathews. Her statement was not taken for several months, and the court did not find what she said likely to be credible. The other two witnesses were two men, Seal and McNair, who said that they were in vehicles in the street in Luton and saw the green van leave the bank car park and head off at speed towards the station. The substance of their evidence was that the driver of the van could not have been Murphy because it was a much older man, and McNair had identified a photograph of Mathews as resembling the driver when shown photographs.

In accordance with the normal practice at the time the names of Seal and McNair had been supplied to the defence, but when approached prior to trial they had said that they could not help. The Court of Appeal was prepared to accept that there was therefore a reasonable explanation for the failure to adduce at trial evidence which, if heard, might have been credible, but it went on to say that the evidence did not afford any ground for allowing the appeal. At page 11F of the transcript Fenton Atkinson LJ said –

"Once one assumes, as we all do, that the jury clearly must have reckoned that Mathews was very much more deeply in this than he said, and whether he was the actual driver of the van, or exactly what part he was playing, does not matter. The vital matter on which clearly the jury believed him and rejected the applicants was that they were the three men who were with him making up this party"

The only ground of appeal pursued on behalf of McMahon and Cooper was that there was no proven intention on their part to kill or to cause grievous bodily harm, and that therefore they should only have been convicted of manslaughter. That possible alternative verdict was properly left to the jury, and the evidence was plainly such as to enable the jury to draw the necessary inference as to their state of mind.

(2) November 1973: Appeal by Murphy (Widgery LCJ, Stephenson and Browne JJ). Sometime during the first half of 1971 the Murphy family found another witness, Terence Edwards, who was not a close associate of Murphy but knew him slightly. The statement of Edwards was submitted to the Secretary of State who then referred the case of Murphy to the Court of Appeal, and in 1973 Edwards gave evidence before the court that on 10^th September 1969 he came home from work at about 3 p.m. It was his younger brother's birthday, so he went out at about 4 p.m. to buy a birthday card. He heard a two-tone car horn. He looked up and saw Murphy driving a red sports car with a white top. He waved to Murphy, but got no response, and the car passed by. Edwards' evidence as to the date was supported by a till receipt in relation to a sewing machine which Edwards said he bought on that day after seeing Murphy. Murphy did, it seems, at that time own a car such as Edwards described. Edwards made a favourable impression upon the court, so Murphy's appeal was allowed. The court expressly refrained from offering any view about the convictions of McMahon and Cooper.

(3) February 1975: First reference by Secretary of State ( Widgery LCJ, James LJ and Ashworth J). Predictably the Secretary of State referred the cases of McMahon and Cooper to the Court of Appeal on the basis that –

"The decision to allow Mr Murphy's appeal raises an issue about the credibility of Mr Mathews' evidence against Mr Cooper and Mr McMahon which makes it desirable that the Court of Appeal should be asked to consider their cases."

A number of issues were then canvassed. The court was invited to order that Mathews be recalled for further cross-examination in the light of material which had become available since the trial, but it refused to follow that course. It did hear from two witnesses, Stephens and Lennard, who claimed to have been taken by Mathews to Luton in 1967 or early 1968 in order to carry out a reconnaissance for just such a robbery as was attempted on 10^th September 1969. Their evidence was found not to be reliable and was rejected. The statements of five witnesses were considered relating to sports cars in Luton on the day of the offence. The court found that all that they showed was that, not surprisingly, there were in Luton on that day a number of red sports cars of different manufacture and models.

Despite what had been said in February 1971 Seal and McNair were called to give evidence. The evidence of Seal was not found to be of any significant weight but McNair was different. He had seen the driver of the green van, and had picked out as resembling that driver two photographs from 42, one of those photographs being a photograph of Mathews. He also failed to identify Murphy, Cooper or McMahon when he attended identification parades on which they were standing. The court found his evidence to be credible and therefore considered its weight. As to that the court adopted the reasoning of Fenton Atkinson LJ in 1971 that it must have been clear to the jury, without the evidence of McNair and Seal, that Mathews was far more involved than he was prepared to admit in his evidence. Counsel submitted that the combined effect of McNair's evidence, that of Seal, and that of Edwards in the Murphy appeal, was to put the credibility of Mathews so much in doubt that the conviction of the appellants ought not to be allowed to stand, but James LJ said at page 11 of the transcript –

"In our judgment this fresh evidence would not have resulted in the jury arriving at a different assessment of Mathews as a witness and it does not touch the crucial question of the reliability of Mathews in respect of his evidence as to who were the members of the team which went to Luton."

Other grounds of appeal related to the evidence relied upon by the prosecution to support the evidence of Mathews, but the court like the trial judge, said that Mathews was crucial. On the same page of the transcript James LJ said –

"In our judgment, if the credibility of Mathews's identification of Cooper and McMahon is so much in doubt as to make the conviction unsafe or unsatisfactory upon the basis of his evidence, then in the case of each appellant the conviction cannot be saved by reliance upon the remainder of the evidence."

At trial the defence had not known that before attending the identification parade at which he identified McMahon, Mrs Crawley had said that she would not be able to identify anyone, and was shown a large number of photographs including a photograph of McMahon. The court found no substance in the suggestion that her identification evidence was thereby undermined.

That left the narrow point which the court described as "the real nub of these appeals". At trial there was no suggestion that Mathews might be mistaken in his identification of those who were with him. Did the evidence of Edwards so undermine his credibility as to render all of the convictions unsafe? There was reference to Baksh v The Queen [1958] AC 167 and counsel for the Crown pointed out that the jury could have acquitted Murphy and convicted the two appellants without it being said that they had reached inconsistent verdicts. The court agreed, and on that basis dismissed the appeal.

(4) 1976: Further reference by Secretary of State (Roskill and Lawton LJJ and Pain J). On 8^th April 1976 the Secretary of State took the unprecedented step of referring the cases of Cooper and McMahon to the Court of Appeal for a second time. There were two reasons for the second reference, namely –

"(1) Fresh alibi evidence in the case of Cooper was available from two men named Slade and Firmstone which, if it were or might be true, would, in the words of Roskill LJ at page 8 of the transcript 'shed grave doubt upon Mathews' credibility, certainly in relation to Cooper and therefore inferentially in relation to McMahon'

(2) Allegations had been made by a prisoner named Humphreys as to the conduct of former Detective Chief Superintendent Drury which, if admissible, might be relevant to the consideration of the cases of the two appellants. Drury had resigned in about 1972, but in 1976 he had yet to be convicted of any offence. He was awaiting trial on charges of corruption of which he was convicted in July 1977, when he was sentenced to eight years imprisonment."

Slade did not prove to be an impressive witness. The court examined his evidence in detail because it said –

"If it is, or may be, true, Mathews' credibility is obviously gravely shaken as regards Cooper, and if false as regards Cooper, clearly cannot safely be relied upon against McMahon."

In 1969 Slade and Cooper's father had been fellow directors of an office cleaning company. Slade claimed that on 10^th September 1969, the day of the murder, he had tea with Cooper in Hackney between 2.30 and 5 p.m., and that he saw Cooper again that evening in the vicinity of an East End amusement arcade. Slade's evidence, if accepted, also provided some support for Cooper's assertion that on that day he visited the tailor Mr Lerman. But Cooper had never mentioned Slade as a possible alibi witness, and Slade could offer no satisfactory explanation to explain his delay in presenting himself. From the outset he knew when Cooper had been arrested, and when the murder had occurred.

Firmstone, as a witness, was found to be good and careful. He had been Mr Lerman's assistant, and recalled a rushed job for Cooper on a day which he thought was a Wednesday in 1969, but he could not give the date and, as the court observed, Mr Lerman himself had in his statement to the police given another date before saying in the witness box that it was on 10^th September 1969 that Cooper called.

In 1976 the court was plainly anxious to hear any credible and relevant evidence that the appellants might wish it to hear and to explore every arguable ground of appeal, so it re-visited the effect upon the safety of the appellants' convictions of the fact that Murphy's appeal had been allowed, and, partly in order to assist in relation to that issue, Mathews was recalled for further cross-examination. He emphatically rejected the suggestion that his identification of Murphy might be mistaken, and, as the court pointed out, Edwards had not provided Murphy with a cast iron alibi. He had merely given evidence which was apparently honest and relevant, and which the trial jury had no opportunity to evaluate. As was pointed out by Roskill LJ the evidence of Edwards "was identification evidence of the kind now known as a fleeting glance, unsupported by any other evidence." As to its impact on the cases of McMahon and Cooper Roskill LJ said at page 26 of the transcript that –

"The highest that this part of the case can be put in their favour (as Mr McNeill accepted in his final speech) is that we know that there is in existence some evidence of the kind we have described inconsistent with Mathews's evidence in relation to Murphy, albeit evidence which, as Lord Widgery pointed out, is at least susceptible of being explained on the basis of honest but mistaken belief in an identification otherwise unsupported by Edwards of Murphy."

Four allegations were made about ex Detective Chief Superintendent Drury, but two were abandoned during the course of the hearing on the express instructions of the appellants. That left, first, an allegation that Drury had "leant on" Mathews to procure his confession of 25^th October 1969, which led to the arrest of Murphy and the two appellants, a confession which, it was alleged Drury knew to be wholly false. The court, having examined the written material and heard oral evidence from both Mathews and Drury rejected that allegation saying –

"Mathews denied it. Drury denied it. There is no evidence whatever to support it."

The second surviving allegation was that Drury, for improper reasons, refrained from putting Mathews on an identification parade, and thus prevented the possible identification by four witnesses, McNair and Seal, Isaacs and Andrews. McNair and Seal might, it was submitted, have identified Mathews as being present at the bank car park where the murder was committed. Drury said that he did not regard McNair or Seal as impressive identification witnesses, and he was anxious not to do anything which might cause Mathews to resile from his confession. As the court pointed out, if McNair and Seal had identified Mathews as the driver of the van "it would only have meant that Mathews was more deeply involved than he was prepared to admit, a consideration which the learned judge placed before the jury in his summing-up." Isaacs and Andrews were only at the station car park, so there was nothing in their evidence to disprove Mathews's denial of having been at the scene of the murder. In the result the court rejected the suggestion that there was anything sinister in the omission to put Mathews on an identification parade.

In May 1970 Mathews received a reward of £2000 from the Post Office. Michael Good and others were also rewarded. In this court in 1976 Mathews was evasive about his receipt of and use of that money, £700 of which was withdrawn from his account in cash the day after the money was received. But the criticisms of Mathews were not assisted by calling a prisoner named Wilkinson to say that in September 1979 Mathews had a red MG sports car, which was a "ringer". Wilkinson turned out to be a "most unsatisfactory, unconvincing and unreliable witness" whose evidence "bore the hall mark of recent prison manufacture designed to help a fellow prisoner in trouble."

Mathews was recognised to be unreliable in certain respects, but he was consistent as to who had been with him, and, as the court noted, if he was framing innocent men it was an odd coincidence that he picked on three men not one of whom could produce cogent evidence of alibi when arrested two months after the murder, and one of whom, McMahon –

"(1) was identified by Mrs Crawley as having been at the bank car park, and –

(2) was said to have confessed in prison to Weyers and to have given to Weyers 'startlingly accurate' information as to the value of the contents of the Luton Post Office safe."

(5) 1978: Secretary of State seeks opinion of Court. In April 1977 solicitors acting for McMahon obtained a statement from a man named Richard Hurn in which Hurn claimed to have seen McMahon in the East End of London twice on one day some years previously. If read together with the statement of Frederick Lawrence it could be inferred that the sightings took place of the day of the murder. The statements were forwarded to the Secretary of State who, under section 17(1)(b) of the Criminal Appeal Act 1968 asked the court to consider whether if the statements were tendered in the course of an appeal arising from a reference under section 17(1)(a) the court would regard itself as required to receive the fresh evidence by virtue of section 23(2) of the Act. The court found that Lawrence had been at court as a potential alibi witness for McMahon at the time of the trial, but had not been called. Furthermore his two statements were inconsistent with each other, and the fresh evidence to which the reference related did not justify an inference that the events referred to took place on 10^th September 1969, so it could not afford any ground for allowing an appeal. Furthermore the fresh evidence described in detail events 8 years ago and, despite intensive efforts to find alibi evidence, it had not previously been forthcoming, so it was considered not likely to be credible.

The structure of this appeal and the law.

14. Mr Emmerson QC submits that there are now five topics which were not considered by the trial jury. Two of them have been considered on previous occasions by this court but he submits they should be reconsidered because the approach adopted by the court on previous occasions was wrong. In any event when the additional material is taken into account it is clear that the convictions of the two appellants can no longer be regarded as safe.
15. The five topics relate to –

(1) The evidence of Michael Good.

(2) The successful appeal of Murphy in November 1973.

(3) The evidence of Seal and McNair (heard in 1975).

(4) The evidence of Mrs Mathews, - and

(5) The conviction of Drury in 1977.

Section 2 of the Criminal Appeal Act 1968, as amended by the Criminal Appeal Act 1995, requires that this court "shall allow an appeal against conviction if they think that the conviction is unsafe". There is no reference in that section to what the trial jury might have thought, or to the thoughts of any other hypothetical jury, but of course the court has to apply the statutory test against the background of a criminal justice system which involves trial by jury, and under which a jury has returned a verdict of guilty. When this case was before this court in 1976 Roskill LJ cited from the speech of Viscount Dilhorne in Stafford v DPP [1974] AC 878 at 893 –

"Parliament has, in terms, said that the court should only quash a conviction if, there being no error of law or material irregularity at the trial…. 'they think' the verdict was unsafe (or unsatisfactory). They have to decide and Parliament has not required them or given them power to quash a verdict if they think that a jury might conceivably reach a different conclusion from that to which they have come. If the court has no reasonable doubt about the verdict, it follows that the court does not think that the jury could have one; and, conversely, if the court says that a jury might in the light of the new evidence have a reasonable doubt, that means that the court has a reasonable doubt."

In Pendleton [2002] 1 WLR 72 the House of Lords was invited to reconsider the decision in Stafford, and at page 83, paragraph 19 Lord Bingham, with whom Lords Mackay, Steyn and Hope agreed, said –

"I am not persuaded that the House laid down any incorrect principle in Stafford, so long as the Court of Appeal bears very clearly in mind that the question for its consideration is whether the conviction is safe and not whether the accused is guilty. But the test advocated by counsel in Stafford and by Mr Mansfield in this appeal does have a dual virtue to which the speeches I have quoted perhaps give somewhat inadequate recognition. First, it reminds the Court of Appeal that it is not and never should become the primary decision-maker. Secondly, it reminds the Court of Appeal that it has an imperfect and incomplete understanding of the full processes which led the jury to convict. The Court of Appeal can make its assessment of the fresh evidence it has heard, but save in a clear case it is at a disadvantage in seeking to relate that evidence to the rest of the evidence which the jury heard. For these reasons it will usually be wise for the Court of Appeal, in a case of any difficulty, to test their own provisional view by asking whether the evidence, if given at the trial, might reasonably have affected the decision of the trial jury to convict. If it might, the conviction must be thought to be unsafe."

At page 89, paragraph 35, Lord Hobhouse, having set out the words of the statute, said –

"The sole criterion which the Court of Appeal is entitled to apply is that of what it thinks is the safety of the conviction. It has to make the assessment. That is made clear by the use of the words 'if they think'. The change in the language of the statute has reinforced the reasoning in Stafford v DPP and shows that appeals are not to be allowed unless the Court of Appeal has itself made the requisite assessment and has itself concluded that the conviction is unsafe."

Since Pendleton was decided it has been frequently cited in this court, for example in Mills and Poole (No 2) [2003] EWCA Crim 1753, to which our attention was invited, and in Hakala [2002] CLR 578, where Judge LJ said at paragraph 11 of the transcript –

"In Pendleton itself Lord Bingham's conclusion that it was not possible to be sure of the safety of the conviction followed an analysis of the fresh evidence in its factual context. The judgment in 'fresh evidence' cases will inevitably therefore continue to focus on the facts before the trial jury, in order to ensure that the right question – the safety or otherwise, of the conviction – is answered. It is integral to the process that if the fresh evidence is disputed, this court must decide whether and to what extent it should be accepted or rejected, and if it is to be accepted, to evaluate its importance, or otherwise, relative to the remaining material which was before the trial jury: hence the jury impact test."

With those authorities in mind we turn to look at the five topics we are now asked to consider.

The evidence of Michael Good.

16. Michael Good was not a particularly significant witness at the trial. He was called only to say that the gun found at Luton Station had been his, but after his arrest in May 1969 for armed robbery he had arranged for it to be sold, and that, having been convicted on 2^nd October 1969 he was serving a sentence of seven years imprisonment. That could be regarded as supporting the evidence of Jackson, a serving prisoner to whom McMahon was alleged to have confessed. Part of what Jackson said that he was told was that the gun had been obtained from the brother-in-law of a man doing seven years for armed robbery.
17. Good had been at liberty on 10^th September 1969 when the Luton murder was committed, and on 25^th September 1969 he was arrested in a public house in connection with that murder. On 8^th December 1969, when in custody as a result of his conviction on 2^nd October 1969, he made a statement to Detective Sergeant Fallon in which he identified the gun as having been in his possession prior to his arrest in May 1969. That statement was served at the committal, and formed the basis of his evidence.
18. What those acting for the appellants at the trial did not know, and what is now agreed, is that Good was a police informant, and that DS Fallon, based in north east London, was his handler. It appears that on the day of the murder which was committed at about 6.10 p.m., or possibly on the next day, Good telephoned DS Fallon and told him that he was aware that a gun that was his had been used, and that he was concerned that he would be implicated. The gun was recovered on 11^th September 1969 and then DS Fallon and DCS Drury met Good, who identified the gun. He was then to give DS Fallon the names of the suspects. The first two names were Hughes and Rogers, not the appellants. On 12^th September 1969 Good named Murphy, who was an associate of Hughes, and who owned a red Triumph sports car. On 15^th September 1969 he named McMahon, J. Disher (i.e. Cooper), Byrne and Sullivan. When he was arrested on 25^th September 1969 that was an arrest by arrangement to protect his status as an informant. He then, on 27^th September 1969, made his statement to DCS Drury which was not disclosed to the defence, and which he has subsequently stated to be untrue. In that statement he said that he had asked his wife to get rid of the gun after he was arrested in May 1969. When he asked his brother-in-law Terry Langston what had happened to it he was told that it had been sold to Patsy Murphy for £20, and from other people he heard that it had gone to McMahon. That version of events was supported by Mrs Good, but was contradicted by statements obtained from Terry and Sheila Langston. On 3^rd October 1969 DCS Drury added the name of Good and two others to the list of suspects. Three days later he added the name of Langston.
19. In April 1970 Good was released from custody, his appeal against conviction having been successful. On 21^st May 1970, on the recommendation of DCS Drury, he received a reward of £500 from the Post Office. The cheque was paid into DS Fallon's personal bank account, and was withdrawn in cash. Good later alleged that DS Fallon retained the money, possibly to pass it on to or share with DCS Drury. DS Fallon and DCS Drury denied that.
20. Good's first short statement of 27^th September 1969 was shown to the judge when Good gave evidence at the trial, and the judge ordered that the first five lines be read out. They dealt with his possession of the gun prior to his arrest. The rest was hearsay, dealing with his understanding of what happened to the gun thereafter.
21. Plainly, as it seems to us, the information which was not disclosed, and which today would be regarded as disclosable, would have been of considerable assistance to the defence. Mr Richard Latham QC for the Crown accepts that the information would today be disclosable, at least to the trial judge, but he submits that there was not a lot which the appellants could have done with it. We disagree. It would have shown that Good knew which gun had been used within 24 hours of the murder, and possibly even before the weapon was recovered, which would have supported a suggestion that he was himself involved, or at least he knew who was responsible. He then named a number of suspects, not at first including the appellants. Of course the jury knew that he was seeking to distance himself from the weapon which he conceded that he had once owned and on which he knew his fingerprints might be found, but what was to be made of his position in the light of the additional information, and of the conflict between the Goods and the Langstons as to how the gun was disposed of? As to the reward payment we are not prepared to speculate, even though we recognise that it is reminiscent of an allegation involving Mathews and ex-DCS Drury which was abandoned in 1976, but, as Mr Emmerson and Mr Latham both submit, the inescapable conclusion now must be that at the trial, without judicial approval, the evidence of Good was presented by the prosecution in a way that was deliberately misleading. As we have said, he was made to appear to be of no particular significance, but if the truth were known his evidence could have been of considerable significance. It would certainly have raised questions which the jury would have had to consider.

The impact of Murphy's appeal.

22. This is a topic which, as we have explained, has been fully canvassed in this court on two occasions and, as is clear from the decision in Pendleton, although the law has been clarified it has not been changed. We are therefore reluctant to go over the ground again, but in a case such as this it would not be right to foreclose argument on any aspect of the matter on which counsel for the appellants reasonably wishes to rely. Mr Emmerson criticised the reasoning of the Court in 1975. He submitted that, given the way the matter was contested at trial, it was not appropriate for the Court to accept the Crown's submission that it was open to the jury to find Mathews to have been mistaken about Murphy. For that involved an unacceptable degree of speculation on a point never canvassed at the trial. Even if it was permissible for the Court to speculate about mistaken identification, if Mathews may have been mistaken about Murphy then the jury might consider that they could not rely on his evidence in relation to Cooper and McMahon.
23. He further submitted that the Court fell into error when it said that "… the jury … could well have found Murphy not guilty and Cooper and McMahonn guilty …" in that given the credibility of Edward's evidence the jury "could well have found" Cooper and McMahon not guilty. If, so, the convictions of Cooper and McMahon were unsafe.
24. Finally, he submitted that applying Pendleton the appropriate question the Court might have asked itself was – might Edwards' evidence have affected the jury's view of Mathews' evidence against Cooper and McMahon? As Lord Devlin put it in "Afterword" in a book "Wicked Beyond Belief – The Luton Murder Case", edited by Ludovic Kennedy and published in 1979 (at page 144) :-

"If Edwards had been called at the trial, would any judge have directed the jury that in considering whether or not Cooper and McMahon were correctly identified, they must disregard Edward's evidence? What then are the different circumstances that are to prevail when the same question is being considered by an appellate court?"

25. We accept Mr Emmerson's submissions that once it is accepted that there is credible evidence [Edwards] which might have caused the jury to be unsure whether Mathews' evidence against Murphy was reliable, it is undeniable that such evidence might reasonably have affected its view of Mathews' evidence against Cooper and McMahon. The Court of Appeal in 1973 concluded that Edwards' evidence in relation to Murphy was credible.
26. We, with the distinct advantage and benefit of the decision of the House of Lords in Pendleton in 2001, respectfully disagree with the Court of Appeal's assessment in 1975. If the jury at the trial had had the benefit of Edwards' evidence exculpating Murphy, we accept that it would have been directed by the trial judge that such evidence was also relevant to the jury's assessment of the truthfulness of Mathews, not only in respect of Murphy, but also in respect of Cooper and McMahon. In our opinion it is impossible to say that such evidence would have made no impact on the jury in respect of Mathews' veracity when considering Cooper and McMahon. On the contrary, it might have made a heavy impact.
27. In 1976 Mathews gave evidence before a differently constituted Court of Appeal. He roundly asserted that there was no question of mistake in his identification of Murphy, Cooper or McMahon. The Court explained why Murphy's appeal had been allowed, and went on to consider the evidence from Mathews. It found that "he was clearly telling the truth". The Court did not hear Edwards although it is fair to point out that (a) he had previously been found to have been an honest and credible witness and (b) the Court of Appeal in 1976 was anxious and would have been prepared to listen to any witness whom the court or the defence had wished to call. Nevertheless, it is the submission of Mr Emmerson that the Court of Appeal in fact conducted the very exercise which Pendleton, many years later, said was impermissible. If Edwards was to the Court of Appeal in 1973 a credible witness, it was not permissible for the Court of Appeal in 1976 to say that they believed Mathews. For, he submitted, that was deciding the very matter which in Pendleton in 2001 it was made clear the Court of Appeal could not decide, namely the guilt or innocence of the defendant.
28. In a case in which fresh evidence is introduced in this Court, which raises doubts as to the safety of the conviction of one defendant, it remains to be considered whether the introduction of that fresh evidence renders unsafe the convictions of other defendants. It is not enough to show that the primary evidence against all of the defendants was the same prosecution witness. There may, for example, be other evidence indicative of the guilt of those defendants to whom the fresh evidence does not directly relate. In this case there was some such evidence which was identified by the trial judge as capable of corroborating the evidence of Mathews. But the prosecution always accepted that the weight of that other evidence was such that the credibility of Mathews remained critical. If his credibility was seriously undermined then none of the convictions could be regarded as safe. In 1975 and again in 1976, the function of this court was to evaluate the fresh evidence, and to set it along side the evidence heard at the trial, in order to decide whether it thought that the conviction recorded in 1970 remained safe. Once the impact of the evidence of Edwards was, as we accept, to some extent misunderstood, the task of the Court became increasingly difficult. That was because, as it turned out, giving Mathews an opportunity to deal with the possibility of mistaken identity (a possibility not canvassed at the trial) came very near to putting the Court in 1976 in the position of the primary decision maker in relation to a different case. In our judgment Mr Emmerson's criticisms of the way in which this whole topic has been handled since it was agreed that the evidence of Edwards should be admitted are well founded.

The evidence of Seal and McNair.

29. Seal on 10^th September 1969 was a 25 year old plant fitter driving his firm's small van along High Town Road when a green Ford Transit van emerged from a road on his offside and a little way ahead. It turned left, but in doing so caused Seal some inconvenience, and in his first statement which he made two days later he described the driver as being 40-45 years of age and having a weather-beaten complexion. At first Seal thought that he "may be able to recognise" the driver but by 20^th September 1969 he was "pretty sure" that he would recognise him. On 29^th September 1969 he was shown a green van at the police garage, and described it as identical to the one he had seen on 10^th September. He was shown a number of photographs, one of which he described as very similar to the man driving the van. On 14^th October 1969 he was shown more photographs, one of which he described as similar in facial appearance to the van driver. On three subsequent occasions in late October and early November 1969 he attended identification parades, but did not identify anyone. Mathews was not standing on any of those parades, nor was either of the photographs selected by Seal a photograph of Mathews.
30. Although Seal's statements were relied upon when leave to appeal was sought in 1971 he did not then give evidence. He did however give evidence in 1975. The court was not impressed. James LJ said –

"His description of the movement of the van is difficult to reconcile with the statement of McNair. On his own version the opportunity for registering in his mind a reliable impression of the driver's appearance was very limited."

In 2003 we see no reason why this court should take any different view, but with McNair the position was and is different. He was a 29 year old motor dealer who was driving along Welbeck Street. As James LJ said –

"McNair saw a van driven out of the bank car park into Welbeck Street. The van stopped behind him. He described the driver of the van by age, facial appearance and head-dress. He also described a person who was hanging on to the van at the driver's door and who later moved off hurriedly in the direction of High Town Road. On the 29^th September he picked out from forty two photographs two photographs of persons resembling the driver of the van. One of those he picked out was a photograph of Mathews."

A little later in the judgment James LJ said –

"We see no reason to regard McNair as other than a credible witness and therefore the question is what weight should be given to his evidence? If it is accorded the fullest weight in favour of the appellants and fed into the evidence which was before the jury the maximum effect it could have is that the jury would believe that Mathews played a part in the offence as the driver of the van on its journey from the bank car park, and that Mathews' evidence to the part he played was false. But, as was pointed out in the passage already cited from the judgment of Lord Justice Fenton Atkinson on the hearing of the application, it must have been clear to the jury, without the evidence of McNair and Seal, that Mathews was far more involved than he was prepared to admit in his evidence."

Mr Emmerson submits that it was not enough for the court to say in effect that the jury must have been alive to the possibility that Mathews was more deeply involved than he admitted, because if he was the driver of the van as it left the car park, significant parts of his evidence cannot have been reliable. For example, he would not have only learnt of the shooting as a result of hearing a conversation between Murphy and McMahon. Only three men were seen to run away from the victim's car after he was murdered, and if Mathews was one of those men the Crown would never knowingly have discontinued criminal proceedings against him and allowed him to give evidence against his co-accused.

31. Although for present purposes it would have been tidier if Mathews had been asked to stand on an identification parade attended by McNair and perhaps also by Seal, we can well understand DCS Drury's professed anxiety not to do anything which might cause Mathews to withdraw his co-operation. In the result we, like our predecessors, do not regard the evidence of McNair, to some extent supported by Seal, as decisive, but it cannot be completely ignored when other matters are weighed in the balance.

The evidence of Mrs Mathews.

32. Mrs. Florence Mathews was the wife of Alfred Mathews. She gave evidence about Cooper visiting their flat on 9th and 10th September, and again on the evening of the 10th when she heard him say, "Yes. I was at the back of it all, right at the back". Her evidence therefore broadly corroborated that of her husband that Cooper had arrived late on 10th September during the 10 pm television news and confessed to the murder. Her witness statement of 4th December 1969 was tendered in evidence at the committal proceedings and made an exhibit at the trial on the application of the prosecution. Mrs. Mathews' evidence was identified in the summing-up as capable of being corroboration of Mathews' evidence in the case of Cooper.
33. However, two earlier witness statements made by Mrs. Mathews on 24th October and 5th November were not disclosed to the defence. They contained no mention of the late-night visit on 10th September. In fact, in them Mrs. Mathews suggested that her husband may not have been at home on the evening of the 10th September. At p2 of her statement of 24th October she said:

"Although Alf had been ill for sometime, he looked very white and ill when he came in. He asked me to make a cup of tea but this was what he usually did. He seemed very depressed on this evening.

I don't know if he went out again that night. He may have gone to his, brother Reg, but I don't know, I'm not sure.

I remember hearing something on the news about the murder at Luton. This was the 10pm news. I'm not sure if Alf was here or not. I remember either saying or thinking how terrible it was.

There was thus no mention of the crucial visit on the evening of 10^th September during which Cooper was alleged to have admitted being present at the murder "right at the back".

34. Mr. Emmerson submits that had this evidence been available to the defence, Mrs. Mathews and Alfred Mathews could (and would) have been cross-examined on the basis that they had collaborated in order to fabricate evidence against Cooper. The evidence could have been used to show that Mathews was lying about his own involvement and the involvement of the appellants. Its non-disclosure was therefore a material irregularity, which seriously prejudiced the appellants.
35. Mr. Latham agreed that the much more limited account in Mrs. Mathews' statements of 24th October and 5th November did not support the account in Mr. Mathews' witness statement. Mr. Latham accepted that the two undisclosed statements would have provided material for the cross-examination of both Mr. and Mrs. Mathews on the issue of fabrication of evidence relating to his movements and associations on the afternoon and evening of the murder and that these earlier statements should have been disclosed to the defence prior to the trial.
36. In our judgment Mr Latham was right to concede that Mrs. Mathews' two earlier statements of 24th October and 5th November should have been disclosed to the defence prior to trial. Such disclosure would be required under modern practice and in our view such disclosure was required at the time. In Regina v Ward [1993] 1WLR 619 at 643 the Court of Appeal presided over by Glidewell LJ cited from Archbold, Criminal Pleading Evidence & Practice 38^th edition, page 202:

"...Where a witness whom the prosecution call or tender gives evidence in the box on a material issue, and the prosecution have in their possession an earlier statement from that witness substantially conflicting with such evidence, the prosecution should, at any rate, inform the defence of that fact: Rex v Howes (unreported), 27 March 1950, C.C.A...In certain cases, particularly where the discrepancy involves detail, as in identification by description, it may be difficult effectively to give such information to the defence without handing to them a copy of the earlier statement: Rex v Clarke (1930) 22 Cr. App. R. 58; see also Baksh v The Queen [1958] AC 167."

37. It is not necessary to consider whether the failure of the prosecution to disclose the two statements of 24th October and 5th November taken alone would have rendered the convictions unsafe. In our judgment the deployment of the two earlier statements in cross-examination (had they been disclosed as they should have been) would have had a significant impact upon the course of the trial.
38. For completeness we add that 'notes of interview' between Drury and Mathews allegedly recorded between October 22nd and 24th were not revealed to the defence until Mathews was being cross-examined during the 1976 appeal. Until then, the defence had been informed merely that Mathews, until he had made his first statement, denied ever having been in Luton on the day of the murder, and denied that he had made out a false change of ownership card for his Vauxhall car. The defence were unaware of Mathews' attempt to put up a false alibi by using his wife and brother, and of his highly detailed account (accepted by the prosecution to be a complete fabrication) of selling his car and threatening visits that he claimed he was receiving (presumably from non-existent people). After 11.15 on 24th October Mathews switched to what was revealed to the defence in isolation at trial, namely an account of his activities in Luton in the company of Cooper, McMahon and Murphy. (CCRC Statement of Reasons 19.13.3 and 4). Mrs. Mathews first (undisclosed) statement was dated 24th October.

The conviction of Drury.

39. Drury was convicted in July 1977 of five counts of corruption. He was found to have accepted sums of money, airline tickets and an evening's entertainment as a reward for favours bestowed, but none of the charges have anything to do with the investigation of the Luton Post Office murder. Indeed the first offence in time of which he was convicted was committed on 29^th March 1971, just over a year after the jury returned its verdicts in the present case, and it seems to us that this court must not be too ready to hold that subsequent bad behaviour even by a police officer has a retroactive effect. It may in some cases cast fresh light on credibility, but the credibility of DCS Drury was not a large issue in this case. He did not, for example, contend that either appellant made significant admissions to him. Nevertheless Mr Emmerson contends that, at least in this court, the conviction of Drury enables him to present some of his other complaints with more force.
40. First, Mr Emmerson points to the way in which the trial court was misled as to the position of Good and as to the consistency of Mrs Mathews. The extent of disclosure in relation to those two witnesses can be traced back to the officer in charge of the case. Then, secondly, there was the decision not to proceed against Mathews, and to use him as a prosecution witness. That was approved by the DPP, but Drury's report to the DPP overstated the case for Mathews, saying, as we have already noted, that "almost everything he has said is corroborated somewhere or another by a completely independent and impartial witness." The report went on to say in paragraph 200 –

"There is no evidence available to dispute what he has said about his not being present at the actual scene of the murder."

That does not seem to be right, having regard to the statements of Seal and McNair. Those statements are referred to elsewhere in the report, but their significance may not have been immediately apparent to anyone reading that report. We have already referred to the decision not to put Mathews on an identification parade to see if he could be identified by Seal or McNair, and remain of the view that despite the conviction of Drury nothing sinister should be read into that. Mr Emmerson points out that Mathews was treated as a potential prosecution witness from a very early stage. He identified Cooper, Murphy and McMahon on identification parades held in early November 1969, and on 3^rd December 1969 he made a witness statement which was not under caution. He was also permitted to have contact with his wife while in custody, and prior to his first statement.

41. Finally, there were the allegations about shared rewards, particularly those paid to Mathews and to Good. We do not and cannot make any finding that reward monies were improperly recommended or passed on in whole or in part to serving police officers. Even if they were, that would not directly impact upon the central issue of who went to Luton with Mathews on the day of the murder, but the conviction of DCS Drury does enable Mr Emmerson to underline the points that he has made.

Conclusion.

42. So, as it seems to us, there are now a number of matters which can be described as causes for genuine concern, when evaluating the safety of these convictions.

(1) Good knew which weapon had been used at such an early stage that if not himself involved he must have been very well informed, yet when he began to name suspects he did not at first name the appellants, and he could not satisfactorily account for his disposal of the gun in a way which others who were said to be involved were prepared to support. None of that information was available to the court at trial because the evidence of Good was presented so as to mislead.

(2) The evidence of Edwards in relation to Murphy does, if accepted as reliable, call into question the credibility of Mathews not only in relation to Murphy but also in relation to others he said were with him, namely the appellants. If Murphy could have been re-tried, and on re-trial had been convicted, Edwards would have become irrelevant so far as the present appellants were concerned but because of the passage of time no re-trial was possible.

(3) The trial jury had no opportunity to evaluate the evidence of McNair and Seal, which if accepted indicated that Mathews, or at least a man older than those he said were with him, was present at the scene of the murder. The account given by Mathews did not allow for the presence of himself or any older man, so again his credibility is called into question.

(4) The inadequate disclosure made in relation to the statements of Mrs Mathews meant that her inconsistencies, which were of particular relevance to Cooper, were never fully exposed.

(5) The subsequent conviction of the officer in charge of the investigation on charges of corruption adds to the sense of unease when considering –

(a) the relationship between that officer and Mathews, and –

(b) the recommendation in relation to payments of reward money and the destination of the money when paid.

43. For present purposes it is unnecessary to say that one of those matters, or any combination of them, is decisive. It is sufficient to say that in their totality they persuade us that these convictions are no longer safe, and that the appeals against conviction must be allowed.