1. LORD JUSTICE KENNEDY:

2. On 23^rd October 1998 in the Crown Court at Maidstone this appellant was convicted of two offences of murder and one offence of attempted murder, and was sentenced on each count to life imprisonment. He appeals against conviction by leave of this court.

Facts

3. At about 4 pm on Tuesday 9^th July 1996, after a swimming gala at Goodnestone School, Mrs Lin Russell, then aged 45, and her two daughters Megan, aged 6, and Josie, aged 9, set off to walk home with their family dog. The walk should have taken about 45 minutes. At about 4.25 pm, as they were walking along Cherry Garden Lane, a quiet unmade track, they were attacked. Their attacker tied them up with torn towels and shoe laces, blindfolded them, and beat them over their heads with a hammer. Mrs Russell and Megan died, but although seriously injured Josie survived. There was evidence to suggest that the attacker was searching for money. When his wife and children did not return home Dr Russell raised the alarm, and at 12.30 am the same night the bodies were discovered. It was also discovered that Josie was alive, and she was rushed to hospital, where she received treatment which enabled her to make a partial recovery.

Josie's evidence.

4. Josie was not able to identify the attacker. She was interviewed on two occasions several months after the incident, and an agreed compilation of significant passages from those two interviews was admitted in evidence to avoid her having to attend at the trial. In part her recollection was demonstrably wrong, and for present purposes we need say no more about her evidence.

Arrest of Appellant and Evidence of Daley.

5. On 14^th July 1997, just over a year after the murders, the appellant was arrested at his mother's home at Gillingham and from that date onwards he was detained in custody. As presented to the jury the prosecution case was based on the evidence of Damien Daley, aged 23, who, in September 1997, was detained in Canterbury Prison on remand awaiting trial. He was in a cell on the bottom floor in the Segregation Unit, and said that the appellant was put into an adjoining cell. Other prisoners were shouting at the appellant and Daley told them to be quiet. Daley had something of a reputation as a hard man in the prison, so his instruction was obeyed. The appellant then, according to Daley, spoke to Daley by means of a pipe which ran along the rear wall of both adjoining cells. It is common ground that if he said what Daley alleged that he said that amounted to an admission that he was the murderer. On behalf of the appellant it was put to Daley at trial that he had invented the confession, using information he obtained from a newspaper supplemented by a little inspired guesswork. Daley denied that he had done any such thing, and when he came to sum-up the trial judge told the jury that although there was other background evidence as to what people saw and heard at the time of the murders at the scene and in the area such evidence was “not sufficient to found a conviction. A conviction must rest on the evidence of Daley.” The trial judge continued -

“If you are not sure that Daley was giving a truthful report of what happened when the defendant arrived in the Segregation Block that is the end of the case against him, he must be acquitted.”

6. In the light of that direction, which is accepted for the purposes of this appeal to have been an appropriate direction in the light of the evidence led at the trial, it is unnecessary for us to rehearse the evidence which the judge described as insufficient to found a conviction.

The evidence of Jennings.

7. No witnesses claimed to have overheard what Daley said that the appellant said to him, but two other prisoners were called by the prosecution to testify as to other occasions when the appellant said things which were claimed to be significant. Mark Jennings was one of them.

The evidence of Thompson.

8. In 1997 Barry Thompson was nearing the end of a two year sentence, and for a few days he was employed as a prisoner trustee cleaner in the Segregation Unit at HMP Elmley where the appellant was then held as a prisoner on remand. Thompson gave evidence that on a few occasions he spoke briefly to the appellant. The appellant told him that he was in for robbing a man of £1000 and for burglary, and that he was going on an Identification Parade on Friday. He also said that the Police were rushing through some tests, and that Thompson should not judge him until the results came back. Next day whilst exercising in the yard the appellant went over to Thompson who was in the kitchen and, said Thompson, “he looked menacingly at me with his eyes rolling into the back of his head and said 'I made a mistake with her I won't make the same fucking mistake with you'”. It was put to Thompson in cross-examination that these conversations, and in particular the final conversation never took place, but when Thompson was released he went and made a statement to the police and set out in substance the evidence which he later gave at the trial.

At the Trial.

9. The appellant did not give or call evidence at the trial, and the case was presented to the jury by the judge in the way we have indicated - by pointing out that a conviction must rest on the evidence of Daley, but, as the judge went on to say, in evaluating the evidence of Daley the evidence of Jennings and Thompson might have a part to play. For present purposes we can focus on what the judge said about Thompson. He said that Thompson gave evidence -

“Of a threat which the prosecution say implied so clear a reference to Josephine's escape that it supports, if true, the evidence of Daley.”

10. That evidence, like the evidence of Jennings, the judge described as evidence which if the jury accepted it they could take into account when asking themselves if they could properly convict on Daley's evidence.

11. So, as counsel before us accept, the jury may well have looked to the evidence of Thompson when deciding whether or not to accept the critical evidence of Daley, and the problem which now faces us is that as a result of developments since the trial the evidence of Thompson has been shown to be hopelessly unreliable.

Post-Trial.

12. On 24^th October 1998, the day after the jury returned their verdicts, Thompson began to contact national newspapers telling journalists from the Mirror, the Sun, and later the Daily Mail that he had given false evidence at the trial. He said that it was true that he had met the appellant in prison, but the appellant said nothing incriminating, and he had only said otherwise at the trial because of pressure which had been put upon him by the police. In November 1998 Thompson was arrested on suspicion of perjury, but in the end that charge was not pursued. The Hampshire Police then conducted a thorough investigation which resulted in a very full report. Because that investigation was in progress there was some delay in the conduct of this appeal. By June 1999 Thompson was saying that if required to attend in this court he would say that his evidence at trial was true, and that in October 1998 he lied to achieve publicity. The journalists from the Mirror and the Sun co-operated with the Hampshire Police enquiry, but the relevant journalist from the Daily Mail took a different stance. On Monday 5^th February 2001 after hearing submissions we ruled that tape recordings, transcripts and notes held by the Daily Mail journalist which related to conversations which she had had with Thompson must be disclosed and she willingly complied with that order. On Tuesday 6th February the respondent served upon counsel for the appellant and upon the court a document which reads -

(1) The Respondent accepts the evidence of John Peacock, Adrian Shaw, Arnold Slater, Antonella Lazzeri and Neil Syson (all journalists) as to the statements made by Barry Thompson to them since trial.

(2) Having considered all the material now available to the Court for the purpose of this appeal, the Respondent cannot seek to rely upon Barry Thompson as a witness of truth and would not have called him at trial had the material that is now available been known to the Respondent at the time.

(3) The Respondent concedes that, in summing up, the learned judge correctly identified the linkage then made by the Crown between the evidence of Thompson and the truth of the confession made to Damien Daley and therefore cannot argue against the proposition that the jury's decision as regards Daley's evidence may have been influenced by their assessment of Thompson's.

Notice of Appeal and Conclusion

13. In the Notice of Appeal ground 1 reads -

“The conviction of the appellant is unsafe as a consequence of the witness Thompson retracting his evidence.”

14. In our judgment because of the way the case was presented in the Crown Court and in the summing-up that ground of appeal is made out. We have therefore not considered, nor have we been asked to consider, the remaining three grounds of appeal, which raise issues in relation to disclosure of police officers' day books, scientific evidence, and disclosure of a letter sent by Jennings to Detective Sergeant Bowler after Jennings had given his evidence. It follows that in our judgment this appeal against conviction must be allowed, and we so ordered last Thursday.

Re-Trial

15. We turn now to the question whether to order a re-trial. The power to order a re-trial is to be found in section 7(1) of the Criminal Appeal Act 1968 which reads -

“Where the Court of Appeal allow an appeal against conviction and it appears to the Court that the interests of justice so require, they may order the appellant to be re-tried.”

16. In R v Graham and others [1997] 1 Cr App R 302, a case concerned with criminality of a very different kind, Lord Bingham CJ said at 318 B in relation to section 7 -

“It is apparent that conditions which permit the Court to order a re-trial are two fold: the Court must allow the appeal and consider that the interests of justice require a re-trial. The first condition is either satisfied or is not. The second requires an exercise of judgment, and will involve the consideration of the public interest and the legitimate interests of the defendant. The public interest is generally served by the prosecution of those reasonably suspected on available evidence of serious crime, if such prosecution can be conducted without unfairness to or oppression of the defendant. The legitimate interests of the defendant will often call for consideration of the time which has passed since the alleged offence, and any penalty the defendant may have already paid before the quashing of the conviction.”

17. We accept that we are called upon to exercise our judgment in that way.

Basic submission.

18. Mr Clegg's basic submission in relation to this aspect of the matter is that because of the publicity generated by the case the appellant cannot now receive a fair trial. An order for a re-trial would therefore, it is said, amount to unfairness to the defendant.

Strength of Case.

19. Mr Clegg submits that whether publicity renders a re-trial unfair cannot be considered in a vacuum. The decision must be related to the facts of the case. If the case against the defendant appears weak then the impact of publicity may be crucial, and so persuade the court not to order a re-trial. Conversely if the prosecution appear to have a very strong case the impact of publicity would be reduced and a re-trial would be ordered.

20. Whilst we accept that a re-trial should not be ordered where, taking into account the result of the appeal, the prosecution appears to be left without a viable case, and we also accept that a very powerful prosecution case may weigh in favour of an order for re-trial, we consider that otherwise this court should not speculate as to how strong the prosecution case may appear to be at a re-trial.

Gravity of Charge.

21. A much more relevant factor, in our judgment, is the gravity of the charge, because where there is a viable case against a defendant in relation to a grave matter the public interest does require the verdict of a jury, provided that can be achieved without unfairness to or oppression of the defendant. In the present case the charges could not be more grave.

Publicity in this case.

22. The publicity to which our attention has been drawn took place for the most part in the months immediately after the verdict was returned in October 1998, when many matters were revealed which had not been known to the jury. There was a trickle of publicity thereafter, and it erupted again immediately prior to and during this appeal. Nothing new was said recently, but it can be argued that newspaper readers and television viewers would have been reminded by recent publications of what they were told over two years earlier. Whether they were reminded of the detail of what they had been told is quite another matter. In summary form the early allegations can be said to have been made under eight heads -

23. Not all of these revelations were made in any one newspaper, but many of them featured prominently in tabloid newspapers with wide circulations, such as in particular the Sun on 24^th October 1998, and also various editions of the Mirror and the Daily Mail. Whether or not the revelations were accurate is for present purposes wholly immaterial. Mr Clegg that asserts many of them were false.

24. Mr Sweeney, for the respondent, pointed out that the publicity was not all adverse to the appellant, and that the publicity which was adverse, in the sense of disclosing what a re-trial jury would not normally know, occurred almost entirely at the end of 1998. In 1999 and 2000 there was some discussion of how to deal with those suffering from severe personality disorders, and the House of Commons Select Committee on Home Affairs recommended legislation to provide a service separate from but linked to prisons and hospitals, and a new civil court order for detention. This legislative initiative was triggered by the Chillenden murders, but alongside that publicity, reminding readers of the appellant's personality, there was publicity questioning whether he was rightly convicted, now that Thompson was shown to be a liar. Questions were also raised as to the reliability of confessions to other prisoners. Mr Sweeney calculated that since the verdict there 85 newspapers articles which were adverse to the appellant, of which between 50 and 60 were published in 1998, and there were 95 articles which could be described as adverse to the prosecution case. Mr Sweeney was rightly cautious about the value of such arithmetic, which was derided by Mr Clegg in reply, but it does emphasise two important points - first that the vast bulk of the publicity with which we need to be concerned occurred in late 1998, and secondly, that the publicity has certainly not been all one way. We are conscious that there has been considerable publicity during the course of this hearing, not included in our bundle, which has been distinctly favourable to the appellant.

25. Mr Clegg, who makes no complaint in relation to pre-trial publicity, or even in relation to saturation publicity during the trial, submits that although interest in the appellant and in the Chillenden murders after the trial may at times appear to have declined it has always been there, ready to be rekindled by a new book about Josie, by library material stored on the Internet, or even by a discussion between two Queen's Counsel on a late night television programme. Mr Clegg's submission is that the recurrent publicity, coupled with the shocking nature of both the crime and allegations made about the appellant, has ensured that even if there is to be no re-trial until September 2001 some at least of the jury then assembled will recall matter which they should not know and thus the appellant will not have a fair trial. Even though jurors will no doubt try to be fair, the effect of the publicity, like bias, is likely to be insidious, and taint the procedure. We were reminded that in R v Gough [1993] 97 Cr App R 188 Lord Goff said at 191 that “bias is such an insidious thing that, even though a person may in good faith believe that he was acting impartially, his mind may unconsciously be affected by bias.”

Date of possible re-trial.

26. At the very end of the hearing we ascertained that if there is to be a re-trial the appellant understandably wants to be represented by Mr Clegg, who represented him at his trial as well in this court. Mr Clegg has commitments which make it impossible for him to represent the appellant at a re-trial before September 2001, and the appellant is content to wait until then in order to be represented by counsel of his choice even though the court could accommodate a re-trial in late April. Obviously in a case of this kind it is right for us to accede to the appellant's wishes in relation to representation, so the result is that if a re-trial is to be ordered it will not now start for many months and that is a matter which we think is right to take into account in deciding whether there should be a re-trial.

Relationship of Publicity to Trial.

27. Publicity can impact on a trial at three stages - pre-trial, during a trial, or after a trial so as to affect a decision by the Court of Appeal as to whether or not to order a re-trial. If prejudicial material is published either before or during a trial then the publisher will be at risk of proceedings for contempt of court and a defendant in the trial may be able to seek an order that the action be stayed. If his application is rejected and he is convicted his appeal may be allowed on the basis that the verdict is unsafe. For the reasons which we have already explained we are not here concerned with the publicity prior to the original trial, or with publicity during that trial. We are concerned with the post trial publicity of a kind which could not give rise to proceedings for contempt, and so far as we have been able to ascertain there are no decisions, either in England and Wales or in relation to Article 6 of the European Convention on Human Rights, which focus on adverse publicity after trial. As Mr Sweeney pointed out, this is a case where, but for the publicity, we would inevitably order a retrial. Mr Clegg recognises that even without Thompson the prosecution has a clear case, whatever arguments there may be about its merits, and with offences of such gravity that case ought to be heard, so Mr Clegg's submission can realistically be regarded as a foretaste of the submission he would otherwise make to the trial judge at the start of the re-trial. It is obviously sensible to make that submission now but in reality what we are looking at, viewed from the point of view of the proposed re-trial, is pre-trial publicity, and even by analogy publicity long before committal, which, it is said, has been such as to render a fair trial impossible. That is why we are able to derive some assistance from decisions in relation to pre-trial publicity, but less assistance from decisions in relation to publicity during a trial, which may be particularly objectionable because of the way in which it impacts upon the course of the current trial and the way in which the defence case is being presented. Our decision does not of course bind the trial judge at the re-trial in relation to any pre-trial publicity hereafter, and we recognise that we do not have the advantage which the trial judge will have of knowing precisely the shape of the revised prosecution case, but we think it important to analyse the type of publicity with which we are concerned before we turn to look at the authorities to which we have been referred. We propose to look at them in chronological order because that casts some light on the development of this branch of the law.

English Authorities.

28. The first case we were asked to consider was R v Savundra [1968] 52 Cr App R 637 where the complaint was in relation to publicity 11 months prior to trial, before criminal proceedings had actually begun. There the Court of Appeal found that in the particular circumstances of the case there was no real risk that the jury would be influenced by that publicity and referred, as Mr Clegg pointed out, to the case for the Crown being overwhelming. But of course in that case this court could properly express a view as to the strength of the Crown's case because it had all the relevant material before it. There was no possibility of the shape of the case changing on re-trial.

29. In R v Kray [1969] 53 Cr App R 412 Ronald Kray had been convicted of murder on 4th March 1969, and on 15th April 1969 he and a number of others were facing a second indictment charging them with murder and other offences. His counsel sought to challenge prospective jurors for cause on the ground that the previous trial had been extensively reported, and that prejudice to Kray resulting therefrom would be likely to influence the minds of the jurors in the second trial. At 414 Lawton J. upheld the right to report the first trial and said -

“What is more, the mere fact that a newspaper has reported a trial and a verdict which was adverse to the person subsequently accused ought not in the ordinary way to produce a case of probable bias against jurors empanelled in a later case.”

30. Then at 415 the judge continued -

“The situation, however, is, in my judgment, entirely different when newspapers, knowing that there is going to be later trial, dig up from the past of the convicted who have to meet further charges discreditable allegations which may be either fact or fiction, and those allegations are then publicised over a wide area. This does, in my judgment, lead to a prima facie presumption that anybody who may have read that kind of information might find it difficult to reach a verdict in a fair-minded way. It is, however, a matter of human experience, and certainly a matter of the experience of those who practise in the criminal courts, first, that the public's recollection is short, and, secondly, that the drama, if I may use that term, of a trial almost always has the effect of excluding from recollection that which went before. A person summoned for this case would not, in my judgment, disqualify himself merely because he had read any of the newspapers containing allegations of the kind I have referred to; but the position would be different if, as a result of reading what he had, his mind had become so clogged with prejudice that he was unable to try the case impartially.”

31. In R v Coughlan and Young [1976] 63 Cr App R 33 there was publicity following an unsuccessful plea of autrefois convict which tended to disclose the earlier conviction. Lawton LJ giving the judgment of this court said at 37 -

“It is our experience that juries in general understand the responsibility which rests upon them. They know that they have to be sure of guilt on the evidence before returning a verdict of guilty. The fault with juries nowadays lies not in convicting when they should acquit but in acquitting when they should convict. Juries are capable of disregarding that which is not properly before them. They are expected to disregard what one accused says about another in his absence. If they can do that, which is far from easy, they can disregard what has been said in a newspaper.”

32. As Mr Clegg points out, if that proposition is taken to its extreme an application to stay for abuse of process based on adverse publicity could never succeed, provided that the jury was warned to disregard the offending material, but the point being made in the judgment is that to a very large extent juries are trusted by our system to concentrate on what is relevant and to ignore irrelevant and prejudicial matters even when they know of them.

33. In R v McCann and others [1991] 92 Cr App R 239 the defendants were allegedly members of the IRA who had been found near to the home of Secretary of State for Northern Ireland. They were charged with conspiracy to murder. They did not give evidence. After

34. closing speeches the Home Secretary announced the Government's proposal to change the law so as to permit adverse inferences to be drawn when defendants choose to remain silent, and there was then widespread publicity in relation to that proposal which included comments in favour of it from the Secretary of State for Northern Ireland and a former Master of the Rolls. The judge declined to discharge the jury, but in this court Beldam LJ said at 253 -

“We are left with the definite impression that the impact which the statements in the television interviews may well have had on the fairness of the trial could not be overcome by any direction to the jury, and that the only way in which justice could be done and be obviously seen to be done was by discharging the jury and ordering a retrial.”

35. The important things to note in relation to that case are first, that the court was dealing with a sudden tide of publicity at a critical moment during the trial. It was not dealing with pre-trial publicity. Secondly, the court considered that if the jury had been discharged then it would have been possible to conduct a re-trial in a matter of a month or two. At that time this court could not order a re-trial.

36. In October 1989 the Court of Appeal allowed the appeals of three men and a woman convicted of the notorious IRA offences at Guildford and Woolwich (the Guildford Four). The court was satisfied that certain investigating police officers “must have lied”. They were then prosecuted, but proceedings against them were stayed as an abuse of process in the Magistrates' Court. The Director of Public Prosecutions challenged that decision in the Divisional Court in R v Bow Street MSM ex parte DPP [1992] 95 Cr App R 91, where Neill LJ said at 14 -

“The freeing of the Guildford Four and the comments made by the Court of Appeal attracted immediate and very wide spread publicity. We have seen, as did the magistrate, copies of the press reports. We have also been provided with video-recordings of television programmes and television news reports which were shown on October 19, 1989 and in the succeeding weeks. In addition we have had an opportunity to see video-recordings of later programmes broadcast in the course of 1990. It is right to say that much of the contemporary publicity was sensational, critical of the police and in some cases clearly hostile to the police. The comments that the respondents were liars provided headline news. I shall have to return later to consider the possible effect of this publicity on the prospects of a fair trial.”

37. At page 18 Neill LJ said -

“In my judgment a clear distinction can be drawn between the publicity in the period immediately after the release of the Guildford Four and the reports and broadcasts after December 1989. The earlier material could have been prejudicial to a trial in, say, the first part of 1990. The later material on the other hand could not be regarded in my view as prejudicial in a relevant sense. Even in relation to the earlier material, however, I am quite satisfied that none of the publicity which I have seen could affect a fair trial in, at the earliest, the Autumn of 1992. A jury would be perfectly capable of deciding the case on the evidence without regard to what they might have seen or read three years or so before.”

38. Mr Sweeney submitted that we should adopt a similar approach. No doubt it would not have been appropriate to order a re-trial in late 1998, soon after the tide of publicity which followed the verdicts, but that is now well over two years ago. In our judgment there is force in that submission.

39. In R v Central Criminal Court ex parte the Telegraph plc and others [1994] 98 Cr App R 91 this court considered an application to vary a judge's order restricting reporting in relation to series of related drugs trials, and at page 98 Lord Taylor CJ giving the judgment of the court said -

“In determining whether publication of matter would cause a substantial risk of prejudice to a future trial, a court should credit the jury with the will and ability to abide by the judge's direction to decide the case only on the evidence before them. The court should also bear in mind that the staying power and detail of publicity, even in cases of notoriety, are limited and that the nature of the trial is to focus the jury's mind on the evidence put before them rather than on matters outside the courtroom: see Kray ......”

40. Other authorities as well as Kray were cited. The case is of significance because it shows that the approach adopted by Lawton J cannot be discounted as being out of date.

41. That brings us to a decision upon which Mr Clegg places considerable reliance, namely R v Taylor and Taylor [1994] 98 Cr App R 361. In June 1991 Mrs Shaughnessy was stabbed to death at home. In July 1992 the Taylor sisters were convicted of that murder. The main ground of appeal was that an investigating police officer had suppressed an inconsistent statement made by a highly material witness, but there was also complaint about press coverage during the trial, which this court accepted was “unremitting, extensive, sensational, inaccurate and misleading”. The judge gave appropriate warnings to the jury to decide the case on the evidence alone, but this court found it impossible to say that the jury were not influenced by what they read in the press. Surprisingly there had been no application to discharge the jury because of the press coverage. Prosecuting counsel explained that in this court by saying that -

“Asking for a re-trial puts defence counsel in a hopeless situation, where young girls had spent considerable time in custody, and where to dispel the publicity, it would be necessary to postpone the trial for a further long period.”

42. This court accepted that explanation, but does not explain why it found the explanation acceptable. We can understand the reluctance of defence counsel to subject his young clients to a further period in custody, but if the press coverage was such as to render unsafe any convictions that might be recorded against them, then, as it seems to us, counsel's first duty must have been to ask the court to stop the trial. Thereafter it would have been possible to consider whether there could safely be a re-trial, if so when, and what should happen to the defendants meanwhile. The decision in McCann was referred to, but not what was said in that case about the possibility of a re-trial, and as to that issue this court in Taylor said simply -

“Moreover, by reason of the view we take of the way in which this case was reported, we do not think that a fair trial could now take place. Hence we do not order a re-trial.”

43. The Court of Appeal gave its decision in June 1993, nearly 11 months after the trial. If a re-trial had been ordered it would not in reality have taken place until, at the earliest, the Autumn of 1993. We recognise that we have only a very limited knowledge of the publicity which occurred during the original trial, but the Court of Appeal does not explain why that publicity would have made it impossible for the defendants to have a fair trial 15 to 18 months later. Perhaps the true explanation is that, as may be gleaned from the explanation given by prosecuting counsel to which we have referred, the prosecution in that particular case was not really pressing for a re-trial.

44. Plainly the decision in Taylor is authority for the proposition that this court can, in an appropriate case, regard past press coverage as a sufficient reason not to order a re-trial, but otherwise, as it seems to us, it must be regarded as a decision on its own facts. That of course is true of all of these cases, of which R v Reade and others, 15^th October 1993 (unreported) is another example. Police officers were prosecuted following the release of those convicted of the Birmingham bombing and Garland J granted a stay of the criminal proceedings against those officers, pointing out that publicity, although a powerful factor, did not stand alone. As the judge said, prejudice engendered by publicity is usually local and temporary, but if the impossibility of having a fair trial becomes national and continuing then there is “quite literally, nowhere to go”. That we accept. The question for judgment in each case is whether that point has been reached, bearing in mind, as Scott Baker J said in ex parte B,17^th February 1994 (unreported) that -

“In most cases, one day's headline news is the next day's firelighter. Most members of the public do not remember in any detail what they have seen on television, heard on the radio or read in the newspaper except for a very short period of time.”

45. The same point was made in the Divisional Court in Attorney General v ITN and others [1995] 1 Cr App R 204 in relation to contempt proceedings against ITN and certain newspapers for disclosing immediately after an arrest that one of those arrested on a charge of murder of a special constable was a convicted IRA terrorist who had escaped. Nine months were going to elapse before the murder trial, and at 218 E Leggatt LJ said -

“During the nine months that passed after anyone had read the offending articles, the likelihood is that he no longer would remember it sufficiently to prejudice the trial. When the long odds against the potential juror reading any of the publications is multiplied by the long odds against any reader remembering it, the risk of prejudice is, in my judgment, remote.”

46. In the present case that may be said to apply not to the offence or even to the identity of the appellant, but to the sort of details about him which were released in some widely read newspapers in late 1998.

47. In 1995 Phillips J, as he then was, had to consider the adverse publicity accorded to Kevin and Ian Maxwell before they appeared for trial. The judge said -

“No stay should be imposed unless the defendant shows on the balance of probabilities that owing to the extent and the nature of the pre-trial publicity he will suffer serious prejudice to the extent that no fair trial can be held. I would accept this test, so far as it goes, but it remains necessary to identify the essential aspects of a fair trial for the purpose of the test. If it were enough to render a trial unfair that publicity has created the risk of prejudice against the defendant our system of criminal justice would be seriously flawed. There will inevitably be cases where the facts are so dramatic that almost everyone in the land will know of them. There will be circumstances when arrests are made of defendants whose guilt will, or may, appear likely. Intense media coverage may well take place before a suspect is identified or apprehended. If in the most notorious cases defendants were to claim immunity from trial because of the risk of prejudice public confidence in the criminal justice system would be destroyed.”

48. After referring to two authorities the judge continued -

“Our system of criminal justice is founded on the belief that the jury trial provides the fairest and most reliable method of determining whether guilt is established. This belief is based on the premise that the jury will do their best to be true to their oath and to try the case according to the evidence. The ability of the jury to disregard extrinsic material has been repeatedly emphasised by judges of great experience.”

49. Phillips J then cited from Kray and concluded -

“It seems to me that the court will only be justified in staying a trial on the ground of adverse pre-trial publicity if satisfied on a balance of probabilities that if the jury return a verdict of guilty the effect of the pre-trial publicity will be such as to render that verdict unsafe and unsatisfactory. In considering this question the court has to consider the likely length of time the jury will be subject to the trial process, the issues that are likely to arise and the evidence that is likely to be called in order to form a view as to whether it is probable that - try as they may to disregard the pre-trial publicity - the jury's verdict will be rendered unsafe on account of it.”

50. That seems to us to be a valuable approach, and Mr Clegg did not suggest otherwise, although he did question the suggestion that there is any burden of proof upon the defence. Given that this is a case where but for the publicity we would certainly order a re-trial, we should do so unless satisfied on a balance of probabilities that if at the re-trial the jury returns one or more verdicts of guilty the effect of the publicity to which we have referred will be such as to render that verdict or those verdicts unsafe.

51. The application of Rosemary West for leave to appeal [1996] 2 Cr App R 374 was also concerned with massive adverse press coverage prior to trial, and at 386 A Lord Taylor, CJ, said -

“The question raised on behalf of the defence is whether a fair trial could be held after such intensive publicity adverse to the accused. In our view it could. To hold otherwise would mean that if allegations of murder are sufficiently horrendous so as inevitably to shock the nation, the accused cannot be tried. That would be absurd.”

52. The judgment goes on to refer to Kray and other cases we have considered. Mr Clegg submits that West is not really of assistance because the court was only considering sensational publicity which foreshadowed what the jury was going to hear, but that is not quite right. For example, some reports referred to Rosemary West as a nymphomaniac and a prostitute. Mr Clegg submits that in the light of decisions such as McCann, Taylor, and Reade it must now be accepted that there can be situations where an appropriate direction will not suffice. It is unreasonable to expect a jury to put the adverse publicity out of their minds. We agree, but that can only happen rarely, and usually in relation to adverse publicity which occurs during the course of the trial.

53. Our attention was also invited to the decision of the Divisional Court in Attorney General v MGN Ltd and others [1997] 1 All E R 456 where Schiemann LJ noted at 461 E the ability of a jury “not to accept as true the contents of a publication just because it has been published.”

European Decisions.

54. Our attention was invited to three decisions of the European Commission and one decision of the European Court from which we derive only limited assistance.

55. In X v Austria [1963] 1476/62 the Commission, at page 43, pointed out that Article 6(1) of the European Convention on Human Rights guarantees a fair hearing to every person who is charged with a criminal offence, and particularly where laymen participate as jurors the guarantee may be seriously impaired by a virulent press campaign against the accused which so influenced public opinion and thereby the jurors that the hearing can no longer be considered a fair hearing within the meaning of Article 6. But the Commission went on to say that in the instant case there was no evidence that the publicity had that desired effect.

56. In X v Norway [1970] 35CD37 the Commission found again no evidence that the jurors or the judges were influenced by the publicity complained of when reaching their decisions more than a year later, or that the applicant was prejudiced by the publicity during the extensive examination of his appeals by the Supreme Court, which sits without a jury. English law is more favourable to a defendant in that it does not require such evidence.

57. In Berns and Ewert v Luxembourg [1991] 13251/89 the Commission, whilst accepting that an atmosphere of agitation or a virulent press campaign could be prejudicial to the fairness of a trial, emphasised the need to examine criminal proceedings as a whole after the proceedings have been concluded, which, in the context of the present case, would mean after the re-trial, and any appeal arising therefrom.

58. Allenet de Ribemont v France [1995] 20 EHRR 557 was a decision of the European Court, but it was not in any way related to the issues with which we are concerned.

59. Our conclusion is that although the European decisions do demonstrate that the right to a fair trial enshrined in Article 6 does encompass a right to a trial the result of which is not distorted by publicity, the English criminal courts have been much more involved than the European Commission or the European Court with how that right, which we all recognise, is to be safeguarded.

Caveat.

60. There is a tendency in a case such as this to equate the right to a fair trial with keeping the jury in ignorance of facts which a jury should not normally know, and the two are not the same. For example a jury is not normally told of a defendant's previous convictions, but there are a variety of ways in which that knowledge may come to the attention of a jury in a particular case without the trial being rendered unfair. The offence may have been committed in prison, or the defendant (like Ronald Kray) may be notorious; he may choose to put his record before the jury, perhaps so as to enable him to attack witnesses for the prosecution without reserve. When asked why in the present case a re-trial would be unfair Mr Clegg replied that the publicity would deprive the appellant of the right to keep certain matters from the jury. But, as we have demonstrated, a defendant in a criminal trial does not always have that right. In addition to the examples we have given of circumstances in which a jury may learn of the previous convictions of a defendant the prosecution may be allowed to lead evidence of similar facts, or a co-defendant may cross examine to establish bad character. So the submission has to take account of all of the possibilities, and in our judgment a better approach to our task is that adopted by Phillips J in Maxwell.

Conclusion.

61. So we ask ourselves whether we can be satisfied, on a balance of probabilities, that if the jury in September or October 2001 returns a verdict of guilty the effect of the pre-trial publicity between October 1998 and February 2001 will be such as to render that verdict or those verdicts unsafe.

62. We accept that the early publicity was sensational, possibly in parts inaccurate, and extended far beyond what a jury in 2001 might normally be told, but, at least in so far as it was accurate it was legitimate. Mr Clegg submits that the volume of publicity is unique. We decline to involve ourselves in comparisons, but we have been referred to other cases where the volume of publicity was considerable. Mr Clegg submits that if a re-trial is to be ordered in this case and there is to be no stay it is difficult to envisage any case where as a result of pre-trial publicity there will be a stay or a refusal of an order to re-try. But, as we have demonstrated, circumstances vary infinitely. The re-trial will not start until nearly three years after the October 1998 publicity, which is the principal target of complaint, and people do forget. Even if they do not forget entirely, the passage of time makes it easier for them to set aside that which they are told to disregard. Here, as Mr Sweeney points out, the central question during a two to three week trial is likely not to be the character of the appellant, but whether Daley can be believed, and in justice to the Russell family and to the wider community there should be a trial to answer that question. The risk of prejudice will be reduced if the trial does not take place in Kent, or even in London, because the impact of these crimes was at its greatest locally and it may be that some further safeguard can be provided by a few careful questions to the jury panel (but as to that see R v Tracey Andrews [1999] CLR 156). But whether or not questions are asked we are not now satisfied that if the jury does convict their verdict will be found by this court to be unsafe by reason of the publicity to which we have referred. That is why last Thursday we ordered a re-trial.