Davis, R v

On 25^th May 2004, at the Central Criminal Court, before His Honour Judge Paget QC and a jury, Iain Davis was convicted of two counts of murder. He appeals against conviction with the leave of a single judge. This is the Davis appeal.

On 18 March 2005, at Birmingham Crown Court, before Goldring J and a jury, Marcus Ellis, Michael Gregory and Nathan Martin were convicted of two counts of murder and three counts of attempt murder, and Rodrigo Simms was convicted of the same counts of murder, and two of the three counts of attempt murder. They appeal against conviction, with the leave of the court. This will be described as the Ellis appeal.

There is no connection between the two cases. Leave was given in the Ellis appeal because it appeared that an important point about the anonymity of witnesses was a common and significant feature of both cases. The appeals were heard consecutively, and after hearing argument in both cases, we decided, with the agreement of all counsel, to address the issues in a single judgment.

On 1 January 2002, towards the end of a New Year’s Eve party in the presence of witnesses, two men, Ashley Kenton and Wayne Mowatt were shot dead. Their deaths resulted from a surge of violence by an individual who discharged a loaded gun to assassinate Kenton, against whom he entertained an apparently short-lived grudge. The second death was “accidental”, in the sense that the victim, Mowatt, happened to be in just the wrong place at the wrong time when the bullet fired from the assassin’s gun went through the Kenton’s neck, and by way of a flimsy internal partition, entered the neck of and fatally injured Mowatt. A single bullet killed both men.

On 8 December 2002, Nathan Martin’s brother Yohanne, was shot dead. Those responsible were believed to be members of a gang in Birmingham known as the Johnson Crew. The appellants, together with a co-accused Tafarwa Beckford, who was acquitted by direction of the judge, were members of the Burger Bar Crew. Both gangs were capable of extreme violence. The Johnson Crew and the Burger Bar Crew were, literally, deadly rivals.

Shortly after 4am on 2^nd January 2003, a party was taking place at the Uniseven Hair Salon in Aston, Birmingham. The guests included members of the Johnson Crew. A number of those attending the party happened to be standing in an alleyway at the rear of the premises. They included Charlene Ellis, Letisha Shakespeare, Sophie Ellis and Cheryl Shaw. A convoy of vehicles, which included a red Mondeo and a silver Vauxhall, drove to the scene. Shots were fired into the crowd. Sharleen Ellis and Latisha Shakespeare received fatal gunshot wounds. Sophie Ellis and Cheryl Shaw were badly injured. A man called Leon Harris, who was sitting in a car further down the alleyway, was shot at close range by a gunman who had alighted from one of the convoy of vehicles.

These shootings, and the consequent deaths and injury, were perpetrated by members of the Burger Bar Crew in revenge for the death of Yohanne Martin. The female victims were not the intended targets of the attack. Unhappily they just happened to be in the wrong place at the wrong time, victims of gang warfare on the streets of Birmingham.

These cases are not untypical. There is compelling evidence of an alarming increase in gun-related crime (see Crime in England and Wales 2003/2004: Supplementary Volume 1; Homicide and gun crime, published in January 2005). In an earlier case which we were asked to consider in argument, R v Bola, unreported, 18 June 2003, Hughes J summarised the evidence then before him in a case where a woman was summarily executed in a drug related feud.

“Since April 2002 there have been no less than thirty seven people injured on the streets of Nottingham in shooting incidents. Not domestic incidents, on the streets. Three of them have been fatalities….. The fear of gun crime has an adverse effect on the public who live in the areas of Nottingham where such things occur and particularly has had a considerable impact on the ability of the police to investigate crime. He says, and I have absolutely no reason to doubt, indeed I am confident that he is right that the experience of the police is that after an incident of this kind witnesses are frequently content to come confidentially to the investigators to tell them what they know, what they saw, to give them leads and help about what the background may be and sometimes to name names, but that such witnesses are to a very large extent frightened to be identified as co-operating with the police and unwell, as a result, to give evidence. They fear similar incidents, and they fear retribution, either from those whom they accuse, or, more generally, from those who take the stance at co-operation with the authorities is anathema…the evidence…is entirely consonant with the experience of anybody who has practiced in the criminal courts in the last few years. It describes a situation which all of us know obtains not even or mainly in Nottingham but in several major cities in England….. ”

These observations are entirely consistent with the judicial experience of each member of the court.

In the Davis appeal there is undisputed evidence from a detective officer who has specialised in murder investigations for the last seven years, and in particular, gun related violence. He said:

“Most people opt not to co-operate and do not get involved. Doors are not opened, arranged meetings result in a witness not turning up, telephone messages go unanswered and messages left at home addresses/work, although discrete are ignored. This is not a problem that exists on an occasional basis…..it is a problem that exists in practically every investigation in one way or another. Such problems exist on a daily basis. I have spoken to witnesses about a reluctance to give evidence. The common factor between all of them is fear.

They are in fear of their lives and that of their families and friends. There is a very real danger to such persons of death or serious injury, either to prevent them from giving evidence, or to punish them for giving evidence and to send a warning to those who may be thinking of assisting the police. This risk I know and the witnesses know, is not necessarily at the hands of the defendants themselves, but at the hands of the associates of the defendant. If the defendant is in custody, it is often the associates who are the physical threat.

In many but not all cases, the witness knows of the defendant and their associates. They know they have easy access to firearms and the “ease” with which they are prepared to use them.”

In summary, quite apart from the ghastly callousness involved in the use of firearms to kill, and the devastation suffered by the families of the deceased, it is not an exaggeration to point out that, whether they are aware of it or not, these gun carrying criminals are challenging the rule of law itself. One common feature of both these cases, and many others like them, is the absence of any or any significant attempt at concealment. People are gunned down in busy crowded areas. Although the offences are witnessed, those who use their guns expect to escape justice. They anticipate that the guns which have been used to kill will also serve to silence, blind and deafen witnesses. Without witnesses, justices cannot be done.

Dealing with it generally, there are in principle, two ways to address these problems. The first is a witness protection programme. However in reality, and certainly for the individual of good character, with established roots, this kind of programme is unacceptable. It requires a complete change of identity, and home, and work, not only for the witness himself or herself, but for his family, and a likely permanent separation between them, and other members of their extended family, and a subsequent life which is dominated by the need for continued security, and constant supervision of that security by police officers. This process is grossly invasive of the right of the witness and his family to private and family life. It is likely to be appropriate when the identity of the witness is already known to the defendant, and may be suitable for the professional criminal who has decided, for reasons of his own, to give evidence against his former colleagues and who is treated as a “supergrass”. The alternative to the witness protection programme is for appropriate steps to protect and reassure the witness about the process leading to and the giving of evidence. This includes voice modulation, and screening, and other special measures, and witness anonymity. It is this last feature of possible steps to protect witnesses, and their lives and the lives of their families, that is engaged in these appeals.

The administration of protective arrangements of this kind is not unduly problematic in itself. The difficulty arises from the potential conflict between them and the over-arching principle of the common law, once memorably described by Lord Bingham of Cornhill CJ as the “birthright” of every British citizen, the right to a fair trial. The difficulties are at their most acute in cases like Davis, when there would be no point in a trial, just because without the evidence of anonymous witnesses there would be insufficient evidence on which to found a safe conviction.

The Court undoubtedly possesses an inherent jurisdiction at common law to control its own proceedings, if necessary by adapting and developing its existing processes, as Lord Morris explained in Connelly v DPP (1064) 2 AC 1254 at 1301 “to defeat any attempted thwarting of its process ”. (see also Taylor v Attorney General (1975) 2NILR 675, R v Murphy (1990) NILR 306 and R v Accused (1992) 1 NILR 257).

The principle provided the basis for some earlier changes in the criminal process which have become familiar, but which were once regarded as unusual. Nowadays, for example, it is commonly ordered that the identity of informants should not be disclosed to the defendant, lest “sources of information would dry up” D v NSPCC (1978) AC171 at 218. Again, the practice of screening vulnerable witnesses from the defendant developed during the 1980s. Thus in R v X, Y and Z (1990) 91 CAR 36 this court upheld the decision of the Common Sergeant, sitting at the Central Criminal Court, that screens should be erected to enable children who had been treated indecently to give evidence screened from the defendant. The judge was required to see that

“ the system operates fairly; fairly not only to the defendant, but also to the prosecution and also to the witnesses. Sometimes he has to make decisions as to where the balance of fairness lies…. We do not need authority to confirm us in the view that what the judge did here in his discretion was a perfectly proper, and indeed a laudable attempt to see that this was a fair trial to all, the defendants, the Crown, and indeed the witnesses…..”

Some, but not all of the steps identified by developing common law principles, have now found their way into statute. Thus, the Youth Justice and Criminal Evidence Act 1999 (ss 16-33) makes provision for a series of special measures. These include screening the witness from the accused, witnesses giving evidence by live link, the provision of aids to communication for an incapacitated witness, and a greater degree of “privacy” in what remains a public hearing, to be applied in sexual cases, or where there is a fear that the witness may be intimidated. These provisions, which nevertheless permit cross-examination of the witness, add to, without detracting from common law principles.

Ss 23 – 26 of the Criminal Justice Act 1988 created new exceptions to the hearsay rule. By Section 23 (1) a statement by a person in a document became admissible as evidence of any fact of which direct oral evidence by him would be admissible, provided the maker of the statement is unable to attend court, because among other reasons, he is dead or unfit to attend, or because, having made a statement, he is unable to give evidence through fear. Where such evidence is admitted, of course, there is no scope for cross-examination. That said, however, safeguards for the defendant are provided in schedule 2 of the Act, which it is unnecessary to detail, but which must be considered in the context of the disclosure obligations on the prosecution. The present purposes we need not address the subsequent statutory provisions, or the replacement of the provisions of sections 23-26 of the 1988 Act by sections 116 and 117 of the Criminal Justice Act 2003.

In Scott and anor v R: Barnes and others v R (1989) 89CAR 153 the Privy Council was faced with decisions of the Court of Appeal of Jamaica, dismissing appeals against convictions for capital murder. In Scott, a special constable was shot with his own revolver in a bar, and subsequently died of his wounds. The only evidence identifying Scott and his co-accused, Walters, was contained in the deposition of a witness who had died before trial. In Barnes and others the deceased was shot after stopping his van and his money was stolen. Three defendants were charged with his murder. The only evidence identifying them was given by a witness, who gave evidence at the preliminary inquiry, but who was murdered before the trial. In each case, the evidence of the missing witness was read as his evidence. In the Privy Council, Lord Griffiths giving the judgment of the Board in effect summarised the common law. He underlined the discretion of the judge to exclude such evidence. However he pointed out:

“ If the courts are too ready to exclude the deposition of a deceased witness it may well place the lives of witnesses at risk particularly in a case where only one witness has been courageous enough to give evidence against the accused or only one witness has had the opportunity to identify the accused”.

Lord Griffiths continued by indicating a number of precautions which could be taken, adding:

“no rules can usefully be laid down to control the detail to which a judge should descend in the individual case…. This much however can be said that neither the inability to cross examine, nor the fact that the deposition contains the only evidence against the accused, nor the fact that identification evidence will of itself be sufficient to justify the exercise of the discretion.”

At common law this approach is unremarkable. In R v Dragic (1996) 2CAR 232 Lord Taylor CJ, dealing with a case in which a witness was too ill to give oral evidence, observed:

“The fact that there is no ability to cross-examine, that the witness who is absent is the only evidence against the accused and that his evidence is identification evidence is not sufficient to render the admission of written evidence from that witness contrary to the interests of justice or unfair to the defendant per se. What matters in our judgment, is the content of the statement and the circumstances of the particular case….”

So Scott, and Dragic provide examples of cases where notwithstanding the clear disadvantages faced by the defendant, evidence of critical prosecution witnesses was adduced in statement form.

The deployment of such evidence without the opportunity of cross-examination has recently been addressed in the context of the jurisprudence of the European Court of Human Rights in R v Sellick (2005) 1WLR 327 and R v Al Khawaja (2006) 1 WLR 1078.

The express language of article 6 (3)(d) of the Convention provides that:

“Everyone charged with a criminal offence has the following minimum rights:

……

(d)

to examine or have examined witnesses against him and to obtain the attendance and examination of witnesses on his behalf under the same conditions as witnesses against him…. ”

Nevertheless in Sellick, exercising his powers under sections 23 and 26 of the Criminal Justice Act 1988, and subject to section 28 and schedule 2, the judge gave leave for the written statements of four witnesses for the prosecution to be read to the jury as their evidence. The conviction was challenged on the basis that the order deprived the defendants of an adequate opportunity to challenge their evidence. As the judgment recalls, reliance was placed on a decision of the European Court in Luca v Italy (2001) 36 EHRR 807 where the court observed:

“ ….it may prove necessary in certain circumstances to refer to depositions made during the investigative stage….if the defendant has been given an adequate and proper opportunity to challenge the depositions, either when made or at a later stage, their admission in evidence will not in itself contravene article 6(1) and (3)(d). The corollary of that, however, is that where a conviction is based solely or to a decisive degree based on deposition that had been made by a person whom the accused had had no opportunity to examine or have examined, whether during the investigation or at the trial, the rights of the defence are restricted to an extent that is incompatible with the guarantees provided by article 6.”

One of the main submissions in Sellick was that the statements were indeed the decisive or sole evidence against the appellants. Accordingly, permission to read the statements should not have been given. Save that it arises in the context of anonymous witnesses rather than the reading of a witness statement, this is the essential argument for the present appellants.

In the course of a detailed analysis of eight decisions of the European Court, Waller LJ commented,

“If a defendant, through fear, keeps a witness away from an English trial, then it is, as it would seem to us, the defendant who is depriving himself or herself of the right and the only right that the defendant has to examine that witness. How, we ask rhetorically, can it be said to be an infringement of the defendant’s article 6 rights for him to deprive himself of that opportunity?...the question whether article 6 has been infringed is very fact sensitive. Despite that fact – sensitivity, certain general principles have been stated from time to time…there is, however, no case as far we know where the Strasbourg court has actually had to deal with a key witness being kept away by fear by a defendant from a trial in the English jurisdiction. ”

After examining the authorities, Waller LJ suggested, at paragraph 50, that four propositions could be discerned. After summarising them, he turned to the question whether there was a fifth proposition “to the effect that where the circumstances would otherwise justify the reading of the statement where the defendant has had no opportunity to question the witness at any stage of the trial process, the statement must not be allowed to be read if it is the sole or decisive evidence against the defendant”. He reminded himself of the passage in Luca v Italy, and other authorities in the European Court, which lent support to the proposition that if witness statements were read, then if they were the sole or decisive evidence, a breach of article 6 of the Convention would be established. However, he concluded on examination that although the Court undoubtedly had “extreme circumstances in mind”, neither in Luca v Italy, nor any of the other authorities, was the European Court directly engaged with a case where the witness “whose identity was well known to the defendant was being kept away by fear”.

The court posed itself this question:

“If the European court were faced with the case of an identified witness, well known to a defendant, who was the sole witness of a murder, where the national court could be sure that that witness had been kept away by the defendant, or by persons acting for him, is it conceivable that it would hold that there were no “counter-balancing” measure the court could take which allow that statement to be read?”

Waller LJ addressed the problem, first, on the basis that the Court was sure that the defendant “denied himself the opportunity of examining the witnesses.” If so, he could not complain of an infringement of article 6(3)(d) provided precautions were taken to ensure the fairness of the trial process. Even if the situation were not quite as clearly established to the court’s satisfaction, he suggested,

“In our view, having regard to the rights of victims, their families, and the safety of the public in general, it still cannot be right for there to be some absolute rule that, where compelling evidence is the sole or decisive evidence, an admission in evidence of a statement must then automatically lead to a defendant’s article 6 rights being infringed. That would lead to a situation in which the more successful the intimidation of the witnesses, the stronger the argument becomes that the statements cannot be read. If the decisive witnesses can be “got at” the case must collapse. The more subtle and less easily established intimidation provides defendants with the opportunity of excluding the most material evidence against them. Such an absolute rule cannot have been intended by the European court in Strasbourg. ”

In R v Al Khawaja, the sole witness to an indecent assault died, after she had made a statement for use by the prosecution. Without her statement, the prosecution would have been obliged to abandon the allegation of indecent assault on her. The statement was admitted in evidence and the appellant was convicted. The argument in the appeal relied strongly on Luca v Italy. The appeal was dismissed. After examining the decision in Sellick, the court concluded that there was a strong public interest in the admission of the statement made by the witness prior to her death. Although the public interest could not be “allowed to over-ride the requirement that the defendant should have a fair trial”, the court did not consider that the decisions of the European Court required the inevitable conclusion that in the circumstances which obtained in Al Khawaja, the trial was unfair.

“The provision in article 6(3)(d) that a person charged shall be able to have the witness against him examined is one specific aspect of a fair trial: but if the opportunity is not provided, the question is “whether the proceedings as a whole, including the way in which the evidence was taken, were fair: see Doorson v The Netherlands…. ”

We can now come to the specific problem which arises in the present appeals, non-disclosure of the true identity of witnesses, together with voice modulation and screens, usually summarised as witness anonymity. The disadvantages for the defendant are not identical to the disadvantages created when cross-examination is not available. We gratefully adopt as illustrative of common law principle, as well as Strasbourg jurisprudence, the analysis in Kostovski v Netherlands (1989) 12 EHRR 434:

“If the defence is unaware of the identity of the person it seeks to question, it may be deprived of the very particulars enabling it to demonstrate that he or she is prejudiced, hostile or unreliable. A testimony or other dec1aration inculpating an accused may well be designedly untruthful or simply erroneous and the defence will scarcely be able to bring this to light if it lacks the information permitting it to test the author’s reliability or cast doubt on his credibility. The dangers inherent in such a situation are obvious”.

These disadvantages are immediately ameliorated provided the defendant retains, as he normally does, the ability of his counsel to pursue a substantial degree of cross-examination of the witnesses before the jury. This familiar safeguard, directly reflected in article 6(3)(d) of the Convention, is reinforced by another with which we are equally familiar in this jurisdiction, but which perhaps needs emphasis, the proper discharge of the prosecution’s obligations in relation to disclosure of material which may assist the defence or damage the prosecution case, and in particular which may serve to undermine the creditworthiness of the anonymous witness. We simply add, for completeness, that in broad terms the same considerations would apply to witnesses called by or on behalf of the defendant. In the Davis appeal, we heard evidence from an anonymous witness, described as “Tony Pink”, not his real name, called on behalf of the appellant, not the Crown. As we shall describe, there were a number of profoundly unsatisfactory aspects of his evidence, but the basis on which he asked to be allowed to give his evidence anonymously was his own genuine fear of the consequences if his identity became known.

There is clear jurisdiction at common law to admit incriminating evidence given against the defendant by anonymous witnesses. (see R v Watford Magistrates Court ex parte Lenman (1993) CLR 388, R v Taylor and Crabb, unreported, 22^nd July 1994, and R v Jack, 7^th April 1998, unreported). In Al Fawwaz v Governor of Brixton Prison 1 (2002) AC 556, in extradition proceedings, the decision of the examining magistrate to receive evidence of two anonymous witnesses in support of the allegation of terrorism was upheld in the House of Lords. In R (D) v CamberwellGreen Court (2005) 1 WLR 393, in the context of section 21 of the Youth Justice and Criminal Evidence Act 1991, the principle was again affirmed in the House of Lords. It may therefore be wondered why we should examine the jurisprudence of the European Court of Human Rights, on which so much of the argument on behalf of the appellants depended. The point, simply expressed, is that to a significant extent in the Davis appeal, but not on close analysis to any extent at all in the Ellis appeal, the conviction was based on evidence given by anonymous witnesses.

We begin by noting the argument for Al Fawwaz, presented by Mr Edward Fitzgerald QC. Basing himself on article 6 of the European Convention of Human Rights, he contended that it was “fundamentally unfair” for the evidence of an anonymous witness to be accepted. He pointed out that the subject of an extradition request “is entitled to challenge the credibility of a prosecution witness before the magistrate”, adding, that in the United States of America, his anonymity would not be preserved at trial. We understand this to follow from the Sixth Amendment to the United States Constitution which, in 1791, provided the defendant in a criminal trial with the right “to be confronted with the witnesses against him.” Mr Fitzgerald also argued that reliance on “anonymous statements as sufficient evidence to found a conviction” violated article 6 (3)(d) of the European Convention. He drew attention to and relied on Kostovski v The Netherlands. He also submitted that there was a breach of the principle of fairness which applied by virtue of article 5(4) of the Convention. In further argument the attention of the House of Lords was also directed to Doorsonv The Netherlands (1996) 22 EHRR 330 and Fitt v United Kingdom (2000) 30 EHRR 480.

In the context of witness anonymity three rights protected by the Convention are engaged. Article 2 and article 8 are closely related to the rights, among others, of witnesses, and article 6 is specifically concerned with the protection of the defendant’s right to a fair trial. Article 2 provides that the right to life “shall be protected by law”, and article 8 requires respect to be paid to the private and family life of every individual.

The starting point is article 2 which imposes a positive obligation on the state to take appropriate steps to safeguard the lives of those within its jurisdiction. In Osman v United Kingdom (1999) 1 FLR 3, the court suggested that the authorities should “do all that could reasonably be expected of them to avoid a real and immediate risk to life of which they have or ought to have knowledge. This is a question which can only be answered in the light of all the circumstances of any particular case”. When a witness’s fear for his own life or for that of a member of his family is justified, the state’s obligation to take positive steps to safeguard them is engaged. Indeed, the proper administration of justice requires that witnesses should not be exposed to danger or retribution for coming forward, in accordance with their duties as citizens, to offer and give evidence. When they do, the right to respect for private and family life under article 8 is also engaged. One available protective measure is witness anonymity, supported by voice modulation and screens, which, provided the witness can be examined by or on behalf of the defendant, is not prohibited by any express words in article 6.

As far as we can discover, the only case which has reached the Commission from the United Kingdom, on what we can fairly describe as the anonymity issue (X v UK 15 EHRR CD 113) was dismissed as manifestly ill founded and is of no assistance. In the remaining “anonymity” cases, arising from proceedings in civil law jurisdictions as far as we can discern, statements were obtained from anonymous witnesses who were not examined by or on behalf of the person charged at the hearing at which guilt or innocence was determined; and sometimes, indeed, at none of the stages in the proceedings. In Kostovski vNetherlands the applicant was convicted of armed robbery after a trial before the District Court on the basis of the evidence of two anonymous witnesses who provided statements to the police which incriminated him. One confirmed the truth of his statement before an examining magistrate. Under Dutch law, only that statement was evidence in the case. Neither the applicant nor his legal advisors was present when the statements were made. Neither witness was called at the trial. Evidence was given by the examining magistrates, and one of the police officers who had interviewed the witnesses, to the effect that they had created a favourable impression upon them. Dutch procedure did not permit the defence to ask them questions designed to test the credibility of the anonymous witnesses. It was conceded that the applicant’s conviction was based “to a decisive extent” on the anonymous statements (para 44).

The Court observed that “as a rule these rights (i.e. the rights under article 6) require that an accused should be given an adequate and proper opportunity to challenge and question a witness against him, either at the time the witness was making his statement or at some later stage of the proceedings”. The observation plainly admits of exceptions to the express rights provided in article 6, but as the witnesses could not at any stage be directly questioned by the applicant or on his behalf, their anonymity “compounded the difficulties facing the applicant”. In the circumstances, the Court concluded that the applicant had not received a fair trial.

In Windisch v Austria (1990) 13EHRR 281, the applicant was convicted of burglary and sentenced to 3 years imprisonment by the Innsbruck Regional Court. The evidence against him consisted largely of two statements made by anonymous witnesses to the police. The Court heard evidence from the police officers, but not from the witnesses. Despite repeated requests, neither the applicant nor his advocate was allowed an opportunity to examine them (para 27). The Court held that the applicant had not received a fair trial.

Ludi v Switzerland (1993) 15EHRR 173, was very similar. The applicant was convicted of a drug trafficking offence and sentenced to 3 years imprisonment by the Laufen District Court. The evidence against him consisted mainly of a report by an anonymous undercover agent and transcripts of telephone intercepts of calls between the agent and the applicant. Neither the applicant nor his advocates were given the opportunity directly to question the agent. They wished to demonstrate the extent to which the applicant had been induced or entrapped to commit the offence. The Court held that the rights of the defence were restricted to such an extent that the applicant did not have a fair trial. Significantly, it observed at para 49,

“Yet it would have been possible to do this in a way which took into account the legitimate interest of the police authorities in a drug trafficking case in preserving the anonymity of their agent (our emphasis), so that they could protect him and also make use of him again in the future”.

In short, in each of these three decision, the trial was unfair, not because important witnesses gave evidence anonymously, but because in addition to giving their evidence anonymously, the entitlement to cross-examine was denied. Of itself, it was not suggested that the preservation of witness anonymity made the trial unfair.

In Doorson vNetherlands (1996) 22EHRR 330, the need to take into account the right to life of the witness under article 2 and to family and private life under article 8, when assessing the fairness of a trial under article 6 was expressly taken into account. In complex proceedings before the Amsterdam Regional and Appellate Courts, the applicant was convicted of drug trafficking and sentenced to 15 months imprisonment, and his appeal against conviction was rejected. His conviction was based in significant part upon the statements of two anonymous witnesses, who were questioned by the investigating judge, in the absence of the applicant’s advocate. They were not available to be questioned at the trial. They were questioned by the applicant’s advocate at the appeal hearing, which rectified the lack of opportunity to question them earlier. The Court observed:

“It is true that article 6 does not explicitly require the interests of witnesses in general, and those of victims called upon to testify in particular, to be taken into consideration. However, their life, liberty or security of person may be at stake, as may interests coming generally within the ambit of Article 8 of the Convention. Such interests of witnesses and victims are in principle protected by other, substantive provisions of the Convention, which imply that contracting States should organise their criminal proceedings in such a way that those interests are not unjustifiably imperilled. Against this background, principles of fair trial also require that in appropriate cases the interests of the defence are balanced against those of witnesses or victims called upon to testify”. (para 70)

The Court approved as “a relevant reason to allow them anonymity” the need to obtain evidence from them, while at the same time protecting them against the possibility of reprisals. Although there was no evidence that the applicant himself had threatened the witnesses, the Court stated that “regard must be had to the fact, as established by the domestic courts and not contested by the applicant, that drug dealers frequently resorted to threats or actual violence against persons who gave evidence against them”. The Court added:

“The maintenance of the anonymity of the witnesses Y15 and Y16 presented the defence with difficulties which criminal proceedings should not normally involve. Never the less, no violation of article 6 (1) taken together with article 6 (3)(d) of the Convention can be found if it is established that the handicaps under which the defence laboured were sufficiently counter balanced by the procedures followed by the judicial authorities”. (para 72)

The Court decided that the questioning of the anonymous witnesses at the appeal stage by counsel was a sufficient “counter balancing” procedure.

Mr Malcolm Swift QC, counsel for Davis, and Mr Nigel Rumfitt QC, on behalf of Ellis, underlined the passage at para 76, echoing Luca v Italy that,

“Finally, it should be recalled that, even when “counter balancing” procedures are found to compensate sufficiently the handicaps under which the defence labours, a conviction should not be based either solely or to a decisive extent on anonymous statements”.

In this part of the judgment, the language used by the Court (at least in its English version) is ambiguous: “Anonymous statements” may mean statements made otherwise than in evidence at trial; or it may extend to anonymous evidence given at trial. The fact that the Court contemplated the preservation of anonymity in Ludi v Switzerland suggests that the latter was not mind. We shall return to this point shortly. For the moment, we record that the essential question for the court was whether the proceedings were unfair. Notwithstanding the significance to the conviction of the evidence of the two anonymous witnesses, in Doorson the conclusion was that they were not.

In Van Mechelen v Netherlands (1997) 25 EHRR 647 the applicants were convicted of attempted manslaughter and robbery and sentenced to 10 years imprisonment by the s-Hertogenbosch Regional Court. The only evidence positively identifying the applicants as perpetrators of the crime was provided by statements of anonymous police officers, who were never questioned by or on behalf of the applicants before their conviction. On appeal the Court ordered that an opportunity should be given for the applicant’s lawyers to question the police officers under the supervision of an investigating judge. The investigating Judge, the witness and a registrar were present in one room, the defendants, their lawyers and the Advocate General in another. The two were connected by a voice-only link. Questions were asked and answered through this link. The defence were unaware of the identity of the police witnesses, and were unable directly to test their reliability.

The Court observed that “the balancing of the interests of the defence against arguments in favour of maintaining the anonymity of witnesses raises special problems if the witnesses in question are members of the police force of the State”. It noted that their interests, and those of their families, deserved protection under the Convention, adding that “it must be recognised that their position is to some extent different from that of a disinterested witness or a victim”, because they owe a general duty of obedience to the State’s executive authorities and have links with the prosecution and can expect that their duties will involve giving evidence in open court. That said, the court acknowledged that, provided the rights of the defence were respected, preservation of the anonymity of an undercover agent may be necessary for his own and his family’s protection, and so as not to impair his usefulness for future operations. “Any measures restricting the rights of the defence should be strictly necessary. If a less restrictive measure can suffice then that measure should be applied”. The Court stated:

“It has not been explained to the Court’s satisfaction why it was necessary to resort to such extreme limitations on the right of the accused to have the evidence against them given in their presence, or why less far reaching measures were not considered”. (para 60)

The measures identified by the court included disguise and the prevention of eye contact, both of which would have improved the prospect of successful concealment of the identity of the witnesses.

The Court was unpersuaded that sufficient effort had been made to assess the threat of reprisals against the police officers or their families. In those circumstances, the Court held that the right to a fair trial under article 6 had been infringed. The court also “recalled”…… “that a conviction should not be based either solely or to a decisive extent on anonymous statements”, commenting, in a passage, understandably relied on here by the appellants,

“Moreover, the only evidence relied on by the Court of Appeal which provided positive identification of the applicants as the perpetrators of the crimes were the statements of the anonymous police officers. That being so the conviction of the applicants was based “to a decisive extent” on these anonymous statements”. (para 63)

Again, in the English text, the reference to “statements” is ambiguous. In the French text, it is not. In first sentence, the word used is “déclarations” In the second sentence, it is “dépositions”. Taken together with the suggestion that the protective measures went much further than necessary and that steps could have been taken to conceal the identities of the witnesses, it appears that the objectionable feature of the proceedings was not anonymity as such, but reliance on inadequately tested depositions.

In Visser v Netherlands 1 4/2/O2 the applicant was convicted, in April 1991, on appeal, of a kidnapping committed in 1988 (having been acquitted at the Utrecht Regional Court in June 1989). The foundation of his conviction was an identification made to police officers by an anonymous witness, one of a number of people who had made enquiries about the whereabouts of the victim before the kidnapping. At that stage the witness had not been examined before the court. On further appeal to the Supreme Court, that Court directed that the witness be heard by an investigating judge. That was done in September 1993. The witness reaffirmed his wish to remain anonymous. The judge and the witness were in one room. Counsel for the applicant was permitted to submit questions, from another room, in writing to him, and to submit further questions when the witness’s statement had been made. At no stage did he have the opportunity to question him directly or to observe his demeanour. In reliance on this evidence, the Court of Appeal convicted him afresh.

In 2002 the European Court held that the applicant’s article 6 rights had been infringed. It reminded itself of “its rulings in a series of cases concerning reliance on witness testimony which was not adduced before the Trial Court, that article 6(3)(d) only required the possibility to cross examine such witnesses in situations where this testimony played a main or decisive role in securing the conviction”. The reason for holding that the applicant’s rights were violated was that the investigating judge had not assessed the reasonableness of the witness’s fear in 1988 or in 1993. No adverse finding based on the applicant’s claim that his conviction was based to a decisive extent on “the anonymous testimony” was made.

We note Birutis v Lithuania 28 June 2002, when the Court dealt with a now familiar problem: the conviction of an applicant on the basis of anonymous statements which were not tested by examination at trial. Although the Court criticised the means adopted by the authorities “in handling the anonymous evidence”, the decision provides no assistance in determining the circumstances in which the evidence of an anonymous witness, given in person at trial, may be relied on.

In PS v Germany 36EHRR 61 the applicant was convicted of sexual abuse of a child. The evidence against him consisted of a statement made by the child’s mother about what her daughter had told her, and evidence by a police officer who had questioned the daughter shortly after the offence. The daughter was not available to be questioned by the Court or on behalf of the defendant. On appeal, the Regional Court ordered a psychological assessment of the daughter’s credibility. The expert reported that her statements were credible. Her parents refused to bring her to the Appeal Court for questioning. The appeal was dismissed.

The European Court noted that these procedures could not “be considered as having enabled the defence to challenge the evidence of (the daughter), reported in Court by third persons, one of them a close relative” and accordingly held that there had been a violation of article 6.

The Court said, at paragraphs 21-24,

“All the evidence must normally be produced at a public hearing, in the presence of the accused, with a view to adversarial argument. There are exceptions to this principle, but they must not infringe the rights of the defence. As a general rule, the accused must be given an adequate and proper opportunity to challenge and question a witness against him, either when he makes his statement or at a later stage”.

In passing, we observe that this was a plain acknowledgement of permissible exceptions to the entitlements under article 6 (3)(d), and that the absence of an opportunity to cross-examine an incriminating witness would not, of itself, constitute an unfair trial. The court continued,

“In appropriate cases, principles of fair trial require that the interests of the defence are balanced against those of witnesses or victims called upon to testify, in particular where life, liberty or security of a person are at stake, or interests coming generally within the ambit of article 8 of the Convention. However only such measures restricting the rights of the defence which are strictly necessary are permissible under article 6. Moreover in order to ensure that the accused receives a fair trial, any difficulties caused to the defence by a limitation on its rights must be sufficiently counter balanced by the procedures followed by the judicial authorities.

Where a conviction is based solely or to a decisive degree on depositions that have been made by a person whom the accused has had no opportunity to examine or have examined, whether during the investigation or at the trial, the rights of the defence are restricted to an extent that is incompatible with the guarantees provided by article 6”.

The last paragraph appears to clarify the ambiguity inherent in the use of the phrase “anonymous statements” in some of the earlier decisions. The specific concern was the absence of a proper opportunity for the defence to examine, or have examined, the witness who made an incriminating, and effectively decisive, deposition, rather than the fact that the witness giving evidence critical to the conviction remained anonymous.

The Court used less restrictive language in its most recent reported (but not as yet final) decision, Krasniki v Czech Republic 28 February 2006. The applicant was convicted of retail drug trafficking on the evidence of two anonymous witnesses, “Jan” and “Jana”. Both were questioned during pre-trial investigations by the police and a defence lawyer. At the trial before the District Court in Teplice, “Jan” was questioned by the presiding judge, outside the court room and out of sight of the applicant and his counsel. They were only able to put questions through the presiding judge. “Jana” could not be found and her statement was read. The applicant was convicted. His conviction was based solely on the evidence of the anonymous witnesses. The Court reminded itself of its case law (Doorson and Van Mechelen). It could not determine from the case file how the investigating officer or the trial judge assessed the reasonableness of the personal fear of the witnesses vis a vis the applicant and that, in effect, meant that the protective measures were not apparently justified. The conviction was based solely or to a decisive extent on the “anonymous testimonies”, but as we have seen, without direct cross-examination on the defendant’s behalf of one crucial witness, and no cross-examination at all of the second crucial witness.

This unavoidably lengthy review of the decided cases underlines the need to identify precisely the “evidence” under consideration by the European Court when the issue of anonymous witnesses was being addressed. The “evidence” may be “depositions”, a seemingly clear reference to out of court statements, “statements” and “declarations”, words which are potentially ambiguous, and “testimonies”, apt to describe evidence actually given at a trial. It is also noteworthy however, that whenever the issues of witness anonymity arise, the European Court refers back to its two seminal decisions, Doorson and VanMechelen. Later cases appear to be no more than an application of established principles.

The Court accepted that the Convention rights of witnesses, included, where necessary, the preservation of their anonymity, and in our judgment decided that the concealment of the identity of witnesses is not inconsistent with the right to a fair trial, provided, first, the need for anonymity is clearly established, second, that cross-examination of the witness by an advocate for the defendant is permitted, and finally, the ultimate test, that the trial should be fair. In agreement with the decision in R v Sellick, but applying it to the problem of anonymous witnesses in justifiable and genuine fear, whose testimony can be tested in the adversarial process, we do not accept that the jurisprudence of the Court leads to the conclusion that the trial will inevitably be considered unfair, and the conviction unsafe, simply because the evidence of the anonymous witnesses may be decisive to the outcome. R v Sellick addressed the problem raised by the deprivation of an express right provided by article 6. That does not apply in the context of anonymous witnesses whose evidence may be tested by examination at trial. It may be, but we need not decide, more troublesome to permit a conviction to depend to a decisive extent on evidence admitted in direct contradiction of what would otherwise be an express Convention entitlement than it would if the admission of the evidence did not contravene any such right, but in the final analysis, any discussion about that distinction may be less important than a close examination whether the trial overall was a fair one. That, as all the jurisprudence demonstrates, is the absolute fundamental (Doorson v Netherlands), and the question whether, looked at overall the trial is unfair is, as the detailed factual judgments from the Court show, fact specific.

We can now return to the decision of the House of Lords in Al Fawwaz which, as we remind ourselves, had the advantage of argument in which attention was drawn to the decision of the European Court in Kostovski, and Doorson.

Lord Slynn of Hanley observed that

“ It is obvious that at trial …. The court should be cautious about admitting anonymous statements or affidavits. But that there is jurisdiction to admit them is clear. On the basis of what was said in R v Taylor (Gary) The Times 17 August 1994, and by the European Court of Human Rights in Doorson v TheNetherlands (1996) 22EHRR 330, both for the purposes of the Convention and the purposes of considering whether the legal proceedings have been fairly conducted at common law, there may have to be a balance of the interests of the defence and of the protection of witnesses. It seems to me that the magistrates and the Divisional Court considered this matter carefully and were satisfied that the protection of the witness…. made it necessary in all the circumstances to preserve his anonymity and that the interests of society in prosecuting required that the evidence be taken into account….”

Lord Hutton examined the judgment of this court in R v Taylor (Gary), and considered R v X and ex p Lenman. He suggested that there was a degree of inconsistency between the analysis of principle in Rv Taylor (Gary) and the judgments in R v X and ex p Lenman, pointing out that notwithstanding the importance to the process of the accused knowing the identity of his accuser, there were cases in which the balance of fairness might come down in favour of the prosecution “notwithstanding that the circumstances could not be described as rare and exceptional”.

Lord Rodger of Earlsferry directly addressed the observations of the European Court in Doorson, and concluded that the use of the evidence of anonymous witnesses was not, under the jurisprudence of the European Court itself, invariably to be treated as incompatible with Convention rights. He also drew attention to Fitt v United Kingdom (2000) 30 EHRR 480, where it was pointed out that :

“ In any criminal proceedings there may be competing interests, such as national security or the need to protect witnesses at risk of reprisals or keep secret police methods of investigation of crime, which must be weighed against the rights of the accused (see, for example, the Doorson v The Netherlands….) In some cases it may be necessary to withhold certain evidence from the defence so as to preserve the fundamental rights of another individual or to safeguard an important public interest. However only such measures restricting the rights of the defence which are strictly necessary are permissible under Article 6(1) (see Mechelen v The Netherlands…). Moreover in order to ensure that the accused receives a fair trial, any difficulties caused to the defence by a limitation on its rights must be sufficiently counter-balanced by the procedures followed by the judicial authorities.”

When the House of Lords returned to the issue of witness anonymity in R(D) v Camberwell Youth Court, in the context of the statutory provisions which enable child witnesses to give evidence by video recorded interview and live television link, attention was drawn to decision of the European Court in Kostovsi, Doorson, van Mechelen and Visser. After reviewing these and other authorities, Lord Rodger explained that the distinction in the popular mind between the British and Continental systems of criminal justice was sufficiently accurate “to make one anticipate that the introduction of article 6(3) (d) will not have added anything of significance to any requirements of English law for witnesses to give their evidence in the presence of the accused. An examination of the case law of the European Court tends to confirm that much of the impact of article 6(3) (d) has been on the procedures of Continental systems which previously allowed an accused person to be convicted on the basis of evidence from witnesses who he had not had an opportunity to challenge”. Thereafter, after considering the process by which the defendant could challenge and question witnesses giving incriminating evidence against him, Lord Rodger suggested that the overall impact of the decisions of the European Court was

“that the defence should have an adequate and proper opportunity to challenge and question a witness on his statement at some stage. The requirements of the Convention are satisfied even if that opportunity is afforded before trial”.

Reflecting on the impact of the Sixth Amendment of the Constitution of the United States, he observed,

“Whatever its merits, this line of thought never gave rise to a corresponding requirement in English law….. nor has Article 6(3) (d) of the Convention been interpreting as guaranteeing the accused a right to be in the same room as the witness giving evidence. What matters, as Kostovski v The Netherlands shows, is that the defence should have a proper opportunity to challenge and question the witnesses against the accused…. these requirements can be satisfied even where, for good reason, the accused is not physically present at the questioning”.

Baroness Hale underlined that a face to face confrontation between the defendant and the witnesses against him was not obligatory, and in the context of anonymous witnesses, she added,

“Even then the Strasbourg Court has accepted that exceptions may be made, provided that sufficient steps are taken to counter-balance the handicaps under which the defence laboured and the conviction is not based solely or decisively on anonymous statements”

In context, this observation appears to reflect the concerns expressed by the European Court about the dangers of over-reliance on statements by anonymous witnesses, where the defendant, or his advocate, has no opportunity for cross examination.

Thus the House of Lords has now twice addressed the issue of anonymous witnesses in the light of the decisions of the European Court. In our view it would be quite unrealistic to consider that the observations of the House of Lords in Al Fawwaz, arising in the context of extradition proceedings, and Camberwell Green, in which the use of “special measures” introduced by statute was under consideration, should, for the purpose of considering witness anonymity problems arising in the Crown Court when the statutory “special measures” provisions do not apply, be treated as obiter dicta.

The principle of judicial precedence requires us to apply the reasoning of the House of Lords. (Kay & Others v Lambeth London Borough Council: Leeds City Council v Price (2006) 2WLR 570. In any event, we can detect no conflict between the decisions of the European Court and the observations of the House of Lords on the issue of witness anonymity. In our judgment the discretion to permit evidence to be given by witnesses whose identity may not be known to the defendant is now beyond question. The potential disadvantages to the defendant require the court to examine the application for witness anonymity with scrupulous care, to ensure that it is necessary and that the witness is indeed in genuine and justified fear of serious consequences if his true identity became known to the defendant or the defendant’s associates. It is in any event elementary that the court should be alert to potential or actual disadvantages faced by the defendant in consequence of any anonymity ruling, and ensure that necessary and appropriate precautions are taken to ensure that the trial itself will be fair. Provided that appropriate safeguards are applied, and the judge is satisfied that a fair trial can take place, it may proceed. If not, he should not permit anonymity. If he does so, and there is a conviction, it is not to be regarded as unsafe simply because the evidence of anonymous witnesses may have been decisive.

Among the safeguards, first, is the decision of the trial judge whether to exercise his discretion to allow some or all the witnesses against to the defendant to give their evidence anonymously. If the only evidence against the defendant consists of wholly unsupported anonymous witnesses, whose evidence is demonstrably suspect, the judge may decide, as Hughes J did in R v Bola, that the Crown should not adduce it. Again, if the decisive evidence comes from an unidentified witness who cannot be cross-examined (for example, anonymous witnesses whose evidence was admitted in documentary form only, and who will not be called to give oral testimony), the judge may decide that the evidence should not be admitted. As we emphasise, these are issues for judicial decision in case specific situations, after allowing for the disclosure process, any PII decisions, and the ability to cross-examine together with the deployment of material helpful to the defendant in the course of cross-examination, or even when cross-examination may not be possible. For example, the judge may be satisfied that a wholly independent, understandably terrified witness, a stranger to the defendant, and with no possible motive to implicate him, may have made a note of a crucial car number plate at the scene of the crime, If satisfied that this witness is indeed independent, but unfit to give live evidence, the judge may admit his or her evidence, anonymously, and in statement form.

At the end of the prosecution case, the judge may decide that it would be unsafe for the evidence of the anonymous witnesses to be considered further by the jury, or indeed, that the case as a whole should be withdrawn from their consideration. We are not seeking to formulate a scheme, merely to identify appropriate safeguards currently in place to ensure the fairness of the trial. Thereafter, if the case is fit to go to the jury, when the evidence is concluded, and to enable the jury properly to approach its task, the judge must give appropriate directions in his summing up, sufficient to identify the particular disadvantages under which the defence may have been labouring. (for examples of what may or may not be appropriate, see R v Sellick and R v Al-Khawaja, referring to R v McCoy). The judge would probably suggest that the jury should consider whether there is any independent, supporting evidence, tending to confirm the credibility of the anonymous witnesses, and the incriminating evidence they have given. We are not reinstating outdated principles relating to corroboration, nor implying that such independent evidence is a pre-requisite to conviction. We are simply reflecting the obvious consideration that independent evidence consistent with the defendant’s guilt would be likely to increase confidence in the truthfulness and accuracy of incriminating anonymous witnesses. We should perhaps finally underline that these appeals themselves demonstrate the effectiveness of the safeguards. As we shall see, in the context of the Davis appeal, the defendant in a linked trial, Harvey, was acquitted by the jury after anonymous witnesses gave evidence incriminating him, and in the Ellis appeal, the cross-examination of the anonymous witnesses wholly undermined their evidence.

The role of the Court of Appeal, Criminal Division

In this constitution we have recently addressed the principles relating to the exercise of the trial judge’s discretion under section 78 of the Police and Criminal Evidence Act 1984. For reasons connected with an on-going trial, publication of that judgment as a whole has been postponed until the conclusion of the trial itself. However, on the basis of the elementary principle at common law summarised in the “birthright” of the defendant to a fair trial, and repeated in Article 6 of the European Convention, we observed

“Trial judges should not admit evidence if notwithstanding the robust safeguards provided by the trial process itself, the effect of doing so would produce a trial which could properly be stigmatised at its end as an unfair trial. If this court were satisfied that the admission of the evidence would or did produce an unfair trial, the decision to admit it would not be upheld”

In short, adapting those observations to the issues in the present appeals, the first question is whether the court should interfere with the judgment of the trial judge to allow for witness anonymity. In most cases, as that involves a balance to be drawn between conflicting factors in a fact specific situation this court will not interfere with the judge’s decision unless, at the time when he made it, it was plainly wrong. However, once leave to appeal is given, it may not be sufficient for the court to dispose of the appeal on the basis that there are no grounds to justify interfering with a decision made by the judge at the time when he made it. After conviction, on appeal, this court may conclude that the order for witness anonymity had the effect of producing what in all the circumstances was an unfair trial. If so, even if the original decision of the trial judge would not be open to criticism, the conviction would be unsafe, and it would be quashed. To highlight this point, we note that the decision of Goldring J in the Ellis trial to permit witness anonymity was itself the subject of an unsuccessful appeal. In a constitution presided over by Rose LJ, the Vice President, this court upheld his decision. We are bound by the decision, and should not and cannot interfere with the exercise of Goldring J’s discretion. Leave to appeal having been given after conviction, a careful investigation of the fairness of the trial and the safety of the convictions in the Ellis appeal is required. That is what has taken place.

Counsel for the Defendant

Having made orders for anonymity, the trial judges in both cases were content for counsel to be able to see the witness giving evidence, and to observe their demeanour whilst doing so. However, like everyone else in court, counsel were bound by the trial judge’s anonymity ruling, and indeed would be in contempt of court if they disobeyed it. Plainly, if the anonymity of the witness were ordered, it would be wholly inappropriate for counsel to provide his client with a physical description of the witness. The effect of doing so would be to dilute, if not altogether extinguish the effect of the order.

So far, so obvious. The potential conflict arises from the delicate nature of the relationship between counsel and their clients. Again the principles are well understood. Thus in R v Davis, Johnson and Rowe (1993) 1WLR 613 in the context of the disclosure of information imparted to defence counsel in the course of an ex parte application, Lord Taylor of Gosforth CJ concluded that counsel could not be required to give an undertaking that he would not reveal to his client or solicitors any information coming into his possession in the course of the ex parte hearing.

“It would wholly undermine counsel’s relationship with his client if he were privy to issues in court but could reveal neither discussion nor even the issues to his client… whatever happens in court with defending counsel present would, in our view, have to be disclosable to his clients”.

(See also R v Preston (1994) 2 AC130, and R v Botmeh (2002) 1 WLR 531)

In R v H and others (2004) 2AC 134, Lord Bingham of Cornhill underlined the critical importance of the nature of the relationship between client and lawyer, and the need for “the quality of confidence” between them. This feature was recently emphasised in R v G and anor (2004)1 WLR 2932, where Rose LJ identified the inevitable damage to the relationship between the lawyer “in the know” and the client because

“….in addition to preventing frankness and fettering the free flow of information between lawyer and client, the order would be likely to nurture in the client a belief that his lawyers are putting other interests …above his own; and the client’s perception of the relationship is a matter of importance…. ”

This analysis highlights the critical question, how to deal with the potential conflict of duty for counsel bound by order of the court to preserve witness anonymity, and the fulfilment of a proper relationship between him and his client, the defendant.

We cannot provide an answer to every possible combination of circumstances which may arise. Our pragmatic solution to the problems which arise after an order for anonymity is that counsel should identify the issues for his client, and explain the alternatives. The choice is stark, and should be made by the client, on advice. Counsel should inform him that if he, counsel, were able to observe the witness he may be better able to cross-examine him. That is a potential advantage which he would be well able to explain to his client. However, he would also have to explain the pre-condition to such cross-examination, that he could not flout or ignore the order of the court, and could not, at any stage of the trial or afterwards, provide his client with any description of the witness. That, in effect, was what happened in the Ellis trial.

The alternative is that the defendant may nevertheless instruct his counsel that he should not see the witness if the opportunity to do so is subject to the prohibition on disclosure of his identity, or discussion of his physical description, or, as happened in Davis, the defendant instructs counsel that he should inform him of the appearance of the witness. Alternatively, counsel may believe that the professional relationship with his client would be damaged if he were unable to communicate information which his client wanted from him. In situations like these, the court will order that the anonymous witnesses must be screened from counsel. In this event, again as happened here, counsel for the Crown should be in the same position as counsel for the defendant.

Mr Swift invited us to consider an extreme case, or perhaps more accurately, an example of an extreme case, where counsel for the defendant received express instructions from his client, with a description given by the client of the witness believed by the client to be the prosecution witness against him. Counsel, observing the witness for himself, might very quickly establish from the descriptions, and comparing it with the actual appearance of the witness, that the defendant’s belief was completely wrong. What asked Mr Swift rhetorically, should counsel do then? The short answer is that we should expect counsel to ask the court to adjourn into chambers for the problem to be discussed with the judge, and, after hearing both sides, for the judge to give the appropriate ruling. We cannot legislate in advance for all the possible case specific problems which may arise.

We should perhaps add two important footnotes. First, whether or not counsel decides to observe the witness personally, at the trial each member of the jury, as well as the judge, remains able to see and observe the witness for himself. They are also able to hear the unmodulated voices of the witnesses, even if they are mechanically disguised for the defendant and public, as well as counsel. Second, if for reasons similar to those referred to in paragraph 71 the decision is made that counsel shall not observe the witness, the defendant cannot subsequently make a complaint to this court based on his, or his counsel’s rejection of the opportunity to do so. The alternatives are clear, and the defendant has to accept the consequences of his decision, or as in the Davis case, the instructions he has given to counsel.

At the start of the New Year celebrations which culminated in the deaths of Kenton and Mowatt there was no trouble at a well organised party. At its height some 60 or so people attended. They included the appellant, who was present at the party for a substantial time throughout the evening and indeed even on his own eventual case at trial, until 8.00-8.30am. Kenton was the disc jockey responsible for the music, and the guests included some of his close friends. Towards morning a little ill temper became apparent and some of the visitors left. At about 8.00am the organisers decided to bring the party to an end, and the disc jockey was instructed to stop the music, which he did.

The killings occurred at about 9.20am. The bullet which killed both men was a 7.62mm Tokarev copper jacketed lead bullet, and the firing marks indicated that it was most likely to have been fired from a Tokarev self-loading pistol. The gun was fired at a close range to Kenton’s neck. The bullet was recovered from Mowat’s body. Neither the cartridge case nor the gun was ever recovered.

Obvious confusion was caused by the fact that one shot had caused two deaths. One anonymous witness, a close friend of Kenton, gave evidence about the incident without incriminating the appellant. It gives, and other witnesses gave, alarming evidence about the ease with which the gun came to be used. Towards the end of the party, while a group of friends, including Kenton were leaving, a man blocked the doorway, saying something to the effect that nobody was going anywhere. Something of a verbal skirmish developed, and Kenton put down the box containing his musical equipment, and asked what was wrong. The man grabbed Kenton round the collar and Kenton grabbed him back. The man tried to throw a punch which missed; Kenton punched him back and struck him, causing the man’s head to hit the door of the next flat. This flew open. The man stumbled back. He emerged from the doorway, and stepped forward two paces and fired a shot. The witness did not see how he had come into possession of the gun, but he had reached behind him, and afterwards went straight downstairs. The witness did not see the gun itself, only the flash of the shot. He had not seen the gun before.

The evidence of three of Kenton’s friends identified the appellant as the individual who fired the gun. A number of other anonymous witnesses gave evidence which added little, if anything, to the case against him. There was some confusion about the identity of the man who supplied Davis with the gun, and at a later trial, the man alleged by the Crown to have done so, Martin Harvey, was acquitted. It is perhaps worthy of notice that this acquittal followed a trial in which, like the trial with which we are concerned, significant evidence for the prosecution against Martin was given by anonymous witnesses. Of itself this perhaps underlines that the fairness of the trial process, and the ability of the jury to subject anonymous evidence to careful thought and judgment is not destroyed by an order that significant evidence against the defendant may be given by anonymous witnesses.

We need not narrate all the evidence given by the anonymous witnesses. Three of them, known as Richard Heath, Steven Bennett and Cecil Blackburn, gave direct evidence that they actually saw the appellant shoot Kenton. Of the remaining anonymous witnesses one, Susan Jones, was not initially called by the Crown. Huge pressure was exercised by the defence to have her called and much complaint directed at the initial failure or inability of the Crown to do so. Indeed it was argued that the trial could not properly continue if she was not called, and that the jury should be discharged and a retrial ordered. The other anonymous witnesses added nothing of significance to the outcome of the trial.

This forensic position was based on an essential feature of the defence case. In essence it was contended that the appellant was the victim of a corrupt plot to implicate him in these killings when he was, and was known by those who implicated him, to be an innocent man. This plot was said to be organised by a woman called Anthea Smith, a former girlfriend, who was seeking revenge for the break-up of the relationship. In earlier incidents she had thrown a bottle at the applicant’s then new girlfriend and sprayed CS gas into his car. In 2001 he had had a one night stand with her, at a time when she had become involved with Richard Heath, and this lead Heath to send a man named Delroy round to his partner’s address to threaten him with a gun.

At the time when it was thought that the witness Susan Jones could not be found, the woman allegedly responsible for the plot was positively identified to the jury as Anthea Smith. However when Susan Jones was eventually called, she declined to answer questions which might revel her identity, although the defence maintained, and the jury was asked to consider his case as if she was the witness with a personal grudge against the appellant, an allegation which Susan Jones strongly denied. In the result, notwithstanding the allegation against her, this witness did not give any incriminating evidence against the appellant.

Susan Jones said that she was present at the premises when she heard the gun shot, but she did not see who was responsible for it. That left, if the defence case was correct, a bizarre situation. The individual said to be responsible for the plot to convict the appellant gave no incriminating evidence against him, while the individuals who did incriminate him were very good friends of the deceased, who were undoubtedly present at the scene, and who, with the possible exception of Heath, had no reasons of their own to protect the man actually responsible for his death by implicating a man who was not.

Heath attended the party with his girlfriend, Susan Jones. He denied any personal grievance against the appellant, or, if the man Delroy had ever threatened the appellant, that he was responsible. He denied that he had been told what to say by Susan Jones or that she had arranged for him to be a witness. He also denied that Jones had visited him when he was serving a sentence of imprisonment abroad. Complaint was made about the fact that although the defence was given information about the circumstances of this particular conviction, it was not aware that the crime had been committed abroad, and that the subsequent sentence was served there. In the context which we are considering, the complaint is trivial.

Steven Bennett was asked if he knew Anthea Smith, and said that he was not sure. He did have a friend named Anthea, but did not know if she was at the party. He did not say anything about the appellant being a boyfriend of Anthea. He denied that he was trying to help the family of his friend, Mowatt, identify someone who they, Mowatt’s family, believed was responsible. The defence case was that this witness’s true name was Hughie Duffers. If he was, the defence sought to make something of the fact that either Heath, or Bennett, or both, were giving untruthful evidence about an incident in which the death of a drug dealer identified as “Blue”, who was shot in 1997 in a drug-related incident, and a link between Anthea and that man. Again, in the overall context of the evidence, this point does not bear the weight suggested by Mr Swift.

We must now turn to further significant features of the evidence, beginning with the appellant’s departure from this country shortly after the killing, to Jamaica, and then onwards on to the United States of America, using a false passport in the name of Baldock, stolen in January 2002, and showing a picture of him wearing a beard. In due course he was extradited from the United States. Before his extradition he explained that the reason he had fled England was that he was in fear of his neighbours. On his return, when interviewed, he made no comment. We shall shortly return to the defendant’s evidence at trial about the reasons for his departure.

In due course the defence case statement was served. It is an important document, highly material to the decision which the trial judge had to make when considering the possible anonymity of the Crown’s witnesses. The relevant part reads that the defendant “was present during the party at Flat 19, Shore Road, however he at no stage had a gun and was not responsible for the shooting. He believes that the witnesses who purport to identify him are not only wrong but are likely to have a motive to implicate him. The defendant will provide details in the skeleton argument (due 7/4/04) as to the anonymity of the witnesses”. The statement is silent on the critical issue of alibi which, again, emerged in the appellant’s evidence at trial. Equally surprisingly, it was not until after the trial began that the details of the alleged improper motivation were seriously ventilated. There was no such suggestion in the course of the appellant’s interview with the police (where he confined himself to “no comment”) nor in his defence case statement, nor indeed in the defence skeleton argument on the anonymity issue. On the first day of the trial the Crown revealed the difficulties it was having in producing Susan Jones at court. It was only then that her apparent centrality to the defence case, and her alleged motive for orchestrating a conspiracy to pervert the course of justice against him emerged.

We were expressly told by counsel for the Crown that the prosecution were not, and are not, in possession of any information which suggests that the family of either deceased were seeking to “orchestrate” witnesses in the sense of persuading them to come to court to give false evidence. We were also told that the police investigation team have never been, and at the start of the appeal were not in possession of any information which suggested that any prosecution witness ever had a motive falsely to implicate either Davis, or indeed Harvey, in the murder. All the relevant information available to the police, save that relating to the identity of the witnesses, on any links between them, was disclosed to the defence. Finally, it is convenient to record here that in answer to Mr Swift’s request for information, we were told that neither before nor during the trial were material rewards sought by the anonymous witnesses, nor offered to any of them by anyone connected to the prosecution.

Before trial the defence were provided with what are described as “disclosure packages” in relation to each anonymous witness which contained details of all their previous convictions, as well as any links the police were able to discover between each of them and any other prosecution witness, as revealed by a trawl through the HOLMES computer system. In reality there was vast disclosure of relevant material including, for example, a list of all the party goers who were known to the police, the previous convictions of each prosecution witness, and the identity of all the information available to the police of each and every individual said to be “implicated” in this shooting. One was the appellant, and the other Harvey, later, as we have seen, acquitted. It is however fair to record that some of the material relevant to anonymity was put before the judge (and the defence) in a somewhat piecemeal way, and Mr Swift suggested that, given the importance to the disclosure process in every trial, but with particular emphasis where an order for witness anonymity was made, this fact, of itself, undermined the safety of the conviction.

When in due course the defendant gave evidence he denied responsibility for the shooting. He said that he had left the party at about 8.00-8.30am, after saying goodbye to a girl called Cyrona. None of that, and indeed nothing in the defence case statement directly suggested that he was absent from the premises at the time of the shooting. The reason he fled to the United States was quite unconnected with this killing. He was avoiding outstanding proceedings against him for assault, for which he believed a prison sentence would be imposed. These accounts, not previously suggested, were of course was left to the jury to evaluate. It was equally open to the jury reject them on the basis that they were virtually newly minted to bolster an untrue defence.

We must now embark on a more detailed analysis of the disclosure process criticised by Mr Swift. With the encouragement of both sides, we invited the Attorney General to appoint a Special Advocate to assist us with the issue of disclosure in the context of public interest immunity matters raised with Judge Paget in the absence of the defence and which, in the assessment of the Special Advocate, may give rise to arguable grounds of appeal. The appellant submitted that the appointment of such an advocate was a particularly important step, since special counsel did not play any part in the public interest immunity (“PII”) issues during the trial. In the event, Mr Penny was instructed for this purpose, and we are grateful to him for his industry and assistance.

Mr Penny was provided with a considerable body of documentation (some 21 volumes of documents), that included:

The appellant also submitted we should give careful attention to whether any matters have arisen since the trial which affect the validity of Judge Paget’s anonymity ruling, and generally we were asked to scrutinise whether the judge was provided with all of the documentation necessary for that purpose.

Special counsel, in addition to considering the documentation set out above, was asked to focus on particular issues which were helpfully set out by Mr Swift. In order to demonstrate the care with which this disclosure process has been reviewed in the course of this appeal by special counsel, and, with his assistance, the court, it is instructive to set out in full the detailed issues Mr Swift particularly identified in his document of 3 February 2006:

The list of issues set out above was of considerable help to Mr Penny and the court in determining whether there were material flaws in the disclosure process resulting in prejudice to the appellant.

Mr Swift submitted that as the PII argument developed in the court below, it became clear that the judge had not necessarily seen everything (either at all or until after he delivered some or all of his rulings on anonymity) that in the appellant’s submission he should, including:

Particularly bearing in mind that the judge apparently was without some or all of this material, it was submitted by the appellant the judge should not have embarked on this exercise of ruling on the anonymity issues at the time when he did, and that having been selected to preside over this trial at a relatively late stage, he should have taken further steps to ensure that he was fully conversant with all of the potentially relevant material before he made any decisions on this important issue. As an ingredient of that submission, Mr Swift argued that the judge should have been equipped with more comprehensive information to ensure he was able to determine whether the defence “guesses” as to the real identity of the anonymous witnesses were correct.

Given those concerns, Mr Penny reviewed all of the material that was before the judge, both before and after anonymity was granted. Whilst it would be inappropriate, for self-evident reasons, to set out the submissions made to us by Mr Penny, it is clear that although the judge did not have, for instance, the complete lists of convictions of the relevant witnesses when he gave his ruling, by the time the prosecution called each of them (when, we observe, the judge was well-placed to review his decision as regards the proposed anonymity) he had been provided with all the relevant criminal antecedents. Moreover, save for Susan Jones, the appellant’s lawyers had been furnished with the lists of convictions when this issue was originally considered: her criminal record was disclosed when, in circumstances already outlined, she was called to give evidence on 18 May. As regards the other specific points, including the editing of some of the witness statements and the suggested lack of research into criminal intelligence, following the thorough review conducted by Mr Penny, nothing has been revealed to cause us concern. Whilst the judge only knew the true identity of two of the anonymous witnesses, he was provided with all the information relevant for the purposes of making his decision. Moreover, in our judgment, given the fear expressed by the witnesses and unsurprisingly accepted by the judge, whether or not the defence accurately guessed their identity was essentially irrelevant to his final decision. Additionally, in the course of the trial the judge was fully informed of the editing of the witness statements. We have reviewed this material too, and, given the judge’s decision on anonymity, the editing was appropriate.

It is also clear that the prosecution and the judge were responsive to the changing nature of the case. By way of example, when Mr Swift (during the submissions in the court below on this issue) revealed for the first time the alleged significance and role of Susan Jones – that she was attempting (so it was said) amongst other things, to procure false testimony – disclosure was reviewed so as to ensure that this additional dimension to the case was taken into account.

We accept that some material relevant to anonymity was disclosed in a piecemeal way. Therefore, having reviewed the relevant documentation, with Mr Penny’s assistance, we have carefully revisited each of the judge’s decisions. We have also borne in mind Mr Swift’s submission that he was, on occasion, “wrong-footed” because, statements were edited without this fact being revealed until late on (for example, the statement of Bennett), the relevant information was provided at essentially the last moment (for example, the convictions of Susan Jones) and material was supplied to the defence, which resulted – at least on one occasion – in cross examination proceeding on a false basis (for example the circumstances of one of Heath’s previous convictions). Such occurrences are unfortunate and should not happen.

The appellant is entitled to a fair trial, and simultaneously, the arrangements for the trial must give appropriate effect to the fears expressed by the witnesses (which the judge unsurprisingly accepted as genuine) about the dangers if their true identities were revealed. There is sometimes a tension between these competing considerations which is unlikely to lead to an entirely wrinkle free process. As we have commented elsewhere, a further complication was that key aspects of the defence case, that the defendant was not present at the premises at the moment of the shooting, and that Susan Jones was manipulating witnesses to give false evidence, were only revealed at the last moment. However, by the time each anonymous witness was called the judge was fully aware of each of the factors that properly had a bearing on his decision as to the concealment of identity and that in the result the judge’s (ultimate) decision as regards each of them was reached by him on the basis of a full understanding of all of the relevant facts. As the judge made abundantly clear, this was not a once-and-for-all decision on anonymity, but one which he kept under review throughout – indeed as he said on 27 April 2004:

“I am in no doubt that I need to keep my decision under review and that I must be alive to the fact that different considerations may apply to different witnesses, so that I have to look not at the nine (witnesses) as one unit but at each of the nine individually.”

In the end what matters is not whether there was an entirely wrinkle free process, but whether the trial was fair. We are confident that no error occurred during the PII proceedings which caused any significant disadvantage to the conduct of the appellant’s defence. In particular the problems which have been identified did not have and would not have had any impact on the assessment by the jury of the credibility, or otherwise, of the anonymous witnesses and ultimately, the verdict. Accordingly none of the complaints identified in argument put the fairness of the process or the safety of the conviction in doubt.

Having examined all this material in detail, notwithstanding that Mr Swift was able to point out that the anonymity order had created specific disadvantages for the defence, our conclusion is that the decision to order anonymity was amply justified. We have examined the safety of the convictions in the light of the main defence, the alleged spiteful motivation of Jones, and the rather more nebulous suggestion that witnesses may have been anxious to support the Mowatt family’s belief that the appellant was indeed responsible for Mowatt’s death, in order to see whether the anonymity ruling precluded proper cross-examination of the anonymous witnesses. Without for one moment suggesting that counsel was not faced with unwanted difficulties, and accepting that there were one or two matters which might have been investigated more closely, our conclusion is that the anonymity ruling did not prevent proper investigation with the witnesses, and before the jury, of the essential elements of the defence case. . There was significant independent evidence supportive of the evidence of the anonymous witnesses, and the jury chose to accept their evidence after cross-examination, and to reject that of the defendant. There was ample evidence to sustain the conviction. We can discern nothing in the trial process overall to lead us to the conclusion that this trial may properly be stigmatised as an unfair trial, or that the convictions were unsafe. Subject to any fresh evidence, the appeal will be dismissed.

In support of a discrete ground of appeal Mr Swift applied to call Tony Pink, who was, he suggested a witness to certain important events at about the time of the fatal shooting. Mr Swift argued that his evidence importantly undermines that given at the appellant’s trial by Richard Heath (whom the appellant suggests was also known as “Crusoe”), in that if Crusoe and Heath are one and the same person, then Tony Pink contradicts Heath’s account that he saw the appellant fire the fatal shot.

Pink was called by the defence during the trial of Martin Harvey on Friday 15 July 2005. “Tony Pink” is not his real name, and although the prosecution have been aware of his true identity since before his evidence last year, he has been allowed to use this pseudonym throughout these criminal proceedings in order to preserve his anonymity from the public as well as from the appellant. Mr Swift submitted the appellant only came to know of the availability of Pink as a witness after he was called during the Harvey trial. In those circumstances, Mr Worsley did not oppose the application to call Pink: the provisions of s. 23 Criminal Appeal Act 1968 (relating to fresh evidence) were satisfied and, accordingly, we granted permission for Pink to give evidence. Additionally we agreed to arrangements (a screen and voice modulation equipment) to ensure only the members of the court could see the witness and hear his true voice.

Pink made three written statements, dated respectively 6 June 2005, 6 July 2005 and 20 March 2006 which were produced and considered during the evidence before us. He refused to give a statement to the police when asked shortly after the incident.

In essence his account was that he arrived at the party with an unnamed friend at some stage after midnight and thereafter he divided his time between flats 18 and 19. He drank a couple of glasses of champagne and throughout the entirety of New Year’s Eve and the early hours of the following morning he consumed four or five “joints” of marijuana, half a tablet of the drug ecstasy and “a few lines” of cocaine. He said these stimulants made him feel “nice” but they did not affect his ability to recall events.

Whilst in flat 18 – the home of two people called Nilson and Drew – he met “Crusoe”, whom he knew from when they were both serving prison sentences. He said whilst they were together in that flat he heard the sounds of a fight on the landing outside; the front door to the flat “caved in”; and he went to see what was happening. He put his head outside the front door, encouraged those outside to calm down and then returned to the living room of flat 18. Not long afterwards he heard the shot. At this time there were 5 or 6 people in flat 18: the witness, Crusoe, two women and one or both of Nilson and Drew. Having “digested” what had happened, everyone then went towards the front door (which was closed, although Pink could not say whether the lock was engaged, on the latch or broken) and he was met by what he described as pandemonium. One of the victims was lying immediately outside the door and the other injured man was round the corner in flat 19. Pink said he then did all he could to “save the life” of the man on the landing whilst someone else, a nurse (who was a “black fella”), helped the man lying in flat 19. His evidence was that it took what seemed like an hour (or more) for the police and the ambulance crew to arrive and apart from ringing the emergency services, he was otherwise engaged in helping the injured man on the landing: this was for about ¾ of the time it took for the emergency services to arrive. The nurse told Pink to hold the man in a particular position because friends and family were trying to make him sit up and to hug him, both of which had a suffocating tendency (although at times in his evidence Pink also said he did not hold the man or he was not doing so “exactly”). Furthermore, importantly on Pink’s account, he said he put his hand over the area where the man had been shot and from where he was bleeding, namely the side of his head, or somewhere on his head. He also saw a white man trying to help but he was not sure what he did or for how long.

Pink stayed until the police arrived and he then went to hospital.

Pink is a man with significant previous criminal convictions. He agreed that he has given evidence in his own defence on five or more occasions in the past when he has been on trial, and that each time the jury has not believed him.

In rebuttal of this evidence, the prosecution applied (unopposed by Mr Swift) to call Martin Nilson, one of the men who lived at number 18 at the time and who gave evidence at both the appellant’s and Harvey’s trials. His account was significantly at variance with the testimony of Pink. He told us that he shared flat 18 with Mark Drew and they both attended the party; indeed, during the course of the night they invited some of the partygoers to their flat for drinks. At the time the door burst open, shortly before the shooting, Mark Drew together with two women (and possibly Errol, a black man who had “Rasta hair” and/or Dane, a well-built black man with short hair) were in flat 18. Although someone else might have been in the bathroom, he had no memory of either a man called Crusoe or a further person who could have been Pink being present inside number 18 at this stage.

Furthermore, he said that as soon as he saw someone lying in the hallway, he stepped out in order to help and he turned the injured man (Ashley Kenton) over onto his back. He had served in the Swedish Marines as a mobile paramedic and, assisted by a “frantic” woman, he tried to clear Kenton’s airways using both a towel and his hands. However, he did not seem to be alive and he could see what he described as pieces of his body lying in the doorway. He undertook mouth-to-mouth resuscitation but to no effect. Thereafter he was asked to assist with the other injured man (Wayne Mowatt) and he gave those tending him some advice. He then went back to Kenton (who had remained in his view throughout) and he continued to give him first-aid. He said that at no stage did another man give any assistance to Mr Mowatt.

Nilson made a statement to the police on the night of these events which was essentially wholly consistent with the evidence he gave during this appeal. Although he was possibly in error on one or two matters of detail (such as whether Drew was in the flat when the shooting occurred), the certain and unanimous view of this court is that his account was wholly honest as well as being reliable on all important issues. Apart from the happenchance that he lived next door to the where the party was taking place, he was uninvolved with the people present that night, save for Mark Drew; he returned to this country from Sweden, at short notice and at great personal inconvenience, to provide his account for a third time and without any apparent advantage to himself; and he gave evidence in a markedly convincing and compelling manner.

As a qualified paramedic Nilson was the likely person to give first-aid to the injured man outside his flat, and on his account all of what Pink said about how he purportedly tended Ashley Kenton, doing all he was able to save his life, could not have been true. Moreover, Pink critically misdescribed the injury over which he said he placed his hand: Ashley Kenton had been shot in the neck and not in the head, and, moreover, if Pink had really assisted him in the way he described he could not have failed to notice he was already apparently dead and that part of his body had been shot away. Therefore, Pink’s description was not only contradicted by Nilson, but it was wholly at variance with the pathological findings. Although not determinative, we cannot disregard Pink’s previous convictions, as well as the occasions in the past when he has been disbelieved when he has given evidence on oath, to reinforce our view that his evidence is not creditworthy. In these circumstances it has no impact on the safety of these convictions.

This appeal is accordingly dismissed.

Yohanne Martin was a member of the Burger Bar Crew. He was shot on 8 December 2002, when he was sitting in a hired Mercedes car which was parked in West Bromwich. Another car drove slowly past and nine shots were fired into the Mercedes, killing Yohanne Martin. Members of his gang were convinced that those responsible for his death were members of or associated with the Johnson Crew.

The Mercedes had been hired by Yohanne Martin and his brother, the appellant Martin. They were both linked with linked with a black Vauxhall Vectra car which Martin was using on 31 December 2002. That car, containing a driver and a passenger, was driven via the Aston Expressway and M6 motorway to Roade in Northamptonshire, where a red Mondeo car, with a high performance engine and factory tinted windows, was purchased. The purchasers, using a mobile phone identified as 819 telephoned the seller before their arrival. The purchaser gave a false name and address. The red Mondeo and the Vectra returned to Birmingham. The factory tinted windows were given an additional tinting on the Mondeo’s arrival in Birmingham by Ramesh Jhalli, an employee of a firm known as Quicksilver. During this return journey mobile telephones were used, number 588 and number 619, one from each vehicle. The Crown’s case was that the two telephones were being used by Martin and Gregory. The evidence associating them with these numbers was clear. Although Gregory did not give evidence at trial, Martin accepted that he had participated in the purchase of the Mondeo for a friend.

On the night of 1/2 January trouble was plainly brewing. In a nightclub in Solihull what was described as “chat” that is, public assertion extolling the power and strength of the Johnson crew was marked. In the result, however, the jury acquitted Jermaine Carty, who was said to be one of those most implicated in the “chat”, of firearms offences arising from the developing trouble at the nightclub. We need not examine this evidence further.

The story of the shooting can be taken from the accounts of three victims. Sophie Ellis was present at the nightclub. It was announced that there would be an “after party” at Uniseven. She went there with others. Inside Uniseven it became very hot. She saw a car drive past slowly, which she thought was a burgundy coloured car and for a split second she saw two guns, which looked the same, and were pointed out of the window of the car. She was shot and fell to the ground.

Cheryl Shaw spent most of her time at the Uniseven itself. She went out twice, and on the second occasion she was standing with others, when she noticed a red Mondeo with a gun pointing outside the passenger’s side window and the person inside wearing a balaclava. She heard two different kinds of shooting, machine gun fire, and single shots.

Leon Harris was sitting in a parked Orion car when he heard shots which sounded like machine gun fire. He then saw a car driving towards his car. Someone got out of the driver’s side of a Vectra and fired a number of shots at the Orion. The gunman was wearing a black balaclava, with slits for his eyes and his mouth, together with a hood. Leon Harris escaped from his car and ran away.

These were very stark, unembellished descriptions of the incident. Bullets and cartridge cases were recovered from the scene of the crime. They were fired from three different weapons, one a smooth bore sub-machine gun, and the other two pistols. The pattern of thirteen bullet marks from the machine gun on an adjacent wall, from waist to head height, suggested that the weapon was deliberately fired in a controlled manner. There were no signs of return fire. The Orion car driven by Harris had been struck by six bullets, fired from two different weapons, and it had a deflated tyre which was consistent with a seventh shot. 23 x 9 MM cartridges were recovered. They may have been fired from two separate self-loading weapons. A Llama pistol was recovered in August 2003 from an address in Ladywood. Some of the shots had been fired from that pistol.

Timed CCTV recordings from the scene were subjected to analysis and imagery examination. It was concluded from this evidence alone, that one of the convoy of vehicles at the scene was highly likely to be a red Mondeo, and the other was possibly a Vectra. This, of course, was consistent with eye witness evidence. Flashes of light from the vehicles were attributable either to the firing of weapons, or the opening of a car door. (shortly before 2pm on 31 December, the 588 phone called a mobile number used by Ramesh Jhalli)

A good deal of material relating to telephone evidence and telephone cell sites was available. This was important evidence linking all four appellants. We shall summarise it as briefly as possible. Phone 819 was said to have been used by someone involved in the purchase of the Mondeo, and then again by someone in the Mondeo just before and after the shooting. Indeed there was no dispute that various calls by or to mobile phones 588,819 and 836 were made throughout the period of the journey by the black Vectra to Roade, and on the return journey in both the Vectra and the Mondeo. 819 was active from 30^th December 2002 until shortly before 3pm on 2^nd January 2003. It called phone 801 and 129. The subscriber to 809 was Gregory’s mother, and 129 was Gregory’s sister. 129 was also called by 836 some twelve times. 588 called the same number about twenty times, and it was also included in the billing records for the landline of Martin’s girlfriend. Shortly before 2pm of 31 December, the 588 phone called a mobile number used by Ramesh Jhalli. 819 called 573, Gregory’s girlfriend, four times, and 573 called 819 twice. At about 6.53 on 2^nd January, 819 called 573, and 573 called it back at about 1.15 pm. 573 was also stored in the memory of a phone at 5-18 Ryland Street, the home of the Gregory family. 819 called each of the brothers of Gregory’s girlfriend, his aunt and his uncle once.

A taxi firm regularly collected a fare from Ryland Street, and took the individual to an address where Gregory’s girlfriend lived. There was contact by phone 819 on 31^st December. There was additional evidence linking 819 to Michael Gregory, but the details need no recitation.

Martin was arrested on 24^th February 2003. The sim card for phone 836 was found loose inside a phone recovered from a compartment to the right hand side of the steering wheel. Call analysis related 836 to Martin’s home address. It was also linked to him by telephone calls made to his girlfriend. Ten calls were made from this number to the landline at Gregory’s address. At Martin’s address, bearing his fingerprints, a box for a phone with the number ending 227 was found. The subscriber’s name was Watkinson, which was the name of the business partner of the trader who sold the Mondeo. On 28^th December 836 called 227. Watkinson denied that 227 was a number associated with him, but conceded that it was possible that he was the subscriber when the phone was first registered in 1998. In June 2002, he was still listed as the subscriber. The number stored in the memory of 227 also linked Martin with 836.

588 was active from 30^th December to 4^th January. In that short period, among other calls, it called 819 on fifty seven occasions. The evidence suggested that both 836 and 588 were being used together, throughout 2^nd January 2003. It was accepted by Martin that 836 and 588 were indeed his telephones.

The three best serving signals for Gregory’s address were identified. Between 30^th December and 2^nd January 35% of the calls were made to or from 819 were served by those three cell sites between them, with an indication that the handset was used in the vicinity of Gregory’s address at Ryland Street. Thirty six calls went through a site which served the address where Martin lived, as well as a second address in Smethwick, also associated with Martin. Over a period of roughly 24 hours between 31^st December and 1^st January. The 819 phone made thirty four calls routed through the cell site best serving Gregory’s address, and during a similar period, both the 836 and 588 phones made or received calls routed through the same sites. The Crown drew attention to the fact that there was a time when 819 did not call either 836 or 588, suggesting that the owners of these telephones were or must have been together.

From 1.30am on 2^nd January inward calls to 836 were on divert until 10.40 am that day. There was evidence linking Ellis with the 633 phone number, and Simms with the number 845. At the material time calls were made on the 633 phone consistent with the use of the site closest to Uniseven in Churchill Parade. Between 12.54 and 2.55am on 2^nd January phone 845 made calls which placed the handset in the vicinity of Uniseven, consistently using the same cell site. From 3.07am until about 6.30am on the same date, phone 819 made forty three calls which were handled by eight different cell sites, which indicated that this phone was on the move. At about 3.52 819 called 845, using the cell site close to Ellis’ address for fourteen seconds. Simms phone used a cell site which was dominant for Uniseven, and in a sequence of calls from 819 to 845 (but none the other way) 819 moved in such a way that by 4.01am it was being served by the same cell site as Simms phone in the vicinity of Uniseven, indeed there was positive expert evidence that the calls from Simms number were consistent with him being inside Uniseven when the calls were made. At 4.09am there was a five second call to Ellis’s phone at 633, which indicated that Simms telephone had moved into a cell site area to the west of Uniseven.

After the shootings a number of calls were made or received by 819 and 633, which were routed via the site which served Ellis’ address. The phone belonging to Simms received a call from 819 at 4.22 and he then received calls which routed through cell sites, one of which served Gregory’s address. During the period when the red Mondeo was set alight, 819 called 633. The cell site which served the Maryhill area, where this car was burned, handled a call between the 633 and 819 phones about 45 minutes before the fire started, and if 633 was indeed at Ellis’ home address at the time, it must, for whatever reason, been incapable of using one of the four other cell sites which would have given stronger signals. The data available to the expert was consistent with the 819 phone being in the vicinity of the site of the burning, and the last call using the corresponding cell site took place at 6.08am. That was inconsistent with 819 being at Gregory’s flat. The expert also accepted that it was not possible to discount the possibility that Ellis had been using his phone at home (because of the proximity of the cell site).

This evidence about the use of mobile phones demonstrably belonging to the four appellants at critical times all through the night leading up to and after the shooting, taken as a whole, and when linked with the remaining evidence, provided a formidable case against them.

On 13 January 2003, the home of Ellis was searched. In the garden, inside plastic bags, some clothing, including a hooded top and jeans were found. The top was closely linked by DNA evidence to Ellis. There was insufficient material on the jeans for them to be directly associated with any individual.

The spent ammunition and this clothing were examined together. The prima residue in the spent ammunition contained lead, barium and antimony. The hooded top found in Ellis’ garden bore two particles of firearm discharge residue, one from the front and left sleeve containing primer residue of lead, barium and antimony, with a low level silicone, and the second, from the right front and sleeve, also containing lead, barium, antimony, as well as tin with a low level silicone. The leg area of the jeans yielded one particle of prima residue of lead, barium and antimony with a low level of silicone. There was a trace of nitro-glycerine from the right of the hooded top. This evidence did not prove directly that the residues found on the clothes must have come from the shooting outside Uniseven on 2^nd January. It was however consistent with it, and it was, of course, open to the jury to consider whether any inferences should properly be drawn from the additional fact that the hooded top and jeans were not found where they would normally be, in their owner’s house rather than the garden.

After arrest the four appellants were interviewed. Martin denied any involvement in the offences, and offered an alibi. He told untruths about his link with the Vectra which had travelled to Northampton, and the purchase of the red Mondeo. In general terms, Gregory declined to answer any questions, stating that he did not want to remember where he had been at the New Year because of the death of his brother. Ellis denied involvement, and at first exercised his right of silence. When he was later interviewed, he denied any involvement with the red Mondeo, and at a subsequent interview, claimed that gunshot residue found on his clothing had come from fireworks or through other innocent contact. As to the telephone calls, he claimed that it was impossible for him so long after the events to recollect any specific details. However he did not believe that his phone had been in anyone else’s possession. In his interviews, Simms effectively made no comment, but on the last occasion when he was interviewed he denied any involvement in the shooting or in the obtaining of the red Mondeo. He admitted that he was present at Uniseven at the material time, along with several members of his family, who he would not have endangered. He agreed that the phone 845 was his telephone, and asserted that he had received telephone calls while at Uniseven from different people asking him about the quality of the music, but as the music was loud there were several missed calls. In view of the lapse of time, he could not remember any other details. In due course alibi notices were served by Martin, Gregory and Ellis.

Goldring J gave leave for four witnesses to give their evidence anonymously. As we have already recorded, his decision was upheld in this court. At that date all the information, with the exception of one prison file relating to one of the witnesses, Brown was available. By the time Brown did give evidence, the prison file had been disclosed in full. Criticisms of the disclosure process were repeatedly and fully argued before the judge. In our view, they were makeweight, and we note that it is conceded by one of the appellants, at any rate, that there was a just and appropriate level of disclosure.

The judge kept his ruling under constant review. There were no less than thirty eight PII hearings during the trial itself. On 10^th January 2005, he once again closely reviewed his decision. In the result, of the four witnesses who gave evidence anonymously, only one gave any evidence which purported to incriminate any appellant. Jarrett described the scene, without identifying anyone in the Mondeo, or saying anything which might have led to the identification of anyone in the car. Similarly, with Beattie, who indeed had to be treated as a hostile witness by the Crown. James declined to give any evidence along the lines of his witness statements. The judge observed to the jury, when he came to sum the case up, that his evidence was worthless, and whatever evidence he did give, the judge decided that it was inappropriate to remind the jury of it.

That left Brown who like the other anonymous witnesses, was seen and observed by counsel, but not the appellants or the public. Unlike counsel, they heard a modulated voice. He purported to identify some of those in the red Mondeo, but his evidence was not entirely consistent on the subject, and in any event, faced the major difficulty, that other witnesses had suggested identification was not practicable.

By the time Brown gave evidence, there had been full disclosure of all the material available to the Crown. He was cross-examined for five days, by four leading counsel. His credibility was severely damaged.

When considering submissions that there was no case to answer, where there was no other evidence other than that of Brown, as in the case of Tafarwa Beckford, the judge directed an acquittal. He observed that it was plain that Brown had lied about many matters, and that his credibility was severely dented. In particular, the judge was concerned about Brown’s assertion that the faces of the occupants of the Mondeo were visible because that assertion was contradicted by every other witness to the shooting, if Brown was indeed in the position that he said he was at the time. Despite the various complaints carefully marshalled by the different counsel on behalf of the appellants, the cross-examination of Brown illustrates that notwithstanding his anonymity, the process could be and indeed was extremely effective. In fact, Brown gave no evidence implicating Gregory, and so far as Simms was concerned, his evidence provided no more than a little confirmation of facts already independently demonstrated.

Following the judge’s ruling, neither Ellis nor Gregory gave evidence. Gregory called a character witness. Ellis called one alibi witness. She could not remember whether he had received any phone calls at all, but she would have remembered if the phone had rung constantly. Martin gave evidence. He admitted his involvement in the purchase of the Mondeo. He admitted the correctness of the attribution of telephone numbers to him. He offered an alibi defence. Simms admitted his presence at Uniseven. He was not involved in the crime. He remembered receiving the 819 calls. He was sure that it was a “withheld” number, by a man who refused to give his name, and simply wanted to know who was at the party. After the shooting, there was another call, and he was instructed to say nothing about the calls he had received. He had not mentioned this when interviewed because he was terrified of the man who had rung him, who was plainly a murderer who knew his address.

The case was summed up to the jury with meticulous care. The directions of law were accurate. Appropriate warnings were given. The evidence was closely analysed. The summing up was comprehensive, balanced and fair.

We have reconsidered the safety of these convictions in the light of events subsequent to the earlier decision of this court upholding Goldring J’s decision on witness anonymity. We cannot detect any reason for doubting the integrity of the trial process, or the safety of these convictions. These appeals are dismissed.