AL- KHAWAJA AND TAHERY v. THE UNITED KINGDOM - 26766/05 [2011] ECHR 2127 (15 December 2011)

JUDGMENT STRASBOURG 15 December 2011 This judgment is final but may be subject to editorial revision. In the case of Al-Khawaja and Tahery v. the United Kingdom, The European Court of Human Rights, sitting as a Grand Chamber composed of: Françoise Tulkens, President, Nicolas Bratza, Jean-Paul Costa, Christos Rozakis, Peer Lorenzen, Elisabet Fura, Alvina Gyulumyan, Danutė Jočienė, Dragoljub Popović, Ineta Ziemele, Mark Villiger, Giorgio Malinverni, András Sajó, Mirjana Lazarova Trajkovska, Işıl Karakaş, Nebojša Vučinić, Kristina Pardalos, judges, and Michael O’Boyle , Deputy Registrar , Having deliberated in private on 19 May 2010 and on 9 November 2011, Delivers the following judgment, which was adopted on the last mentioned date: PROCEDURE The cases originated in two applications (nos. 26766/05 and 2228/06) against the United Kingdom of Great Britain and Northern Ireland lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by a British national, Mr Imad Al-Khawaja (“the first applicant”), on 18 July 2005 and by an Iranian national, Mr Ali Tahery (“the second applicant”), on 23 May 2006. The first applicant was represented by Mr A. Burcombe and Mr D. Wells, lawyers practising in London with Wells Burcombe LLP Solicitors, assisted by Mr J. Bennathan QC, counsel. The second applicant was represented by Mr M. Fisher, a lawyer practising in London with Peter Kandler & Co. Solicitors, assisted by Ms R. Trowler, counsel. The United Kingdom Government (“the Government”) were represented by their Agent, Mr J. Grainger of the Foreign and Commonwealth Office. The first applicant alleged that his trial for indecent assault had been unfair because one of the two women who made complaints against him died before the trial and her statement to the police was read to the jury. The second applicant alleged that his trial for wounding with intent to commit grievous bodily harm had been unfair because the statement of one witness, who feared attending trial, was read to the jury. The applications were allocated to the Fourth Section of the Court (Rule 52 § 1 of the Rules of Court). On 20 January 2009, following a hearing on admissibility and the merits (Rule 54 § 3), a Chamber of that Section composed of Judges Casadevall, Bratza, Bonello, Traja, Mijović, Šikuta and Hirvelä, together with T. L. Early, Section Registrar, decided unanimously to join the applications, to declare each application admissible and to find a violation of Article 6 § 1 read in conjunction with Article 6 § 3(d) of the Convention in respect of each applicant. On 1 March 2010, pursuant to a request by the Government dated 16 April 2009, the Panel of the Grand Chamber decided to refer the case to the Grand Chamber in accordance with Article 43 of the Convention. The composition of the Grand Chamber was determined according to the provisions of Article 27 §§ 2 and 3 of the Convention and Rule 24 of the Rules of Court. At the final deliberations, Alvina Gyulumyan, Işıl Karakaş and Nebojša Vučinić, substitute judges, replaced Corneliu Bîrsan, Ireneu Cabral Barreto and Sverre Erik Jebens, who were unable to take part in the further consideration of the case (Rule 24 § 3). Jean-Paul Costa, Christos Rozakis and Giorgio Malinverni, whose terms of office expired in the course of the proceedings, continued to sit in the case (Article 23 § 7 of the Convention and Rule 24 § 4). The applicants and the Government each filed written observations on the merits. In addition, third-party comments were received from the London-based non-governmental organisation JUSTICE, which had been given leave by the President to intervene in the written procedure (Article 36 § 2 of the Convention and Rule 44 § 2). The parties replied to those comments (Rule 44 § 5). A hearing took place in public in the Human Rights Building, Strasbourg, on 19 May 2010 (Rule 59 § 3). There appeared before the Court: (a) for the Government Mr J. Grainger, Agent , Mr D. Perry QC, Mr L. Mably, Ms V. Ailes, Counsel , Mr C. Munro, Mr. N. Gibbs, Advisers ; (b) for the applicants Mr J. Bennathan QC, Counsel , Mr D. Wells, Adviser. The Court heard addresses by Mr Bennathan and Mr Perry and their answers in reply to questions put by the Court. THE FACTS I. THE CIRCUMSTANCES OF THE CASES A. Mr Al-Khawaja The first applicant was born in 1956 and lives in Brighton. The facts which gave rise to his application are as follows. While working as a consultant physician in the field of rehabilitative medicine, the first applicant was charged on two counts of indecent assault on two female patients while they were allegedly under hypnosis. The first count in the indictment alleged he had indecently assaulted a woman called ST during a consultation on 3 June 2003. The second count in the indictment alleged he had indecently assaulted a woman called VU also during a consultation, this time on 12 June 2003. For reasons unrelated to the alleged assault, ST committed suicide before the trial. However, several months after the alleged assault, she had made a statement to the police. She had also told two friends, BF and SH, that the applicant had indecently assaulted her. On 22 March 2004, a preliminary hearing was held to determine whether ST’s statement should be read to the jury. At that hearing, the defence indicated that the defence to each count of the indictment was essentially the same, namely that ST and VU’s allegations were wholly untrue. The judge at the hearing decided that ST’s statement should be read to the jury at trial. He observed that the first applicant was very likely to feel that he had no realistic alternative other than to give evidence in order to defend himself on the second count relating to VU. Therefore, the reading of ST’s statement would not have the effect of making it very difficult for the applicant not to give evidence. The judge also noted that collusion between ST and VU was not alleged, and so that issue did not need to be investigated by cross-examination of each woman. Having regard to the contents of ST’s statement, the judge also observed that it was crucial to the prosecution on count one as there was no other direct evidence of what took place during the consultation on 3 June 2003. He said: “putting it bluntly, no statement, no count one.” He went on to observe that the real issue was whether it was likely to be possible for the first applicant to controvert the statement in a way that achieved fairness to the defendant. The judge found that the first applicant could give evidence as to what happened during the consultation. It was also the intention of the prosecution to call ST’s friends to give evidence as to what she had said to them. There were inconsistencies between their statements and ST’s, which provide a route by which ST’s statement could be challenged. An expert witness would also be called by the prosecution to give evidence on altered perception during hypnosis and cross-examination of that witness might also serve to undermine ST’s credibility. At the trial, once ST’s statement had been read, the jury heard evidence from BF and SH, ST’s friends. ST’s general practitioner also gave evidence as to a letter he had written on ST’s behalf to the hospital authorities, which outlined ST’s allegations against the first applicant. In respect of the second count, the indecent assault upon VU, evidence was given by VU and by the police officers who had investigated the case. Evidence was then given by two women who alleged that the first applicant had made improper suggestions to them during hypnosis consultations. That evidence was relied on by the prosecution as “similar fact evidence” to support the evidence of ST and VU. As the prosecution had indicated, expert evidence was given as to the effects of hypnosis. The defence was given the opportunity to cross-examine all the witnesses who gave live evidence. The first applicant gave evidence in his own defence. He also called a number of witnesses, who gave evidence as to his good character. In his summing up, the trial judge directed the jury, on two separate occasions, as to how they should regard ST’s statement. First, he stated: “ It is very important that you [the jury] bear in mind when considering her [ST’s] evidence that you have not seen her give evidence; you have not heard her give evidence; and you have not heard her evidence tested in cross-examination [by counsel for Mr Al-Khawaja], who would, undoubtedly, have had a number of questions to put to her.” He later stated: “ ... bear in mind...that this evidence was read to you. The allegation is completely denied ... you must take that into account when considering her evidence.” When referring to the evidence of ST’s friends, the trial judge reminded the jury that there was a inconsistency between ST’s account of the consultation and the account given by SH (in her statement ST said the applicant who touched her face and mouth; SH gave evidence that it was ST herself and not the first applicant who had touched her face and mouth). The trial judge continued: “ It is for you to decide the extent to which the evidence of [BF] and [SH] helps you in deciding whether or not [ST] has spoken the truth in her statement. But bear in mind the evidence as to what [ST] said to [BF] and [SH] is not independent evidence as to the truth of her allegations.” The trial judge also instructed the jury that they were entitled to consider the evidence of VU, and of the other two women who had given evidence as to the improper suggestions made by the first applicant, when deciding whether ST’s statement was true. However, the jury first had to discount the possibility of collusion between the four women. Second, they had to ask themselves whether it was reasonable that four people independently making similar accusations could all either be lying or mistaken or have all suffered similar hallucinations or false memory. If the jury though that incredible, then could be satisfied that ST and VU had spoken the truth. The trial judge also directed the jury that the greater the similarity between the allegations, the greater the likelihood that the four women were telling the truth. He added that the jury also had to consider whether the women could have consciously or unconsciously been influenced by hearing of the complaints of the others. In the course of their deliberations, on two occasions the jury asked for clarification of points raised in the statement of ST. On 30 November 2004, the first applicant was convicted by a unanimous verdict of the jury on both counts of indecent assault. He was sentenced to a 15 month custodial sentence on the first count and a 12 month custodial sentence on the second count, to run consecutively. The first applicant appealed against his conviction to the Court of Appeal. The appeal centred on the pre-trial ruling to admit ST’s statement as evidence. It was also submitted that, in his summing up, the trial judge did not give adequate directions to the jury as to the consequential disadvantage of this evidence to the first applicant. The appeal was heard and dismissed on 6 September 2005. In its written judgment handed down on 3 November 2005 the Court of Appeal concluded that the first applicant’s right to a fair trial had not been infringed. With regard to the admission in evidence of the statement of ST, the court held that it was not necessarily incompatible with Article 6 §§ 1 and 3 (d) of the Convention. Relying on Doorson v. the Netherlands , judgment of 26 March 1996, Reports of Judgments and Decisions 1996 II, the court held that admissibility of evidence is primarily a matter of domestic law. It then found: “ 25. The important factors in the present case are the following. The witness, ST, could not be examined on behalf of the [first applicant] because she had died. She was the only witness whose evidence went directly to the commission of an indecent assault on her by the appellant. If her statement had been excluded, the prosecution would have had to abandon the first count. The [applicant] was able to attack the accuracy of [ST’s] statement by exploring the inconsistencies between it and the witnesses, [BF] and [SH], and through the expert evidence relating to ‘altered perception’ under hypnosis. The relevant sections of the 1988 Act [see paragraph 41 below] contained provisions designed to protect defendants, which were properly considered by the judge, before the statement was admitted in evidence. Lastly, the tribunal of fact, here the jury, could and should take proper account of the difficulties which the admission of a statement might provide for the [applicant], which should be provided by an appropriate direction to the jury. 26. Where a witness who is the sole witness of a crime has made a statement to be used in its prosecution and has since died, there may be a strong public interest in the admission of the statement in evidence so that the prosecution may proceed. That was the case here. That public interest must not be allowed to override the requirement that the defendant have a fair trial. Like the court in Sellick [see paragraph 48 below] we do not consider that the case law of the European Court of Human Rights requires the conclusion that in such circumstances the trial will be unfair. The provision in Article 6(3)(d) that a person charged shall be able to [have] the witnesses 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 the evidence was taken, were fair’ – Doorson, paragraph 19. This was not a case where the witness had absented himself, whether through fear or otherwise, or had required anonymity, or had exercised a right to keep silent. The reason was death, which has a finality which brings in considerations of its own, as has been indicated at the start of this paragraph.” Turning to the issue of the trial judge’s summing up, the court stated: “ We consider that it would have been better if the judge had stated explicitly that the [first applicant] was potentially disadvantaged by the absence of [ST] and that in consequence of the inability to cross-examine her and of the jury to see her, her evidence should carry less weight with them. Nonetheless, in the circumstances of this case it must have been wholly clear to the jury from the directions the judge did give, that this was the purpose of his remarks. We therefore consider that the jury were given an adequate direction as to the consequences of [ST’s] statement being in evidence in her absence, and that this is not a factor which might make the [first applicant’s] trial unfair and in breach of Article 6. We should also say that overall the evidence against the [first applicant] was very strong. We were wholly unpersuaded that the verdicts were unsafe.” The Court of Appeal refused leave to appeal to the House of Lords but certified that a point of law of general public importance was involved in the decision. On 30 November 2005 the first applicant petitioned the House of Lords on the point of law certified by the Court of Appeal. On 7 February 2005 the House of Lords refused the petition. B. Mr Tahery The second applicant was born in 1975. His application arises from his conviction for wounding with intent. The background to that conviction is as follows. On 19 May 2004, S, a member of the Iranian community living in London, was involved in an altercation with some Kurdish men. The second applicant interposed himself between S and the Kurdish men in order to protect S. In the small hours of the morning of 20 May 2004, S and the second applicant met again outside an Iranian restaurant in Hammersmith, London. The second applicant asked S to have a word with him and led him into a nearby alleyway. The men began discussing the earlier altercation. Although S denied throwing the first punch, he conceded while giving evidence at the second applicant’s trial that he had punched the second applicant. In the fight, the second applicant pushed S back and, at this stage, S became aware of a burning sensation in his back, which proved to be the result of three stab wounds to his back. S and the second applicant had been face-to-face and S’s account was that he neither saw the second applicant stab him, nor was he aware of the second applicant going behind him or reaching round his back, so as to stab him. During the fight other men were present, including the Kurdish men from the earlier altercation. A friend of S, another member of the Iranian community called T, was there, as were two of T’s friends and the second applicant’s uncle. S could not say which of the men were behind him. S eventually realised he had been stabbed. He then saw a knife lying on the ground. In his evidence at the second applicant’s trial (see paragraph 32 below), he stated that he went to pick it up but that either the second applicant or T had picked it up and thrown it towards the restaurant. S assumed it had been the second applicant who had stabbed him. According to S, the second applicant immediately denied this. He told S to sit down beside him and attempted to staunch the blood flow from S’s wounds until an ambulance arrived; when it did, he accompanied S to the hospital. At the hospital, the second applicant told the police that he had seen two black men stab S. When witnesses were questioned at the scene, no one claimed to have seen the second applicant stab S. Two days later, however, T made a statement to police that he had seen the second applicant stab S. In his statement T recounted that, when the second applicant and S had begun fighting in the alleyway, T had tried to separate them. He then saw the second applicant hold S by the neck, hold up the knife and stab S twice in the back. As T moved towards the second applicant, the second applicant tried to stab T in the neck. According to T, the second applicant then dropped the knife and shouted “don’t tell the police”. On 3 November 2004, the second applicant was arrested and taken to Hammersmith Police Station. In interview, he denied stabbing S and again stated that two black men were responsible. He was charged with wounding with intent and also with attempting to pervert the course of justice for telling the police, at the hospital and at the police station, that he had seen two black men stab S. On 25 April 2005 the second applicant’s trial began at Blackfriars Crown Court. That day, he pleaded guilty in respect of the charge of attempting to pervert the course of justice but maintained his not guilty plea in respect of the charge of wounding with intent. S gave evidence for the prosecution. He recounted how he and the second applicant had fought in the alleyway. After a minute he realised he had been injured in the back. He had not seen who stabbed him. The second applicant had made him sit down and had covered the wound. S had asked the second applicant who had stabbed him and the second applicant had denied that it was him. When cross-examined S accepted that he had not seen the second applicant go behind him and that they had been face to face. He also testified that he had heard someone shout to him “Tell him it was the blacks”; the voice did not belong to the second applicant. After S had given evidence, the prosecution made an application for leave to read T’s statement pursuant to section 116(2)(e) and (4) of the Criminal Justice Act 2003 (the “2003 Act”: see paragraphs 43–45 below). The prosecution argued under the 2003 Act that T was too fearful to attend trial before the jury and that he should qualify for special measures. The trial judge heard evidence from a police officer conducting the case who testified that the Iranian community was close-knit and that T’s fear was genuine. T also gave evidence to the trial judge (but not the jury) from behind a screen. He told the judge that he was in fear for himself and his family because of visits and telephone calls he had received, none of which were said to have been from the second applicant. He did not say who had been responsible for the visits and telephone calls. In ruling that leave should be given for the statement to be read to the jury, the trial judge stated: “ I am satisfied in those circumstances upon the criminal standard of proof that this witness is genuinely in fear; and I base that not only on his oral testimony, but also upon my opportunity while he was in the witness box to observe him. I therefore have to go on to consider the questions posed in [section 116(4) of the 2030 Act]. Subsection 4(a) requires me to look at the statement’s contents. I have done so. It is submitted by the defence that they may be unreliable; there being some inconsistencies with the evidence that was given by [S]. There will always be cases, whether it be oral evidence or evidence that is read, where there are inconsistencies. It is always for the jury to come to a conclusion, based upon submissions of counsel and the evidence that they have heard, as to whether the evidence is reliable or not. And they will receive from me the appropriate warning when the time comes as to how they should view that statement. It is further submitted that in looking at the statement, it is a statement of great importance; in that it is from a person who purports to witness the incident and consequently goes to the heart of the matter. In my view, it is precisely this type of witness who is likely to be put in fear, and consequently that must have been what Parliament had in mind when it enacted this particular section. I therefore have to look, having looked at the contents of the statement, to any risk its admission or exclusion will result in unfairness to any party to the proceedings. I am satisfied that there would be an unfairness caused by its exclusion; but I am equally satisfied that no unfairness would be caused by its admission. And in doing so, I have taken into account the words of [the 2003 Act]; in particular how difficult it will be to challenge the statement if the relevant person does not give oral evidence. Challenge of a statement does not always come from cross-examination. Challenge of a statement can be caused by evidence given in rebuttal; by either the defendant, if he chooses to do so, or by any other bystander – and we know that there were some – who were on the street at that time. Consequently I am satisfied that the defendant’s evidence, if he chooses to give it, would be sufficient to rebut and to challenge the evidence that is contained in that statement. I have further considered other relevant factors, and I have also offered to the witness whilst he was in the witness box behind screens the possibility of him giving evidence with the same special measures in place. He told me his position would not change; that he could not give evidence before a jury, and the reason that he could not was because he was in fear. Having taken all those matters into account in those circumstances, I am satisfied that this is the type of case which Parliament envisaged might require a statement to be read.” T’s witness statement was then read to the jury. Evidence was also given by the doctor who had treated S at hospital as to the nature of the wound and by a forensic scientist who had tested the blood found on the second applicant’s clothes and confirmed it matched that of S (though no firm conclusions could be drawn as to how it had been deposited on the clothes). The record of the second applicant’s interview by the police was also tendered by the prosecution (see paragraph 30 above). The second applicant gave evidence in his defence. He stated that he had been present at the earlier altercation with the Kurdish men. When he and S later met outside the restaurant, he had taken S by the hand and suggested that they go and talk, but S had begun to punch him. He had defended himself by grabbing S by the collar and pushing him. T had then tried to intervene and a number of other members of the Iranian community had restrained S. T had been standing between S and the second applicant and, at this stage, the second applicant noticed the knife on the ground. He had picked it up and thrown it, not knowing at this point that S had been stabbed. When S blamed him for the stabbing, the second applicant had told him to sit down and had successfully calmed him down. He placed his hand on the wound on S’s back. S then appeared to accept that the second applicant had not stabbed him. The second applicant also gave evidence that he had told the police two black men were responsible because this was what his uncle had told him to say. Finally, the second applicant gave evidence that, before he had been interviewed by the police, T had told him that he, T, knew that the second applicant had not stabbed S. The judge, in his summing up, warned the jury about the danger of relying on the evidence of T. He stated: “ That evidence, as you know, was read to you under the provisions that allow a witness who is frightened, it is not a question of nerves it is a question of fright, fear, for his statement to be read to you but you must be careful as to how you treat it. It is right, as has been pointed out by the defence, that they were deprived of an opportunity to test that evidence under cross-examination. It is right also that you did not have the advantage of seeing the witness and his demeanour in Court. You did not have the opportunity for him to think back and say ‘possibly because of things I saw I put two and two together and made five’, as counsel for the defence invites you to say. In other words, you must always be alert to [the fact] that he could put things that he did see together and come to the wrong conclusion. That is a way of examining the statement. You must ask yourselves ‘can we rely upon this statement? Is it a statement which we find convincing?’ It is only, if you are satisfied so that you are sure, that what is in the statement has accurately depicted what happened that night and what the witness saw, that you could rely upon it. That goes for any witness. It is only if you find that the evidence is compelling and satisfies you, so that you are sure, that you act upon it. So you must always ask yourselves ‘is the statement he made reliable?’ You must bear in mind also, importantly, that it is agreed and acknowledged that it is not the defendant who is responsible for putting the witness in fear.” On 29 April 2005, the second applicant was convicted by a majority verdict of wounding with intent to cause grievous bodily harm, for which he was later sentenced to nine years’ imprisonment to be served concurrently with a term of fifteen months’ imprisonment for the charge of attempting to pervert the course of justice to which he had pleaded guilty. The second applicant appealed to the Court of Appeal, arguing that the inability to cross-examine T infringed his right to a fair trial. The Court of Appeal acknowledged that the Crown accepted that T’s statement was “both important and probative of a major issue in the case... had it not been admitted the prospect of a conviction would have receded and that of an acquittal advanced.” The court upheld the reasoning of the trial judge, stating that there was available not only cross-examination of other prosecution witnesses but also evidence from the second applicant himself and the potential for evidence from other bystanders in order to prevent unfairness. It was also stated that the trial judge had explicitly warned the jury in detail as to how they should treat this evidence and properly directed them as to how they should consider it in reaching their verdict. Although the second applicant maintained that even a proper direction by the judge could not cure the unfairness, the Court of Appeal held that the jury was informed of all matters necessary to its decision-making process. Leave to appeal on conviction was refused on 24 January 2006. The Court of Appeal did, however, give the second applicant leave to appeal against his sentence and reduced the sentence of nine years’ imprisonment to seven years’ imprisonment. II. RELEVANT DOMESTIC LAW AND PRACTICE A. Common law principles relevant to both cases Hearsay evidence is any statement of fact other than one made, of his own knowledge, by a witness in the course of oral testimony (see paragraph 20 of judgment of Lord Phillips in R. v. Horncastle and others , summarised at paragraphs 57–62 below). As a general rule it is inadmissible in a criminal case unless there is a common law rule or statutory provision which allows for its admission. The relevant statutory provisions applicable to each applicant are set in the following section. Those statutory provisions are supplemented by three common law principles. First, there is an additional discretion at common law for a trial judge to exclude any evidence if its prejudicial effect outweighs its probative value. This, in turn, is supplemented by section 78 of the Police and Criminal Evidence Act 1984, which provides the court with a discretion to exclude evidence if its admission would have such an adverse effect on the fairness of the trial that it ought not to be admitted. Second, if hearsay evidence is admitted and the jury have heard it, the trial judge, in his summing up, must direct the jury as to the dangers of relying on hearsay evidence. Third, in a jury trial, the jury must receive the traditional direction as to the burden of proof i.e. that they must be satisfied of the defendant’s guilt beyond reasonable doubt. B. Primary legislation 1. Primary legislation applicable in Al-Khawaja’s case At the time of the first applicant’s trial, the relevant statutory provisions were to be found in sections 23 to 28 of the Criminal Justice Act 1988. Section 23 of the 1988 Act provides for the admission of first hand documentary hearsay in a criminal trial: “ 23. — ... a statement made by a person in a document shall be admissible in criminal proceedings as evidence of any fact of which direct oral evidence by him would be admissible if— (2)(a) the person who made the statement is dead or by reason of his bodily or mental condition unfit to attend as a witness ... 25. —(1) If, having regard to all the circumstances— (a) the Crown Court— (i) on a trial on indictment; (ii) on an appeal from a magistrates’ court; or (iii) on the hearing of an application under section 6 of the [1987 c. 38.] Criminal Justice Act 1987 (applications for dismissal of charges of fraud transferred from magistrates’ court to Crown Court); or (b) the criminal division of the Court of Appeal; or (c) a magistrates’ court on a trial of an information, is of the opinion that in the interests of justice a statement which is admissible by virtue of section 23 or 24 above nevertheless ought not to be admitted, it may direct that the statement shall not be admitted. (2) Without prejudice to the generality of subsection (1) above, it shall be the duty of the court to have regard— (a) to the nature and source of the document containing the statement and to whether or not, having regard to its nature and source and to any other circumstances that appear to the court to be relevant, it is likely that the document is authentic; (b) to the extent to which the statement appears to supply evidence which would otherwise not be readily available; (c) to the relevance of the evidence that it appears to supply to any issue which is likely to have to be determined in the proceedings; and (d) to any risk, having regard in particular to whether it is likely to be possible to controvert the statement if the person making it does not attend to give oral evidence in the proceedings, that its admission or exclusion will result in unfairness to the accused or, if there is more than one, to any of them. 26. — Where a statement which is admissible in criminal proceedings by virtue of section 23 or 24 above appears to the court to have been prepared, ..., for the purposes- (a) of pending or contemplated criminal proceedings; or (b) of a criminal investigation, the statement shall not be given in evidence in any criminal proceedings without the leave of the court, and the court shall not give leave unless it is of the opinion that the statement ought to be admitted in the interests of justice; and in considering whether its admission would be in the interests of justice, it shall be the duty of the court to have regard- (i) to the contents of the statement; (ii) to any risk, having regard in particular to whether it is likely to be possible to controvert the statement if the person making it does not attend to give oral evidence in the proceedings, that its admission or exclusion will result in unfairness in the accused or, if there is more that one, to any of them; and (iii) to any other circumstances that appear to the court to be relevant...” Schedule 2 to the Act allows for the admission of evidence relating to the credibility and consistency of the maker of the statement, where such evidence would have been admissible had he or she given evidence in person, or where the matter could have been put to him in cross examination. The Schedule also provides that , in estimating the weight, if any, to be attached to such a statement regard shall be had to all the circumstances from which any inference can reasonably be drawn as to its accuracy or otherwise. 2. Primary legislation applicable in Tahery’s case (a) The Law Commission’s report In its report of April 1997, entitled Evidence in Criminal Proceedings: Hearsay and Related Topics , the Law Commission recommended a series of reforms to the law of hearsay in England and Wales, including the provisions contained in the 1988 Act. In addition to recommending clarification of the conditions under which a witness statement could be admitted at trial (including the existing grounds of death and fear), the Commission proposed that there should be a limited discretion to admit hearsay evidence that did not fall within any other statutory or common law exception (recommendation 28). In its earlier consultation paper, published in January 1995, the Commission had reviewed the case-law of this Court on Article 6 § 3(d) and concluded that there was a risk of a breach of the Convention where a person stood to be convicted on hearsay evidence alone. The Commission considered that this risk was sufficiently serious to warrant requiring the trial court to stop the case where hearsay is the only evidence of an element of the offence (paragraph 9.5 of the consultation paper). After criticisms of this proposal (principally that it was unduly cautious and was beset with practical difficulties), in its 1997 report the Commission decided not to maintain its proposal (see paragraphs 5.33-5.41 of the report). It concluded instead that the adequate protection would be provided by the safeguards it proposed, in particular its recommendation 47, which proposed giving the trial judge the power to stop a case if hearsay evidence was unconvincing (see paragraph 45 below). (b) The Criminal Justice Act 2003 Part 11, Chapter 2 of the Criminal Justice Act 2003 entered into force in April 2005. It was intended to reform substantially the law governing the admission of hearsay evidence in criminal proceedings on the basis of the draft bill proposed by the Law Commission. Under section 114 of the 2003 Act, hearsay evidence is only admissible in criminal proceedings if one of a number of “gateways” applies. Although it was not relied upon in the second applicant’s case, one such gateway is section 114(1)(d) which allows for the admission of hearsay if the court is satisfied that it is in the interests of justice for it to be admissible. Section 114(2) provides: “ In deciding whether a statement not made in oral evidence should be admitted under subsection (1)(d), the court must have regard to the following factors (and to any others it considers relevant)— (a) how much probative value the statement has (assuming it to be true) in relation to a matter in issue in the proceedings, or how valuable it is for the understanding of other evidence in the case; (b) what other evidence has been, or can be, given on the matter or evidence mentioned in paragraph (a); (c) how important the matter or evidence mentioned in paragraph (a) is in the context of the case as a whole; (d) the circumstances in which the statement was made; (e) how reliable the maker of the statement appears to be; (f) how reliable the evidence of the making of the statement appears to be; (g) whether oral evidence of the matter stated can be given and, if not, why it cannot; (h) the amount of difficulty involved in challenging the statement; (i) the extent to which that difficulty would be likely to prejudice the party facing it.” The “gateway” relied on the second applicant’s case was section 116, which allows for the admission of statements of absent witnesses. Section 116 where relevant provides: “ (1) In criminal proceedings a statement not made in oral evidence in the proceedings is admissible as evidence of any matter stated if— (a) oral evidence given in the proceedings by the person who made the statement would be admissible as evidence of that matter, (b) the person who made the statement (the relevant person) is identified to the court’s satisfaction, and (c) any of the five conditions mentioned in subsection (2) is satisfied. (2) The conditions are— a) that the relevant person is dead; (b) that the relevant person is unfit to be a witness because of his bodily or mental condition; (c) that the relevant person is outside the United Kingdom and it is not reasonably practicable to secure his attendance; (d) that the relevant person cannot be found although such steps as it is reasonably practicable to take to find him have been taken; (e) that through fear the relevant person does not give (or does not continue to give) oral evidence in the proceedings, either at all or in connection with the subject matter of the statement, and the court gives leave for the statement to be given in evidence. (3) For the purposes of subsection (2)(e) ‘fear’ is to be widely construed and (for example) includes fear of the death or injury of another person or of financial loss. (4) Leave may be given under subsection (2)(e) only if the court considers that the statement ought to be admitted in the interests of justice, having regard— (a) to the statement’s contents, (b) to any risk that its admission or exclusion will result in unfairness to any party to the proceedings (and in particular to how difficult it will be to challenge the statement if the relevant person does not give oral evidence), (c) in appropriate cases, to the fact that a direction under section 19 of the Youth Justice and Criminal Evidence Act 1999 (c. 23) (special measures for the giving of evidence by fearful witnesses etc) could be made in relation to the relevant person, and (d) to any other relevant circumstances. (5) A condition set out in any paragraph of subsection (2) which is in fact satisfied is to be treated as not satisfied if it is shown that the circumstances described in that paragraph are caused— (a) by the person in support of whose case it is sought to give the statement in evidence, or (b) by a person acting on his behalf, in order to prevent the relevant person giving oral evidence in the proceedings (whether at all or in connection with the subject matter of the statement).” By section 121 of the Act, section 116(2)(e) can only be relied upon in respect of first hand hearsay; it cannot be relied only to allow the admission of multiple hearsay. In addition, section 124 allows the admission of evidence to challenge the credibility of the absent witness, for example, through the admission of evidence of his bad character, including previous convictions, a propensity to be untruthful and so on. It also allows the admission of inconsistent statements that the witness has made. Section 124(2) allows the admission of evidence to challenge the credibility of the absent witness in circumstances where it would not be admissible in respect of a live witness, for example, when it relates to a collateral issue in the case. Where a case is based wholly or partly on hearsay evidence, section 125 requires the trial judge to stop the case (and either direct an acquittal or discharge the jury) if, considering its importance to the case against the defendant, the hearsay evidence is so unconvincing that a conviction would be unsafe. This enacted the Law Commission’s recommendation 47 (see paragraph 42 above). Section 126 preserves both the common law discretion and the section 78 discretion of the trial judge to exclude hearsay evidence (see paragraph 40 above). It also provides a statutory discretion to exclude hearsay evidence if the trial judge is satisfied that “the case for excluding it, taking account of the danger that to admit it would result in undue waste of time, substantially outweighs the case for admitting it, taking account of the value of the evidence”. 3. The Coroners and Justice Act 2009 The conditions under which witnesses can give evidence anonymously in criminal proceedings are now regulated by the Coroners and Justice Act 2009. Formerly this was regulated by similar provisions in the Criminal Evidence (Witness Anonymity) Act 2008, which was enacted following the House of Lords’ judgment in R v. Davis (see paragraphs 49 and 50 below). Under the 2009 Act, witnesses can only give evidence anonymously when, upon the application of either the prosecution or a defendant in the proceedings, the trial judge makes a “witness anonymity order”. Section 87 requires that the trial judge be informed of the identity of the witness. Sections 88(2)-(6) and 89 lay down the conditions for the making of a witness anonymity order. In deciding whether those conditions are met, the court must have regard, inter alia , to whether evidence given by the witness might be the sole or decisive evidence implicating the defendant (section 89(2)(c)). 4. The Human Rights Act 1998 Section 2 of the Human Rights Act 1998 provides that, in determining any question that arises in connection with a Convention right, courts and tribunals must take into account any case-law from this Court so far as, in the opinion of the court or tribunal, it is relevant to the proceedings in which that question has arisen. Section 6(1) provides that it is unlawful for a public authority to act in a way which is incompatible with a Convention right. C. Relevant case-law from England and Wales 1. R v. Sellick and Sellick This Court’s judgment in Lucà v. Italy , no. 33354/96, § 40, ECHR 2001 II was considered by the Court of Appeal in R. v. Sellick and Sellick [2005] EWCA Crim 651 , which concerned two defendants who were alleged to have intimidated witnesses. Leave was given by the trial judge to have the witnesses’ statements read to the jury. The defendants appealed on the ground that the admission of the statements breached Article 6 § 1 read with 3 (d) of the Convention. The Court of Appeal dismissed the appeal. In considering the relevant case-law of this Court, at paragraph 50 of its judgment, it stated that what appeared from that case-law were the following propositions: “ i) The admissibility of evidence is primarily for the national law; ii) Evidence must normally be produced at a public hearing and as a general rule Article 6(1) and (3)(d) require a defendant to be given a proper and adequate opportunity to challenge and question witnesses; iii) It is not necessarily incompatible with Article 6(1) and (3)(d) for depositions to be read and that can be so even if there has been no opportunity to question the witness at any stage of the proceedings. Article 6(3)(d) is simply an illustration of matters to be taken into account in considering whether a fair trial has been held. The reasons for the court holding it necessary that statements should be read and the procedures to counterbalance any handicap to the defence will all be relevant to the issue, whether, where statements have been read, the trial was fair. iv) The quality of the evidence and its inherent reliability, plus the degree of caution exercised in relation to reliance on it, will also be relevant to the question whether the trial was fair.” The Court of Appeal then stated: “ The question is whether there is a fifth proposition to the effect that where the circumstances 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. Certainly at first sight paragraph 40 of Luca seems to suggest that in whatever circumstances and whatever counterbalancing factors are present if statements are read then there will be a breach of Article 6, if the statements are the sole or decisive evidence. Furthermore there is some support for that position in the previous authorities. But neither Luca nor any of the other authorities were concerned with a case where a witness, whose identity was well-known to a defendant, was being kept away by fear, although we must accept that the reference to Mafia-type organisations and the trials thereof in paragraph 40 shows that the court had extreme circumstances in mind.” 2. R. v. Davis In R. v. Davis [2008] UKHL 36 the House of Lords considered an appeal against conviction by a defendant who had been convicted of two counts of murder by shooting. Three witnesses had given evidence at trial identifying the defendant as the gunman. They gave evidence anonymously, testifying behind a screen so that they could be seen by the judge and jury but not the defendant. The House of Lords unanimously allowed the defendant’s appeal. It found that the witnesses’ testimony was inconsistent with the long-established principle of the English common law that, subject to certain exceptions and statutory qualifications, the defendant in a criminal trial should be confronted by his accusers in order that he may cross-examine them and challenge their evidence, a principle which originated in ancient Rome (Lord Bingham at paragraph 5). Moreover, this Court had not set its face absolutely against the admission of anonymous evidence in all circumstances. However, it had said that a conviction should not be based solely or to a decisive extent on anonymous statements. In any event, on the facts in Davis’s case, this Court would have found a violation of Article 6: not only was the anonymous witnesses evidence the sole or decisive basis on which Davis had been convicted, but effective cross-examination had been hampered. Lord Mance, concurring in judgment, considered the relevant authorities of this Court on Article 6. Having done so, he doubted that there was an absolute requirement that anonymous testimony should not be the sole or decisive evidence against a defendant. Instead, the extent to which such testimony is decisive might be no more than a very important factor to balance in the scales. He also considered that R. v. Sellick and Sellick (among other authorities) served as a caution against treating the Convention, or apparently general statements by this Court, as containing absolutely inflexible rules. 3. R. v. Horncastle and others The Chamber’s judgment of 20 January 2009 in the present cases was considered by the Court of Appeal and the Supreme Court of the United Kingdom in R. v. Horncastle and others . The case concerned the appeals of four defendants who had been convicted on the basis of statements of absent victims, which were read at trial under section 116 of the 2003 Act. For two of the defendants, the maker of the statement had died. For another two, the witness was too fearful to attend trial. Their appeals were heard together with that of a fifth defendant, Carter, who had been convicted on the basis of business records, which were introduced at his trial. (a) The judgment of the Court of Appeal On 22 May 2009, the Court of Appeal unanimously dismissed the appeals of the first four defendants ( [2009] EWCA Crim 964 ). It accepted that Article 6 § 3(d) had a content of its own but, given that it did not create any absolute right to have every witness examined, the balance struck by the 2003 Act was legitimate and wholly consistent with the Convention. There could be a very real disadvantage in admitting hearsay evidence and it needed cautious handling. However, having regard to the safeguards contained in the 2003 Act, which were rigorously applied, there would be no violation of Article 6 if a conviction were based solely or to a decisive degree on hearsay evidence. Where the hearsay evidence was demonstrably reliable, or its reliability could properly be tested and assessed, the rights of the defence would be respected, there would be sufficient counterbalancing measures and the trial would be fair. It was not appropriate that there should be a rule that counterbalancing measures could never be sufficient where the evidence was sole or decisive. This had been considered and rejected by the Law Commission and Parliament when it enacted the 2003 Act. There were also principled and practical difficulties with a sole or decisive rule. First, as a principled difficulty, the test assumed that all hearsay evidence was unreliable in the absence of testing in open court and second, it further assumed that the fact finder (such as the jury) could not be trusted to assess the weight of the evidence. Neither assumption was justified. For the first, the Court of Appeal gave examples of hearsay evidence that would be reliable such as a victim who, before dying, revealed the name of his or her murderer. For the second, the Court of Appeal found that juries were perfectly able to understand the limitations of written statements and, under section 124 of the 2003 Act, would be provided with material about the maker of the statement (see paragraph 45 above). The mere fact that the evidence was an essential link in the chain of evidence against the accused did not alter that conclusion. For example, forensic evidence might depend on work done by unidentified laboratory assistants (and to that extent was hearsay). However, it was not necessary for every member of the laboratory who had worked with the evidence to be called in order for the strength of the evidence to be tested. The court also found that there were practical difficulties with the sole or decisive rule as a test of admissibility of evidence. It observed: “ It is clear from the judgment in Al-Khawaja v UK that the ECtHR took the view that the error had lain in admitting the hearsay evidence: see in particular paragraphs 37, 40, 42 and 46. Any test of admissibility must be one which can be applied in advance of the giving of the evidence, let alone of the outcome of the trial. A routine test of admissibility of evidence which can only be applied in retrospect, after the outcome of the trial is known, makes the trial process little more than speculative. Judge, jury, witnesses and parties may find themselves engaged in shadow-boxing without knowing whether the solemn result of the trial will stand to be reversed on the grounds that, as things have turned out, the test of admissibility was not met. Nor can any defendant decide how to conduct his case, and indeed whether or not to plead guilty, if he does not know what evidence can and cannot be relied upon. It will no doubt often be possible to identify in advance a case in which the hearsay evidence in question is the sole evidence against the accused. An obvious case is that of the single eye-witness case, with no suggested support from any other source. But this frequently may not be clear from the outset; there may be other evidence which the prosecution intends to present, but which, on hearing, turns out not to incriminate the accused, or is effectively demolished. Conversely, what appears at the outset to be hearsay evidence standing alone may sometimes come to be supported by other material as the evidence develops. A witness may add something of great significance not previously mentioned, or the significance of something always said may become apparent when apparently unconnected other evidence is given. In any case where there is more than one accused jointly charged, it is common experience that the evidence of one may well shed enormous light upon the guilt or innocence of another. So even the concept of ‘sole’ evidence is an impractical test for admissibility. It is, however, the second limb of the suggested test which is apt to cause the greatest difficulty. No one can know what evidence is decisive until the decision making process is over. On no view can it be possible to rule in advance, at the stage when admissibility is in question, which evidence will be decisive ... The application of the test is made the more difficult yet if the meaning of ‘decisive’ is extended to encompass any evidence of which it can be said that, if it were absent, ‘the prospect of a conviction would have receded and that of an acquittal advanced’ (see paragraph 21 of Al-Khawaja ). Indeed, if that is the test of what is decisive, virtually all evidence would qualify; evidence which does not, if accepted by the jury, advance the prospect of conviction will ordinarily be excluded as irrelevant.” The Court of Appeal also observed that the test would also be impossible to apply at a trial of two or more defendants, where one defendant sought to introduce hearsay evidence for his defence. In such a case, the trial judge would be bound to allow the defendant to introduce that evidence even if it might also incriminate one of the other defendants and indeed be decisive evidence against that other defendant. The Court of Appeal also considered that the safeguard in section 125 of the 2003 Act (the power of a trial judge to stop the case if the hearsay evidence is unconvincing: see paragraph 45 above) provided for a “proportionate assessment of the reliability” of hearsay evidence and it would not serve justice if that power were to be trammelled by a requirement that it be exercised in every case in which the hearsay evidence were the sole or decisive evidence. Sole or decisive hearsay could be wholly convincing and, equally, evidence which was neither sole nor decisive might have such a potential influence on the jury that the judge would be persuaded that a conviction was unsafe. Where there was a legitimate argument that that hearsay was unconvincing and important to the case, the trial judge was required to make up his own mind as to whether a conviction would be safe; this involved assessing the reliability of the hearsay evidence, its place in the evidence as a whole, the issues in the case as they had emerged, and all the other individual circumstances of the case. Finally, the other safeguards contained in the 2003 Act were rigorously applied and the difficulties faced by defendants when hearsay was admitted were well understood by the courts. The Act did not equate hearsay with first hand evidence; on the contrary it required cautious handling. The Court of Appeal also gave guidance as to assessing when it would be appropriate to allow hearsay evidence to be introduced because a witness was in fear. There was, in the case law of this Court, no requirement that the fear had to be attributable to the defendant: the essential questions were whether there was a justifiable reason for the absence of the witness supported by evidence and whether the evidence was demonstrably reliable or its reliability could be properly tested and assessed. The Court of Appeal added: “ It is, however, important that all possible efforts are made to get the witness to court. As is clear, the right to confrontation is a longstanding requirement of the common law and recognised in Article 6(3)(d). It is only to be departed from in the limited circumstances and under the conditions set out in the [2003 Act]. The witness must be given all possible support, but also made to understand the importance of the citizen’s duty, and indeed that the violent and intimidatory will only flourish the more if that duty is not done, whilst they will normally back down in the face of determination that it be performed. For this reason it is of especial importance that assurances are never given to potential witnesses that their evidence will be read. Unless the defendant consents, it is only the court applying the strict conditions of the [2003 Act] based on evidence that can admit such a statement. Any indication, let alone an assurance, can only give rise to an expectation that this will indeed happen, when if it does the impact of the evidence will be diminished and the disadvantage to the accused may result in it not being given at all. It may well be that in the early stage of police enquiries into a prominent crime the investigators need to seek out information on a confidential basis: that is a matter for practical policing and not for us. But no person who is becoming not simply a source of information but a witness should be told that his evidence will be read, or indeed given any indication whatsoever that this is likely. The most that he can be told is that witnesses are expected to be seen at court, that any departure from that principle is exceptional, and that the decision whether to depart from it is one for the Judge and not for the police. In the case before us of Marquis and Graham [two of the appellants], as we set out at paragraphs 127 and 132, the Judge found that the investigating police officer had significantly contributed to the fear of the witness by referring repeatedly to a notorious local example of witnesses being hunted down, although relocated, and killed. Although notorious, that incident was an extreme and very unusual case. The need for police officers to tender careful advice to potential witnesses in order to discharge their duty of care towards them should not lead to such frightening information being laboured out of defensiveness. Whilst the [2003 Act] requires fear to be construed broadly, it is not to be expected that fear based upon inappropriate assurances by police officers will result in the evidence being read and the case proceeding on the basis of it to the jury. If the evidence can really only be assessed by the jury by seeing the witness, as will often be the case, it may not be admitted. If it is admitted and central to the case, there is a significant possibility that at the end of the trial the Judge may have to rule under s.125 that a conviction relying upon it would be unsafe.” (b) The judgment of the Supreme Court On 9 December 2009, the United Kingdom Supreme Court unanimously upheld the Court of Appeal’s judgment ( [2009] UKSC 14 ). Lord Phillips, giving the judgment of the Supreme Court, found that, although domestic courts were required by the Human Rights Act 1998 to take account of the Strasbourg jurisprudence in applying principles that were clearly established, on rare occasions, where a court was concerned that the Strasbourg judgment did not sufficiently appreciate or accommodate some aspect of English law, it might decline to follow the judgment. The Chamber’s judgment was such a case. Lord Phillips considered that a defendant should not be immune from conviction where a witness, who had given critical and apparently reliable evidence in a statement, was unavailable to give evidence at trial through death or some other reason. In analysing the relevant case-law of this Court on Article 6 § 3(d), Lord Phillips concluded that, although this Court had recognised the need for exceptions to the strict application of Article 6 § 3(d), the manner in which it approved those exceptions resulted in a jurisprudence which lacked clarity. The sole or decisive rule had been introduced into the Strasbourg case-law in Doorson v. the Netherlands , 26 March 1996, Reports of Judgments and Decisions 1996 II without discussion of the underlying principle or full consideration of whether there was justification for imposing the rule as an overriding principle applicable equally to continental and common law jurisdictions. Indeed the rule seemed to have been created because, in contrast to the common law, continental systems of criminal procedure did not have a comparable body of jurisprudence or rules governing the admissibility of evidence. Lord Phillips, in agreement with the Court of Appeal, found that the rule would create severe practical difficulties if applied to English criminal procedure. First, it was not easy to apply because it was not clear what was meant by “decisive”: under English criminal procedure no evidence could be admitted unless it was potentially probative and, in theory, any item of probative evidence could make the difference between conviction and acquittal. Second, it would be hard enough to apply that test at first instance but it would be impossible for national appellate courts or this Court to decide whether a particular statement was the sole or decisive basis for a conviction. In a jury trial, the only way the rule could be applied would be to exclude all hearsay evidence. Lord Phillips also observed: “ The sole or decisive test produces a paradox. It permits the court to have regard to evidence if the support that it gives to the prosecution case is peripheral, but not where it is decisive. The more cogent the evidence the less it can be relied upon. There will be many cases where the statement of a witness who cannot be called to testify will not be safe or satisfactory as the basis for a conviction. There will, however, be some cases where the evidence in question is demonstrably reliable. The Court of Appeal has given a number of examples. I will just give one, which is a variant of one of theirs. A visitor to London witnesses a hit and run road accident in which a cyclist is killed. He memorises the number of the car, and makes a statement to the police in which he includes not merely the number, but the make and colour of the car and the fact that the driver was a man with a beard. He then returns to his own country, where he is himself killed in a road accident. The police find that the car with the registration number that he provided is the make and colour that he reported and that it is owned by a man with a beard. The owner declines to answer questions as to his whereabouts at the time of the accident. It seems hard to justify a rule that would preclude the conviction of the owner of the car on the basis of the statement of the deceased witness, yet that is the effect of the sole or decisive test.” Lord Phillips instead concluded that the 2003 Act made such a rule unnecessary in English criminal procedure because, if the 2003 Act were observed, there would be no breach of Article 6 § 3(d) even if a conviction were based solely or to a decisive extent on hearsay evidence. To demonstrate this point, Annex 4 to the judgment analysed a series of cases against other Contracting States where this Court had found a violation of Article 6 § 1 when taken with Article 6 § 3(d). In each case, had the trial taken place in England and Wales, the witness’s testimony would not have been admissible under the 2003 Act either because the witness was anonymous and absent or because the trial court had not made sufficient enquiries to ensure there was good reason for the witness’s absence. Alternatively, had the evidence been admitted, any conviction would have been quashed on appeal. Lord Brown delivered a concurring judgment in which he stated: “ These appeals are of the utmost importance. If the Strasbourg case law does indeed establish an inflexible, unqualified principle that any conviction based solely or decisively on evidence adduced from an absent or anonymous witness is necessarily to be condemned as unfair and set aside as contrary to articles 6(1) and 6(3)(d) of the Convention, then the whole domestic scheme for ensuring fair trials – the scheme now enshrined (as to hearsay evidence) in the Criminal Justice Act 2003 and (as to anonymous evidence) in the Criminal Evidence (Witness Anonymity) Act 2008 [see paragraph 46 above] – cannot stand and many guilty defendants will have to go free. It is difficult to suppose that the Strasbourg Court has in fact laid down so absolute a principle as this and, indeed, one exception to it, at least, appears to be acknowledged: the fairness of admitting hearsay evidence from a witness absent as a result of the defendant’s own intimidation. ... Nor can Strasbourg readily be supposed to have intended the sort of practical problems and anomalies identified by the Court of Appeal (paragraphs 61-63 and 68 71 [of its judgment]) that must inevitably flow from any absolute principle of the kind here contended for. Obviously, the more crucial the evidence is to the proof of guilt, the more scrupulous must the Court be to ensure that it can be fairly adduced and is likely to be reliable. In this connection there can be no harm in using the concept of ‘sole or decisive’ so long as it is used broadly – as it is in the 2008 Act with regard to anonymous witnesses and, indeed, in the control order context where it relates rather to the allegations made against the suspect than the evidence adduced in support. Understood and applied inflexibly, however, the concept would involve insoluble problems of detailed interpretation and application. The better view may therefore be that no such absolute principle emerges from the Strasbourg Court’s judgment in Al-Khawaja and Tahery v United Kingdom ...” III. RELEVANT COMPARATIVE LAW 1. Scotland . Subject to certain statutory exceptions, in Scots criminal law a person cannot be convicted of a crime or statutory offence on the uncorroborated testimony of one witness, however credible (see Morton v. HM Advocate 1938 JC 52, quoted in Campbell v. HM Advocate 2004 S. L. T. 135). Hearsay is regulated by section 259 of the Criminal Procedure (Scotland) Act 1995, which allows for its admissibility under certain conditions, including when the person who made the statement is dead. Section 259(4) permits evidence to be admitted which is relevant to the credibility of the person as a witness. In N v. HM Advocate 2003 S. L. T. 761 the High Court of Justiciary, sitting as an appeal court, reluctantly reached the conclusion that section 259 deprived the court of discretion it previously enjoyed at common law to exclude such evidence if it was unreliable. The Lord Justice Clerk observed that, notwithstanding section 259, the long recognised dangers in hearsay evidence remained. He added: “ Where a general provision such as s 259 applies, there are bound to be cases in the circumstances of which hearsay evidence would be so prejudicial to the fairness of the trial that the only just and proper course would be to exclude it. This, I think, is such a case. I am not impressed by the three safeguards to which the trial judge referred ( HM Advocate v N , at p 437C–E). The requirement of corroboration is a neutral consideration. It is a safeguard that applies to prosecution evidence in any form. I cannot see what worthwhile safeguard the principle of corroboration provides if the primary evidence sought to be corroborated is per se unfair to the accused. Moreover, the leading of evidence bearing upon the credibility of the maker of the hearsay statement may be at most an exercise in damage limitation where clearly prejudicial evidence has already been led. As for the safeguard of the judge’s directions, I think that there may be cases where the hearsay evidence is so prejudicial that no direction, however strong, could make adequate amends for the unfairness of its having been admitted. ... In English provisions governing the admissibility of a statement made in a document, Parliament has expressly conferred a discretion on the court to exclude such a statement if it is of the opinion that in the interests of justice it ought not to be admitted. One specific consideration to which the court must have regard is the risk that the admission of the evidence will result in unfairness to the accused (Criminal Justice Act 1988, ss 25(1), 25(2)(d) and 26(ii); cf R v Gokal ). These, in my view, are prudent provisions. If provisions of this kind had been available to the trial judge in this case, they could have enabled him to exclude the hearsay at the outset.” The High Court of Justiciary also considered the compatibility of the introduction of the evidence of an absent witness with Article 6 § 3(d) in McKenna v. HM Advocate . In that case, a murder trial, the prosecution sought to introduce statements made to the police by a possible co-accused who had since died. In a previous ruling given before trial (2000 SLT 508), the High Court of Justiciary, sitting as an appeal court, had found that it was only in extreme circumstances that an accused could contend in advance of trial that the introduction of the hearsay evidence would be so prejudicial to the prospects of a fair trial that the court could determine the issue in advance. It therefore allowed the trial to proceed. When the statements were introduced at trial and the accused was convicted, he appealed against his conviction. In its judgment dismissing the appeal (2003 SLT 508), the High Court of Justiciary found that, while the statements were important evidence, having regard to the other evidence led at trial (which included admissions by the accused and forensic evidence), it could not be said that the appellant’s conviction was based to a decisive extent on them. The jury had also been adequately and satisfactorily directed as to how to approach the absent witness’s statement. A similar conclusion was reached by the same Court in HM Advocate v. M 2003 SLT 1151. The compatibility of the admission of hearsay evidence with Article 6 § 3(d) was further considered by the High Court of Justiciary in Campbell v. HM Advocate , cited above. The court observed that, in the relevant case law of this Court, many of the violations of Article 6 § 3(d) which had been found had arisen in jurisdictions which did not apply the Scots law rule of corroboration: “ Most of the situations in which it has been held by the court that there had been a violation of art 6(1) and (3)(d) could not arise in Scotland. Against the requirement for corroboration of all crucial facts, a conviction could not be based solely on the evidence of a single witness, whether in primary or in secondary form. Violations of the Convention right have been established where the principal witness against the accused has not been made available for questioning or, in circumstances where there have been a number of principal witnesses, where none of them has been made so available. No case was cited to us in which a violation was held to have occurred in circumstances where the accused had had an opportunity to question or have questioned the complainer or other direct or central witness and other supporting evidence was in statement form. ‘To a decisive extent’, as used in the European authorities, appears to be concerned with the significance of the evidence as a matter of weight. It is not concerned with any rule that a conviction cannot be based on a single source of evidence. The fact that the hearsay is required to meet the rule about corroboration does not of itself render that hearsay ‘decisive’ in the European sense. In these circumstances we are not persuaded that in every case in which hearsay evidence is a necessary ingredient of the Crown’s corroborated proof there will be a violation of art 6(1) and (3)(d). It will, however, be necessary for the trial judge to address, in the context of the whole evidence in the case, the significance of any hearsay evidence relied on by the Crown and to take appropriate action to ensure that the accused’s entitlement to a fair trial is not violated thereby.” The court added that, in directing a jury, the guidance given by the Lord Justice Clerk in N v. HM Advocate , quoted above, should be borne in mind. The High Court allowed the appeal of the first appellant in Campbell because of the inadequacy of the trial judge’s direction to the jury. It dismissed the second appellant’s appeal insofar as it related to Article 6 of the Convention, finding that the hearsay evidence was not decisive and that the principal evidence against him came from a witness who had testified in court. Campbell was applied in HM Advocate v. Johnston 2004 S. L. T. 1005 where, in a ruling given during the trial, the Lord Ordinary allowed the admission of a witness statement made to the police by a witness who subsequently died before trial; the statement was admissible inter alia because it could not be “decisive”. In Humphrey v. HM Advocate [2008] HCJAC 30 , the High Court of Justiciary observed that it had “great difficulty” in understanding the meaning of the word “decisive” in the context of a case based on circumstantial evidence but, in that case, the evidence in the form of a police statement of a witness who had died was not “remotely decisive” and there was sufficient evidence to support the conviction without it. Similar results were reached in Allison v. HM Advocate [2008] HCJAC 63 and Harkins v. HM Advocate [2008] HCJAC 69. 2. Ireland In its March 2010 consultation paper on hearsay in civil and criminal cases, the Law Reform Commission of Ireland provisionally recommended that, subject to existing common law and statutory inclusionary exceptions, hearsay should continue to be excluded in criminal proceedings. It also provisionally recommended that there should be no statutory introduction of a residual discretion to include hearsay evidence and that the concepts of reliability and necessity should not form the basis for reform of the hearsay rule because they lacked clarity. The Commission also noted: “ Article 38.1 of the Constitution of Ireland protects the right to cross-examination and that the free admissibility of hearsay evidence in criminal proceedings would infringe this constitutionally protected right. There are dangers associated with allowing evidence of unavailable witnesses: it undermines the defendant’s right to a fair trial and creates the potential of miscarriages of justice arising if evidence adduced from the following categories of witnesses is admitted: Where the witness is dead (with the exception of dying declarations); Where a witness because of a bodily or mental infirmity cannot give evidence; Where the witness is outside of the jurisdiction; Where the witness cannot be found.” The Commission declined to recommend that the statutory scheme in England and Wales, and in particular section 114 of the 2003 Act, be adopted in Ireland. It observed that: “ This model of reform relaxes the rule in such a manner as to potentially render the rule against hearsay redundant. The categories of admissible hearsay under this model are extended significantly and, in light of the constitutional protection afforded to the right to cross-examination, the Commission is of the provisional opinion that to allow in untested evidence from frightened and unavailable witnesses would undermine this right. The Commission notes that it has provisionally recommended that the courts should retain a discretion to develop the hearsay rule if the necessity exists.” The Commission also found that (subject to possible reservations concerning the ultimate outcome in the present cases), the case law of this Court appeared broadly in line with the approach taken in Irish law. 3. Australia A strict approach to the exclusion of out of court statements has always been taken by the Australian courts (see, for example, Bannon v. The Queen (1995) 185 CLR 1). The admission of hearsay in federal criminal proceedings in Australia is now regulated by the Evidence Act 1995. Section 65 of the Act allows the admission of evidence as to previous representations (for example statements) when a witness is not available to give evidence about an “asserted fact”. Such evidence will be admissible inter alia when: (i) the representation was made when or shortly after the asserted fact occurred in circumstances that make it unlikely that the representation is a fabrication (section 65(2)(b)); or (ii) when the representation was made in circumstances that make it highly probable that the representation was reliable (section 65(2)(c)). The use of these provisions to admit the statement of a witness who died before trial was considered by the Federal Court of Australia in Williams v. The Queen (2000) 119 A Crim R 490. The court found that, in the circumstances of the case, the statement made to the police was not sufficiently reliable, particularly when the witness had reasons to tell the police what they wanted to hear. Equivalent statutory provisions were considered by the Supreme Court of New South Wales in Harris v. The Queen (2005) 158 A Crim R 454), where the deceased complainant’s statement was found to be sufficiently reliable to be admitted, given that inter alia the complainant knew that the police would interview other witnesses in the case. 4. Canada The law of hearsay in Canada has been reformed as a result of three principal decisions of the Supreme Court of Canada, which created a “principled approach” to the admissibility of hearsay evidence. First, in R v. Khan [1990] 2 S. C. R. 531, the Supreme Court found that the trial judge had erred in refusing to allow the victim, a three and a half year old girl who alleged she had been sexually assaulted, to give unsworn evidence and his refusal to allow the Crown to introduce statements the child had made to her mother fifteen minutes after the assault. If unsworn evidence could not be given there would be danger that offences against very young children could never be prosecuted. In respect of the statements to the mother, the Supreme Court found that it was appropriate to take a more flexible but “principled” approach to hearsay. Despite the need for caution, hearsay could be admitted where the two general requirements of necessity and reliability were met. In determining admissibility the trial judge was required to have regard to the need to safeguard the interests of the accused. Concerns as to the credibility of the evidence remained to be addressed by submissions as to the weight to be accorded to it and submissions as to the quality of any corroborating evidence. Second, in R v. Smith [1992] 2 S. C. R. 915 the Court approved the “principled approach” taken in Khan and found that two telephone calls made by the deceased to her mother shortly before her death were admissible. However, the mother’s evidence as to the contents of a third telephone call should have not been admitted as the conditions under which that call was made did not provide the “circumstantial guarantee of trustworthiness” which would justify its admission without the possibility of cross-examination. . Khan and Smith were followed in R. v. Rockey [1996] 3 S. C. R. 829. The Supreme Court was satisfied that any reasonable trial judge would have found it necessary to admit pre-trial statements made by a child, who was five years of age at trial, in which he complained he had been sexually assaulted. It was observed by McLachlin J (concurring in judgment) that the case against the appellant was strong: the child’s statements were entirely consistent and supported inter alia by medical evidence and evidence of behavioural changes in him after the assault and the absence of any plausible explanation of someone other than the appellant perpetrating the assault. The third and most significant development in the Supreme Court’s case-law was its judgment in R. v Khewalon [2006] 2 S. C. R. 787, which took a stricter approach to reliability. The case concerned a complaint of assault made by an elderly resident at a retirement home against one of the home’s employees. The patient, S, gave the police a videotaped statement; the statement was unsworn. After S’s statement other residents came forward to give statements that they too had been assaulted by the accused. By the time of trial, all those who had made statements, including S, had either died or were no longer competent to testify. Some of the statements were admitted by trial judge because of the striking similarity between them. The Court of Appeal for Ontario excluded all the statements and acquitted the accused. The Supreme Court dismissed the Crown’s appeal from that decision and affirmed the acquittal. The Supreme Court clarified its previous case-law on reliability and stated that the reliability requirement would generally be met by showing: (i) that there was no real concern about whether the statement was true or not because of the circumstances in which it came about; or (ii) that no real concern arose because the truth and accuracy of the statement could nonetheless be sufficiently tested by means other than cross-examination. It was for the trial judge to make a preliminary assessment of the “threshold” reliability of the statement and to leave the ultimate determination of its worth to the fact finder/jury. All relevant factors had to be considered by the trial judge including, in appropriate cases, the presence of supporting or contradictory evidence. Charron J, giving the unanimous judgment of the court, stated (at paragraph 49 of the judgment): “ In some cases, because of the circumstances in which it came about, the contents of the hearsay statement may be so reliable that contemporaneous cross-examination of the declarant would add little if anything to the process. In other cases, the evidence may not be so cogent but the circumstances will allow for sufficient testing of evidence by means other than contemporaneous cross-examination. In these circumstances, the admission of the evidence will rarely undermine trial fairness. However, because trial fairness may encompass factors beyond the strict inquiry into necessity and reliability, even if the two criteria are met, the trial judge has the discretion to exclude hearsay evidence where its probative value is outweighed by its prejudicial effect.” The court concluded that S’s statements were not admissible. It observed that the charges against Khelawon in respect of S were “entirely based” on the truthfulness of the allegations contained in S’s statements (paragraph 101 of the judgment). The absence of any opportunity to cross examine him had a bearing on the question of reliability and there were no adequate substitutes for testing the evidence. The principled exceptions to the hearsay rule did not provide a vehicle for founding a conviction on the basis of a police statement, videotaped or otherwise, without more (paragraph 106 of the judgment). Nor could the reliability requirement be met by relying on the inherent trustworthiness of the statement; this was not a case such as Khan where the cogency of the evidence was such that it would be pedantic to insist upon testing by cross examination. S was elderly and frail; his mental capacity was at issue and there was medical evidence that his injuries could have arisen from a fall. It was also not clear that he had understood the consequences of his statement for the accused. In the circumstances, S’s unavailability for cross examination posed significant limitations on the accused’s ability to test the evidence and, in turn, on the trier of fact’s ability to properly assess its worth. 5. Hong Kong In its report of November 2009, the Law Reform Commission of Hong Kong proposed substantial reforms to the admission of hearsay evidence in criminal proceedings in that jurisdiction. It proposed that the present rule in Hong Kong against the admission of hearsay evidence should be retained but there should be greater scope to admit hearsay evidence in specific circumstances. The Commission rejected the English statutory scheme contained in the 2003 Act, observing that Hong Kong had none of the statutory mechanisms for excluding hearsay which applied in England and Wales, such as section 126(1) of the 2003 Act and section 78 of PACE (see paragraphs 40 and 45 above). The Commission also observed that grounds for admitting hearsay under section 116 of the 2003 Act, although offering a fair degree of certainty and consistency in decision–making, had “an over-inclusive effect by allowing in all types of relevant evidence, including unreliable hearsay evidence” (paragraph 8.25 of the report). The Commission instead proposed a model of reform, which was based on the approach taken by the New Zealand Law Commission (see paragraph 81 below) and the Canadian courts since Khelawon . It recommended that hearsay evidence should be admissible inter alia if the trial judge was satisfied that it was necessary to admit the hearsay evidence and that it was reliable. Assessment of reliability by a trial judge should include consideration of whether the hearsay evidence was supported by other admissible evidence. The Commission also recommended that, at any stage of criminal proceedings after hearsay evidence had been admitted, the court should have the power to direct the acquittal of the accused if the trial judge considered that it would be unsafe to convict. In reaching that decision, the court should have regard inter alia to the importance of such evidence to the case against the accused. The Commission also found that these recommendations meant that its model of reform would comply with the Chamber’s judgment in the present cases. 6. New Zealand The New Zealand Law Commission, in its 1999 report on evidence, recommended that the admissibility of hearsay should be based on two considerations: reliability and necessity. That recommendation was enacted in the Evidence Act 2006, which came into force in 2007. Section 18(1) of the Act provides: “ A hearsay statement is admissible in any proceeding if— (a) the circumstances relating to the statement provide reasonable assurance that the statement is reliable; and (b) either— (i) the maker of the statement is unavailable as a witness; or (ii) the Judge considers that undue expense or delay would be caused if the maker of the statement were required to be a witness.” “Circumstances” for these purposes are defined in s 16(1) as including: (a) the nature of the statement; (b) the contents of the statement; (c) the circumstances that relate to the making of the statement; (d) any circumstances that relate to the veracity of the person; and (e) any circumstances that relate to the accuracy of the observation of the person. Section 8(1) of the Act provides that evidence must be excluded if its probative value is outweighed by the risk that the evidence will (a) have an unfairly prejudicial effect on the proceeding; or (b) needlessly prolong the proceeding. Section 8(2) provides that in determining whether the probative value of evidence is outweighed by the risk that the evidence will have an unfairly prejudicial effect on a criminal proceeding, the judge must take into account the right of the defendant to offer an effective defence. 7. South Africa Although jury trials are no longer used in South Africa, it too has substantially reformed its law of hearsay. Section 3(4) of Law of Evidence Amendment Act 1988 defines hearsay as evidence, whether oral or in writing, the probative value of which depends upon the credibility of any person other than the person giving evidence. By section 3(1) such hearsay evidence is not admissible unless (i) it is admitted by consent; (ii) the person upon whose credibility the evidence depends testifies; or (iii) the court is of the opinion that the evidence should be admitted in the interests of justice. Section 3(1)(c) provides that, in forming its opinion, the court must have regard to: (i) the nature of the proceedings; (ii) the nature of the evidence; (iii) the purpose for which the evidence is tendered; (iv) the probative value of the evidence; (v) the reason why the evidence is not given by the person upon whose credibility the probative value of such evidence depends; (vi) any prejudice to a party which the admission of such evidence might entail; and (vii) any factor which should in the opinion of the court be taken into account. In State v. Ramavhale [1996] ZASCA 14, the Supreme Court observed that, notwithstanding the wording of section 3(1), there remained an “intuitive reluctance to permit untested evidence to be used against an accused in a criminal case”. It also endorsed previous authority that a court should “hesitate long” in admitting or relying on hearsay evidence which plays a decisive or even significant part in convicting an accused, unless there are compelling reasons for doing so. Despite the reforms enacted by section 3(1), the Supreme Court found that reliability was a factor that should continue to be considered in determining the admissibility of hearsay. Moreover, in that case, in assessing what prejudice would arise to the accused if the hearsay were admitted, the trial judge had erred in relying upon the fact that the accused had other avenues open to him to counter the evidence, one of which was to give evidence himself; the logical conclusion of this approach was that the State, by introducing “flimsy” hearsay could force the accused to testify in a case where the absence of direct evidence was such that he would be entitled not to testify. Compelling reasons for admitting hearsay evidence, even though it was decisive, were found to exist in State v. Ndhlovu and others [2002] ZASCA 70, not least because the guarantees as to the reliability of the evidence were high and the hearsay evidence interlinked powerfully with the other evidence in the case. T he concerns expressed in Ramavhale as to admitting or relying on hearsay evidence which played a decisive or even significant part in convicting an accused were reiterated by the Supreme Court in State v. Libazi and others [2010] ZASCA 91, State v. Mpungose and another [2011] ZASCA 60, and State v. Mamushe [2007] ZASCA 58. In the latter, the Supreme Court observed that it stood to reason that a hearsay statement which would only serve to complete a “mosaic pattern” would be more readily admitted than one which was destined to become a vital part of the State’s case. 8. United States of America The Sixth Amendment to the Constitution of the United States guarantees the accused in all criminal prosecutions the right to be confronted with the witnesses against him (“the confrontation clause”). In Ohio v. Roberts 448 U. S. 56 (1980) the Supreme Court of the United States held that evidence with “particularized guarantees of trustworthiness” was admissible without confrontation. That was overruled in Crawford v. Washington 541 U. S. 36, where the court ruled that the confrontation clause applied to all evidence which was testimonial in nature and there was no basis in the Sixth Amendment for admitting evidence only on the basis of its reliability. Where testimonial statements were at issue, the only indicium of reliability which was sufficient was the one prescribed by the Constitution: confrontation. Consequently, testimonial evidence was thus inadmissible unless the witness appeared at trial or, if the witness was unavailable, the defendant had a prior opportunity for cross-examination (see also Melendez-Diaz v. Massachusetts 129 S. Ct. 2527 (2009) and Bullcoming v. New Mexico 131 S. Ct. 2705 (2011)). The right to confront witnesses will be forfeited if it can be demonstrated that the defendant has frightened the witness into not testifying ( Davis v. Washington 547 U. S. 813 (2006) and Giles v. California 554 U. S. 353 (2008). THE LAW I. ALLEGED VIOLATION OF ARTICLE 6 §§ 1 AND 3 (d) OF THE CONVENTION Article 6 §§ 1 and 3 (d) of the Convention, which reads as follows: “ 1. In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law... 3. 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.” A. The Chamber’s conclusions For the Chamber the provisions of Article 6 § 3(d), as with the other elements of Article 6 § 3, was one of the minimum rights which had to be accorded to anyone who was charged with a criminal offence. The Chamber also observed that as minimum rights, the provisions of Article 6 § 3 constituted express guarantees and could not be read as illustrations of matters to be taken into account when considering whether a fair trial had been held. The Chamber then observed that whatever the reason for the defendant’s inability to examine a witness, whether absence, anonymity or both, the starting point for the Court’s assessment of whether there was a breach of Article 6 §§ 1 and 3(d) was set out in Lucà , cited above, at § 40: “ 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 on depositions that have been made by a person whom the accused has had no opportunity to examine or to 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 [references omitted].” The Chamber noted that both parties were content to approach the matter on the basis that each conviction was based solely or to a decisive degree on the two witnesses concerned and it proceeded on the same basis. The Chamber then considered the counterbalancing factors which the Government relied on in each case. First, it considered the counterbalancing factors that, in each case, the trial judge correctly applied the relevant statutory test and that the Court of Appeal reviewed the safety of each conviction. It found these factors to be of limited weight since the very issue in each case was whether the trial judges and the Court of Appeal had acted compatibly with Article 6 §§ 1 and 3 (d) of the Convention. In respect of Mr Al-Khawaja, the Chamber continued: “ 41. In examining the facts of Mr Al-Khawaja’s case, the Court observes that the counterbalancing factors relied by the Government are the fact that ST’s statement alone did not compel the applicant to give evidence; that there was no suggestion of collusion between the complainants; that there were inconsistencies between ST’s statement and what was said by other witnesses which could have been explored in cross-examination of those witnesses; the fact that her credibility could be challenged by the defence; and the warning to the jury to bear in mind that they had neither seen nor heard ST’s evidence and that it had not been tested in cross-examination. 42. Having considered these factors, the Court does not find that any of them, taken alone or together, could counterbalance the prejudice to the defence by admitting ST’s statement. It is correct that even without ST’s statement, the applicant may have had to give evidence as part of his defence to the other count, count two. But had ST’s statement not been admitted, it is likely that the applicant would only have been tried on count two and would only have had to give evidence in respect of that count. In respect of the inconsistencies between the statement of ST and her account as given to two witnesses, the Court finds these were minor in nature. Only one such inconsistency was ever relied on by the defence, namely the fact that at one point during the alleged assault, ST had claimed in her statement that the applicant had touched her face and mouth while in the account given to one of the witnesses she had said that she had touched her own face at the instigation of the applicant. While it was certainly open to the defence to attempt to challenge the credibility of ST, it is difficult to see on what basis they could have done so, particularly as her account corresponded in large part with that of the other complainant, with whom the trial judge found that there was no evidence of collusion. The absence of collusion may be a factor in domestic law in favour of admissibility but in the present case it cannot be regarded as a counterbalancing factor for the purposes of Article 6 § 1 read with Article 6 § 3(d). The absence of collusion does not alter the Court’s conclusion that the content of the statement, once admitted, was evidence on count one that the applicant could not effectively challenge. As to the judge’s warning to the jury, this was found by the Court of Appeal to be deficient. Even if it were not so, the Court is not persuaded that any more appropriate direction could effectively counterbalance the effect of an untested statement which was the only evidence against the applicant.” The Chamber therefore found a violation of Article 6 §§ 1 read in conjunction with Article 6 § 3(d) of the Convention in respect of Mr Al Khawaja. In respect of Mr Tahery, the Chamber observed: “ 45. In this case, the Government relied on the following principal counterbalancing factors: that alternative measures were considered by the trial judge; that the applicant was in a position to challenge or rebut the statement by giving evidence himself and by calling other witnesses; that the trial judge warned the jury that it was necessary to approach the evidence given by the absent witness with care; and that the judge told the jury that the applicant was not responsible for T’s fear. 46. The Court does not find that these factors, whether considered individually or cumulatively, would have ensured the fairness of the proceedings or counterbalanced the grave handicap to the defence that arose from the admission of T’s statement. It is appropriate for domestic courts, when faced with the problem of absent or anonymous witnesses, to consider whether alternative measures could be employed which would be less restrictive of the rights of the defence than admitting witness statements as evidence. However, the fact that alternative measures are found to be inappropriate does not absolve domestic courts of their responsibility to ensure that there is no breach of Article 6 §§ 1 and 3 (d) when they then allow witness statements to be read. Indeed, the rejection of less restrictive measures implies a greater duty to ensure respect for the rights of the defence. As regards the ability of the applicant to contradict the statement by calling other witnesses, the very problem was that there was no witness, with the exception of T, who was apparently able or willing to say what he had seen. In these circumstances, the Court does not find that T’s statement could have been effectively rebutted. The Court accepts that the applicant gave evidence himself denying the charge, though the decision to do so must have been affected by the admission of T’s statement. The right of an accused to give evidence in his defence cannot be said to counterbalance the loss of opportunity to see and have examined and cross-examined the only prosecution eye-witness against him. 47. Finally, as to the trial judge’s warning to the jury, the Court accepts that this was both full and carefully phrased. It is true, too, that in the context of anonymous witnesses in Doorson , cited above, § 76, the Court warned that “evidence obtained from witnesses under conditions in which the rights of the defence cannot be secured to the extent normally required by the Convention should be treated with extreme care”. In that case, it was satisfied that adequate steps had been taken because of the express declaration by the Court of Appeal that it had treated the relevant statements “with the necessary caution and circumspection”. However, in the case of an absent witness such as T, the Court does not find that such a warning, including a reminder that it was not the applicant who was responsible for the absence, however clearly expressed, would be a sufficient counterbalance where that witness’s untested statement was the only direct evidence against the applicant.” The Chamber therefore also found a violation of Article 6 § 1 read in conjunction with Article 6 § 3(d) of the Convention in respect of Mr Tahery. B. The parties’ submissions 1. The Government (a) The Chamber’s judgment The Government considered that Article 6 § 3(d) was an express guarantee but not an absolute rule from which no derogation was possible: the focus was on the proceedings as a whole, the safeguards available to the accused, the reliability of the evidence and, in appropriate cases, the interests of witnesses or victims. An inflexible sole or decisive rule was incompatible with the Court’s general approach to Article 6 § 3(d). The Government submitted that the manner in which the Court had applied and developed the test lacked clarity in a number of respects: there had been no adequate discussion of the principle underlying the rule; there had never been a full consideration of whether it should be applied equally to civil and common law jurisdictions; prior to A. M. v. Italy , no. 37019/97, ECHR 1999 IX, there was no support in the Court’s case-law for applying the test in respect of an absent witness whose identity was known to the defence; and it was not possible to discern any consistency in the Court’s case-law as to when evidence was to be regarded as decisive. The Government further submitted that the sole or decisive rule, as applied by the Chamber in its judgment, was predicated on the false assumption that all hearsay evidence which was critical to a case was either unreliable or, in the absence of cross-examination of the witness, incapable of proper assessment. In fact, sole or decisive hearsay might be perfectly reliable and cross-examination might add little or nothing to the assessment of its reliability. The rule could produce arbitrary results; it could operate to exclude evidence simply because it was important, irrespective of its reliability or cogency. The Chamber had not explained whether or how the issue of reliability was relevant to the application of the rule. It had not conducted a full analysis of the safeguards available in England and Wales nor had it appreciated the important difference between common law trial procedures and those of other Contracting States. It had not explained when evidence would be decisive with sufficient precision to enable a trial court to apply the sole or decisive rule in practice nor given adequate consideration to the practical problems which would arise by application of the rule in England and Wales. The test would also serve as an incentive to intimidate witnesses, particularly when only one witness had been courageous enough to come forward. It would have greatest impact in sexual abuse cases, where the offending usually took place in private and thus the evidence of the victim was likely to be “sole or decisive”. For these reasons, the Government invited the Court to adopt the approach taken by the Supreme Court in Horncastle and others . The Supreme Court’s judgment demonstrated that this Court’s case-law permitted a more flexible approach than the apparently hard-edged sole or decisive rule set out by the Chamber. In the light of the Supreme Court’s conclusion in Horncastle and others that the sole or decisive rule would give rise to severe practical difficulties in England and Wales, the Government invited the Court to make clear that the importance of the untested evidence was better regarded as one factor among others which were to be taken into account when deciding whether the proceedings as a whole were fair. Alternatively, the Government suggested that any sole or decisive rule should not apply when there were good reasons for the unavailability of the witness such as those set out in section 23 of the 1988 Act and section 116(2) of the 2003 Act. (b) Domestic law and practice The Government underlined that general safeguards protected the defendant against unfairness as a result of the admission of hearsay evidence. The trial judge acted as “gatekeeper”: he had a duty at common law and under section 78 of the Police and Criminal Evidence Act 1984 to prevent the jury hearing evidence which would have such an adverse effect on proceedings that it ought not to be received. The trial judge had to be satisfied that the prosecution could not adduce the evidence by calling the witness and he had to direct the jury on the dangers of relying on hearsay evidence. The jury had to be satisfied of the defendant’s guilt beyond reasonable doubt and there remained the possibility of appeal to the Court of Appeal, which would grant the appeal if the conviction were unsafe. In the 1988 Act, section 23 made provision for the admissibility of a written statement in certain, enumerated circumstances and it applied equally to the prosecution and defence. The trial judge was required to subject the need to admit the evidence to rigorous scrutiny. Section 25 allowed the judge to refuse to admit the statement if he was of the opinion that it ought not to be admitted in the interests of justice (a reliability and due process safeguard). Section 26, which applied to statements prepared for the purpose of criminal proceedings, required the judge to have regard to the nature, source and contents of the statement, the availability of other evidence, the relevance of the evidence and the extent to which its admission would result in unfairness to the accused. Schedule 2 to the Act also allowed the admission of evidence relating to the credibility and consistency of the maker of the statement. In respect of the 2003 Act, the Government underlined that this had been enacted after a detailed review of the hearsay rule by the Law Commission, which had concluded that many of the assumptions underpinning the traditional exclusion of hearsay were not justified: for example, it had found that hearsay evidence was not necessarily inaccurate or unreliable and that, in many cases, the rule had led to the arbitrary exclusion of cogent evidence. As with the 1989 Act, the 2003 Act allowed both the prosecution and defence to apply to adduce hearsay evidence. Section 116(2)(e) had been included in the 2003 Act in order to tackle crime by providing special measures to protect witnesses who have a genuine fear of intimidation and repercussions. When application was made because the witness was in fear, the trial judge could only give leave when the admission of the statement was in the interests of justice. Moreover, and contrary to many other member States of the Council of Europe, the Act did not allow for the admission of anonymous hearsay evidence: a witness could not be both anonymous and absent. Finally, the Government underlined the further protections provided by sections 124-126 of the Act (see paragraph 45 above). (c) The facts of each case In respect of Mr Al-Khawaja, the Government submitted that the relevant facts were that ST had not been called to give evidence for a justifiable reason (death). The trial judge accepted the admission of her statement was in the interests of justice and, in reaching that conclusion, had taken into account the disadvantages that might be caused to the first applicant. The defence had accepted that they would be in a position to rebut the statement. ST’s statement was not sole or decisive. There was other evidence in the case supporting it, including evidence that she had made a prompt complaint, evidence of her demeanour when she made the complaint and strikingly similar allegations made by other women. The first applicant had been able to cross-examine the other witnesses, give evidence in his own defence and address the jury in relation to the case against him. The jury were carefully directed both in relation to ST’s statement and the burden of proof in the case. The 1988 Act had been properly applied and there was no basis for contradicting the assessment of the trial judge and the Court of Appeal that admission of the statement was fair. In Tahery’s case, T’s fear was a justifiable reason for not calling him. As in the case of the first applicant, the trial judge had concluded that the admission of the evidence was in the interests of justice and, in reaching that conclusion, had taken into account the disadvantages that might be caused to defence. T’s evidence was not sole or decisive: there was other evidence supporting it, including the second applicant’s admission that he was present at the time of the offence, had handled the knife used to stab the victim, had been involved in an altercation with S and had lied to the police. The second applicant had been able to cross-examine the other witness, give evidence in his own defence and address the jury in relation to the case against him. The jury were carefully directed both in relation to T’s statement and the burden of proof in the case. The 2003 Act had been properly applied and there was no basis for contradicting the assessment of the trial judge and the Court of Appeal that admission of the statement was fair. 2. The applicants (a) The Chamber’s judgment The applicants submitted that there were three possible approaches to Article 6 § 3(d). The first was the rigid and literal approach taken by the United States Supreme Court in respect of the similar provisions of the Sixth Amendment, which the applicants did not urge the Court to adopt. The second, the approach taken by the Court, did not treat the words of Article 6 § 3(d) as inflexible but interpreted them as setting a minimum irreducible core of fairness. This approach rested on the sole or decisive rule and was the correct one. The third approach was that taken by the United Kingdom Government and the Supreme Court in Horncastle and others , which, in the applicants’ submission, reduced the guarantees offered by Article 6 § 3 to matters only to be considered in deciding whether an accused had had a fair trial. For these reasons, the applicants submitted that the domestic courts and the Government had not given appropriate weight to the right of a defendant to examine the witnesses against him. That right was not to be dismissed as formalistic or historic; there were a number of reasons why a conviction based solely or decisively on the evidence of an absent witness would be unsafe and unfair. For example, the person’s demeanour could be observed in court; there could be enquiries into the witness’s perception, memory or sincerity, without which there was the possibility of mistakes, exaggerations or deliberate falsehoods going undetected. It was also one thing for a person to make a damaging statement in private and quite another to repeat the statement in the course of a trial. Moreover, the rationale for the rule was clear: in assessing whether, when hearsay evidence had been admitted, the procedures followed had been sufficient to counterbalance the difficulties caused to the defence, due weight had to be given to the extent to which the relevant evidence had been decisive in convicting the applicant. If the evidence was not decisive, the defence was handicapped to a much lesser degree and, if the evidence was decisive, the handicap was much greater. The applicants submitted that the rule had been stated in clear and simple terms in Lucà v. Italy , no. 33354/96, § 40, ECHR 2001 II; it was well-established in the Court’s case-law and a principled, practical and sensible interpretation of Article 6 § 3(d). The applicants submitted that the Supreme Court was incorrect to take the approach it had taken in Horncastle and others . It had erred in finding that the approach taken by the Chamber would lead to the exclusion of evidence that was cogent and demonstrably reliable. There were inherent dangers in admitting untested evidence which appeared to be demonstrably reliable: the law was full of cases where evidence had appeared overwhelming only for it to be shown later that this was not the case. Lord Phillips had suggested that Article 6 § 3(d) was intended to address a weakness in civil law, as opposed to common law systems. However, the applicants submitted that, whatever the historic roots of that Article, English law had moved away from the traditional rule against hearsay. It was also impermissible for different standards to apply to different legal systems: this would undermine the very nature of the Convention system. The applicants also considered that the Supreme Court was wrong to find that there would be practical difficulties in applying the sole or decisive rule in England and Wales. “Decisive” clearly meant evidence that was central to the case and, in applying the rule, the Court itself had shown that it was possible to decide what evidence was sole or decisive. Before the Chamber the Government had had no difficulty in conceding that the evidence of the absent witness in each case was sole or decisive. Trial judges in the United Kingdom routinely assessed complicated factual circumstances and assessed whether legal tests such as “fairness” or “sufficient evidence” had been met. In addition, the test had been included in the Coroners and Justice Act 2009 in respect of anonymous witnesses (see paragraph 46 above). The applicants also observed that the Supreme Court had laid great emphasis on demonstrating that the outcome of cases decided by this Court would have been the same if decided under the 1988 and 2003 Acts. The corollary of this proposition was that the implementation of the sole or decisive rule would lead to the same result in many cases before the domestic courts prior to any such implementation. (b) Domestic law and practice The applicants also submitted that, contrary to the Governments’ submissions and the observations of the Supreme Court in Horncastle and others , the Chamber had understood the procedural safeguards contained in the 1988 and 2003 Acts. Those procedures could only provide adequate protection if they operated as part of a scheme where the core values of Article 6 prevailed; they could not do so until the domestic courts had proper regard to the right of an accused to examine witnesses against him. (c) The facts of each case For the applicants, the sole or decisive nature of the untested evidence in each case meant that no procedures short of questioning the relevant witnesses could have secured a fair trial. In the first applicant’s case, there were obvious questions which could have been posed to ST, not least whether she had heard of VU’s allegations before she made her statement and why she had waited nearly four months before making the statement. In the second applicant’s case, the obvious questions which could have been posed to T were why he had not spoken to the police on the night of the incident, what view he had of the very quick incident, whether he knew the other men at the scene and whether he had any reason to protect them. In the absence of cross-examination, therefore, the procedural safeguards contained in the 1988 and 2003 Acts were insufficient. The fact that the trial judge in each case had applied the “interests of justice” test took the question to be determined no further: that test had to be applied in accordance with Article 6 and whether the trial judges had correctly decided that the admission of each statement was in the interests of justice was, in effect, the issue for the Court to determine. Nor was it of any relevance that the applicants were entitled to introduce evidence challenging the credibility of the absent witnesses: there was no such material available for each witness and the statutory provisions provided no opportunity for challenging the credibility or truthfulness of the witnesses as to the central allegation in each case. The applicants also submitted that no warning from the trial judge could compensate for the lack of opportunity for the jury to see and hear a witness of such importance, nor prevent the danger that the evidence would have a probative value that it did not deserve. The right of appeal against conviction was of no effect since the Court of Appeal in each case had refused to apply the sole or decisive rule as set out in Lucà , cited above. 3. The third party intervener JUSTICE (see paragraph 7 above) submitted that the right to examine witnesses was an ancient one. It was rooted in Roman law and the historic common law right of confrontation. That common law right recognised the dangers of admitting hearsay, including its inherent unreliability. It had, in turn, shaped the fair trial guarantees enshrined in the Sixth Amendment to the United States Constitution, the criminal procedure of all common law countries and international human rights law. Although the hearsay rule had become too rigid in English law, it was incorrect to state, as the Supreme Court had in Horncastle and others , that the Law Commission’s proposed reforms had been “largely implemented” in the 2003 Act. For example, the 2003 Act had included in section 114 a more general discretion to admit hearsay when it was “in the interests of justice” than the Law Commission had envisaged (see paragraphs 42 and 43 above). Moreover, even if the hearsay rule was in need of reform the common law right of confrontation was not: the right enshrined the principle that a person could not offer testimony against a criminal defendant unless it was given under oath, face to face with the defendant and subject to cross examination. . The Supreme Court had also been wrong to suggest that there was no justification for imposing the sole or decisive rule equally on continental and common law jurisdictions. The Supreme Court had placed too much faith in the possibility of counterbalancing measures as a means of overcoming the manifest unfairness of convicting a person wholly or largely on the basis of unchallenged testimony. The essence of the common law right of confrontation lay in the insight that cross-examination was the most effective way of establishing the reliability of a witness’s evidence. It was instructive to remember the warning of Megarry J (in John v. Rees [1970] Ch 345) that “the path of the law [was] strewn about with examples of open and shut cases which, somehow, were not; of unanswerable charges which, in the event, were completely answered”. In addition JUSTICE submitted that it was, as Sedley LJ had observed in Secretary of State for the Home Department v. AF and others [2008] EWCA Civ 1148 , “seductively easy to conclude that there can be no answer to a case of which you have only heard one side”. It was for this reason that the paradox referred to by Lord Phillips was not a paradox at all. Hearsay did not become more cogent simply because it might be the sole or substantive evidence of a person’s guilt. The need for the exclusion of apparently “decisive” hearsay evidence was justified by the centuries old fear of common law judges that juries might give it undue weight. The sole or decisive rule was not difficult to apply in English criminal procedure as trial judges already had to consider the potential consequences of admitting evidence in various statutes, including parts of the 2003 Act. Section 125(1) for example, obligated a trial judge to direct an acquittal where the case against an accused was based “wholly or partly” on unconvincing hearsay. JUSTICE accepted that some clarification of what was meant by “decisive” was required but considered that the test was still workable and indeed had been applied in England and Wales in the context of witness anonymity orders (see the Coroners and Justice Act 2009 at paragraph 46 above). In the context of criminal proceedings “decisive” should be understood narrowly: it should not mean merely “capable of making a difference” but instead “likely to be determinative or conclusive”. Finally, the Supreme Court’s analysis of comparative law was, at best, a partial one. On the available evidence (see, inter alia , paragraphs 63 87 above), it was simply not possible to conclude, as the Supreme Court had, that other common law jurisdictions would find a criminal conviction which was based solely or decisively on hearsay evidence to be compatible with the right to a fair trial. C. The Grand Chamber’s assessment 1. The general principles The Court recalls that the guarantees in paragraph 3(d) of Article 6 are specific aspects of the right to a fair hearing set forth in paragraph 1 of this provision which must be taken into account in any assessment of the fairness of proceedings. In addition, the Court’s primary concern under Article 6 § 1 is to evaluate the overall fairness of the criminal proceedings (see, as a recent authority, Taxquet v. Belgium [GC], no. 926/05 , § 84, 16 November 2010, with further references therein ). In making this assessment the Court will look at the proceedings as a whole having regard to the rights of the defence but also to the interests of the public and the victims that crime is properly prosecuted (see Gäfgen v. Germany [GC], no. 22978/05, § 175, ECHR 2010 ....) and, where necessary, to the rights of witnesses (see, amongst many authorities, Doorson , cited above, § 70). It is also recalled in this context that the admissibility of evidence is a matter for regulation by national law and the national courts and that the Court’s only concern is to examine whether the proceedings have been conducted fairly (see Gäfgen , cited above, § 162, and references therein). Article 6 § 3(d) enshrines the principle that, before an accused can be convicted, all evidence against him must normally be produced in his presence at a public hearing with a view to adversarial argument. Exceptions to this principle are possible but must not infringe the rights of the defence, which, as a rule, require that the accused should be given an adequate and proper opportunity to challenge and question a witness against him, either when that witness makes his statement or at a later stage of proceedings (see Lucà , cited above, § 39; Solakov v. “the former Yugoslav Republic of Macedonia” , no. 47023/99, § 57, ECHR 2001 X). A similar and long-established principle exists in the common law of England and Wales (see Lord Bingham’s observation at paragraph 5 of R. v. Davis, summarised at paragraph 49 above). Having regard to the Court’s case-law, there are two requirements which follow from the above general principle. First, there must be a good reason for the non-attendance of a witness. Second, when 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 to have examined, whether during the investigation or at the trial, the rights of the defence may be restricted to an extent that is incompatible with the guarantees provided by Article 6 (the so-called “sole or decisive rule”). The Court will examine below whether the latter rule is to be considered as an absolute rule whose breach automatically leads to a finding that the proceedings have not been fair in violation of Article 6 § 1 of the Convention. 2. Whether there is a good reason for the non-attendance of a witness The requirement that there be a good reason for admitting the evidence of an absent witness is a preliminary question which must be examined before any consideration is given as to whether that evidence was sole or decisive. Even where the evidence of an absent witness has not been sole or decisive, the Court has still found a violation of Article 6 §§ 1 and 3(d) when no good reason has been shown for the failure to have the witness examined (see, for example, in Lüdi v. Switzerland , 15 June 1992, Series A no. 238 , Mild and Virtanen v. Finland , no. 39481/98 and 40227/98, 26 July 2005; Bonev v. Bulgaria , no. 60018/00, 8 June 2006; and Pello v. Estonia , no. 11423/03, 12 April 2007). This is because as a general rule witnesses should give evidence during the trial and that all reasonable efforts will be made to secure their attendance. Thus, when witnesses do not attend to give live evidence, there is a duty to enquire whether that absence is justified. There are a number of reasons why a witness may not attend trial but, in the present cases, it is only necessary to consider absence owing to death or fear. It is plain that, where a witness has died, in order for his or her evidence to be taken into account, it will be necessary to adduce his or her witness statement (see, for example, Ferrantelli and Santangelo v. Italy , 7 August 1996, § 52, Reports of Judgments and Decisions 1996 III; Mika v. Sweden (dec.), § 37, no. 31243/06, 27 January 2009). Absence owing to fear calls for closer examination. A distinction must be drawn between two types of fear: fear which is attributable to threats or other actions of the defendant or those acting on his or her behalf and fear which is attributable to a more general fear of what will happen if the witness gives evidence at trial. When a witness’s fear is attributable to the defendant or those acting on his behalf, it is appropriate to allow the evidence of that witness to be introduced at trial without the need for the witness to give live evidence or be examined by the defendant or his representatives – even if such evidence was the sole or decisive evidence against the defendant. To allow the defendant to benefit from the fear he has engendered in witnesses would be incompatible with the rights of victims and witnesses. No court could be expected to allow the integrity of its proceedings to be subverted in this way. Consequently, a defendant who has acted in this manner must be taken to have waived his rights to question such witnesses under Article 6 § 3(d). The same conclusion must apply when the threats or actions which lead to the witness being afraid to testify come from those who act on behalf of the defendant or with his knowledge and approval. In the Horncastle and others case the Supreme Court observed that it was notoriously difficult for any court to be certain that a defendant had threatened a witness. The Court does not underestimate the difficulties which may arise in determining whether, in a particular case, a defendant or his associates have been responsible for threatening or directly inducing fear in a witness. However, the case Tahery itself shows that, with the benefit of an effective inquiry, such difficulties are not insuperable. The Court’s own case law shows that it is more common for witnesses to have a general fear of testifying without that fear being directly attributable to threats made by the defendant or his agents. For instance, in many cases, the fear has been attributable to the notoriety of a defendant or his associates (see, for example, Dzelili v. Germany (dec.), no. 15065/05, 29 September 2009). There is, therefore, no requirement that a witness’s fear be attributable directly to threats made by the defendant in order for that witness to be excused from giving evidence at trial. Moreover, fear of death or injury of another person or of financial loss are all relevant considerations in determining whether a witness should not be required to give oral evidence. This does not mean, however, that any subjective fear of the witness will suffice. The trial court must conduct appropriate enquiries to determine first, whether or not there are objective grounds for that fear, and, second, whether those objective grounds are supported by evidence (see, for example, Krasniki v. the Czech Republic , no. 51277/99, §§ 80–83, 28 February 2006, where the Court was not satisfied that the domestic courts had carried out an examination of the reasons for the witnesses’ fear before granting them anonymity). Finally, given the extent to which the absence of a witness adversely affects the rights of the defence, the Court would emphasise that, when a witness has not been examined at any prior stage of the proceedings, allowing the admission of a witness statement in lieu of live evidence at trial must be a measure of last resort. Before a witness can be excused from testifying on grounds of fear, the trial court must be satisfied that all available alternatives, such as witness anonymity and other special measures, would be inappropriate or impracticable. 3. The sole or decisive rule (a) General Considerations The Court notes that the present cases concern only absent witnesses whose statements were read at trial. It is not the Court’s task to consider the operation of the common law rule against hearsay in abstracto nor to consider generally whether the exceptions to that rule which now exist in English criminal law are compatible with the Convention. As the Court has reiterated (at paragraph 118 above), Article 6 does not lay down any rules on the admissibility of evidence as such, which is primarily a matter for regulation under national law. The Court further observes that, in the present cases, it is not concerned with testimony that is given at trial by witnesses whose identity is concealed from the accused (anonymous testimony). While the problems raised by anonymous and absent witnesses are not identical, the two situations are not different in principle, since, as was acknowledged by the Supreme Court, each results in a potential disadvantage for the defendant. The underlying principle is that the defendant in a criminal trial should have an effective opportunity to challenge the evidence against him. This principle requires not merely that a defendant should know the identity of his accusers so that he is in a position to challenge their probity and credibility but that he should be able to test the truthfulness and reliability of their evidence, by having them orally examined in his presence, either at the time the witness was making the statement or at some later stage of the proceedings. The seeds of the sole or decisive rule are to be found in the Unterpertinger v Austria judgment, 24 November 1986, § 33, Series A no. 110, which also provides the rationale for the test to be applied: if the conviction of a defendant is solely or mainly based on evidence provided by witnesses whom the accused is unable to question at any stage of the proceedings, his defence rights are unduly restricted. As was pointed out by the Supreme Court, in the earlier case-law of the Court, where the rule had been adumbrated in cases of absent or anonymous witnesses, the violation found of Article 6 § 3(d) had been based at least in part on the fact that there was no justification for not calling or identifying the witness in question. It was in Doorson , cited above, that the Court first held that, even in a case where there was a justification for the failure to call a witness, a conviction based solely or to a decisive extent on evidence of that witness would be unfair. (b) Objections to the sole or decisive rule Drawing on the judgment of the Supreme Court in Horncastle and others , the Government challenge the sole or decisive rule, or its application by the Chamber in the present cases, on four principal grounds. First, it is argued that the common law, through its rules of evidence outlawing the admission of hearsay, which long preceded the coming into force of the Convention, protected that aspect of fair trial that Article 6 § 3(d) was designed to ensure, without the necessity of applying a sole or decisive rule. By contrast, civil law countries had no such rules of evidence. Therefore, much of the impact of Article 6 § 3(d) was on procedures in continental systems which previously allowed an accused person to be convicted on the basis of evidence from witnesses whom he had not had an opportunity to challenge. Second, it is said that its application gives rise to practical difficulties, in that the Chamber did not explain when evidence would be decisive with sufficient precision to enable a trial court to apply the sole or decisive rule in practice. Equally, no adequate consideration has been given to the practical problems which would arise by the application of the rule in a common law system such as that of England and Wales. Next, it is said that there has been no adequate discussion of the principle underlying the rule, which is predicated on the false assumption that all hearsay evidence which is critical to a case is either unreliable or, in the absence of cross-examination of the witness, incapable of proper assessment. Finally, it is said that the Chamber applied the rule with excessive rigidity and that it failed to conduct a full analysis of the safeguards available in England and Wales or to appreciate the important difference between common law trial procedures and those of other Contracting States. The Court will address each of these arguments in turn. As to the first argument, the Court accepts that the sole or decisive rule may have been developed in the context of legal systems which permitted a defendant to be convicted on evidence of witnesses whom he did not have an opportunity to challenge, a situation which would not have arisen if the strict common law rule against hearsay evidence had been applied. However, the Court notes that the present cases have arisen precisely because the legal system in England and Wales has abandoned the strict common law rule against hearsay evidence. Exceptions to the rule have been created, notably in the 1988 and 2003 Acts, which allowed for admission of ST’s statement in Al-Khawaja’s case and T’s statement in Tahery’s case (see paragraphs 41 and 44 above). The Court recognises that these dilutions of the strict rule against hearsay have been accompanied by statutory safeguards and, accordingly, the central question in the present cases is whether the application of these safeguards was sufficient to secure the applicants’ rights under Article 6 §§ 1 and 3(d). Against this background, while it is important for the Court to have regard to substantial differences in legal systems and procedures, including different approaches to the admissibility of evidence in criminal trials, ultimately it must apply the same standard of review under Article 6 §§ 1 and 3(d), irrespective of the legal system from which a case emanates. Second, as to the alleged lack of precision of the rule, the Court notes that the word “sole”, in the sense of the only evidence against an accused (see, for example, Saïdi v. France , 20 September 1993, Series A no. 261 C), does not appear to have given rise to difficulties, the principal criticism being directed to the word “decisive”. “Decisive” (or “déterminante”) in this context means more than “probative”. It further means more than that, without the evidence, the chances of a conviction would recede and the chances of an acquittal advance, a test which, as the Court of Appeal in Horncastle and others pointed out (see paragraph 54 above), would mean that virtually all evidence would qualify. Instead, the word “decisive” should be narrowly understood as indicating evidence of such significance or importance as is likely to be determinative of the outcome of the case. Where the untested evidence of a witness is supported by other corroborative evidence, the assessment of whether it is decisive will depend on the strength of the supportive evidence; the stronger the corroborative evidence, the less likely that the evidence of the absent witness will be treated as decisive. It is also argued that the sole or decisive rule cannot be applied without excessive practical difficulties in a common law system. In Horncastle and others, the Supreme Court observed that a duty not to treat a particular piece of evidence as decisive was hard enough for a professional judge to discharge, but that if the rule were to be applied in the context of a jury trial the only practical way to apply it would be as a rule of admissibility: the trial judge would have to rule inadmissible any witness statement capable of proving decisive, which would be no easy task. As for the Court of Appeal or the European Court, it was said that it would often be impossible to decide whether a particular statement was the sole or decisive basis of a conviction in the absence of reasons being given for a jury’s verdict. The Court accepts that it might be difficult for a trial judge in advance of a trial to determine whether evidence would be decisive without having the advantage of examining and weighing in the balance the totality of evidence that has been adduced in the course of the trial. However, once the prosecution has concluded its case, the significance and weight of the untested evidence can be assessed by the trial judge against the background of the other evidence against the accused. In common law systems, at the conclusion of the prosecution case, trial judges are frequently asked to consider whether there is a case to answer against the accused. As part of that process they are often asked to assess the strength and reliability of the evidence for the prosecution. Indeed, the Court notes that section 125 of the 2003 Act expressly requires the trial judge to stop the case if, considering its importance to the case against the defendant, the hearsay evidence is so unconvincing that a conviction would be unsafe. The Court is further not persuaded that an appellate court in a common law system, where a jury gives no reasons for its verdict, will be unable to determine whether untested evidence was the sole or decisive evidence in the conviction of the defendant. Appellate judges are commonly required to consider whether evidence was improperly admitted at trial and, if it was, whether the conviction is still safe. In doing so, they must consider inter alia the significance of the impugned evidence to the prosecution’s case and the extent to which it prejudiced the rights of the defence. An appellate court is thus well placed to consider whether untested evidence could be considered to be the sole or decisive evidence against the accused and whether the proceedings as a whole were fair. The Court observes that the comparative materials before it support this conclusion as regards the application of the rule in different common law jurisdictions (see paragraphs 63–87 above and, in particular, the approach of the Scottish High Court of Justiciary). The Court also notes in this context that, in the case of R. v. Davis (see paragraphs 49 and 50 above), the House of Lords appeared to foresee no apparent difficulty in the application of the sole or decisive rule in the context of anonymous witnesses. Lord Bingham observed that a conviction which was based solely or to a decisive extent on statements or testimony of anonymous witnesses resulted from a trial which could not be regarded as fair and that “this [was] the view traditionally taken by the common-law of England” (see paragraph 25 of the Davis judgment). The House of Lords concluded in the Davis case that not only had the evidence of the anonymous witness been the sole or decisive basis on which the defendant could have been convicted but that effective cross-examination had been hampered. The decision in the Davis case led to the introduction into the Coroners and Justice Act 2009 of the very requirement that, in deciding whether to make a witness anonymity order, a judge must have regard, inter alia , to whether evidence given by the witness might be the sole or decisive evidence implicating the defendant (see paragraph 46 above). The Court further notes in this regard that, in the context of the drawing of adverse inferences from a defendant’s silence, the Court has applied the rule that it would be incompatible with the right to silence to base a conviction solely or mainly on the accused’s silence or on a refusal to answer questions or to give evidence himself (see John Murray v. the United Kingdom , 8 February 1996, § 47, Reports of Judgments and Decisions 1996 I; Condron v. the United Kingdom , no. 35718/97, § 56, ECHR 2000 V). The Court similarly cannot accept the third argument that the sole or decisive rule is predicated on the assumption that all hearsay evidence which is crucial to a case is unreliable or incapable of proper assessment unless tested in cross-examination. Rather, it is predicated on the principle that the greater the importance of the evidence, the greater the potential unfairness to the defendant in allowing the witness to remain anonymous or to be absent from the trial and the greater the need for safeguards to ensure that the evidence is demonstrably reliable or that its reliability can properly be tested and assessed. In Kostovski v. the Netherlands , 20 November 1989, Series A no. 166, where it was accepted that the applicant’s conviction was based to a decisive extent on the evidence of both anonymous and absent witnesses, the Court emphasised (at § 42 of its judgment): “ Testimony or other declarations 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.” The Court further found that while the trial courts had observed caution in evaluating the statements in question, this could scarcely be regarded as a proper substitute for direct observation. It thus concluded that the use of the evidence involved limitations on the rights of the defence which were irreconcilable with the guarantees contained in Article 6. In the subsequent case of Doorson , cited above, the Court observed that the anonymity of two witnesses in the case presented the defence with “difficulties which criminal proceedings should not involve” but that no violation could be found if the handicaps under which the defence laboured were sufficiently counterbalanced by the procedures followed (§ 72 of the judgment). It then considered that, in contrast to Kostovski , the defence was able to challenge the reliability of the anonymous witnesses (§§ 73 and 75). Moreover, even after its statement that a conviction should not be based either solely or to a decisive extent on anonymous witnesses, the Court emphasised that “evidence obtained from witnesses under conditions in which the rights of the defence cannot be secured to the extent normally required by the Convention should be treated with extreme care” (§ 76). With respect to the Government’s final argument, the Court is of the view that the two reasons underpinning the sole or decisive rule that were set out in the Doorson judgment remain valid. For the first, the Court finds no reason to depart from its finding in Kostovski , that inculpatory evidence against an accused may well be “designedly untruthful or simply erroneous”. Moreover, unsworn statements by witnesses who cannot be examined often appear on their face to be cogent and compelling and it is, as Lord Justice Sedley pointed out, “seductively easy” to conclude that there can be no answer to the case against the defendant (see paragraph 115 above). Experience shows that the reliability of evidence, including evidence which appears cogent and convincing, may look very different when subjected to a searching examination. The dangers inherent in allowing untested hearsay evidence to be adduced are all the greater if that evidence is the sole or decisive evidence against the defendant. As to the second reason, the defendant must not be placed in the position where he is effectively deprived of a real chance of defending himself by being unable to challenge the case against him. Trial proceedings must ensure that a defendant’s Article 6 rights are not unacceptably restricted and that he or she remains able to participate effectively in the proceedings (see T. v. the United Kingdom [GC], no. 24724/94, § 83, 16 December 1999; Stanford v. the United Kingdom , 23 February 1994, § 26, Series A no. 282 A). The Court’s assessment of whether a criminal trial has been fair cannot depend solely on whether the evidence against the accused appears prima facie to be reliable, if there are no means of challenging that evidence once it is admitted. For these reasons, the Court has consistently assessed the impact that the defendant’s inability to examine a witness has had on the overall fairness of his trial. It has always considered it necessary to examine the significance of the untested evidence in order to determine whether the defendant’s rights have been unacceptably restricted (see, as early examples, Unterpertinger , cited above; and Bricmont v. Belgium , 7 July 1989, Series A no. 158 and, more recently, Kornev and Karpenko v. Ukraine , no. 17444/04 , §§ 54-57, 21 October 2010; Caka v. Albania , no. 44023/02, §§ 112-116, 8 December 2009; Guilloury v. France , no. 62236/00, §§ 57-62, 22 June 2006; and A. M. , Krasniki , Lucà , and Saïdi , all cited above). The Commission had taken the same approach in its case-law (see, among the earliest cases of the Commission, X. v. Austria , no. 4428/70, decision of 1 June 1972, Collection 40, p. 1; X v. Belgium , no. 8417/78, decision of 4 May 1979, D. R. 16, p. 205; X. v. the Federal Republic of Germany , no. 8414/78, decision of 4 July 1979, DR. 17, p. 231; and S. v. the Federal Republic of Germany , no. 8945/80, decision of 13 December 1983, D. R. 39, p. 43). At the same time, however, the Court has always interpreted Article 6 § 3 in the context of an overall examination of the fairness of the proceedings. (see, as a recent authority, Salduz v. Turkey [GC], no. 36391/02, § 50, 27 November 2008). Traditionally, when examining complaints under Article 6 § 1, the Court has carried out its examination of the overall fairness of the proceedings by having regard to such factors as the way in which statutory safeguards have been applied, the extent to which procedural opportunities were afforded to the defence to counter handicaps that it laboured under and the manner in which the proceedings as a whole have been conducted by the trial judge (see, for example, John Murray v. the United Kingdom , cited above). Also, in cases concerning the withholding of evidence from the defence in order to protect police sources, the Court has left it to the domestic courts to decide whether the rights of the defence should cede to the public interest and has confined itself to verifying whether the procedures followed by the judicial authorities sufficiently counterbalance the limitations on the defence with appropriate safeguards. The fact that certain evidence was not made available to the defence was not considered automatically to lead to a violation of Article 6 § 1 (see, for example, Rowe and Davis v. the United Kingdom [GC], no. 28901/95, ECHR 2000 II). Similarly, in the case of Salduz , cited above, § 50, the Court reiterated that the right to legal assistance, set out in Article 6 § 3 (c) was one element, amongst others, of the concept of a fair trial in criminal proceedings contained in Article 6 § 1. The Court is of the view that the sole or decisive rule should also be applied in a similar manner. It would not be correct, when reviewing questions of fairness, to apply this rule in an inflexible manner. Nor would it be correct for the Court to ignore entirely the specificities of the particular legal system concerned and, in particular its rules of evidence, notwithstanding judicial dicta that may have suggested otherwise (see, for instance, Lucà , cited above, at § 40). To do so would transform the rule into a blunt and indiscriminate instrument that runs counter to the traditional way in which the Court approaches the issue of the overall fairness of the proceedings, namely to weigh in the balance the competing interests of the defence, the victim, and witnesses, and the public interest in the effective administration of justice. (c) General conclusion on the sole or decisive rule The Court therefore concludes that, where a hearsay statement is the sole or decisive evidence against a defendant, its admission as evidence will not automatically result in a breach of Article 6 § 1. At the same time where a conviction is based solely or decisively on the evidence of absent witnesses, the Court must subject the proceedings to the most searching scrutiny. Because of the dangers of the admission of such evidence, it would constitute a very important factor to balance in the scales, to use the words of Lord Mance in R. v. Davis (see paragraph 50 above) , and one which would require sufficient counterbalancing factors, including the existence of strong procedural safeguards. The question in each case is whether there are sufficient counterbalancing factors in place, including measures that permit a fair and proper assessment of the reliability of that evidence to take place. This would permit a conviction to be based on such evidence only if it is sufficiently reliable given its importance in the case. (d) Procedural safeguards contained in the 1988 and 2003 Acts The Court will therefore examine the counterbalancing measures in place in English law at the relevant time. The Court observes that under the terms of the 1988 and 2003 Acts the absence of witnesses must be justified and fall within one of the defined categories (see section 23 of the 1988 Act and section 116 of the 2003 Act at paragraphs 41 and 44 above). Whatever the reasons for the absence of a witness, the admission of statements of a witness who is not only absent but anonymous is not admissible. Moreover, where the absence is due to fear, under the 2003 Act, the trial judge may only give leave if he considers the admission of the statement to be in the interests of justice and he must decide whether special measures can be put in place to allow the witness to give live evidence. In such a case, the trial judge must have regard to the difficulty faced by the defendant in challenging the witness statement if the maker of the statement is not called. The 2003 Act also provides that, whatever the reason for the absence of a witness, evidence relevant to the credibility or consistency of the maker of the statement may be admitted even where the evidence would not have been admissible had the witness given evidence in person. The trial judge retains a specific discretion to refuse to admit a hearsay statement if satisfied that the case for its exclusion substantially outweighs the case for admitting it. Of particular significance is the requirement under the 2003 Act that the trial judge should stop the proceedings if satisfied at the close of the case for the prosecution that the case against the accused is based “wholly or partly” on a hearsay statement admitted under the 2003 Act, provided he or she is also satisfied that the statement in question is so unconvincing that, considering its importance to the case against the accused, a conviction would be unsafe. The Court also notes that, in addition to the safeguards contained in each Act, section 78 of the Police and Criminal Evidence Act 1984 provides a general discretion to exclude evidence if its admission would have such an adverse effect on the fairness of the trial that it ought not to be admitted. Finally, the common law requires a trial judge to give the jury the traditional direction on the burden of proof, and direct them as to the dangers of relying on a hearsay statement. The Court considers that the safeguards contained in the 1988 and 2003 Acts, supported by those contained in section 78 of the Police and Criminal Evidence Act and the common law, are, in principle, strong safeguards designed to ensure fairness. It remains to be examined how these safeguards were applied in the present cases. 4. The present cases In turning to the present cases, the Court begins by observing that, in the course of the hearing before the Grand Chamber, a question was put to the parties as to whether it was accepted that the testimony of ST was the sole or decisive evidence in respect of Mr Al-Khawaja and that the testimony of T was the sole or decisive evidence in respect of Mr Tahery. In reply to that question, the Government departed from the position they had taken before the Chamber and submitted that neither ST nor T’s testimony was sole or decisive (see paragraphs 104 and 105 above). The Court will therefore consider three issues in each case: first, whether it was necessary to admit the witness statements of ST or T; second, whether their untested evidence was the sole or decisive basis for each applicant’s conviction; and third, whether there were sufficient counterbalancing factors including strong procedural safeguards to ensure that each trial, judged as a whole, was fair within the meaning of Article 6 §§ 1 and 3(d). (a) Al-Khawaja’s case The Court observes that it is not in dispute that ST’s death made it necessary to admit her witness statement if her evidence was to be considered. Notwithstanding the Government’s submission that ST’s statement was not sole or decisive because there was other evidence supporting it, the Court notes that the judge who admitted ST’s statement was well placed to evaluate its significance. He was quite clear when he observed “no statement, no count one” (see paragraph 13 above). It is not for the Court, so far removed from the trial proceedings, to gainsay such an evaluation. The Court is therefore compelled to conclude that ST’s statement was decisive. Nevertheless, as the Court has indicated (see paragraph 147 above) the admission of the statement in evidence cannot be considered as conclusive as to the unfairness of the trial, but as a very important factor to be placed in the balance alongside the procedural safeguards noted above and other counterbalancing factors present in the case. The interests of justice were obviously in favour of admitting in evidence the statement of ST, which was recorded by the police in proper form. The reliability of the evidence was supported by the fact that ST had made her complaint to two friends, BF and SH, promptly after the events in question, and that there were only minor inconsistencies between her statement and the account given by her to the two friends, who both gave evidence at the trial. Most importantly, there were strong similarities between ST’s description of the alleged assault and that of the other complainant, VU, with whom there was no evidence of any collusion. In a case of indecent assault by a doctor on his patient, which took place during a private consultation where only he and the victim were present, it would be difficult to conceive of stronger corroborative evidence, especially when each of the other witnesses was called to give evidence at trial and their reliability was tested by cross-examination. It is true that the judge’s direction to the jury was found to be deficient by the Court of Appeal. However, the Court of Appeal also held that it must have been clear to the jury from that direction that, in consequence of the applicant’s inability to cross-examine ST and the fact that they were unable to see and hear her, her statement should carry less weight with them (see paragraph 22 above). Having regard to this direction, and the evidence offered by the prosecution in support of ST’s statement, the Court considers that the jury were able to conduct a fair and proper assessment of the reliability of ST’s allegations against the first applicant. Against this background, and viewing the fairness of the proceedings as a whole, the Court considers that, notwithstanding the difficulties caused to the defence by admitting the statement and the dangers of doing so, there were sufficient counterbalancing factors to conclude that the admission in evidence of ST’s statement did not result in a breach of Article 6 § 1 read in conjunction with Article 6 § 3 (d) of the Convention. (b) Tahery’s case The Court considers that appropriate enquiries were made to determine whether there were objective grounds for T’s fear. The trial judge heard evidence from both T and a police officer as to that fear. The trial judge was also satisfied that special measures, such as testifying behind a screen, would not allay T’s fears. Even though T’s identity as the maker of the incriminating statement was publicly disclosed, the conclusion of the trial judge that T had a genuine fear of giving oral evidence and was not prepared to do so even if special measures were introduced in the trial proceedings, provides a sufficient justification for admitting T’s statement. The Court notes that when those present at the scene of the stabbing were originally interviewed, no-one claimed to have seen the applicant stab S, and S himself had not seen who had stabbed him, although initially he presumed it was the second applicant. T had made his statement implicating the applicant two days after the event. He was the only witness who had claimed to see the stabbing. His uncorroborated eyewitness statement was, if not the sole, at least the decisive evidence against the applicant for that reason. It was obviously evidence of great weight and without it the chances of a conviction would have significantly receded. Even though the testimony may have been coherent and convincing on its face it cannot be said to belong to the category of evidence that can be described as “demonstrably reliable” such as a dying declaration or other examples given by the Court of Appeal and Supreme Court in their Horncastle and others judgments (see paragraphs 53 and 60 above). Such untested evidence weighs heavily in the balance and requires sufficient counterbalancing factors to compensate for the consequential difficulties caused to the defence by its admission. Reliance is placed by the Government on two main counterbalancing factors: the fact that the trial judge concluded that no unfairness would be caused by the admission of T’s statement since the applicant was in a position to challenge or rebut the statement by giving evidence himself or calling other witnesses who were present, one of whom was his uncle; and the warning given by the trial judge to the jury that it was necessary to approach the evidence given by the absent witness with care. However, the Court considers that neither of these factors, whether taken alone or in combination, could be a sufficient counterbalance to the handicap under which the defence laboured. Even if he gave evidence denying the charge, the applicant was, of course, unable to test the truthfulness and reliability of T’s evidence by means of cross-examination. The fact is that T was the sole witness who was apparently willing or able to say what he had seen. The defence was not able to call any other witness to contradict the testimony provided in the hearsay statement. The other evidence was that given by the victim S who did not know who had stabbed him, although initially he presumed it was the applicant. His evidence was circumstantial in nature and largely uncontested by the applicant. He gave evidence of the fight and the applicant’s actions after the stabbing (see paragraph 32 above). While this evidence corroborated some of the details of T’s testimony, it could only provide at best indirect support for the claim by T that it was the applicant who had stabbed S. It is true that the direction in the judge’s summing up to the jury was both full and carefully phrased, drawing attention to the dangers of relying on untested evidence. However the Court does not consider that such a warning, however clearly or forcibly expressed, could be a sufficient counterbalance where an untested statement of the only prosecution eyewitness was the only direct evidence against the applicant. The Court therefore considers that the decisive nature of T’s statement in the absence of any strong corroborative evidence in the case meant the jury in this case were unable to conduct a fair and proper assessment of the reliability of T’s evidence. Examining the fairness of the proceedings as a whole, the Court concludes that there were not sufficient counterbalancing factors to compensate for the difficulties to the defence which resulted from the admission of T’s statement. It therefore finds that there has been a violation of Article 6 § 1 read in conjunction with Article 6 § 3 (d) of the Convention in respect of Mr Tahery. II. APPLICATION OF ARTICLE 41 OF THE CONVENTION Article 41 of the Convention provides: “ If the Court finds that there has been a violation of the Convention or the Protocols thereto, and if the internal law of the High Contracting Party concerned allows only partial reparation to be made, the Court shall, if necessary, afford just satisfaction to the injured party.” As no violation has been found in respect of the first applicant, just satisfaction need only be considered in respect of the second applicant. A. Damage The second applicant maintained his submission that, unlike the majority of Article 6 cases, it could properly be said that his conviction for the offence in question and his subsequent sentence of imprisonment would not have occurred had it not been for the violation. Based on the additional time he served in prison and with reference to comparable domestic awards for unlawful detention, the second applicant claimed GBP 65,000 (approximately EUR 83,830). The Government maintained their submissions that there was no causal connection between the alleged violation and the conviction of the applicant, that domestic case-law was of limited relevance and that any finding of a violation amounted to sufficient just satisfaction. Alternatively, a much lower sum, such as the 6,000 EUR awarded in Visser v. the Netherlands , no. 26668/95, § 56, 14 February 2002, would be appropriate. In its judgment, the Chamber accepted that domestic case-law was of limited relevance to the question of non-pecuniary damage in proceedings before it. It found, however, that the criminal proceedings against the applicant, at least in respect of the charge which based on T’s statement, was not conducted in conformity with the Convention and that the applicant had inevitably have suffered a degree of distress and anxiety as a result. It awarded the sum of EUR 6,000 by way of compensation for non-pecuniary damage. The Grand Chamber sees no reason to depart from the Chamber’s assessment and accordingly makes the same award of EUR 6,000. B. Costs and expenses Before the Chamber, the second applicant claimed a total of GBP 7,995 for costs and expenses, which is approximately EUR 9,079. This comprised GBP 5,571.47 (inclusive of VAT) for approximately forty-five hours’ work by Ms Trowler, which included attendance at the Chamber hearing and travelling time to Strasbourg. Solicitor’s costs and expenses were GBP 2,423.56 (inclusive of VAT) which covered costs of GBP 1,734.16 for approximately fifteen hours’ work and GBP 689.40 in expenses. Before the Grand Chamber, the second applicant claimed a total of GBP 3,614.82 in respect of solicitor’s costs and expenses, which covered proceedings before the Chamber and Grand Chamber. He also claimed an additional GBP 3,643 (inclusive of VAT) for a further thirty-seven hours’ work by Ms Trowler. This included seventeen hours’ work to cover her anticipated attendance at the Grand Chamber hearing and travelling time to Strasbourg. She was, in the event, unable to attend the hearing. The Government considered the hourly rates which had been charged to be excessive. The Court finds the amounts claimed are not excessive in light of the complexity of the case. Thus, with the exception of the seventeen hours’ attendance and travelling time to the Grand Chamber hearing (which were not actually incurred), the remainder of the second applicant’s costs and expenses should be met in full. It therefore awards EUR 13,150, inclusive of VAT, less EUR 1,150 already received in legal aid from the Council of Europe, to be converted into pounds sterling on the date of settlement. C. Default interest The Court considers it appropriate that the default interest should be based on the marginal lending rate of the European Central Bank, to which should be added three percentage points. FOR THESE REASONS, THE COURT Holds , by fifteen votes to two, that there has been no violation of Article 6 § 1 read in conjunction with Article 6 § 3(d) of the Convention in respect of Mr Al-Khawaja; Holds , unanimously, that there has been a violation of Article 6 § 1 read in conjunction with Article 6 § 3(d) of the Convention in respect of Mr Tahery; Holds , unanimously, (a) that the respondent State is to pay the second applicant, Mr Tahery, within three months, the following amounts, to be converted into pounds sterling at the rate applicable at the date of settlement: (i) EUR 6,000 (six thousand euros), plus any tax that may be chargeable, in respect of non-pecuniary damage; (ii) EUR 12,000 (twelve thousand euros), plus any tax that may be chargeable to the second applicant, in respect of costs and expenses; (b) that from the expiry of the above-mentioned three months until settlement simple interest shall be payable on the above amounts at a rate equal to the marginal lending rate of the European Central Bank during the default period plus three percentage points; Dismisses , unanimously, the remainder of the applicants’ claim for just satisfaction. Done in English and in French, and delivered at a public hearing in the Human Rights Building, Strasbourg, on 15 December 2011. Michael O’Boyle Françoise Tulkens Deputy Registrar President In accordance with Article 45 § 2 of the Convention and Rule 74 § 2 of the Rules of Court, the following separate opinions are annexed to this judgment: (a) Concurring opinion of Judge Bratza; (b) Joint partly concurring and partly dissenting opinion of Judges Karakaş and Sajó. F. T. M. O. B. CONCURRING OPINION OF JUDGE BRATZA In his judgment in the case of R. v. Horncastle and others , with which the other members of the Supreme Court agreed, Lord Phillips declined to apply this Court’s “sole or decisive test” rather than the provisions of the Criminal Justice Act 2003. Those provisions, in his view, struck the right balance between the imperative that a trial had to be fair and the interests of victims in particular and society in general that a criminal should not be immune from conviction where a witness, who had given critical evidence in a statement that could be shown to be reliable, had died or could not be called to give evidence for some other reason. While stating that he had, in reaching this view, taken careful account of the jurisprudence of the Court, Lord Phillips concluded by expressing the hope that “in due course the Strasbourg Court may also take account of the reasons that have led me not to apply the sole or decisive test in this case” ( Horncastle judgment, § 108). The present case affords, to my mind, a good example of the judicial dialogue between national courts and the European Court on the application of the Convention to which Lord Phillips was referring. The Horncastle case was decided by the Supreme Court after delivery of the judgment of the Chamber in the present case, to which I was a party, and it was, in part, in order to enable the criticisms of that judgment to be examined that the Panel of the Grand Chamber accepted the request of the respondent Government to refer the case to the Grand Chamber. As the national judge in a case brought against the United Kingdom, I had the uncomfortable duty under the Convention of sitting and voting again in the Grand Chamber. The judgment of the Grand Chamber, in which I concur, not only takes account of the views of the Supreme Court on the sole or decisive test and its application by the Chamber but re-examines the safeguards in the 2003 Act (and its predecessor, the Criminal Justice Act 1998) which are designed to ensure the fairness of a criminal trial where hearsay evidence is admitted. While, as is apparent from the judgment, the Court has not been able to accept all the criticisms of the test, it has addressed what appears to be one of the central problems identified by the Supreme Court, namely the inflexible application of the test or rule, as reflected in the Chamber’s Lucà v. Italy judgment, whereby a conviction based solely or to a decisive degree on the statement of an absent witness is considered incompatible with the requirements of fairness in Article 6, notwithstanding any counterbalancing procedural safeguards within the national system. I share the view of the majority that to apply the rule inflexibly, ignoring the specificities of the particular legal system concerned, would run counter to the traditional way in which the Court has, in other contexts, approached the issue of the overall fairness of criminal proceedings. While, as the Court has now held, in assessing the fairness of the proceedings, the fact that a conviction is based solely or to a decisive extent on the statement of an absent witness is a very important factor to weigh in the scales and one which requires strong counterbalancing factors, including the existence of effective procedural safeguards, it should not automatically result in a breach of Article 6 § 1 of the Convention. Having re-examined the two cases in the light of these principles, I agree with the majority that, contrary to my original view in the Chamber, Article 6 § 1 was not violated in the case of Al-Khawaja but that there was a violation of that Article in the case of Tahery. JOINT PARTLY DISSENTING AND PARTLY CONCURRING OPINION OF JUDGES SAJÓ AND KARAKAŞ We were invited by the Supreme Court of the United Kingdom ( R v. Horncastle and others [2009] UKSC 14 ) to clarify the principles behind the exclusionary rule in cases where hearsay evidence is the sole or decisive evidence. Such requests, which reflect genuine concerns about, and apparent inconsistencies within, our case-law, deserve due consideration to enable a bona fide dialogue to take place . Apparently, the issue to be clarified is as follows. When 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 to have examined, whether during the investigation or at the trial, the rights of the defence are generally considered to be restricted to an extent that is incompatible with the guarantees provided by Article 6 (the so-called “sole or decisive rule”). The question is whether the latter rule is to be considered as an absolute rule whose breach automatically leads to a finding that the proceedings have not been fair, in violation of Article 6 § 1 of the Convention (see paragraph 119 of the judgment). But there is a much more important issue at stake, namely the relationship between the fundamental human rights of the accused and society’s legitimate interest in imposing punishment – after a fair trial. (Fairness also implies that the innocent shall not be punished.) In fact, the issue is to what extent the right to a fair trial, which is an institutional concern and a matter of striking a fair balance between the conflicting interests of the accused and the administration of justice, can absorb or undermine specific individual rights which are defined in the Convention in absolute and categorical terms. To clarify the principles in this case it is necessary to start from the express protection of the defence’s rights under Article 6 §§ 1 and 3 (d). It should be stressed that Article 6 §§ 1 and 3 (d) exist in the context of an inherent power imbalance between the accused and the State. The impetus behind Article 6 is the premise that the interests of justice will not properly safeguard the accused against the dangers of an unfair trial and improper conviction. Because prosecutorial power is subject to abuse as well as to the bureaucratic pressure to single out and punish a perpetrator, the defence should not be unduly impeded in countering the State’s allegations. It is sometimes said that the defendant’s rights must be “balanced” against the public interest in administering justice, and in particular against the Convention rights of victims and witnesses. But the protection of the defence’s rights, including the right to examine adverse witnesses, is already embedded as fundamental to a fair trial in the administration of justice, prior to such considerations. When the Convention singled out paragraph 3 rights this meant that these basic rights of the defence were necessary to counterbalance the dominant power of the prosecution, in the interests of fairness. To balance these rights a second time against the interests of the administration of justice, as the Government have sought to do in Al-Khawaja and Tahery , is to give the prosecution and the interests of administering justice (namely, to punish) a clear advantage. This Court has never held that “Article 6(3)(d) is simply an illustration of matters to be taken into account in considering whether a fair trial has been held” as Waller LJ claimed, reviewing Strasbourg case-law in R v. Sellick [2005] EWCA Crim 651 , quoted in Horncastle , paragraph 79 . The judgment (paragraph 143) states as follows: “At the same time, however, the Court has always interpreted Article 6 § 3 in the context of an overall examination of the fairness of the proceedings (see, as a recent authority, Salduz v. Turkey [GC], no. 36391/02, § 50, 27 November 2008)” (emphasis added). The term “overall examination” is new in the context of Article 6. It is true that the Court has consistently assessed the impact that the defendant’s inability to examine a witness has had on the overall fairness of his trial. In fact, the Court has recognised that even where an element of a specific named defence right has been restricted, this can be counterbalanced and the fairness of the trial can be achieved. But in applying the holistic approach (now presented as “an overall examination”) in order to determine the fairness of the trial, this Court has never stated that fairness can still be achieved if one of the fundamental rights is deprived of its essence. With regard to the right to cross-examine witnesses and the related but broader equality-of-arms principle, the Court has systematically and consistently drawn a bright line, which it has never abandoned, in the form of the sole or decisive rule. Today this last line of protection of the right to defence is being abandoned in the name of an overall examination of fairness. In the light of our case-law it is undeniable that the rights listed under Article 6 § 3 are subject to interpretation in the context of the concept of fair trial. Reference is made in that regard to Salduz (see paragraph 143 of the judgment). But the differences in the reasoning in Salduz , which follows a long line of cases, are quite telling. The term “overall examination” is noticeably absent. The Court found a violation “of Article 6 § 3 (c) of the Convention in conjunction with Article 6 § 1. ” Of course, the terminology itself is not particularly telling but the change in terminology is quite instructive. What was relevant in Salduz (and is applicable mutatis mutandis in the present case, given the structural similarity of Article 6 § 3 (c) and (d) rights) for the nature of the relationship under consideration is the following: the reliance on Article 6 § 1 in the interpretation of the right to counsel was dictated by the needs of an extensive interpretation of the specific right at stake. The Convention grants the right to legal assistance for a person’s defence (the right to “defend himself”); it does not say at which stage of the proceedings this right is to be granted. It could be argued that it is limited to the trial stage. However, the Court did not accept such a narrow construction, and it interpreted the right to counsel in terms of its consequences for a fair trial, and extended its scope. Neither in Salduz nor elsewhere has the Court argued that a fair trial absorbs the specific rights enumerated in Article 6 § 3. Undeniably, there is a line of cases in our jurisprudence in which the rights enumerated under Article 6 § 3, including the right to examine witnesses, were interpreted restrictively and where the requirements of a fair trial were nevertheless found to have been satisfied. Article 6 § 1 was always used as a fallback position; the restriction was found not to have been fatal to the fairness of the trial where it was counterbalanced to some extent, at least in theory. The restrictions were found to be dictated by the practical needs of an effective investigation and trial (the needs of the administration of justice) or by conflicts with the Convention rights of others (for example, where the right to life of a witness had to be protected). In all these cases additional safeguards were demanded and the essence of the right to cross-examine in the sense of equality of arms was always considered to have been respected (as in the case of “special counsel”, or where special confrontation techniques were used for the protection of vulnerable witnesses). This Court’s case-law supports four propositions concerning the rights of the defence in the context of a fair trial. Firstly, those rights are premised on the principles of equality of arms and adversarial proceedings, both of which require that the defence should not be disadvantaged relative to the prosecution and that a criminal defendant should be able to test the evidence underlying his conviction. Secondly, although some derogation is permitted with respect to these rights, it must be justified and sufficiently counterbalanced by procedural safeguards. Thirdly, the extent of disadvantage to the defence is a factor in determining whether the trial as a whole has been fair. Finally, where statements by witnesses whom the defence has had no chance to examine before or at trial underpin the conviction in a decisive manner, the disadvantage is of such a degree as to constitute in itself a violation of Article 6 which no procedural safeguards can effectively counterbalance. It is the fourth proposition that has come to be known as the sole or decisive rule. But, as we suggest below, it is perhaps more sensible to understand the second, third, and fourth propositions together, as carving out a narrow exception to the prohibition of hearsay. Once again, it is up to States to determine what kind of evidence is admissible, and the Court has never imposed a blanket prohibition in that regard. But it is obvious in the light of its case-law that hearsay remains a problematic source of evidence which requires special caution because of the general inability to test the reliability of hearsay statements. As the Court has put it: “The dangers inherent in such a situation are obvious” (see Kostovski v. the Netherlands , 20 November 1989, § 42, Series A no. 166). While the Court calls for “extreme” care in the treatment of untested evidence, the reality is that either evidence is used or it is not . The Convention does not list grounds for restricting the rights of the defence. Rather, it is often understood as singling these rights out for protection in the context of ensuring a fair trial, as being essential for fairness. Where practical difficulties arise, the measures taken to remedy them are to be assessed by determining whether the rights were adequately protected, not whether the rights were outweighed by other legitimate interests. One must not allow prosecutorial interests to prevail simply because they appear in the guise of witness protection or the need to convict the accused (which is presented as the interest of justice). This Court has reiterated in its prior case-law that the admissibility of evidence is a matter of national law and that the Court’s task is to rule not on whether a piece of evidence was correctly admitted but on whether the proceedings as a whole were fair (see Kostovski , cited above, § 39). To determine the fairness of a trial, the Court must nevertheless assess, inter alia , the manner in which evidence for the defence and prosecution was given. In principle, Article 6 § 3 (d) demands that witness statements be made at a public hearing in the presence of the accused with a view to adversarial argument. In the absence of confrontation at trial, examination at the pre-trial stage may suffice (see Kostovski , cited above, § 39). Concerns around Article 6 § 3 (d) will be triggered above all in cases like Al-Khawaja and Tahery , where the defence was afforded no opportunity to examine key witnesses in either the pre-trial or trial proceedings. The point of departure is thus that the defence shall be given a chance to contest the witness’s statement in court and, unless compelling reasons dictate otherwise, in the very court judging the case. The fact that the Court will not review issues of admissibility of evidence (except for evidence resulting from a violation of Article 3) does not mean that it is not aware of its problematic nature. The need for cross-examination is well founded, particularly where testimony is central to the prosecution. The value of testimony hinges on a witness’s credibility. To challenge that credibility requires knowledge of the witness’s identity. Personal demeanour is of great importance, as is the direct impression of the evidence on the judge or jury hearing the case. Even experienced trial judges may erroneously give undue weight to evidence by witnesses whom the defence has not cross-examined. A fortiori , these factors are more significant in a jury system, in so far as a professional judge is better positioned than a layperson to evaluate information obtained by or in the presence of another judge conducting a prior hearing. Asking members of a jury to weigh evidence in the light of its being untested demands far more in the way of judicial competence than asking judges to rule on admissibility in the light of the potential value of the evidence to the prosecution’s case. To conform with Article 6 § 3 (d), special care may therefore be necessary to ensure that untested evidence does not go before the jury if that evidence is likely by itself to decide the case. In this regard this Court has to ask: (a) is the care applied in the English and Welsh legal system in the light of the present cases sufficient; or (b) are the risks associated with an otherwise carefully counterbalanced admission of hearsay as sole or decisive evidence such that they put the defence rights and hence the fairness of the trial in jeopardy to an extent that undermines effective human rights protection? The paradigmatic injustice that requires the robust protection granted by Article 6 § 3 (d) is a criminal conviction based (decisively) on hearsay. The issue cannot be, then, whether the court is otherwise satisfied that the untested evidence is reliable. Justice Scalia’s cautionary reminder is instructive in this regard: “Dispensing with confrontation because testimony is obviously reliable is akin to dispensing with jury trial because a defendant is obviously guilty” (see Crawford v. Washington , 541 U. S. 36, 62 (2004)) . The Government’s assessment of cogency and reliability in Al-Khawaja and Tahery is dubious for the very reason that the means by which that assessment was reached restricted the defence’s right to examine key witnesses. Once we recognise the relationship in principle between the rights of the defence and a fair trial, the sole or decisive test looks like an exception to the total prohibition of hearsay provided by a separate rule. As an exception, it should be interpreted strictly and narrowly. Where this Court has previously found no violation of Article 6 § 3 (d), it has also noted the presence of corroborating evidence capable of supporting the conviction (see, for example, Ferrantelli and Santangelo v. Italy , 7 August 1996, § 52, Reports of Judgments and Decisions 1996-III; Doorson v. the Netherlands , cited above, § 72; and Gossa v. Poland , no. 47986/99, § 61, 9 January 2007). English law as applied in Al-Khawaja and Tahery amounts to asking for an exception to what is already the exception. The question before the Court is, hence, whether the principles underlying Article 6 §§ 1 and 3 (d) can survive this further retrenchment. The Government asserted that the evidence in these cases was trustworthy because, in effect, the common law system could be trusted to assess reliability without confrontation and to protect the defence’s rights independently of the Convention. This reasoning is unpersuasive. The very aim of the Convention is to protect human rights from violation by the State, which includes a judiciary composed of fallible triers of fact, be they as professional as is humanly possible. It is no accident that mistrust with regard to non-confronted evidence continues to prevail in common law systems. The views of the Law Reform Commission of Ireland are particularly instructive: “There are dangers associated with allowing evidence of unavailable witnesses: it undermines the defendant’s right to a fair trial and creates the potential of miscarriages of justice and ... to allow in untested evidence from frightened and unavailable witnesses would undermine this right. The Commission notes that it has provisionally recommended that the courts should retain a discretion to develop the hearsay rule if the necessity exists” (paragraphs 69-71). The dangers of admitting hearsay in a context comparable to Al-Khawaja forced the Supreme Court of Canada to restrict the admissibility of non-contestable witness statements in R. v. Khewalon ([2006] 2 S. C. R. 787). And, as previously noted, the United States, which has a fully functioning common law system, employs a more stringent requirement of confrontation than the one established under the European Convention on Human Rights. Most importantly, the United Kingdom, like other countries, ratified the Convention on the assumption that its legal system was in conformity with the Convention. At the time of ratification the applicable rule of common law was one that precluded hearsay . Populism, the police and the prosecuting authorities subject courts all around the world to pressure to disregard fundamental safeguards of criminal procedure. Sometimes the demands are legitimately grounded in practical difficulties, but this is not a good enough reason to disregard the protection of the rights of the accused, which are decisive for a fair trial and the fair administration of justice. Since the reasons behind a witness’s absence or anonymity may differ (for instance, death, witness intimidation, vulnerability as victims of domestic abuse or child sexual abuse), we should avoid dealing with the cases in a one-size-fits-all fashion. The protection of child witnesses from further trauma, for instance, requires special care. Even in such cases it should be possible for the defence to have questions put to the witness during a pre-trial hearing or preliminary investigation. Such sessions can be videotaped to ensure that the defence is also able to challenge the witness’s credibility before the jury (see, for example, S. N. v. Sweden , no. 34209/96, § 52, ECHR 2002-V). In general, any restrictions on the rights of the defence should be treated with extreme care (see S. N. v. Sweden , cited above, § 53). Here again, the acceptable solution was a special form of testing evidence. But the evidence was tested and there was no reliance on untested sole or decisive witness evidence. In Al-Khawaja and Tahery no pre-trial opportunity to cross-examine the witnesses was provided. Nor were any clear interests alleged as justification for the handicap to the defence, apart from the ever-present interests in ensuring public safety and criminal punishment. S. T.’s suicide in Al Khawaja and T.’s refusal in Tahery to testify in court for fear of being branded an informer in his community are clearly distinguishable from cases involving child abuse victims or organised crime prosecutions, both of which involve an unusual need to shield the witness from the defendant. Special caution may well be needed where key witnesses die or are intimidated as a result of the defendant’s actions. Al-Khawaja and Tahery , however, do not constitute such cases. We need not, therefore, tackle this question in order to resolve them. While we understand the nature of the challenges faced by the prosecution when key witnesses die or refuse to appear at trial out of genuine fear, the protections guaranteed by Article 6 speak only to the rights of the defence, not to the plight of witnesses or the prosecution. The task of this Court is to protect the accused precisely when the Government limit rights under the Convention in order to bolster the State’s own position at trial. Counterbalancing procedures may, when strictly necessary, allow the Government flexibility in satisfying the demands of Article 6 § 3 (d). Our evolving application of the sole or decisive test, however, shows that this exception to the general requirement of confrontation is not itself without limits in principle. In the end, it is the job of the Government to support their case with non-hearsay corroborating evidence. Failure to do so leaves the Government open to serious questions about the adequacy of their procedures and violates the State’s obligations under Article 6 § 1 in conjunction with Article 6 § 3 (d). Today the Court has departed from its previous position according to which, where a witness cannot be cross-examined and the conviction is based on hearsay as the sole or decisive evidence, the rights protected under Article 6 will be violated. The Court relies on cases concerning the withholding of evidence from the defence in order to protect police sources (it cites Rowe and Davis v. the United Kingdom [GC], no. 28901/95, ECHR 2000-II). According to the present judgment, in those cases the Court “has left it to the domestic courts to decide whether the rights of the defence should cede to the public interest and has confined itself to verifying whether the procedures followed by the judicial authorities sufficiently counterbalance the limitations on the defence with appropriate safeguards. The fact that certain evidence was not made available to the defence was not considered automatically to lead to a violation of Article 6 § 1” (see paragraph 145 of the judgment). In other words, the Court asserts that it has already accepted that access to evidence is a matter of striking an appropriate balance between the public and private interests at stake. It is clear that, unlike in the case of Articles 8 11, the language and the general practice of the Court do not allow such a balancing exercise. We simply cannot see how one can rely on this line of cases, which do not deal with sole witness evidence. In fact, Rowe and Davis was about the entitlement to disclosure of relevant evidence. It was found that this was not an absolute right: “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 judgment of 26 March 1996, Reports of Judgments and Decisions 1996-II, p. 470, § 70)” (see Rowe and Davis , cited above, § 61). It was in Doorson that the Court indicated that this counterbalancing exercise applied only where the evidence was not sole or decisive evidence (and particularly, as in the present case, was based on hearsay). Counterbalancing exercises run the risk of falling short of the standard of precision and reliability expected of legal rule , and that risk has been found to be impermissibly high and beyond any counterbalancing measures in the case of sole or decisive evidence. Beyond the reference to fairness based on the misconstruction of Rowe and Davis no reason is offered for departing from the categorical interpretation of the “sole or decisive” rule, except to say that the Court will demand that the counterbalancing measures be such as to satisfy the most searching scrutiny. The uncertainty and inadequate protection of rights resulting from the counterbalancing approach are well illustrated in Al Khawaja , where even the Court of Appeal had to concede that the directions given by the judge were deficient. However, the court assumed that the jury was able to make the necessary distinction, which then enabled the latter to rely on the strong similarities with the complaint of the other alleged victim of assault. We consider such trust to be cavalier, for the reasons given by the Supreme Court in Horncastle . As the Supreme Court stated, one of the themes that have marked the common law approach to a fair criminal trial “has been a reluctance to trust the lay tribunal to attach the appropriate weight to the evidence placed before them” (paragraph 17). This is what was expressly allowed to happen in Al-Khawaja . In other words, the assurances allegedly offered in the counterbalancing system provided by the legislation and applied by the most competent judges trained in the noble tradition of the common law failed their first test, and in the second case ( Tahery ), even this Court had to find that the trial judge had misapplied the allegedly foolproof guarantee. In conclusion, for the above reasons we are of the view that Article 6 § 3 (d) in conjunction with Article 6 § 1 has been violated in Al-Khawaja and we must therefore dissent in that case; in the case of Tahery , meanwhile, we follow the majority but on a different ground, namely because the sole or decisive rule was disregarded. This Court, in its efforts to accommodate legitimate demands in the context of a fair trial, has systematically and consistently drawn a bright line in the form of the sole or decisive rule. Today this position is being abandoned in the name of an overall examination of fairness, in the hope that the fairness of the trial will be achieved by demanding a counterbalancing of the restrictions on the right in question and requiring very stringent reasons for, and genuine proof of, such counterbalancing. Even where this is achieved, it will remain a questionable achievement, as it comes at the price of sacrificing an expressly granted Convention right. Legal history shows that convictions based on untested hearsay evidence are often wrong and certainly a favourite instrument of political abuse. True, this is not the case with the current counterbalancing approach as applied in England and Wales. But the very cases in front of us today, and others reported in Horncastle , show the uncertainties inherent in the counterbalancing approach, which undermines legal certainty in the proceedings and also the foreseeability of the law. The dangers of such an approach were highlighted as late as 2008, a year before the ruling in Horncastle . In R v. Davis (paragraph 8), Lord Bingham, in a noble effort to protect the principles of a fair trial, found it necessary to quote the Court of Appeal of New Zealand in R v Hughes [1986] 2 NZLR 129, in which Richardson J observed (p. 147): “Clearly the accused cannot be assured of a true and full defence to the charge unless he is supplied with sufficient information about his accuser in order to decide on investigation whether his credibility should be challenged.” Lord Bingham quoted further: “We would be on a slippery slope as a society if on a supposed balancing of the interests of the State against those of the individual accused the Courts were by judicial rule to allow limitations on the defence in raising matters properly relevant to an issue in the trial. Today the claim is that the name of the witness need not be given: tomorrow, and by the same logic, it will be that the risk of physical identification of the witness must be eliminated in the interests of justice in the detection and prosecution of crime, either by allowing the witness to testify with anonymity, for example from behind a screen, in which case his demeanour could not be observed, or by removing the accused from the Court, or both. The right to confront an adverse witness is basic to any civilised notion of a fair trial. That must include the right for the defence to ascertain the true identity of an accuser where questions of credibility are in issue” (pp. 148-149) . The sole or decisive rule that has been followed so far was intended to protect human rights against the “fruit of the poisonous tree” (if the source of the evidence (the “tree”) is tainted, then anything gained from it (the “fruit”) is as well). The adoption of the counterbalancing approach means that a rule that was intended to safeguard human rights is replaced with the uncertainties of counterbalancing. To our knowledge this is the first time ever that this Court, in the absence of a specific new and compelling reason, has diminished the level of protection. This is a matter of gravest concern for the future of the judicial protection of human rights in Europe. We agree with the Supreme Court that insufficient guidance may have been given concerning the concept of “decisive evidence”. This shortcoming is partly the result of the case-specific approach of this Court’s jurisprudence, which is intended to respect national legislative and judicial sovereignty, and is partly due to the assumption that such definitions are best developed within the context of the national legal system. This approach is in accordance with the subsidiarity principle. With the intention of responding to the request, the present judgment provides a definition. In our view, “decisive” evidence is reasonably taken to mean evidence without which the prosecuting authorities could not bring a case. A higher standard is likely to revert to an absolute bar on hearsay. A lower standard invites abuse. Moreover, where testimony is such that without it there could be no prosecution, let alone conviction, failure to afford an adequate opportunity to cross-examine the witness restricts the defence to a far greater extent than in cases where other evidence independently supports the conviction. The sole or decisive test operates here as a backstop on Article 6 § 3 (d), so as to guarantee that the exception does not undermine the principle and that any resulting conviction does not rest ultimately or exclusively on hearsay. The travaux préparatoires indicate that, there, paragraphs 1 and 3 were considered separately . Some of the concurring opinions use the expression “Article 6 § 3 (c), read in conjunction with Article 6 § 1”. Another standard formula that is used is: “As the requirements of Article 6 § 3 are to be seen as particular aspects of the right to a fair trial guaranteed by Article 6 § 1, the Court will examine the complaint under Article 6 §§ 1 and 3 (d) taken together” (see Gossa v. Poland , no. 47986/99, § 51, 9 January 2007). See the criticism by Stefan Trechsel in Human Rights in Criminal Proceedings , OUP, 2005, 313. In its submissions to the Grand Chamber the Government mischaracterise the Crawford holding as absolutist and anachronistic. Justice Scalia’s references to the Framers’ concerns about the trial of Sir Walter Raleigh are, on the contrary, where one finds the principle behind the rule. Albeit more exacting than our standard, Crawford is hardly absolutist. The US exclusionary rule applies only to testimonial statements and recognises what were already well-known exceptions at the time of the Republic’s founding (see Davis v. Washington , 547 U. S. 813 (2003)). The unavailability of a witness and an adequate prior opportunity to cross-examine the witness will, taken together, suffice to satisfy the confrontation requirement under current US law. In Davis and Hammon , both domestic violence cases, the Supreme Court explained that “[s]tatements are non-testimonial when made in the course of police interrogation under circumstances objectively indicating that the [interrogation’s] primary purpose ... is to enable police assistance to meet an ongoing emergency,” but they “are testimonial when the circumstances objectively indicate that there is no such ongoing emergency, and that the [interrogation’s] primary purpose is to establish or prove past events potentially relevant to later criminal prosecution" (547 U. S., at 822). See Lord Bingham’s statement: “As my noble and learned friend Lord Rodger of Earlsferry suggested ... ‘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’. It may well be (this was not explored in argument) that the inclusion of article 6(3)(d), guaranteeing to the defendant a right to examine or have examined witnesses against him, reflected the influence of British negotiators. It is in any event clear, as my noble and learned friend observed in the same case, para 11, that ‘An examination of the case law of the European Court of Human Rights 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 whom he had not had an opportunity to challenge’” ( R v. Davis , paragraph 24). See Trechsel, op cit., 313. For reliance on the sense of justice of the judge on the admissibility of untested hearsay see the wording of Part 11, Chapter 2 of the Criminal Justice Act 2003, Section 116: “... (4) Leave may be given under subsection (2)(e) only if the court considers that the statement ought to be admitted in the interests of justice, having regard— (a) to the statement’s contents, (b) to any risk that its admission or exclusion will result in unfairness to any party to the proceedings (and in particular to how difficult it will be to challenge the statement if the relevant person does not give oral evidence).” We would recall the statement of the judge at the preliminary hearing: “putting it bluntly, no statement, no count one.” See further his instructions to the jury (paragraphs 15 and 16). In a sad sequel to the judicial efforts to ensure fairness, 33 days after the Lords’ ruling, Parliament authorised by law what the Lords had held to be eminently contrary to common law and the principles of a fair trial, in a matter that the Lords had considered settled by this Court’s case-law. A year later the Supreme Court was convinced (in a different context and on the basis of earlier laws) that the principles reaffirmed by the late Lord Bingham were satisfied in a comparable context with lesser guarantees, notwithstanding the concerns expressed by him and his fellow judges. So it goes. BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII