SCHATSCHASCHWILI v. GERMANY - 9154/10 (Judgment (Merits and Just Satisfaction) : Court (Grand Chamber)) [2015] ECHR 1113 (15 December 2015)

JUDGMENT STRASBOURG 15 December 2015 This judgment is final but it may be subject to editorial revision. In the case of Schatschaschwili v. Germany, The European Court of Human Rights, sitting as a Grand Chamber composed of: Dean Spielmann, President, Işıl Karakaş, András Sajó, Luis López Guerra, Päivi Hirvelä, Khanlar Hajiyev, Dragoljub Popović, Nona Tsotsoria, Kristina Pardalos, Angelika Nußberger, Julia Laffranque, Helen Keller, André Potocki, Paul Mahoney, Valeriu Griţco, Egidijus Kūris, Jon Fridrik Kjølbro, judges, and Lawrence Early, Jurisconsult , Having deliberated in private on 4 March 2015 and on 8 October 2015, Delivers the following judgment, which was adopted on the last-mentioned date: PROCEDURE 1. The case originated in an application (no. 9154/10 ) against the Federal Republic of Germany lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by a Georgian national, Mr Swiadi Schatschaschwili (“the applicant”), on 12 February 2010. 2. By letter dated 29 December 2013 the applicant’s lawyer notified the Court that the applicant had informed him that his true name was Avtandil Sisvadze. The Court advised the parties on 14 January 2014 that it would continue processing the application under the case name of Schatschaschwili v. Germany . This corresponded to the applicant’s name as referred to in the domestic court proceedings at issue as well as in his application lodged with the Court. 3. The applicant, who had been granted legal aid, was represented by Mr H. Meyer-Mews, a lawyer practising in Bremen. The German Government (“the Government”) were represented by their Agents, Mrs A. Wittling-Vogel, Mr H.-J. Behrens and Mrs K. Behr, of the Federal Ministry of Justice and Consumer Protection. 4. The applicant alleged in particular, relying on Article 6 § 3 (d) of the Convention, that his trial had been unfair as neither he nor his lawyer had been granted an opportunity at any stage of the criminal proceedings against him to examine the victims and only direct witnesses of the offence allegedly committed by him in Göttingen in February 2007, on whose statements the Göttingen Regional Court had relied in convicting him. 5. The application was allocated to the Fifth Section of the Court (Rule 52 § 1 of the Rules of Court). The Government were given notice of the application on 15 January 2013. On 17 April 2014 a Chamber of the Fifth Section composed of Mark Villiger, President, Angelika Nußberger, Boštjan M. Zupančič, Ann Power-Forde, Ganna Yudkivska, Helena Jäderblom and Aleš Pejchal, judges, and Claudia Westerdiek, Section Registrar, unanimously declared the application partly admissible and delivered its judgment. It held, by five votes to two, that there had been no violation of Article 6 § 1 read in conjunction with Article 6 § 3 (d) of the Convention. 6. On 15 July 2014 the applicant requested that the case be referred to the Grand Chamber in accordance with Article 43 of the Convention and Rule 73 of the Rules of Court. On 8 September 2014 the Panel of the Grand Chamber accepted that request. 7. The composition of the Grand Chamber was determined according to the provisions of Article 26 §§ 4 and 5 of the Convention and Rule 24 of the Rules of Court. At the final deliberations, Jon Fridrik Kjølbro and András Sajó, substitute judges, replaced Josep Casadevall and Isabelle Berro, who were unable to take part in the further consideration of the case (Rule 24 § 3). 8. The applicant and the Government each filed a memorial (Rule 59 § 1) on the merits. In addition, third-party comments were received from the Czech Government, which had been given leave by the President on 3 November 2014 to intervene in the written procedure (Article 36 § 2 of the Convention and Rule 44 § 3). 9. The Government of Georgia, having been informed of their right to intervene in the proceedings (Article 36 § 1 of the Convention and Rule 44 §§ 1 and 4), did not indicate that they wished to exercise that right. 10. A hearing took place in public in the Human Rights Building, Strasbourg, on 4 March 2015 (Rule 59 § 3). There appeared before the Court: (a) for the respondent Government Mr H.-J. Behrens , Federal Ministry of Justice and Consumer Protection, Agent , Mr H. Satzger , Professor of Criminal Law at the University of Munich, Mr F. Zimmermann , Legal Assistant at the University of Munich, Mr H. Paetzold , Federal Ministry of Justice and Consumer Protection, Mr C. Tegethoff , Judge, Lower Saxony Ministry of Justice, Advisers ; (b) for the applicant Mr H. Meyer-Mews , Lawyer, Counsel , Mr A. Rotter , Lawyer, Mr J. Lam , Lawyer, Advisers . The Court heard addresses by Mr Meyer-Mews and Mr Behrens and their replies to questions put by judges. THE FACTS I. THE CIRCUMSTANCES OF THE CASE 11. The applicant was born in 1978. When he lodged his application he was being detained in Rosdorf Prison, Germany. He now lives in Khashuri /Surami, Georgia. A. The events in Kassel and Göttingen as established by the domestic courts 1. The offence committed in Kassel 12. On the evening of 14 October 2006 the applicant and an unidentified accomplice robbed L. and I., two Lithuanian nationals, in the women’s apartment in Kassel. 13. The perpetrators were aware that the apartment was used for prostitution and expected its two female occupants to keep valuables and cash there. They passed by the apartment in the early evening in order to make sure that no clients or a procurer were present. Shortly afterwards they returned and overpowered L., who had answered the doorbell. The applicant pointed a gas pistol which resembled a real gun at both women and threatened to shoot them if they did not disclose where their money was kept. While his accomplice watched over the women, the applicant partly collected in the apartment and partly forced the women to hand over to him some 1,100 euros (EUR) and six mobile phones. 2. The offence committed in Göttingen 14. On 3 February 2007 the applicant, acting jointly with several accomplices, robbed O. and P., two female Latvian nationals who were temporarily resident in Germany and working as prostitutes, in their apartment in Göttingen. 15. On the evening of 2 February 2007, the day before the offence, one of the applicant’s co-accused had passed by O. and P.’s apartment in Göttingen together with an accomplice, R., an acquaintance of O. and P. They intended to verify whether the two women were the apartment’s only occupants and whether they kept any valuables there, and discovered a safe in the kitchen. 16. On 3 February 2007 at around 8 p.m. the applicant and a further accomplice, B., gained access to O. and P.’s apartment by pretending to be potential clients, while one of their co-accused waited in a car parked close to the apartment building and another waited in front of the building. Once inside the apartment B. produced a knife that he had been carrying in his jacket. P., in order to escape from the perpetrators, jumped from the balcony located approximately two metres off the ground and ran away. The applicant jumped after her but abandoned the chase after some minutes when some passers-by appeared nearby on the street. He then called the co-accused who had been waiting in front of the women’s apartment building on his mobile phone and told him that one of the women had jumped from the balcony and that he had unsuccessfully chased her. The applicant agreed on a meeting point with his co-accused where they would pick him up by car once B. had left the crime scene and joined them. 17. In the meantime inside the apartment, B., after having overpowered O., threatened to kill her with his knife if she did not disclose where the women kept their money or if she refused to open the safe for him. Fearing for her life, O. opened the safe, from which B. removed EUR 300, and also handed over the contents of her wallet, EUR 250. B. left the apartment at around 8.30 p.m., taking the money and P.’s mobile telephone as well as the apartment’s landline telephone with him, and joined the co-accused. The co-accused and B. then picked up the applicant at the agreed meeting point in their car. At approximately 9.30 p.m. P. rejoined O. in the apartment. 18. O. and P. gave an account of the events to their neighbour E. the morning after the offence. They then left their Göttingen apartment out of fear and stayed for several days with their friend L., one of the victims of the offence committed in Kassel, to whom they had also described the offence in detail the day after it occurred. B. The investigation proceedings concerning the events in Göttingen 19. On 12 February 2007 L. informed the police of the offence committed against O. and P. in Göttingen. Between 15 and 18 February 2007 O. and P. were repeatedly questioned by the police as to the events of 2 and 3 February 2007. In those interviews they described the course of events as set out above. The police, having checked O. and P.’s papers, found their residence and occupation in Germany to be in compliance with German immigration and trade law. 20. As the witnesses had explained during their police interviews that they intended to return to Latvia in the days to come, on 19 February 2007 the prosecution asked the investigating judge to question the witnesses in order to obtain a true statement which could be used at the subsequent trial (“ eine[r] im späteren Hauptverfahren verwertbare[n] wahrheitsgemäße[n] Aussage ”). 21. Thereupon, on 19 February 2007, O. and P. were questioned by an investigating judge and again described the course of events as set out above. At that time, the applicant had not yet been informed about the investigation proceedings initiated against him, so as not to put the investigation at risk. No warrant for his arrest had yet been issued and he was not yet represented by counsel. The investigating judge excluded the applicant from the witness hearing before him in accordance with Article 168c of the Code of Criminal Procedure (see paragraph 56 below) since he was concerned that the witnesses, whom he had found to be considerably shocked and distressed by the offence, would be afraid of telling the truth in the applicant’s presence. The witnesses confirmed at that hearing that they intended to return to Latvia as soon as possible. 22. Witnesses O. and P. returned to Latvia shortly after that hearing. The applicant was subsequently arrested on 6 March 2007. C. The trial before the Göttingen Regional Court 1. The court’s attempts to question O. and P. and the admission of O. and P.’s pre-trial statements 23. The Göttingen Regional Court summoned O. and P. by registered mail to appear at the trial on 24 August 2007. However, both witnesses refused to attend the hearing before the Regional Court, relying on medical certificates dated 9 August 2007 which indicated that they were in an unstable, post-traumatic emotional and psychological state. 24. On 29 August 2007 the Regional Court therefore sent letters by registered mail to both witnesses informing them that the court, while not being in a position to compel them to appear at a court hearing in Germany, nonetheless wished to hear them as witnesses at the trial. The court stressed that they would receive protection in Germany and that all costs incurred in attending the hearing would be reimbursed and, proposing several options, asked in what circumstances they would be willing to testify at the trial. While an acknowledgement of receipt was returned for both letters, no response was obtained from P. O., for her part, informed the Regional Court in writing that she was still traumatised by the offence and would therefore neither agree to appear at the trial in person nor would she agree to testify by means of an audio-visual link. O. further stated that she had nothing to add to the statements she had made in the course of the interviews carried out by the police and the investigating judge in February 2007. 25. The Regional Court nevertheless decided to request legal assistance from the Latvian authorities under the European Convention on Mutual Assistance in Criminal Matters of 20 April 1959, as supplemented by the Convention on Mutual Assistance in Criminal Matters between the Member States of the European Union of 29 May 2000 (see paragraphs 64-66 below), taking the view that O. and P. were obliged under Latvian law to appear before a court in Latvia following a request for legal assistance. It asked for the witnesses to be summoned before a court in Latvia and for an audio-visual link to be set up in order for the hearing to be conducted by the presiding judge of the Regional Court ( audiovisuelle Vernehmung ). It considered, by reference to Article 6 § 3 (d) of the Convention, that defence counsel and the accused, just like the judges and the prosecution, should have the right to put questions to the witnesses for the first time. 26. However, the witness hearing of O. and P. scheduled by the competent Latvian court for 13 February 2008 was cancelled shortly before that date by the presiding Latvian judge. The latter found that the witnesses, again relying on medical certificates, had demonstrated that they were still suffering from post-traumatic disorder as a consequence of the offence and that further confrontation with the events in Göttingen would risk aggravating their condition. O. had further claimed that, following threats by the accused, she feared possible acts of revenge. 27. By letter dated 21 February 2008 the Regional Court, which had obtained copies of the medical certificates the witnesses had submitted to the Latvian court at the Regional Court’s request, informed its Latvian counterpart that, according to the standards of German criminal procedure law, the witnesses had not sufficiently substantiated their refusal to testify. The court suggested to the competent Latvian judge that the witnesses be examined by a public medical officer ( Amtsarzt ) or, alternatively, that they be compelled to attend the hearing. The letter remained unanswered. 28. By decision of 21 February 2008 the Regional Court, dismissing an objection to the admission of the witnesses’ pre-trial statements raised by counsel for one of the co-accused, ordered that the records of O. and P.’s interviews by the police and the investigating judge be read out at the trial in accordance with Article 251 §§ 1 (2) and 2 (1) of the Code of Criminal Procedure (see paragraph 61 below). It considered that, as required by the said provisions, there were insurmountable obstacles which made it impossible to hear the witnesses in the foreseeable future as they were unreachable. It had not been possible to hear witnesses O. and P. in the course of the trial since they had returned to their home country, Latvia, shortly after their interviews at the investigation stage, and all attempts to hear their evidence at the main hearing, which the court had no means of enforcing, had been to no avail. Pointing out that the courts were under an obligation to conduct proceedings involving deprivation of liberty expeditiously, and in view of the fact that the accused had already been in custody for a considerable period of time, the court was of the opinion that it was not justified to further delay the proceedings. 29. The Regional Court emphasised that at the investigation stage there had been no indication that O. and P., who had testified on several occasions before the police and then before the investigating judge, would refuse to repeat their statements at a subsequent trial. It considered that, notwithstanding the resulting restrictions for the defence on account of the admission of O. and P.’s pre-trial statements as evidence in the proceedings, the trial as a whole could be conducted fairly and in compliance with the requirements of Article 6 § 3 (d) of the Convention. 2. The Regional Court’s judgment 30. By judgment of 25 April 2008 the Göttingen Regional Court, considering the facts established as described above, convicted the applicant of two counts of aggravated robbery combined with aggravated extortion involving coercion, committed jointly with other perpetrators in Kassel on 14 October 2006 and in Göttingen on 3 February 2007, respectively. It sentenced the applicant, who had been represented by counsel at the trial, to nine years and six months’ imprisonment. (a) The assessment of the available evidence concerning the offence in Kassel 31. The Regional Court based its findings of fact concerning the offence committed by the applicant in Kassel on the statements made at the trial by the victims L. and I., who had identified the applicant without any hesitation. It further noted that their statements were supported by the statements made at the trial by the police officers who had attended the crime scene and had interviewed L. and I. in the course of the preliminary investigation. In view of these elements, the Regional Court considered that the submissions made by the applicant, who had initially claimed his innocence and had then admitted that he had been in L. and I.’s flat but had only secretly stolen EUR 750, alone, after a quarrel with the women, had been refuted. (b) The assessment of the available evidence concerning the offence in Göttingen (i) O. and P.’s statements 32. In the establishment of the facts concerning the offence in Göttingen, the Regional Court relied in particular on the pre-trial statements made by the victims O. and P., whom it considered to be key witnesses for the prosecution ( maßgebliche[n] Belastungszeuginnen ), in the course of their police interviews and before the investigating judge. 33. In its judgment, which ran to some 152 pages, the Regional Court pointed out that it was aware of the reduced evidentiary value of the records of O. and P.’s pre-trial testimonies. It further took into account the fact that neither the applicant nor counsel for the defence had been provided with an opportunity to examine the only direct witnesses to the offence in Göttingen at any stage of the proceedings. 34. The Regional Court noted that the records of O. and P.’s interviews at the investigation stage showed that they had given detailed and coherent descriptions of the circumstances of the offence. Minor contradictions in their statements could be explained by their concern not to disclose their residence and activities to the authorities and by the psychological strain to which they had been subjected during and following the incident. The witnesses had feared problems with the police and acts of revenge by the perpetrators. This explained why they had not reported the offence immediately after the events and why the police had only been informed on 12 February 2007 by their friend L. 35. The Regional Court further took note of the fact that O. and P. had failed to identify the applicant when confronted with several photos of potential suspects during the police interviews. It observed that the witnesses’ attention during the incident had been focused on the other perpetrator carrying the knife and that the applicant himself had only stayed a short period of time in the apartment. Their inability to identify the applicant also showed that, contrary to the defence’s allegation, the witnesses had not testified with a view to incriminating him. The court further considered that the fact that the witnesses had failed to attend the trial could be explained by their unease at having to recall, and being questioned about, the offence and therefore did not as such affect their credibility. (ii) Further available evidence 36. In its establishment of the facts, the Regional Court further had regard to the following additional evidence: the statements made at the trial by several witnesses to whom O. and P. had reported the offence shortly after it happened, namely the victims’ neighbour E. and their friend L., as well as the police officers and the investigating judge who had examined O. and P. at the pre-trial stage; geographical data and information obtained by tapping the applicant’s and his co-accused’s mobile telephones and by means of a satellite-based global positioning system (“GPS”) receiver in the car of one of the co-accused; the applicant’s admission in the course of the trial that he had been in the victims’ apartment at the relevant time; and the similarity in the way in which the offences in Kassel and Göttingen had been committed. 37. The Regional Court stressed that, once witnesses O. and P. had proved to be unavailable, it had ensured that as many as possible of the witnesses who had been in contact with O. and P. in relation to the events at issue were heard at the trial, in order to verify the victims’ credibility. 38. In the Regional Court’s opinion the fact that the detailed description of the events given in O. and P.’s pre-trial statements was consistent with the account they had given the morning after the offence to their neighbour E. was a strong indication of their credibility and the veracity of their statements. E. had further testified that, on the evening of 3 February 2007 at around 9.30 p.m., another neighbour, an elderly woman who became scared and angry when she saw P. running around in front of her window, had called on her and asked her to accompany her to the women’s apartment to investigate what had happened. O. and P. had, however, not answered the door when the neighbours rang the doorbell. 39. The Regional Court further observed that O. and P.’s description of the events was also consistent with their friend L.’s recollection of her conversations with O. and P. after the offence. 40. In addition, the Regional Court noted that the three police officers and the investigating judge who had examined O. and P. at the pre-trial stage had all testified at the trial that they had found O. and P. to be credible. 41. The Regional Court stressed that since neither the defence nor the court itself had had an opportunity to observe the main witnesses’ demeanour at the trial or during examination by means of an audio-visual link, it had to exercise particular diligence in assessing the evaluation of the witnesses’ credibility by the police officers and the investigating judge. The court further emphasised that, when taking into account the testimonies given by the witnesses’ neighbour E. and their friend L., it had paid special attention to the fact that their statements constituted hearsay evidence and had to be assessed particularly carefully. 42. In this context it had been of relevance that O. and P.’s testimonies as well as the statements of the additional witnesses heard at the trial had been supported by further significant and admissible evidence such as data and information obtained by tapping the applicant’s and the co-accused’s mobile telephones and by means of GPS. The information in question had been gathered in the context of police surveillance measures carried out at the relevant time in the criminal investigation initiated against the accused on suspicion of racketeering and extortion on the Göttingen drug scene. 43. It transpired from the geographical data and the recordings of two mobile telephone conversations between one of the co-accused and the applicant on the evening of 3 February 2007 at 8.29 p.m. and 8.31 p.m. that the latter had been present in the victims’ apartment together with B., and that he had jumped from the balcony in order to chase one of the escaping victims, whom he had failed to capture, while B. had stayed in the apartment. Furthermore, an analysis of the GPS data showed that the car of one of the co-accused had been parked near the crime scene from 7.58 p.m. to 8.32 p.m. on the evening of 3 February 2007, a period that coincided with the timeframe in which the robbery in question had occurred. 44. Furthermore, while the applicant and the co-accused had denied any participation in the robbery as such or any premeditated criminal activity, their own statements at the trial had at least confirmed that one of the co-accused together with R. had visited the victims’ apartment in Göttingen on the evening before the offence and that they had all been present in the car parked close to the victims’ apartment at the time of the offence. The accused had initially stated that a different perpetrator and R. had been in the apartment at the time of the incident the following day. The applicant had subsequently amended his submissions and claimed that it had been he and B. who had gone into the victims’ apartment on 3 February 2007 with a view to making use of the women’s services as prostitutes. He had further conceded that he had followed P. when she escaped over the balcony. He explained that he had done so in order to prevent her from calling the neighbours or the police, since, in view of his criminal record, he had been afraid of getting into trouble and because of the problems he had previously encountered with prostitutes on a similar occasion in Kassel. 45. Finally, the Regional Court considered that the very similar way in which the offences had been committed against two female victims, foreign nationals working as prostitutes in an apartment, was an additional element indicating that the applicant had also participated in the offence committed in Göttingen. 46. In the Regional Court’s view, the body of evidence, taken together, gave a coherent and complete overall picture of events which supported the version provided by witnesses O. and P. and refuted the contradictory versions of events put forward by the applicant and his co-accused in the course of the trial. D. The proceedings before the Federal Court of Justice 47. On 23 June 2008 the applicant, represented by counsel, lodged an appeal on points of law against the judgment of the Göttingen Regional Court. He complained that he had not been able to examine the only direct and key witnesses to the offence committed in Göttingen at any stage of the proceedings, in breach of Article 6 §§ 1 and 3 (d) of the Convention. As the prosecution authorities, contrary to the case-law of the Federal Court of Justice (the applicant referred to a judgment dated 25 July 2000, see paragraphs 58-59 and 62 below), had not requested that defence counsel be appointed for him prior to O. and P.’s hearing before the investigating judge, their statements ought to have been excluded at the trial. 48. In written submissions dated 9 September 2008 the Federal Public Prosecutor General requested that the applicant’s appeal on points of law be dismissed by the Federal Court of Justice as manifestly ill-founded in written proceedings, under Article 349 § 2 of the Code of Criminal Procedure (see paragraph 63 below). The Federal Public Prosecutor General argued that while it was true that the proceedings had been characterised by a “complete loss” of the applicant’s right to examine O. and P. (“ Totalausfall des Fragerechts” ), they had as a whole been fair and there had been no reason to exclude the witness statements of O. and P. as evidence. 49. The Federal Public Prosecutor General considered that the Regional Court had assessed the content of the records of the witnesses’ testimonies read out at the trial particularly carefully and critically. Furthermore, the victims’ statements had been neither the sole nor the decisive basis for the applicant’s conviction by the Regional Court, as the latter had based its findings on further significant evidence. In view of the various layers of corroborating evidence the applicant had had ample opportunity to challenge the credibility of the two prosecution witnesses and to defend himself effectively. 50. Endorsing the Regional Court’s reasoning, the Federal Public Prosecutor General further pointed out that there was nothing to demonstrate that the restrictions on the defence’s right to examine witnesses O. and P. had been imputable to the domestic authorities. The prosecution authorities had not been obliged to appoint counsel for the applicant in order for counsel to participate in the hearing by the investigating judge. In view of the witnesses’ consistent cooperation, the authorities had had no reason to expect that, despite their return to their home country, they would no longer be available for questioning at the trial, especially as they had been obliged under Latvian law to at least participate in a hearing via video link. 51. By decision of 30 October 2008 the Federal Court of Justice, referring to Article 349 § 2 of the Code of Criminal Procedure, dismissed the applicant’s appeal on points of law as manifestly ill-founded. 52. In its decision of 9 December 2008 rejecting the applicant’s complaint concerning a violation of his right to be heard ( Anhörungsrüge ) the Federal Court of Justice pointed out that any decision dismissing an appeal on the basis of Article 349 § 2 of the Code of Criminal Procedure necessarily entailed a reference to the reasoned application by the Federal Public Prosecutor General. E. The proceedings before the Federal Constitutional Court 53. In a constitutional complaint dated 30 December 2008 against the decisions of the Federal Court of Justice of 30 October and 9 December 2008, the applicant complained, in particular, that there had been a breach of his right to a fair trial and of his defence rights under Article 6 § 3 (d) of the Convention. He argued that neither he nor his counsel had had the opportunity to question O. and P. at any stage of the proceedings. 54. By decision of 8 October 2009 the Federal Constitutional Court, without providing reasons, declined to consider the applicant’s complaint (file no. 2 BvR 78/09). II. RELEVANT DOMESTIC LAW AND PRACTICE A. Relevant provisions and practice concerning the conduct of the investigation proceedings 55. Article 160 §§ 1 and 2 of the Code of Criminal Procedure provides that, in investigating the facts relating to a suspicion that a criminal offence has been committed, the public prosecution authorities must investigate not only the incriminating but also the exonerating circumstances and must ensure that evidence which might be lost is taken. 56. Under Article 168c § 2 of the Code of Criminal Procedure, the prosecutor, the accused and defence counsel are authorised to be present during the judicial examination of a witness prior to the opening of the main proceedings. The judge may exclude an accused from being present at the hearing if his or her presence would endanger the purpose of the investigation, in particular if there is a risk that a witness will not tell the truth in the presence of the accused (Article 168c § 3 of the Code of Criminal Procedure). The persons entitled to be present must be given prior notice of the dates set down for the hearings. Notification may be dispensed with if it would endanger the success of the investigation (Article 168c § 5 of the Code of Criminal Procedure). 57. In accordance with Article 141 § 3 of the Code of Criminal Procedure, defence counsel may be appointed during the investigation proceedings. The public prosecutor’s office requests such appointment if, in its opinion, the assistance of defence counsel in the main proceedings will be mandatory. The assistance of defence counsel is mandatory if, inter alia , the main hearing is held at first instance before the Regional Court or the accused is charged with a serious criminal offence (Article 140 § 1 (1) and (2) of the Code of Criminal Procedure). 58. In a leading judgment of 25 July 2000 (published in the official reports, BGHSt , volume 46, pp. 96 et seq . ), the Federal Court of Justice found that Article 141 § 3 of the Code of Criminal Procedure, interpreted in the light of Article 6 § 3 (d) of the Convention, obliged the investigating authorities to consider the appointment of counsel for an unrepresented accused if the key witness for the prosecution was to testify before an investigating judge for the purpose of securing evidence and the accused was excluded from that hearing. 59. The Federal Court of Justice stressed that respect for the right to cross-examination required that the appointed counsel be given an opportunity to discuss the matter with the accused prior to the witness’s examination by the investigating judge, in order to be in a position to ask the relevant questions. The court also noted that it might not be necessary to appoint a lawyer for the accused if there were justifiable reasons not to notify counsel of the hearing before the investigating judge or if the delay caused by appointing and involving a lawyer would endanger the success of the investigation. In the case before it, the Federal Court of Justice further did not have to determine whether it was necessary to appoint counsel for the accused when the purpose of the investigation might be endangered simply as a result of the lawyer discussing the matter with the accused prior to the hearing. B. Relevant provisions and practice concerning the conduct of the trial 60. Article 250 of the Code of Criminal Procedure lays down the principle according to which, where the proof of a fact is based on a person’s observation, that person must be examined at the trial. The examination must not be replaced by reading out the record of a previous examination or a written statement. 61. Article 251 of the Code of Criminal Procedure contains a number of exceptions to that principle. Under Article 251 § 1 (2), the examination of a witness may be replaced by reading out a record of another examination if the witness has died or cannot be examined by the court for another reason within a foreseeable period of time. Article 251 § 2 (1) of the Code of Criminal Procedure provides that, in the event of previous examination by a judge, the examination of a witness may be replaced by reading out the written record of his or her previous examination; this also applies if illness, infirmity or other insurmountable obstacles prevent the witness from appearing at the main hearing for a long or indefinite period. 62. In its above-mentioned judgment of 25 July 2000 (see paragraphs 58-59 above), the Federal Court of Justice found that the failure to appoint counsel for the accused as required by Article 141 § 3 of the Code of Criminal Procedure did not result in the exclusion of the evidence obtained during examination by the investigating judge, but diminished its evidentiary value. Regard had to be had to the proceedings as a whole. As a rule, a conviction could be based on the statement of a witness whom the defence had been unable to cross-examine only if the statement was corroborated by other significant factors independent of it. The trial court was further obliged to assess the evidence with particular care, also having regard to the fact that the statement made by the investigating judge at the trial constituted hearsay evidence. C. Provision concerning appeals on points of law 63. Under Article 349 § 2 of the Code of Criminal Procedure the court deciding on the appeal on points of law may, on a reasoned application by the public prosecutor’s office, dismiss a defendant’s appeal on points of law without a hearing if it considers the appeal to be manifestly ill-founded. The decision must be unanimous. III. RELEVANT INTERNATIONAL LAW 64. Mutual assistance in criminal matters between Germany and Latvia is governed, in particular, by the European Convention on Mutual Assistance in Criminal Matters of 20 April 1959, supplemented by the Convention on Mutual Assistance in Criminal Matters between the Member States of the European Union of 29 May 2000. 65. Article 10 of the Convention on Mutual Assistance in Criminal Matters between the Member States of the European Union of 29 May 2000 provides for the possibility of hearing witnesses by videoconference. Such hearings must be carried out in the presence of a judicial authority of the requested Member State and be conducted by the judicial authority of the requesting Member State. The witness may claim the right not to testify which would accrue to him or her under the law of either the requested or the requesting Member State (Article 10 § 5 of the said Convention). Each Member State must take the necessary measures to ensure that, where witnesses are being heard within its territory and refuse to testify when under an obligation to testify, its national law applies in the same way as if the hearing took place in a national procedure (Article 10 § 8 of the said Convention). 66. Article 8 of the European Convention on Mutual Assistance in Criminal Matters of 20 April 1959 provides that a witness who has failed to answer a summons to appear issued by the requesting Party shall not be subjected to any punishment unless subsequently he voluntarily enters the territory of the requesting Party and is there again duly summoned. THE LAW I. ALLEGED VIOLATION OF ARTICLE 6 §§ 1 AND 3 (d) OF THE CONVENTION 67. The applicant complained that his trial had been unfair and that the principle of equality of arms had been infringed since neither he nor his lawyer had been granted an opportunity at any stage of the criminal proceedings to examine O. and P., the only direct witnesses to and victims of the offence allegedly committed by him in Göttingen in February 2007. He relied on Article 6 of the Convention, which, in so far as relevant, reads as follows: “1. In the determination of ... any criminal charge against him, everyone is entitled to a fair ... hearing ... by [a] ... tribunal ... 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; ...” 68. The Government contested that argument. A. The Chamber judgment 69. The Chamber held that there had been no violation of Article 6 § 1 read in conjunction with Article 6 § 3 (d) of the Convention. 70. Applying the principles established by the Court in its judgment in Al-Khawaja and Tahery v. the United Kingdom ([GC], nos. 26766/05 and 22228/06, ECHR 2011), the Chamber found that there had been a good reason for the witnesses’ non-attendance at the trial. The Regional Court had made reasonable efforts to enable the witnesses to be examined. The fact that all attempts in this regard had remained fruitless was not imputable to that court. The Chamber further considered that while the witness statements at issue may not have been the sole or decisive evidence on which the applicant’s conviction was based, they clearly carried considerable weight in the establishment of his guilt. 71. However, in the Chamber’s view, there had been sufficient counterbalancing factors to compensate for the difficulties under which the defence laboured as a result of the admission of the victims’ witness statements. It considered that the Regional Court had complied with the procedural safeguards under domestic law. Under Article 168c of the Code of Criminal Procedure (see paragraph 56 above), the accused and defence counsel were, as a rule, permitted to be present during the judicial examination of a witness at the pre-trial stage. However, it had been justified to exclude the applicant, who had not yet been assigned a lawyer at that time, from the hearing of O. and P. by the investigating judge in accordance with Article 168c § 3 of the Code of Criminal Procedure. The investigating judge’s concern that the suspects would put pressure on the witnesses once they or counsel were informed of the hearing, thus jeopardising the ongoing investigations, had been well-founded. The Chamber also took note of the Government’s submission that, at the time of the hearing, it had not been foreseeable that the witnesses, who had already given evidence on several occasions, would refuse to testify at the trial. 72. Moreover, the Chamber noted that the Regional Court had thoroughly scrutinised O. and P.’s witness statements in the light of their reduced evidentiary value. That court had also had regard to the statements of two witnesses, E. and L., in whom the victims had confided directly after the incident. The coherent witness testimonies had been supported by factual evidence obtained by telephone tapping, by GPS surveillance and by the applicant’s own admission that he had been in the victims’ apartment at the time of the incident. Moreover, the similarity in the way in which the offences in Kassel and Göttingen had been committed had further corroborated the court’s findings. The proceedings as a whole had therefore been fair. B. The parties’ submissions 1. The applicant 73. In the applicant’s submission his right to a fair trial, including the right to examine witnesses against him under Article 6 §§ 1 and 3 (d) of the Convention, had been breached. He stressed that neither he nor his counsel had had the opportunity, at any stage of the proceedings, to examine the key witnesses O. and P. (a) The applicable principles 74. In his observations before the Grand Chamber the applicant agreed that the principles developed by the Court in Al-Khawaja and Tahery (cited above) were applicable to his case. He stressed that, according to that case-law, failure to give the defence an opportunity to cross-examine a prosecution witness would result in a breach of Article 6 § 3 (d) of the Convention save in exceptional circumstances. (b) Whether there was a good reason for the non-attendance of witnesses O. and P. at the trial 75. In the applicant’s submission there had not been a good reason for the non-attendance of witnesses O. and P. at his trial. The psychological difficulties allegedly caused by the offence in Göttingen had not prevented the witnesses from making statements to the police and the investigating judge at the investigation stage. Moreover, the Göttingen Regional Court had itself considered that there had not been sufficient reason for the witnesses not to attend the trial. Further attempts should have been made by the domestic authorities to obtain the hearing of those witnesses at the trial, notably by means of bilateral negotiations with Latvia at political level. (c) Whether the evidence of the absent witnesses was the sole or decisive basis for the applicant’s conviction 76. In the applicant’s view, his conviction had been based at least to a decisive extent on the evidence given by O. and P., who had been the only eyewitnesses to the events in Göttingen. He could not have been found guilty on the basis of the other available evidence if the evidence provided by witnesses O. and P. had been disregarded. (d) Whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured 77. The applicant took the view that there had not been any counterbalancing factors to compensate for the difficulties caused to the defence as a result of the witnesses’ absence at the trial. 78. The applicant submitted that the Regional Court had not assessed the witness statements made by O. and P. with particular caution. It had not taken into account the fact that the witnesses’ failure to attend the hearing before it without an adequate excuse had affected their credibility. Moreover, the fact that there had been some additional hearsay evidence and that the applicant had had the opportunity to question the investigating judge had not constituted sufficient counterbalancing factors to secure equality of arms in the proceedings. The fact that, under German criminal procedure law, the prosecution was obliged to investigate both the incriminating and the exonerating evidence against the accused (see paragraph 55 above) did not compensate for his lack of opportunity to cross-examine the prosecution witnesses, as the prosecution authorities had not investigated the exonerating evidence in his case. 79. The applicant stressed, in particular, that he had been deprived of a procedural safeguard under domestic law aimed at protecting his defence rights, in that counsel representing him had not been allowed to be present at the hearing of witnesses O. and P. before the investigating judge. Under the applicable provisions of the Code of Criminal Procedure (Article 141 § 3, read in conjunction with Article 140, see paragraph 57 above), as interpreted by the Federal Court of Justice (he referred to that court’s judgment of 25 July 2000, see paragraphs 58-59 and 62 above), the prosecution had been obliged to appoint counsel to represent him at the stage of the investigation proceedings. This should have been done prior to the hearing of the main witnesses for the prosecution by the investigating judge, from which he had been excluded under Article 168c § 3 of the Code of Criminal Procedure. In such circumstances, defence counsel had a right to be present at the witness hearing under Article 168c § 2 of the Code of Criminal Procedure (save in the circumstances enumerated in Article 168c § 5 of the Code of Criminal Procedure, which were not present in his case). He referred to the findings in this Court’s judgment in the case of Hümmer v. Germany (no. 26171/07 , §§ 42 et seq. , 19 July 2012 ) in support of his submission. 80. The applicant stressed that, in practice, witnesses were only heard by the investigating judge in the investigation proceedings, in addition to their examination by the police, if there was a danger of evidence being lost. Records of examinations by an investigating judge could be read out and used as evidence at the trial under less strict conditions than records of interviews by the police (Article 251 §§ 1 and 2 of the Code of Criminal Procedure, see paragraph 61 above). The presence of the accused and counsel at hearings conducted by an investigating judge in accordance with Article 168c § 2 of the Code of Criminal Procedure was thus essential in order to safeguard the accused’s right under Article 6 § 3 (d) of the Convention. 81. In the applicant’s submission, it had not been justified to deny him that right simply because the investigating judge had gained the mistaken impression that the witnesses had been afraid to testify in the applicant’s and even his counsel’s presence, without him having given cause for any such fears. In any event, this would not have justified his and counsel’s exclusion from that hearing as there were various means of allaying such fears. As witnesses O. and P. were due to leave Germany shortly after their hearing by the investigating judge, it would have been possible to appoint counsel for the applicant just before the hearing and also to arrest the applicant immediately before that hearing, thus allowing him or at least his counsel to question the witnesses in person without the latter having to fear any intimidation. 82. In the applicant’s view, the likelihood that witnesses O. and P., who had possibly been liable to punishment under the trade or tax laws because of their work as prostitutes, would no longer be available to testify in Germany in the proceedings against him had been foreseeable for the investigating authorities. He nevertheless stressed that he had not had any reason to request a repetition of the witnesses’ hearing by the investigating judge in his presence following his arrest as he had assumed that he would be able to cross-examine the witnesses at the trial; the witnesses had in any event already left Germany at the time of his arrest. 2. The respondent Government 83. In the Government’s submission, the criminal proceedings against the applicant had complied with Article 6 §§ 1 and 3 (d) of the Convention despite the fact that the applicant had not had the opportunity to cross-examine witnesses O. and P. at any stage of the proceedings. (a) The applicable principles 84. In the Government’s view, there was no reason to tighten or amend the principles established by the Court in its judgment in Al-Khawaja and Tahery (cited above), which were applicable to the present case and according to which the cross-examination of witnesses could be dispensed with in certain circumstances. The Court’s findings in that judgment, made in the context of a common-law system, should be transposed to continental-law systems in a flexible manner. Even if these principles were applied, the scope for exceptions to the principle of cross-examination was liable to be wider in continental-law systems such as the German legal system. The latter relied to a greater extent on professional judges experienced in evaluating the reliability of evidence, and the assessment of the evidence was made far more transparent in the reasoning of the judgments. 85. The Government added that a comparative-law study commissioned by them had shown that in none of the Contracting Parties to the Convention with a criminal-law system comparable to the German system was there an unrestricted right for the defendant to cross-examine prosecution witnesses at the hearing. Moreover, in many other legal systems it was not prohibited to have recourse to the records of previous witness examinations even if the accused had been unable to question the witness concerned at that stage. (b) Whether there was a good reason for the non-attendance of witnesses O. and P. at the trial 86. In the Government’s view, there had been a good reason, as defined by the Court’s case-law, for the non-attendance of witnesses O. and P. at the trial. The Regional Court had made all reasonable efforts to hear the witnesses, who had resided and worked legally in Germany, in person at the trial or to examine them via a video conference with the help of the Latvian courts. It had summoned the witnesses for a hearing. Following the submission of medical certificates by the witnesses, the court had again attempted to secure their presence by informing them that they would be protected and asking them to state in which circumstances they would be prepared to testify. The Regional Court had no jurisdiction to compel the witnesses, who were Latvian nationals residing in Latvia, to attend a hearing in Germany, as coercive measures were prohibited under Article 8 of the European Convention on Mutual Assistance in Criminal Matters (see paragraph 66 above). 87. The Government submitted that the Regional Court had then asked the Latvian authorities, by way of legal assistance in accordance with the applicable rules, to have the witnesses summoned by a court in Latvia so that they could be examined via a video conference. However, the Latvian court had cancelled the hearing following a preliminary discussion with the witnesses, who had again submitted medical certificates. The Regional Court’s request to the Latvian court asking the latter to verify the grounds given by the witnesses for their refusal to testify, or to explore further ways of questioning them, had remained unanswered. There was nothing to indicate that the hearing of the witnesses could have been brought about by different means such as bilateral negotiations on a political level, mentioned for the first time by the applicant in the proceedings before the Court. (c) Whether the evidence of the absent witnesses was the sole or decisive basis for the applicant’s conviction 88. The Government submitted that in the Regional Court’s view, which was decisive in that respect, the witness statements made by O. and P. had been “relevant” (“ maßgeblich ”) in grounding the applicant’s conviction. However, there had also been a number of other weighty items of evidence, including the results of police surveillance measures and the applicant’s own submissions, which had allowed the veracity of the witness statements to be tested. The question whether the witness evidence at issue, in the light of these elements, amounted to “decisive evidence” for the purposes of the Court’s case-law could be left open because, in any event, sufficient counterbalancing factors had been both necessary and present in the applicant’s case to compensate for the defence’s lack of opportunity to question the witnesses. (d) Whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured 89. In the Government’s submission, the impossibility for the applicant to question witnesses O. and P. had been sufficiently compensated for by the Regional Court, which had made a comprehensive and critical assessment of the credibility of the witness statements. The Regional Court had assessed with particular caution the evidence given by the two witnesses for the prosecution, inter alia by comparing the statements made by them during their different examinations. 90. The Government argued that in German criminal proceedings, both the court and the prosecution were obliged by law to investigate both the incriminating and the exonerating evidence against the accused. This partly compensated for an accused’s lack of opportunity to cross-examine a prosecution witness. 91. In testing the veracity of the witness statements, the Regional Court had also had recourse to a large number of corroborating evidentiary elements, including both hearsay witness evidence and reliable physical evidence obtained by means of surveillance of the applicant. The surveillance measures had included, in particular, analysis of the geographical data from the applicant’s mobile phone and the recording of his telephone conversation with one of his co-accused at the time of the offence, in which he had described one of the witnesses jumping down from a balcony and hiding from the applicant, who had pursued her. 92. Moreover, the applicant had had the opportunity to cross-examine and challenge the credibility of almost all the persons who had questioned O. and P. at the investigation stage; the Regional Court had also heard evidence from those persons concerning the witnesses’ conduct and emotional state during questioning. 93. As to the fact that neither the applicant nor his counsel had been given an opportunity to question witnesses O. and P. at the investigation stage, the Government argued that the investigating judge had excluded the applicant from the hearing in accordance with Article 168c § 3 of the Code of Criminal Procedure in order to ensure the witnesses’ protection and the establishment of the truth. The witnesses, who had been very frightened of the perpetrators, would not have made complete and truthful statements about the offence in the presence of the applicant. They had had legitimate grounds for their fear of revenge, given that the applicant had been suspected of committing a similar robbery in Kassel. 94. Moreover, as the witnesses would have had reason to fear that any defence counsel appointed to represent the applicant would inform the latter of the hearing and of their statements made therein, they would not have made any, or accurate, statements in counsel’s presence either. The Government explained that under Article 168c § 5 of the Code of Criminal Procedure the trial court was authorised to dispense with notifying any lawyer appointed for the applicant of the hearing if it considered that notification would endanger the success of the investigation. Therefore, in accordance with the case-law of the Federal Court of Justice (they also referred to that court’s judgment of 25 July 2000, see paragraphs 58-59 and 62 above), the appointment of defence counsel and the latter’s presence at the hearing before the investigating judge were not required. 95. The Government noted that following his arrest the applicant had not requested a repetition of the witnesses’ examination in his presence in the investigation proceedings. They stressed that it had not been foreseeable that O. and P. would not attend the trial since the applicant and his accomplices, who were in detention at that stage, would then have posed less of a threat to them. In any event, the applicant had never lodged any applications stating which questions he would have liked to put to the witnesses, whose identity and whereabouts had been known to him, or on what grounds he might have wished to challenge their credibility. 3. The Czech Government, third-party interveners 96. The Czech Government considered that the present case afforded the Court the opportunity to clarify and refine the principles developed in its judgment in Al-Khawaja and Tahery (cited above). They took the view that the principles on the admission of the evidence of absent witnesses developed therein in the context of a common-law system were not fully applicable to continental-law systems. They considered that the Court should take into account the specific features of the particular legal system concerned. 97. The intervening Government suggested that the Court, prior to examining whether there were good reasons for admitting the evidence of an absent witness (they referred to Al-Khawaja and Tahery , cited above, § 120), should verify whether the impugned evidence had been the sole or decisive evidence grounding the accused’s conviction, as it had done, for instance, in Sarkizov and Others v. Bulgaria (nos. 37981/06 , 38022/06, 39122/06 and 44278/06, § 58, 17 April 2012) and Damir Sibgatullin v. Russia (no. 1413/05 , §§ 54-56, 24 April 2012). They argued that in a situation where a statement by an absent witness was not decisive, proof of a good reason for not affording the defence the possibility to question the witness was unnecessary. Furthermore, the Court should clarify whether it still accepted the principle of impossibilium nulla obligatio est as a good reason for admitting the evidence of an absent witness. This was particularly relevant in cases where a witness had left the national courts’ jurisdiction, as the latter then lacked the coercive power to secure the witness’s presence at a trial. 98. The Czech Government further stressed that it was for the national courts to assess the significance of a witness statement for the outcome of the case. A detailed analysis by the Court of the decisiveness or degree of importance of the evidence in question was liable to come into conflict with the domestic authorities’ margin of appreciation and the Court’s fourth-instance doctrine. 99. In the Czech Government’s submission, the more flexible approach adopted by the Court in Al-Khawaja and Tahery (cited above) in respect of sole or decisive evidence made its case-law less predictable. They proposed that the Court should clarify which counterbalancing factors would be considered sufficient for preventing a breach of Article 6 of the Convention. C. The Grand Chamber’s assessment 1. Recapitulation of the relevant principles (a) The general principles 100. The Court reiterates 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 (see Al-Khawaja and Tahery , cited above, § 118); it will therefore consider the applicant’s complaint under both provisions taken together (see Windisch v. Austria , 27 September 1990, § 23, Series A no. 186, and Lüdi v. Switzerland , 15 June 1992, § 43, Series A no. 238). 101. The Court’s primary concern under Article 6 § 1 is to evaluate the overall fairness of the criminal proceedings (see, inter alia , Taxquet v. Belgium [GC], no. 926/05 , § 84, ECHR 2010 , with further references ). In making this assessment the Court will look at the proceedings as a whole, including the way in which the evidence was obtained, having regard to the rights of the defence but also to the interest of the public and the victims in seeing crime properly prosecuted (see Gäfgen v. Germany [GC], no. 22978/05 , §§ 163 and 175, ECHR 2010) and, where necessary, to the rights of witnesses (see Al-Khawaja and Tahery , cited above, § 118 , with further references, and Hümmer , cited above , § 37 ). 102. The principles to be applied in cases where a prosecution witness did not attend the trial and statements previously made by him were admitted as evidence have been summarised and refined in the judgment of the Grand Chamber of 15 December 2011 in Al-Khawaja and Tahery (cited above). 103. The Court reiterated in that judgment that Article 6 § 3 (d) enshrined the principle that, before an accused could be convicted, all evidence against him normally had to be produced in his presence at a public hearing with a view to adversarial argument (see Al-Khawaja and Tahery , cited above, § 118). 104. The Court must stress, in that context, the importance of the investigation stage for the preparation of the criminal proceedings, as the evidence obtained during this stage determines the framework in which the offence charged will be considered at the trial (see Salduz v. Turkey [GC], no. 36391/02 , § 54, ECHR 2008). Even if the primary purpose of Article 6 of the Convention, as far as criminal proceedings are concerned, is to ensure a fair trial by a “tribunal” competent to determine “any criminal charge”, it does not follow that the Article has no application to pre-trial proceedings. Thus, Article 6 - especially paragraph 3 thereof - may be relevant before a case is sent for trial if and in so far as the fairness of the trial is likely to be seriously prejudiced by an initial failure to comply with its provisions (see Salduz , cited above, § 50, referring to Imbrioscia v. Switzerland , 24 November 1993, § 36, Series A no. 275). 105. However, the use as evidence of statements obtained at the stage of a police inquiry and judicial investigation is not in itself inconsistent with Article 6 §§ 1 and 3 (d), provided that the rights of the defence have been respected. As a rule, these rights require that the defendant be given an adequate and proper opportunity to challenge and question a witness against him - either when that witness is making his statements or at a later stage of the proceedings (see Al-Khawaja and Tahery , cited above, § 118, with further references; see also A. G. v. Sweden (dec.), no. 315/09 , 10 January 2012 , and Trampevski v. the former Yugoslav Republic of Macedonia , no. 4570/07 , § 44, 10 July 2012 ). 106. In its judgment in Al-Khawaja and Tahery the Court concluded that the admission as evidence of the statement of a witness who had been absent from the trial and whose pre-trial statement was the sole or decisive evidence against the defendant did not automatically result in a breach of Article 6 § 1. It reasoned that applying the so-called “sole or decisive rule” (under which a trial was unfair if a conviction was based solely or to a decisive extent on evidence provided by a witness whom the accused had been unable to question at any stage of the proceedings; ibid . , §§ 128 and 147) in an inflexible manner would run counter to the traditional way in which the Court approached the right to a fair hearing under Article 6 § 1, namely to examine whether the proceedings as a whole had been fair. However, the admission of such evidence, because of the inherent risks for the fairness of the trial, constituted a very important factor to balance in the scales (ibid . , §§ 146-47). 107. According to the principles developed in the Al-Khawaja and Tahery judgment, it is necessary to examine in three steps the compatibility with Article 6 §§ 1 and 3 (d) of the Convention of proceedings in which statements made by a witness who had not been present and questioned at the trial were used as evidence (ibid . , § 152). The Court must examine (i) whether there was a good reason for the non-attendance of the witness and, consequently, for the admission of the absent witness’s untested statements as evidence (ibid . , §§ 119-25); (ii) whether the evidence of the absent witness was the sole or decisive basis for the defendant’s conviction (ibid . , §§ 119 and 126-47); and (iii) whether there were sufficient counterbalancing factors, including strong procedural safeguards, to compensate for the handicaps caused to the defence as a result of the admission of the untested evidence and to ensure that the trial, judged as a whole, was fair (ibid . , § 147). 108. As regards the applicability of the above principles in the context of the diverse legal systems in the Contracting States, and in particular in the context of both common-law and continental-law systems, the Court reiterates that, while it is important for it 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 (see Al-Khawaja and Tahery , cited above, § 130). 109. Furthermore, in cases arising from individual applications the Court’s task is not to review the relevant legislation in the abstract. Instead, it must confine itself, as far as possible, to examining the issues raised by the case before it (see, among many other authorities, N. C. v. Italy [GC], no. 24952/94 , § 56, ECHR 2002-X, and Taxquet , cited above, § 83). When examining cases, the Court is of course mindful of the differences between the legal systems of the Contracting Parties to the Convention when it comes to matters such as the admission of evidence of an absent witness and the corresponding need for safeguards to ensure the fairness of the proceedings. It will have due regard in the instant case to such differences when examining, in particular, whether there were sufficient counterbalancing factors to compensate for the handicaps caused to the defence as a result of the admission of the untested witness evidence (compare Al-Khawaja and Tahery , cited above, § 146). (b) The relationship between the three steps of the Al-Khawaja test 110. The Court considers that the application of the principles developed in Al-Khawaja and Tahery in its subsequent case-law discloses a need to clarify the relationship between the above-mentioned three steps of the Al-Khawaja test when it comes to the examination of the compliance with the Convention of a trial in which untested incriminating witness evidence was admitted. It is clear that each of the three steps of the test must be examined if - as in the Al-Khawaja and Tahery judgment - the questions in steps one (whether there was a good reason for the non-attendance of the witness) and two (whether the evidence of the absent witness was the sole or decisive basis for the defendant’s conviction) are answered in the affirmative (see Al-Khawaja and Tahery , cited above, §§ 120 and 147). The Court is, however, called upon to clarify whether all three steps of the test must likewise be examined in cases in which either the question in step one or that in step two is answered in the negative, as well as the order in which the steps are to be examined. (i) Whether the lack of a good reason for a witness’s non-attendance entails, by itself, a breach of Article 6 §§ 1 and 3 (d) 111. As to the question whether the lack of a good reason for a witness’s non-attendance (first step of the Al-Khawaja test) entails, by itself, a breach of Article 6 §§ 1 and 3 (d) of the Convention, without it being necessary to examine the second and third steps of the Al-Khawaja test, the Court observes the following. In its judgment in Al-Khawaja and Tahery , it considered that the requirement that there be a good reason for admitting the evidence of an absent witness was a “preliminary question” which had to be examined before any consideration was given as to whether that evidence was sole or decisive (ibid . , § 120). It further noted that it had found violations of Article 6 §§ 1 and 3 (d) even in cases in which the evidence of an absent witness had been neither sole nor decisive, when no good reason had been shown for the failure to have the witness examined (ibid., with further references). 112. The Court observes that the requirement to provide a justification for not calling a witness has been developed in its case-law in connection with the question whether the defendant’s conviction was solely or to a decisive extent based on evidence provided by an absent witness (see Al-Khawaja and Tahery , cited above, § 128). It further reiterates that the rationale underlying its judgment in Al-Khawaja and Tahery , in which it departed from the so-called “sole or decisive rule”, was to abandon an indiscriminate rule and to have regard, in the traditional way, to the fairness of the proceedings as a whole (ibid . , §§ 146-47). However, it would amount to the creation of a new indiscriminate rule if a trial were considered to be unfair for lack of a good reason for a witness’s non-attendance alone, even if the untested evidence was neither sole nor decisive and was possibly even irrelevant for the outcome of the case. 113. The Court notes that in a number of cases following the delivery of the Al-Khawaja judgment it took an overall approach to the examination of the fairness of the trial, having regard to all three steps of the Al-Khawaja test (see Salikhov v. Russia , no. 23880/05 , §§ 118 et seq., 3 May 2012; Asadbeyli and Others v. Azerbaijan , nos. 3653/05 , 14729/05, 20908/05, 26242/05, 36083/05 and 16519/06 , § 134, 11 December 2012; Yevgeniy Ivanov v. Russia , no. 27100/03 , §§ 45-50, 25 April 2013; and Şandru v. Romania , no. 33882/05 , §§ 62-70, 15 October 2013). However, in other cases, the lack of a good reason for a prosecution witness’s absence alone was considered sufficient to find a breach of Article 6 §§ 1 and 3 (d) ( see Rudnichenko v. Ukraine, no. 2775/07 , §§ 105-110, 11 July 2013 , and Nikolitsas v. Greece , no. 63117/09 , § 35, 3 July 2014; in the latter case the Court nevertheless addressed the further steps of the Al-Khawaja test, see ibid., §§ 36-39 ). In yet other cases a differentiated approach was taken: the lack of good reason for a prosecution witness’s absence was considered conclusive of the unfairness of the trial unless the witness testimony was manifestly irrelevant for the outcome of the case ( see Khodorkovskiy and Lebedev v. Russia , nos. 11082/06 and 13772/05, §§ 709-16, 25 July 2013; Cevat Soysal v. Turkey , no. 17362/03 , §§ 76-79, September 2014; and Suldin v. Russia , no. 20077/04 , §§ 56-59, 16 October 2014). The Grand Chamber, i n the light of the foregoing (see paragraphs 111-112), considers that the absence of good reason for the non-attendance of a witness cannot of itself be conclusive of the unfairness of a trial. This being said, the lack of a good reason for a prosecution witness’s absence is a very important factor to be weighed in the balance when assessing the overall fairness of a trial, and one which may tip the balance in favour of finding a breach of Article 6 §§ 1 and 3 (d). (ii) Whether sufficient counterbalancing factors are still necessary if the untested witness evidence was neither sole nor decisive 114. In its judgment in Al-Khawaja and Tahery the Court addressed the requirement of the existence of sufficient counterbalancing factors to secure a fair and proper assessment of the reliability of the evidence in the context of cases in which convictions were based solely or to a decisive extent on the evidence of absent witnesses (ibid . , § 147). 115. As regards the question whether it is necessary to review the existence of sufficient counterbalancing factors even in cases in which the importance of an absent witness’s evidence did not attain the threshold of sole or decisive evidence grounding the applicant’s conviction, the Court reiterates that it has generally considered it necessary to carry out an examination of the overall fairness of the proceedings. This has traditionally included an examination of both the significance of the untested evidence for the case against the accused and of the counterbalancing measures taken by the judicial authorities to compensate for the handicaps under which the defence laboured (see Gani v. Spain , no. 61800/08 , § 41, 19 February 2013, with many references; see also Fąfrowicz v. Poland , no. 43609/07 , §§ 58-63, 17 April 2012 ; Sellick and Sellick v. the United Kingdom (dec.), no. 18743/06 , §§ 54-55, 16 October 2012 (concerning evidence of absent witnesses characterised as “important”); Beggs v. the United Kingdom , no. 25133/06 , §§ 156-159, 6 November 2012 (concerning absent witness evidence characterised as only one piece of circumstantial evidence) ; Štefančič v. Slovenia , no. 18027/05 , §§ 42-47, 25 October 2012 (concerning evidence by an absent witness classified as one out of several elements on which the applicant’s conviction was based) ; and Garofolo v. Switzerland (dec.), no. 4380/09 , §§ 52 and 56-57 , 2 April 2013 ; but see also Matytsina v. Russia , no. 58428/10 , §§ 164-65, 27 March 2014, and Horncastle and Others v. the United Kingdom , no. 4184/10 , §§ 150-51, 16 December 2014 (in both of which, in view of the low level of importance of the absent witness’s testimony, the existence of counterbalancing factors was not examined). 116. Given that the Court’s concern is to ascertain whether the proceedings as a whole were fair, it must review the existence of sufficient counterbalancing factors not only in cases in which the evidence given by an absent witness was the sole or the decisive basis for the applicant’s conviction. It must also do so in those cases where, following its assessment of the domestic courts’ evaluation of the weight of the evidence (described in more detail in paragraph 124 below), it finds it unclear whether the evidence in question was the sole or decisive basis but is nevertheless satisfied that it carried significant weight and that its admission may have handicapped the defence. The extent of the counterbalancing factors necessary in order for a trial to be considered fair will depend on the weight of the evidence of the absent witness. The more important that evidence, the more weight the counterbalancing factors will have to carry in order for the proceedings as a whole to be considered fair. (iii) As to the order of the three steps of the Al-Khawaja test 117. The Court observes that in Al-Khawaja and Tahery , the requirement that there be a good reason for the non-attendance of the witness (first step), and for the consequent admission of the evidence of the absent witness, was considered as a preliminary question which had to be examined before any consideration was given as to whether that evidence was sole or decisive (second step; ibid . , § 120). “Preliminary”, in that context, may be understood in a temporal sense: the trial court must first decide whether there is good reason for the absence of the witness and whether, as a consequence, the evidence of the absent witness may be admitted. Only once that witness evidence is admitted can the trial court assess, at the close of the trial and having regard to all the evidence adduced, the significance of the evidence of the absent witness and, in particular, whether the evidence of the absent witness is the sole or decisive basis for convicting the defendant. It will then depend on the weight of the evidence given by the absent witness how much weight the counterbalancing factors (third step) will have to carry in order to ensure the overall fairness of the trial. 118. Against that background, it will, as a rule, be pertinent to examine the three steps of the Al-Khawaja -test in the order defined in that judgment (see paragraph 107 above). However, all three steps of the test are interrelated and, taken together, serve to establish whether the criminal proceedings at issue have, as a whole, been fair. It may therefore be appropriate, in a given case, to examine the steps in a different order, in particular if one of the steps proves to be particularly conclusive as to either the fairness or the unfairness of the proceedings (see in this connection, for instance, Nechto v. Russia , no. 24893/05 , §§ 119-25 and 126-27, 24 January 2012 ; Mitkus v. Latvia , no. 7259/03 , §§ 101-102 and 106, 2 October 2012; Gani , cited above, §§ 43-45; and Şandru , cited above, §§ 62-66, in all of which the second step, that is, the question whether the evidence of the absent witness was sole or decisive, was examined before the first step, that is, the question whether there was a good reason for the witness’s absence). (c) Principles relating to each of the three steps of the Al-Khawaja test (i) Whether there was a good reason for the non-attendance of a witness at the trial 119. Good reason for the absence of a witness must exist from the trial court’s perspective, that is, the court must have had good factual or legal grounds not to secure the witness’s attendance at the trial. If there was a good reason for the witness’s non-attendance in that sense, it follows that there was a good reason, or justification, for the trial court to admit the untested statements of the absent witness as evidence. There are a number of reasons why a witness may not attend trial, such as absence owing to death or fear (see Al-Khawaja and Tahery , cited above, §§ 120-25), absence on health grounds (see, for instance, Bobeş v. Romania , no. 29752/05 , §§ 39-40, 9 July 2013 ; Vronchenko v. Estonia , no. 59632/09 , § 58, 18 July 2013 ; and Matytsina , cited above, § 163 ) or the witness’s unreachability. 120. In cases concerning a witness’s absence owing to unreachability, the Court requires the trial court to have made all reasonable efforts to secure the witness’s attendance (see Gabrielyan v. Armenia , no. 8088/05 , § 78, 10 April 2012; Tseber v. the Czech Republic , no. 46203/08 , § 48, 22 November 2012; and Kostecki v. Poland , no. 14932/09 , §§ 65 and 66, 4 June 2013 ). The fact that the domestic courts were unable to locate the witness concerned or the fact that a witness was absent from the country in which the proceedings were conducted was found not to be sufficient in itself to satisfy the requirements of Article 6 § 3 (d), which require the Contracting States to take positive steps to enable the accused to examine or have examined witnesses against him (see Gabrielyan , cited above , § 81; Tseber , cited above , § 48; and Lučić v. Croatia , no. 5699/11 , § 79, 27 February 2014 ). Such measures form part of the diligence which the Contracting States have to exercise in order to ensure that the rights guaranteed by Article 6 are enjoyed in an effective manner (see Gabrielyan , cited above , § 81, with further references ). Otherwise, the witness’s absence is imputable to the domestic authorities (see Tseber , cited above , § 48 , and Lučić , cited above , § 79). . It is not for the Court to compile a list of specific measures which the domestic courts must have taken in order to have made all reasonable efforts to secure the attendance of a witness whom they finally considered to be unreachable ( see Tseber , cited above , § 49). However, it is clear that they must have actively searched for the witness with the help of the domestic authorities including the police (see Salikhov , cited above , §§ 116-17 ; Prăjină v. Romania , no. 5592/05 , § 47, 7 January 2014; and Lučić , cited above , § 79) and must, as a rule, have resorted to international legal assistance where a witness resided abroad and such mechanisms were available (see Gabrielyan , cited above , § 83; Fąfrowicz , cited above , § 56 ; Lučić , cited above , § 80; and Nikolitsas , cited above , § 35). 122. The need for all reasonable efforts on the part of the authorities to secure the witness’s attendance at the trial further implies careful scrutiny by the domestic courts of the reasons given for the witness’s inability to attend trial, having regard to the specific situation of each witness (see Nechto , cited above , § 127 ; Damir Sibgatullin , cited above , § 56 ; and Yevgeniy Ivanov , cited above , § 47 ). (ii) Whether the evidence of the absent witness was the sole or decisive basis for the defendant’s conviction 123. As regards the question whether the evidence of the absent witness whose statements were admitted in evidence was the sole or decisive basis for the defendant’s conviction (second step of the Al-Khawaja test), the Court reiterates that “sole” evidence is to be understood as the only evidence against the accused (see Al-Khawaja and Tahery , cited above, § 131). “Decisive” evidence should be narrowly interpreted 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 supporting evidence; the stronger the corroborative evidence, the less likely that the evidence of the absent witness will be treated as decisive (ibid . , § 131). 124. As it is not for the Court to act as a court of fourth instance (see Nikolitsas , cited above , § 30), its starting point for deciding whether an applicant’s conviction was based solely or to a decisive extent on the depositions of an absent witness is the judgments of the domestic courts (see Beggs , cited above , § 156; Kostecki , cited above , § 67; and Horncastle , cited above , §§ 141 and 150). The Court must review the domestic courts’ evaluation in the light of the meaning it has given to “sole” and “decisive” evidence and ascertain for itself whether the domestic courts’ evaluation of the weight of the evidence was unacceptable or arbitrary (compare, for instance, McGlynn v. the United Kingdom (dec.), no. 40612/11 , § 23, 16 October 2012, and Garofolo , cited above, §§ 52-53). It must further make its own assessment of the weight of the evidence given by an absent witness if the domestic courts did not indicate their position on that issue or if their position is not clear (compare, for instance, Fąfrowicz , cited above , § 58 ; Pichugin v. Russia , no. 38623/03 , §§ 196-200, 23 October 2012; Tseber , cited above , §§ 54-56; and Nikolitsas , cited above , § 36 ). (iii) Whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured 125. As to the question whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured as a result of the admission of untested witness evidence at the trial (third step of the Al-Khawaja test), the Court reiterates that these counterbalancing factors must permit a fair and proper assessment of the reliability of that evidence (see Al-Khawaja and Tahery , cited above, § 147). 126. The fact that the domestic courts approached the untested evidence of an absent witness with caution has been considered by the Court to be an important safeguard (compare Al-Khawaja and Tahery , cited above, § 161; Gani , cited above, § 48; and Brzuszczyński v. Poland , no. 23789/09 , §§ 85-86, 17 September 2013). The courts must have shown that they were aware that the statements of the absent witness carried less weight (compare, for instance, Al-Khawaja and Tahery , cited above, § 157, and Bobeş , cited above, § 46). The Court has taken into account, in that context, whether the domestic courts provided detailed reasoning as to why they considered that evidence to be reliable, while having regard also to the other evidence available (see Brzuszczyński , cited above , §§ 85-86 and 89; Prăjină , cited above , § 59; and Nikolitsas , cited above , § 37). It likewise has regard to any directions given to a jury by the trial judge as to the approach to be taken to absent witnesses’ evidence (see, for instance, Sellick and Sellick , cited above, § 55 ). 127. An additional safeguard in that context may be to show, at the trial hearing, a video recording of the absent witness’s questioning at the investigation stage in order to allow the court, prosecution and defence to observe the witness’s demeanour under questioning and to form their own impression of his or her reliability (see A. G. v. Sweden , cited above; Chmura v. Poland , no. 18475/05 , § 50, 3 April 2012 ; D. T. v. the Netherlands (dec.), no. 25307/10 , § 50, 2 April 2013 ; Yevgeniy Ivanov , cited above, § 49; Rosin v. Estonia , no. 26540/08 , § 62, 19 December ; and Gonzáles Nájera v. Spain (dec.), no. 61047/13 , § 54, 11 February 2014). 128. A further considerable safeguard is the availability at the trial of corroborative evidence supporting the untested witness statement (see, inter alia , Sică v. Romania , no. 12036/05 , §§ 76-77, 9 July 2013 ; Brzuszczyński , cited above , § 87; and Prăjină , cited above , §§ 58 and 60 ). Such evidence may comprise, inter alia , statements made at the trial by persons to whom the absent witness reported the events immediately after their occurrence (see Al-Khawaja and Tahery , cited above, § 156; McGlynn , cited above, § 24; D. T. v. the Netherlands , cited above, § 50; and Gonzáles Nájera , cited above, § 55), further factual evidence secured in respect of the offence, including forensic evidence (see, for instance, McGlynn , cited above, § 24 (DNA evidence)), or expert opinions on a victim’s injuries or credibility (compare Gani , cited above, § 48; Gonzáles Nájera , cited above, § 56; and Rosin , cited above , § 61 ). The Court has further considered as an important factor supporting an absent witness’s statement the fact that there were strong similarities between the absent witness’s description of the alleged offence committed against him or her and the description, given by another witness with whom there was no evidence of collusion, of a comparable offence committed by the same defendant. This holds even more true if the latter witness gave evidence at the trial and that witness’s reliability was tested by cross-examination (compare Al-Khawaja and Tahery , cited above, § 156). 129. Moreover, in cases in which a witness is absent and cannot be questioned at the trial, a significant safeguard is the possibility offered to the defence to put its own questions to the witness indirectly, for instance in writing, in the course of the trial (see Yevgeniy Ivanov , cited above, § 49, and Scholer v. Germany , no. 14212/10 , § 60, 18 December 2014 ). 130. Another important safeguard countering the handicaps under which the defence labours as a result of the admission of untested witness evidence at the trial is to have given the applicant or defence counsel an opportunity to question the witness during the investigation stage (see, inter alia , A. G. v. Sweden , cited above ; Gani , cited above, § 48; and Şandru , cited above, § 67). The Court has found in that context that where the investigating authorities had already taken the view at the investigation stage that a witness would not be heard at the trial, it was essential to give the defence an opportunity to have questions put to the victim during the preliminary investigation (see Rosin , cited above, §§ 57 et seq., in particular §§ 57 and 60, and Vronchenko , cited above, §§ 61 and 63, both concerning the absence at the trial of a minor victim of a sexual offence for the protection of the child’s well-being; and compare Aigner v. Austria , no. 28328/03 , §§ 41-42, 10 May 2012, and Trampevski , cited above, § 45). Such pre-trial hearings are indeed often set up in order to pre-empt any risk that a crucial witness might not be available to give testimony at the trial (see Chmura , cited above, § 51). 131. The defendant must further be afforded the opportunity to give his own version of the events and to cast doubt on the credibility of the absent witness, pointing out any incoherence or inconsistency with the statements of other witnesses (see Aigner , cited above, § 43; D. T. v. the Netherlands , cited above, § 50; Garofolo , cited above, § 56; and Gani , cited above , § 48 ). Where the identity of the witness is known to the defence, the latter is able to identify and investigate any motives the witness may have for lying, and can therefore contest effectively the witness’s credibility, albeit to a lesser extent than in a direct confrontation (see Tseber , cited above , § 63; Garofolo , cited above, § 56; Sică , cited above, § 73; and Brzuszczyński , cited above , § 88 ). 2. Application of these principles to the present case (a) Whether there was a good reason for the non-attendance of witnesses O. and P. at the trial 132. In the present case the Court shall examine, first, whether there was a good reason for the non-attendance of prosecution witnesses O. and P. at the trial from the trial court’s perspective and, as a result, a good reason or justification for that court to admit the untested statements of the absent witnesses as evidence (see paragraph 119 above). 133. In determining whether the Regional Court had good factual or legal grounds for not securing the witnesses’ attendance at the trial, the Court would note at the outset that, as rightly stressed by the applicant, the Regional Court did not accept the witnesses’ state of health or fear on their part as justification for their absence at the trial. 134. This is demonstrated by the fact that the Regional Court, by letter of 29 August 2007, asked the witnesses residing in Latvia to appear at the hearing although they had previously refused to comply with the court’s summons, relying on medical certificates indicating that they were in an unstable post-traumatic emotional and psychological state (see paragraphs 23-24 above). In addition, following the cancellation of the hearing by the Latvian court, before which the witnesses had again relied on medical certificates indicating that they were still suffering from post-traumatic disorder, the Regional Court indicated to the Latvian court that, according to the standards of German criminal procedure law, the witnesses had not sufficiently substantiated their refusal to testify. The Regional Court therefore suggested to the Latvian court that it have the witnesses’ state of health and ability to testify examined by a public medical officer or, alternatively, that it compel them to attend the hearing in Latvia. The Latvian court did not respond to these suggestions (see paragraphs 26-27 above). 135. It was only after these efforts to hear the witnesses in person proved futile that the Regional Court found that there were insurmountable obstacles to its hearing the witnesses in the near future. The Regional Court, relying on Article 251 §§ 1 (2) and 2 (1) of the Code of Criminal Procedure, therefore admitted the records of the witnesses’ examination at the investigation stage as evidence in the proceedings (see paragraph 28 above). The reason for this measure by the Regional Court was therefore the witnesses’ unreachability for the trial court, which lacked power to compel them to appear (that is, a procedural or legal ground), and not their state of health or fear on their part (a substantive or factual ground). 136. As required in cases concerning the absence of prosecution witnesses owing to unreachability, the Court must examine whether the trial court made all reasonable efforts to secure the witnesses’ attendance (see paragraph 120 above). It notes in this regard that the Regional Court took considerable positive steps to enable the defence, the court itself and the prosecution to examine witnesses O. and P. 137. The Regional Court, having critically reviewed the reasons given by each witness for refusing to testify at the trial in Germany, as set out in medical certificates submitted by them, and having, as shown above, considered these reasons insufficient to justify their non-attendance, contacted the witnesses individually, offering them different options in order to testify at the trial, which the witnesses declined. 138. The Regional Court then had recourse to international legal assistance and requested that the witnesses be summoned to appear before a Latvian court in order for the presiding judge of the Regional Court to examine them via a video link and to enable the defence to cross-examine them. However, the hearing was cancelled by the Latvian court, which accepted the witnesses’ refusal to testify on the basis of the medical certificates they had submitted. The Regional Court, having again critically reviewed the reasons given for the witnesses’ inability to attend the trial, as mentioned above, then even suggested to the Latvian court that it have the witnesses’ state of health examined by a public medical officer or that it compel the witnesses to attend the hearing, a suggestion which received no response (see, in detail, paragraphs 23-27 above). 139. In view of these elements the Grand Chamber, sharing the Chamber’s conclusion in this regard, finds that the Regional Court made all reasonable efforts within the existing legal framework (see paragraphs 64-66) to secure the attendance of witnesses O. and P. It did not have any other reasonable means within its jurisdiction, on the territory of Germany, to secure the attendance at the trial of O. and P., Latvian nationals residing in their home country. The Court considers, in particular, that there is nothing to indicate that the trial court would have been likely to obtain a hearing of the witnesses, within a reasonable time, following bilateral negotiations with the Republic of Latvia at political level, as proposed by the applicant. In line with the principle impossibilium nulla est obligatio , the witnesses’ absence was thus not imputable to the domestic court. 140. Accordingly, there was a good reason, from the trial court’s perspective, for the non-attendance of witnesses O. and P. at the trial and, as a result, for admitting the statements they had made to the police and the investigating judge at the pre-trial stage as evidence. (b) Whether the evidence of the absent witnesses was the sole or decisive basis for the applicant’s conviction 141. In determining the weight of the evidence given by the absent witnesses and, in particular, whether the evidence given by them was the sole or decisive basis for the applicant’s conviction, the Court has regard, in the first place, to the domestic courts’ assessment. It observes that the Regional Court considered O. and P. to have been key witnesses for the prosecution ( “maßgebliche [n] Belastungszeuginnen” ), but relied on further available evidence (see paragraphs 32 and 36 above). The Federal Court of Justice, for its part, in dismissing the applicant’s appeal on points of law, made a general reference to the reasoning provided by the Federal Public Prosecutor General before that court. The latter had argued that the said witness statements had been neither the sole nor the decisive basis for the applicant’s conviction as the Regional Court had based its findings on further significant evidence (see paragraph 49 above). 142. The Court finds that the domestic courts, which did not consider O. and P.’s witness statements as the sole (that is to say, only) evidence against the applicant, did not clearly indicate whether they considered the witness statements in question as “decisive” evidence as defined by the Court in its judgment in Al-Khawaja and Tahery (which itself was delivered after the domestic courts’ decisions in the present case), that is, as being of such significance as to be likely to be determinative of the outcome of the case (see paragraph 123 above). The Regional Court’s classification of the witnesses as “ maßgeblich ” (which, in addition to “key”, may also be translated as “important”, “significant” or “decisive”), is not unambiguous in this regard. Moreover, the Federal Court of Justice’s general reference to the reasoning given by the Federal Public Prosecutor General denying that the victims’ statements were the sole or decisive basis for the applicant’s conviction (see paragraph 49 above) cannot be understood as signifying that that court endorsed each and every argument made by the prosecutor. 143. In making its own assessment of the weight of the witness evidence in the light of the domestic courts’ findings, the Court must have regard to the strength of the additional incriminating evidence available (see paragraph 123 above). It observes that the Regional Court had before it, in particular, the following further evidence concerning the offence: the hearsay statements made by the witnesses’ neighbour E. and their friend L. at the trial concerning the account O. and P. had given them of the events of 3 February 2007; the applicant’s own admission in the course of the trial that he had been in O. and P.’s apartment at the relevant time, and had jumped from the balcony to follow P.; the geographical data and recordings of two mobile telephone conversations between one of the co-accused and the applicant at the time of the offence, which revealed that the applicant had been present in an apartment at the scene of the crime and had jumped from the balcony to chase one of the escaping inhabitants; the GPS data revealing that the car of one of the co-accused had been parked near the witnesses’ apartment at the relevant time; and, finally, the evidence relating to the offence committed in Kassel on 14 October 2006 by the applicant and an accomplice. 144. The Court, having regard to these elements of evidence, cannot but note that O. and P. were the only eyewitnesses to the offence in question. The other evidence available to the courts was either just hearsay evidence or merely circumstantial technical and other evidence which was not conclusive as to the robbery and extortion as such. In view of these elements, the Court considers that the evidence of the absent witnesses was “decisive”, that is, determinative of the applicant’s conviction. (c) Whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured 145. The Court must further determine, in a third step, whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured as a result of the admission of the decisive evidence of the absent witnesses. As shown above (see paragraphs 125-131), the following elements are relevant in this context: the trial court’s approach to the untested evidence, the availability and strength of further incriminating evidence, and the procedural measures taken to compensate for the lack of opportunity to directly cross-examine the witnesses at the trial. (i) The trial court’s approach to the untested evidence 146. As regards the domestic courts’ treatment of the evidence of the absent witnesses O. and P., the Court observes that the Regional Court approached that evidence with caution. It expressly noted in its judgment that it had been obliged to exercise particular diligence in assessing the witnesses’ credibility, as neither the defence nor the court had been able to question and observe the demeanour of the witnesses at the trial. 147. The Court observes in that context that the Regional Court was unable to watch, at the trial, a video recording of the witness hearing before the investigating judge, no such recording having been made. It notes that trial courts in different legal systems have recourse to that possibility (compare the examples in paragraph 127 above) which allows them, as well as the defence and the prosecution, to observe a witness’s demeanour under questioning and to form a clearer impression of the witness’s credibility. 148. The Regional Court, in its thoroughly reasoned judgment, made it clear that it was aware of the reduced evidentiary value of the untested witness statements. It compared the content of the repeated statements made by both O. and P. at the investigation stage and found that the witnesses had given detailed and coherent descriptions of the circumstances of the offence. The trial court considered that minor contradictions in the witnesses’ statements could be explained by their concern not to disclose their professional activities to the authorities. It further observed that the witnesses’ inability to identify the applicant showed that they had not testified with a view to incriminating him. 149. The Court further observes that the Regional Court, in assessing the witnesses’ credibility, also addressed different aspects of their conduct in relation to their statements. It took into account the fact that the witnesses had not reported the offence to the police immediately and that they had failed to attend the trial without an adequate excuse. It considered that there were explanations for that conduct - namely the witnesses’ fear of encountering problems with the police or of acts of revenge by the perpetrators, and their unease about having to recall and be questioned about the offence - which did not affect their credibility. 150. In view of the foregoing, the Court considers that the Regional Court examined the credibility of the absent witnesses and the reliability of their statements in a careful manner. It notes in that context that its task of reviewing the trial court’s approach to the untested evidence is facilitated by the fact that the Regional Court, as is usual in a continental-law system, gave reasons for its assessment of the evidence before it. (ii) Availability and strength of further incriminating evidence 151. The Court further observes that the Regional Court, as shown above (see paragraphs 143-44), had before it some additional incriminating hearsay and circumstantial evidence supporting the witness statements made by O. and P. (iii) Procedural measures aimed at compensating for the lack of opportunity to directly cross-examine the witnesses at the trial 152. The Court observes that the applicant had the opportunity to give his own version of the events on 3 February 2007 - an opportunity of which he availed himself - and to cast doubt on the credibility of the witnesses, whose identity had been known to him, also by cross-examining the other witnesses giving hearsay evidence at his trial. 153. The Court notes, however, that the applicant did not have the possibility to put questions to witnesses O. and P. indirectly, for instance in writing. Moreover, neither the applicant himself nor his lawyer was given the opportunity at the investigation stage to question those witnesses. 154. The Court observes in that context that the parties disagreed as to whether or not the refusal to appoint defence counsel for the applicant and to permit counsel to take part in the witnesses’ hearing before the investigating judge had complied with domestic law. The Court considers that it is not necessary for the purposes of the present proceedings for it to take a final stance on that question. It reiterates that in examining compliance with Article 6 of the Convention, it is not its function to determine whether the domestic courts acted in accordance with domestic law (compare Schenk v. Switzerland , 12 July 1988, § 45, Series A no. 140; Jalloh v. Germany [GC], no. 54810/00 , § 94, ECHR 2006-IX; and Heglas v. the Czech Republic , no. 5935/02 , § 84, 1 March 2007), but to evaluate the overall fairness of the trial in the particular circumstances of the case, including the way in which the evidence was obtained (compare paragraph 101 above). 155. The Court considers that in the present case it is sufficient for it to note that, under the provisions of German law, the prosecution authorities could have appointed a lawyer for the applicant (Article 141 § 3 of the Code of Criminal Procedure, read in conjunction with Article 140 § 1). That lawyer would have had a right to be present at the witness hearing before the investigating judge and, as a rule, would have had to be notified thereof (Article 168c §§ 2 and 5 of the Code of Criminal Procedure). However, these procedural safeguards, which existed in the Code of Criminal Procedure and were reinforced by their interpretation by the Federal Court of Justice (see paragraphs 58-59 above), were not used in the applicant’s case. 156. The Court would stress that, while Article 6 § 3 (d) of the Convention concerns the cross-examination of prosecution witnesses at the trial itself, the way in which the prosecution witnesses’ questioning at the investigation stage was conducted attains considerable importance for, and is likely to prejudice, the fairness of the trial itself where key witnesses cannot be heard by the trial court and the evidence as obtained at the investigation stage is therefore introduced directly into the trial (compare paragraph 104 above). 157. In such circumstances, it is vital for the determination of the fairness of the trial as a whole to ascertain whether the authorities, at the time of the witness hearing at the investigation stage, proceeded on the assumption that the witness would not be heard at the trial. Where the investigating authorities took the reasonable view that the witness concerned would not be examined at the hearing of the trial court, it is essential for the defence to have been given an opportunity to put questions to the witness at the investigation stage (compare also Vronchenko , cited above, §§ 60 et seq., and Rosin , cited above, §§ 57 et seq., where the minor victims of a sexual offence were promised at the investigation stage that they would not be asked questions about the offence on any further occasions). 158. The Court notes in this regard that the applicant challenged the Regional Court’s finding that the witnesses’ absence at the trial had not been foreseeable. It agrees with the applicant that the witnesses were heard by the investigating judge because, in view of the witnesses’ imminent return to Latvia, the prosecution authorities considered that there was a danger of their evidence being lost. This is shown by the reasoning of the prosecution’s own request to the investigating judge to hear O. and P. speedily in order to obtain a true statement which could be used at the subsequent trial (see paragraph 20 above). The Court observes in that context that under Article 251 § 1 of the Code of Criminal Procedure, the written records of a witness’s previous examination by an investigating judge may be read out at the trial under less strict conditions than the records of a witness examination by the police (Article 251 § 2 of the Code of Criminal Procedure; see paragraph 61 above). 159. The Court observes that in the present case the authorities were aware that witnesses O. and P. had not pressed charges against the perpetrators immediately for fear of problems with the police and acts of revenge by the perpetrators, that they had been staying in Germany only temporarily while their families remained in Latvia and that they had explained that they wished to return to their home country as soon as possible. In these circumstances, the prosecution authorities’ assessment that it might not be possible to hear evidence from those witnesses at a subsequent trial against the applicant in Germany indeed appears convincing. 160. Despite this, the prosecution authorities did not give the applicant an opportunity - which he could have been given under the provisions of domestic law - to have witnesses O. and P. questioned at the investigation stage by a lawyer appointed to represent him. By proceeding in that manner, they took the foreseeable risk, which subsequently materialised, that neither the accused nor his counsel would be able to question O. and P. at any stage of the proceedings (compare, for the importance of counsel’s presence at the hearing of prosecution witnesses by the investigating judge, Hümmer , cited above, §§ 43 and 48). (iv) Assessment of the trial’s overall fairness 161. In assessing the overall fairness of the trial the Court will have regard to the available counterbalancing factors, viewed in their entirety in the light of its finding to the effect that the evidence given by O. and P. was “decisive” for the applicant’s conviction (see paragraph 144 above). 162. The Court observes that the trial court had before it some additional incriminating evidence regarding the offence of which the applicant was found guilty. However, the Court notes that hardly any procedural measures were taken to compensate for the lack of opportunity to directly cross-examine the witnesses at the trial. In the Court’s view, affording the defendant the opportunity to have a key prosecution witness questioned at least during the pre-trial stage and via his counsel constitutes an important procedural safeguard securing the accused’s defence rights, the absence of which weighs heavily in the balance in the examination of the overall fairness of the proceedings under Article 6 §§ 1 and 3 (d). 163. It is true that the trial court assessed the credibility of the absent witnesses and the reliability of their statements in a careful manner, thus attempting to compensate for the lack of cross-examination of the witnesses, and that the applicant had the opportunity to give his own version of the events in Göttingen. However, in view of the importance of the statements of the only eyewitnesses to the offence of which he was convicted, the counterbalancing measures taken were insufficient to permit a fair and proper assessment of the reliability of the untested evidence. 164. The Court considers that, in these circumstances, the absence of an opportunity for the applicant to examine or have examined witnesses O. and P. at any stage of the proceedings rendered the trial as a whole unfair. 165. Accordingly, there has been a violation of Article 6 §§ 1 and 3 (d) of the Convention. II. APPLICATION OF ARTICLE 41 OF THE CONVENTION 166. 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.” 167. The applicant did not make any claims for just satisfaction in his observations dated 25 June 2013 in reply to the Government’s observations before the Chamber. He claimed 30,000 euros (EUR) in compensation and EUR 10,000 in costs and expenses, both in the application form and at the hearing before the Grand Chamber, without providing any further details or documentary proof. 168. The Government did not comment on the question of just satisfaction in the proceedings before the Court. 169. Under Rule 60 § 2 of the Rules of Court an applicant must submit itemised particulars of all claims, together with any relevant supporting documents, within the time-limit fixed for the submission of the applicant’s observations on the merits. If the applicant fails to comply with these requirements, the Court may reject the claim in whole or in part (Rules 60 § 3 and 71). In its letter dated 15 May 2013 the Court drew the applicant’s attention to the fact that these requirements applied even if he had indicated his wishes concerning just satisfaction at an earlier stage of the proceedings. 170. The Court observes that the applicant failed to submit any just satisfaction claims, together with any relevant supporting documents, within the time-limit fixed therefor in the proceedings before the Chamber. The applicant, who had been granted legal aid in the proceedings before the Court, likewise did not submit new quantified claims, together with the required supporting documents, in respect of additional costs and expenses incurred in the proceedings before the Grand Chamber. The Court, having regard to Rule 60, therefore makes no award under Article 41 of the Convention. FOR THESE REASONS, THE COURT 1. Holds , by nine votes to eight, that there has been a violation of Article 6 §§ 1 and 3 (d) of the Convention; 2. Dismisses , unanimously, the applicant’s 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 2015. Lawrence Early Dean Spielmann Jurisconsult 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) Joint concurring opinion of Judges Spielmann, Karakaş Sajó and Keller; (b) Joint dissenting opinion of Judges Hirvelä, Popović, Pardalos, Nußberger, Mahoney and Kūris; (c) Dissenting opinion of Judge Kjølbro. D. S. T. L. E. JOINT CONCURRING OPINION OF JUDGES SPIELMANN, KARAKAŞ, SAJÓ AND KELLER 1. We agree with the view of the majority as far as the violation of Article 6 §§ 1 and 3 (d) of the Convention is concerned. However, we have some concerns that the majority’s clarification of the judgment rendered in Al-Khawaja and Tahery v. the United Kingdom ([GC], nos. 26766/05 and 22228/06, ECHR 2011) will result in a weakening of the fundamental role of defence rights. 2. We shall first explain our concerns with regard to the new approach to the application of the threefold test developed by the Court in Al-Khawaja and Tahery (I), before highlighting some critical points concerning the application of those principles in the case at hand (II). I. Clarification of the “sole or decisive rule” 3. The crucial issue underlying the present case is the extent to which the Court may apply the criteria of the three-step approach in a different order, and whether the fact that there is no good reason for the non-attendance of a witness will automatically lead to a violation of Article 6 §§ 1 and 3 (d) or whether the other steps still have to be examined. 4. The Court firstly made clear that the lack of a good reason for the absence of a witness during the main trial does not automatically result in a breach of Article 6 § 1 of the Convention (see paragraph 113 of the judgment). Secondly, it considered whether, in cases where it remains unclear if the evidence provided by that witness was the sole or decisive basis for the accused’s conviction, it still has to evaluate the counterbalancing factors. The Court also answered this question in the affirmative, finding that such evaluation was necessary in order to ascertain the overall fairness of the proceedings (see paragraph 116 of the judgment). It took the view that all three steps are interrelated and that the fairness of the proceedings should be measured in the light of all of the criteria (see paragraph 118 of the judgment). 5. We partly disagree with this clarification. We start from the assumption that the aim of the threefold test in Al-Khawaja and Tahery was to lend substance to the overall fairness test in situations in which the accused could not confront witnesses in person but the statements of the witnesses were nonetheless used as incriminating evidence. 6. We understand the three steps outlined in Al-Khawaja and Tahery as three independent - although related - steps. We would have preferred the Court to state that the unjustified absence of a witness amounts to a violation of Article 6 §§ 1 and 3 (d) of the Convention even if his or her statement was not the sole or decisive basis for the accused’s conviction, if it was of some importance to the trial. In other words, if the national authorities fail to provide good reasons for the absence of the witness, the Court need not examine the second and third steps of the Al-Khawaja and Tahery test. This approach has already been applied by the Court in the cases of Gabrielyan v. Armenia (no. 8088/05 , §§ 77, 84, 10 April 2012), Rudnichenko v. Ukraine (no. 2775/07 , §§ 105-110, 11 July 2013), Nikolitsas v. Greece (no. 63117/09 , § 35, 3 July 2014) and Karpyuk and others v. Ukraine (nos. 305832/04 and 32152/04, § 108, 6 October 2015, not yet final at the time of writing). For example, in Rudnichenko v. Ukraine the Court found a violation based solely on the first step and stated clearly: “The foregoing considerations are sufficient to enable the Court to conclude that there were no reasons, let alone good reasons, for the restriction of the applicant’s right to obtain the examination of the witness ... In these circumstances, the Court does not consider it necessary to proceed with the second part of the test...” (§ 109; in the same vein, see, for example, Suldin v. Russia, no. 20077/04 , § 58, 16 October 2014). In other cases the approach seems less clear, but there has at least been a trend towards finding that the unjustified absence of a main witness amounts to a breach of the Convention (see Khodorkovskiy and Lebedev v. Russia , nos. 11082/06 and 13772/05, § 715, 25 July 2013, and Cevat Soysal v. Turkey , no. 17362/03 , §§ 77-78, 23 September 2014). 7. We regret that the majority in the present case proceeded to test the other steps even absent a good reason for the non-appearance of witnesses. In our view, the defence must have the possibility to challenge any witness testimony that is of some importance to the trial. If the national courts cannot advance “good reasons” for the absence of a witness, then there has been a violation of Article 6 §§ 1 and 3 (d) of the Convention. Not only for logical reasons, but also in the interests of the efficiency of the Court’s work and procedural economy, we would opt for finding a violation at this early stage of the judgment under such circumstances. 8. In addition to the above, we consider a further remark to be necessary. The Court’s approach (see paragraphs 123 and 124 of the judgment) reduces the importance of the second step of the threefold test (whether the evidence was “sole or decisive”). The national judge will in reality avoid characterising a hearsay statement as “sole or decisive”. The case at hand is a good example of this problem. The national courts characterised the statements of witnesses O. and P. as “ massgeblich ”, meaning decisive, thus identifying O. and P. as key witnesses. In this regard it is important that the Court look beyond the wording of the national court’s characterisation, in the light of the second step (as done in paragraphs 143 and 144 of the judgment). Otherwise, this approach will in fact lead to a two-stage test, under which it is only necessary to examine whether there were good reasons for the non-attendance of the witness and sufficient counterbalancing factors. 9. Finally, we agree that the Court should scrutinise carefully whether the national authorities applied sufficient counterbalancing measures (see paragraphs 125 et seq.). II. Application of these principles in the case at hand 10. In the present case, the Court first examined whether there had been a good reason for the non-attendance of witnesses O. and P. at the trial. It went on to determine whether the evidence given by those two witnesses had been sole or decisive for the criminal conviction of the defendant by the domestic courts. After responding to these questions in the affirmative, the Court then examined the various counterbalancing factors, ending with an assessment of the overall fairness of the trial. 11. We agree with the majority’s finding in paragraph 140 that there was good reason for the non-attendance of witnesses O. and P. It is unnecessary to determine at this stage whether these reasons were sufficient. In any event, the trial court made all reasonable efforts to ensure the appearance of the witnesses at trial. 12. We fully subscribe to the Court’s finding in paragraph 144. We would like to stress that the simple fact that O. and P. were the only eyewitnesses to the relevant events sufficed for the Court to regard their evidence as “decisive” (see also paragraph 8 above in this connection). 13. Before considering the last part of the Al-Khawaja and Tahery test (counterbalancing factors), a preliminary remark is necessary. There is a recent tendency at the national level to shift procedural measures which belong to the trial stage forward to the investigation stage; the proceedings against the applicant in the present case are a good example of this tendency. When the Court finds itself confronted with such a shift, we see two different possible reactions. On the one hand, the Court could find that bringing procedural steps forward to the investigation stage is entirely incompatible with the Convention. This is not the approach that the Court has adopted. On the other hand, the Court could allow such procedural measures to be taken as early as the investigative stage. However, if the Court considers this shifting of procedural measures to the investigation stage to be in accordance with the Convention, it must make unmistakably clear that the relevant procedural safeguards must be rigorously adhered to. Otherwise, the right to confront a witness at the trial stage would be seriously undermined. In the case of Salduz v. Turkey ([GC], no. 36391/02 , § 54, ECHR 2008), this Court already stressed “the importance of the investigation stage for the preparation of the criminal proceedings, as the evidence obtained during this stage determines the framework in which the offence charged will be considered at the trial...”. 14. That being so, we see some difficulties in the case at issue. German law provides two safeguards for the situation at hand. Under Article 168c § 2 of the Code of Criminal Procedure, the accused (and his or her lawyer) is allowed to be present if a witness is questioned by a judge at the pre-trial stage. In exceptional cases the accused may be excluded if his or her presence would endanger the outcome or purpose of the questioning, in particular if it is reasonably to be feared that a witness will not tell the truth in the presence of the accused (Article 168c § 2 CCP). We have doubts as to whether the exclusion of the accused in the case at issue was in accordance with the requirements of Article 168c § 2 CCP. The accused and the witness already knew each other. In this regard, the case can be differentiated from the case of Pesukic v. Switzerland (no. 25088/07 , 6 December 2012) in which the disclosure of the witness’s identity was at stake. However, if the accused is excluded, his or her defence counsel is entitled to be present. 15. The second safeguard must be considered in the light of the first. According to Article 141 § 3 CCP, the prosecutor can provide defence counsel to assist the accused during the pre-trial stage. Independently of the uncertainty regarding the correct interpretation of this provision (see paragraphs 58 et seq. and paragraph 154 of the judgment), the national authorities failed to give reasons why the standard of Article 141c CCP did not apply to the applicant in the present case. 16. In the light of the special circumstances of the case at hand, the non-application of both safeguards provided for in domestic law must be seen as a serious shortcoming in the pre-trial proceedings. If those safeguards are not rigorously applied in the early stages of the investigations, the rights guaranteed by Article 6 § 3 (d) of the Convention during the trial stage may lose their importance. III. Conclusion 17. If the Grand Chamber allows an overall assessment of the fairness of proceedings to be conducted in the absence of good reasons for the non-attendance of a witness, the right to confront witnesses will become very weak. We agree that within the threefold test there must be a degree of flexibility. However, an approach that unconditionally leads to a final overall examination of the fairness of the proceedings would give too much leeway to the national authorities. This application of the three-step examination would imply that there was no need for the different steps as long as the overall fairness test was fulfilled. 18. The Court’s overly cautious approach is also evident in paragraph 118. Although the order of the three questions is pertinent in principle, the majority stated that it “may ... be appropriate, in a given case, to examine the steps in a different order.” We are not convinced that the Court has given clear guidance to the national authorities as to the appropriate application of the Al-Khawaja and Tahery test. 19. We have a reasonable fear that the clarification provided by the Court in this case (which will be known as the “ Schatschaschwili test” in the future) can be summarised in one single question: were the proceedings fair as a whole? This overall test is not, in our view, a step in the direction of strengthening the rights guaranteed by Article 6 (3) (d) of the Convention. JOINT DISSENTING OPINION OF JUDGES HIRVELÄ, POPOVIĆ, PARDALOS, NUSSBERGER, MAHONEY AND KŪRIS 1. We regret that we are unable to agree with the view of the majority that the applicant’s rights under Article 6 §§ 1 and 3 (d) of the Convention were violated in the present case. A. As to the recapitulation of the relevant principles 2. We should make it clear at the outset that our difference of opinion with the majority of the Grand Chamber does not relate to the recapitulation of the general principles relevant to the case, in respect of which we are in full agreement with the majority. 3. In our view, the Grand Chamber’s judgment in the present case confirms the principles which the Court established in its judgment of 15 December 2011 in the case of Al-Khawaja and Tahery v. the United Kingdom ([GC], nos. 26766/05 and 22228/06, ECHR 2011). It further clarifies the relationship between the three steps of the Al-Khawaja test for examining the compatibility with Article 6 §§ 1 and 3 (d) of proceedings in which statements made by a prosecution witness who was not present and questioned at the trial were used as evidence. 4. The need for clarification, which had become apparent in the Court’s post- Al-Khawaja case-law in cases in which the factual situations differed from that at issue in Al-Khawaja , essentially concerned three points. 5. Firstly, the Grand Chamber clarified that the absence of good reason for the non-attendance of a prosecution witness is not of itself conclusive of the unfairness of a trial. It is, however, a very important factor to be weighed in the balance when assessing the overall fairness of a trial. We agree with the Grand Chamber’s finding in the present case that the rationale underlying the Court’s judgment in Al-Khawaja and Tahery , in which it departed from the so-called “sole or decisive rule”, was to abandon an indiscriminate rule and to have regard, in the traditional way, to the fairness of the proceedings as a whole. It would have amounted to the creation of a new indiscriminate rule if a trial were considered to be unfair for lack of a good reason for a witness’s non-attendance alone, even if the untested evidence was neither sole nor decisive and was possibly even irrelevant for the outcome of the case (see paragraph 112 of the judgment). In line with that finding, a large majority of the Al-Khawaja follow-up cases did indeed not consider the lack of good reason for a prosecution witness’s absence alone to entail an automatic breach of Article 6 §§ 1 and 3 (d) (for references see paragraph 113 of the judgment). 6. Secondly, we agree with the majority that, given that the Court’s concern is to ascertain whether the proceedings as a whole were fair, it must review the existence of sufficient counterbalancing factors also in cases where the untested witness evidence was neither the sole nor the decisive basis for the defendant’s conviction, but carried significant weight (see paragraph 116 of the judgment). 7. Thirdly, we equally consider that the order in which the three steps are to be examined as defined in the case of Al-Khawaja and Tahery is, as a rule, pertinent, even though it may be appropriate in certain circumstances to depart from that order (see paragraph 118 of the judgment). 8. Lastly, we agree with the summary of the principles relating to each of the three steps of the Al-Khawaja test in the present judgment (see paragraphs 119-131 of the judgment). These give guidance, in particular, on how to assess the unreachability of a witness and what kind of efforts are required from the domestic authorities to reach the witness, how to evaluate whether evidence was the sole or decisive basis for a defendant’s conviction and what kind of substantive or procedural counterbalancing factors can serve to compensate for the handicaps under which the defence laboured as a result of the admission of untested witness evidence at the trial. B. As to the application of these principles to the present case 9. Where we part company with the majority is on the question of the application of the relevant principles to the present case. We agree with the majority’s finding that there was a good reason for the non-attendance of witnesses O. and P. at the trial and, as a result, for admitting the statements they had made to the police and the investigating judge at the pre-trial stage as evidence, and with the reasoning given therefor. We can also accept the majority’s conclusion that the evidence of the absent witnesses O. and P. was a decisive, albeit not the sole, basis for the applicant’s conviction in the present case, as O. and P. were the only eyewitnesses to the offence in question. 10. In view of this finding, we consider it necessary to examine whether there were sufficient counterbalancing factors to compensate for the handicaps under which the defence laboured. In contrast to the majority’s finding, we take the view that the counterbalancing factors were sufficient in the present case. 11. As to the assessment of the different counterbalancing factors present, we agree with the majority’s finding that the Regional Court examined the credibility of the absent witnesses and the reliability of their statements in a careful manner and we consider its examination to have been particularly thorough. 12. But, contrary to the majority, we find that the Regional Court had before it very strong and coherent additional incriminating evidence regarding the offence of robbery combined with extortion of which the applicant was convicted. Not only did the evidence comprise a complete account of the events given by two additional witnesses (the witnesses’ neighbour E. and their friend L.), albeit in the form of hearsay evidence only. It was also fully supported by very strong direct and reliable technical evidence. The latter included, in particular, the geographical data and recordings of two mobile telephone conversations proving that the applicant had been present in an apartment at the scene of the crime and had jumped from the balcony to chase one of the escaping inhabitants. Finally, the evidence relating to the offence committed in Kassel on 14 October 2006 by the applicant and an accomplice, in respect of which all the witnesses testified at the trial, bore striking similarities to the offence committed in Göttingen as regards the victims chosen, the place of the offence and the manner in which the perpetrators proceeded. Furthermore, we cannot but note that the applicant himself admitted in the course of the trial that he had been in the witnesses’ apartment at the relevant time and had followed P. when she had escaped over the balcony, arguing that he had done so for fear of problems he had previously encountered with prostitutes on a similar occasion in Kassel (see paragraph 44 of the judgment). 13. As regards the procedural measures aimed at compensating for the lack of opportunity to directly cross-examine the witnesses at the trial, we observe that the domestic courts did not consider it contrary to Article 141 § 3 of the Code of Criminal Procedure, read in conjunction with Article 140 § 1 and as interpreted by the Federal Court of Justice (see paragraphs 28-29, 57-59 and 62 of the judgment), that no defence counsel had been appointed to represent the applicant at the time of the witness hearing before the investigating judge. We take note in that context of the Government’s explanation (see paragraph 94 of the judgment) to the effect that, under Article 168c § 5 of the Code of Criminal Procedure, the trial court was authorised to dispense with giving notice of the hearing to any lawyer appointed to represent the applicant if it considered that notification would endanger the success of the investigation. 14. We agree with the majority that the way in which the prosecution witnesses’ questioning at the investigation stage was conducted attains considerable importance for, and can prejudice, the fairness of the trial itself where key witnesses cannot be heard by the trial court and the evidence as obtained at the investigation stage is therefore introduced directly into the trial. However, we disagree with the majority as regards its finding that the prosecution authorities, at the time of the questioning of the witnesses at the investigation stage in the absence of the applicant and his counsel, proceeded on the assumption that the witnesses could not be heard at the trial. 15. We share the applicant’s view that the witnesses O. and P. were heard by the investigating judge because, owing to the witnesses’ imminent return to Latvia, the prosecution authorities considered that there was a danger of their evidence being lost. This is shown by the reasoning of the prosecution’s own request to the investigating judge to hear evidence from O. and P. speedily. However, the fact that it must be considered as foreseeable that the witnesses would leave Germany shortly after the hearing before the investigating judge cannot be equated, in our view, to a finding that it would have been impossible to hear evidence from them in person at a subsequent trial, at least via a video link. The witnesses were to leave for a State, Latvia, which was bound by international treaties to provide assistance in criminal matters to the German authorities, including the hearing of witness evidence by videoconference. We see our finding further confirmed by the applicant’s own submission that he had assumed that he would be able to cross-examine the witnesses at the trial and had not therefore had any reason to request a repetition of the witnesses’ hearing by the investigating judge (see paragraph 82 of the judgment). 16. To conclude, we agree with the majority that affording the defendant the opportunity to have key prosecution witnesses questioned at least during the pre-trial stage and via their counsel constitutes an important procedural safeguard, the absence of which weighs heavily in the balance in the examination of the overall fairness of the proceedings under Article 6 §§ 1 and 3 (d). Despite this, in the circumstances of the present case there were other strong safeguards permitting the trial court to properly assess the reliability of the evidence before it. In particular, there was very strong and coherent additional incriminating evidence regarding the offence of which the applicant was found guilty. In addition, the trial court made a particularly thorough and careful examination of the credibility of the absent witnesses and of the reliability of their statements. In these circumstances, the absence of an opportunity for the applicant to examine or have examined witnesses O. and P. at any stage of the proceedings did not, in our view, render the trial as a whole unfair. DISSENTING OPINION OF JUDGE KJØLBRO 1. I have some hesitations as regards the Grand Chamber’s clarification of the Court’s case-law and the three criteria established in Al-Khawaja and Tahery v. the United Kingdom ([GC], nos. 26766/05 and 22228/06, ECHR 2011). Furthermore, I disagree with the Grand Chamber’s assessment of the present case, and I voted against finding a violation of Article 6 § 1 of the Convention. Below, I will briefly explain my view on the two issues mentioned. The Grand Chamber’s clarification of the so-called “ Al-Khawaja criteria” 2. Consistency of the Court’s case-law is very important for the credibility and legitimacy of the Court and for the domestic authorities’ compliance with the Convention and application of the Court’s case-law. The Court should not, without good reason, depart from precedents laid down in previous cases (see Micallef v. Malta [GC], no. 17056/06 , § 81, ECHR 2009). This applies in particular to recent Grand Chamber judgments. Furthermore, the Court should not clarify and further develop its case-law unless there are good reasons to do so. 3. In 2011 the Grand Chamber clarified and further developed the Court’s long-standing case-law on the use as evidence of written statements from absent witnesses. In Al-Khawaja and Tahery , cited above, the Grand Chamber established the three criteria to be applied as well as the order of the three tests. First, there has to be “a good reason” for the non-attendance of the witness. Second, it has to be assessed whether the statement of the absent witness is the “sole or decisive” evidence. Third, if the written statement is the sole or decisive basis for convicting the accused, there have to be sufficient “counterbalancing factors”. In my view, the Court could easily have determined the present case on the basis of the criteria established in Al-Khawaja and Tahery , thereby confirming that recent Grand Chamber judgment. 4. The Grand Chamber’s clarifications in the present judgment should not, in my view, be understood as a departure from the three-step test established in Al-Khawaja and Tahery , which should therefore still be applied in similar cases in the future. For that reason I find it necessary to make a few additional remarks on the criteria to be applied. 5. Firstly, if there is no good reason for the non-attendance of a witness the domestic court should, as a main rule, not allow the prosecutor to use the written statement of the absent witness as evidence against the accused (see Al-Khawaja and Tahery , cited above, §§ 120-125). 6. If the statement of the absent witness, in the view of the prosecutor, is of such relevance and importance to the case that it should be used as evidence, the witness should be summoned to appear before the trial court and give evidence unless there is a good reason for non-attendance. If there is no good reason for the non-attendance of the witness in question, the domestic court should not allow the prosecutor to use the written statement as evidence against the accused. 7. Failure to summon a witness, without any good reason, would run counter to the rights of the defence to cross-examine prosecution witnesses. That being said, I agree that the absence of a good reason for the non-attendance of a witness will not necessarily and automatically render the trial unfair (see paragraph 113 of the judgment). However, this clarification of the Court’s case-law cannot be understood as implying a general departure from the main rule. According to that rule, if the statement of an absent witness is of such relevance and importance to the case that the domestic court will allow it to be used as evidence against the accused, there should be a good reason for not summoning the witness to give testimony at the hearing. 8. Secondly, the “sole and decisive evidence” test has, with some minor variations in terminology, been applied consistently since the Court’s judgment in Unterpertinger v. Austria (24 November 1986, § 33, Series A no. 110). Prior to the Court’s judgment in Al-Khawaja and Tahery , the Court would find a violation of Article 6 of the Convention if the written statement of the absent witness was the “sole or decisive” basis for the conviction of the accused. 9. In the Al-Khawaja and Tahery judgment (§ 131), the Court also applied the “sole or decisive” test, while defining what was meant by “sole” and “decisive”. At the same time, the Court further developed its case-law by saying 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” (ibid., § 147). 10. However, both prior to and subsequent to the Al-Khawaja and Tahery judgment, the question to be asked has been whether the written statement of the absent witness was the “sole or decisive” evidence. 11. In the present case the Grand Chamber stated that “it must review the existence of sufficient counterbalancing factors not only in cases in which the evidence given by an absent witness was the sole or the decisive basis for the applicant’s conviction” but also in cases where the Court “finds it unclear whether the evidence in question was the sole or decisive basis but is nevertheless satisfied that it carried significant weight” (see paragraph 116 of the judgment). 12. I find it important to underline that the term “significant weight” is not to be understood as a departure from the “sole or decisive” test, thereby creating three categories: “sole evidence”, “decisive evidence” or “evidence carrying significant weight”. The clarification does not imply a departure from the “sole or decisive” test, but takes account of the fact that sometimes, having regard to the statement in question and the reasoning of the domestic courts, it may be evident that a statement carries “significant weight”, but at the same time it may be difficult to determine whether the statement is “decisive” for a conviction. If that is the case, the written statement should be treated, by the domestic court as well as the Court, as “decisive”. 13. Therefore, in my view, the clarification does not imply a departure from the “sole or decisive” test. 14. Thirdly, the order of the three tests follows clearly from the Al-Khawaja and Tahery judgment. First, there has to be “a good reason” for the non-attendance of the witness. It will only be relevant to assess the other criteria if the first question is answered in the affirmative (see Al-Khawaja and Tahery , cited above, § 120). Second, it has to be assessed whether the written statement of the absent witness is “the sole or decisive evidence”. It will only be relevant to assess the third criterion if the second question is answered in the affirmative (ibid., § 147). Third, if the written statement is the sole or decisive basis for convicting the accused, there have to be “sufficient counterbalancing factors” (ibid . , § 147). 15. There are very good reasons for the order of the tests. The question of the use of written statements from absent witnesses will arise at different stages of the proceedings. First, the question will arise when the trial court assesses a request from the prosecutor to use the written statement of an absent witness as evidence against the accused, or an objection from the defence to such a measure. This assessment will take place during the hearing. Second, the question will arise when the trial court assesses whether there is sufficient basis for convicting the accused. This assessment will take place at the end of the hearing. Third, the question will arise when the fairness of the criminal proceedings is assessed, either by a domestic appeal court or subsequently by the Court. At the point where the trial court decides whether the prosecutor should be allowed to use a written statement from an absent witness as evidence it will often be difficult, if not impossible, to assess whether the evidence will be the sole or decisive basis for a conviction. Therefore, in practice, the three steps will in most cases have to be assessed in the order stated in the Al-Khawaja and Tahery judgment, and, most often, at different moments in time. Furthermore, the principles whereby all evidence against an accused must normally be produced in his presence at a public hearing with a view to adversarial argument, and the accused should be given an adequate and proper opportunity to challenge and question a witness against him, are so important that they should not be departed from unless there is a good reason to do so. Doing so without a good reason will in most if not all cases render the proceedings unfair. 16. Therefore, I would like to underline the importance not only of the three steps in the Al-Khawaja and Tahery judgment, but also of the order of those three steps. That being said, I would not exclude the possibility that there may be situations where the three steps may be assessed in a different order. For example, in some cases it will be clear to the trial court at the outset that the written statement will be the sole or decisive evidence and that it would therefore render the proceedings unfair if the evidence was admitted and used. Likewise, there may be situations where the Court, for practical reasons, finds it appropriate to assess the three steps in a different order. Nevertheless, the three steps should, as a main rule, be assessed in the order prescribed in the Al-Khawaja and Tahery judgment. The Grand Chamber’s assessment of the present case 17. I agree with the majority that there was a “good reason” for the non-attendance of the two prosecution witnesses O. and P. (see paragraphs 132-140 of the judgment). 18. I also concur with the majority that the written statements from the two absent witnesses O. and P. were “decisive” for the applicant’s conviction, within the meaning of the Court’s case-law (see paragraphs 141-144 of the judgment). 19. However, I disagree with the majority as regards the fairness of the trial. In my view and as explained below, there were sufficient “counterbalancing factors” to render the applicant’s trial fair. 20. In Al-Khawaja and Tahery (§ 147) the Court held as follows: “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.” Thus, the purpose of “counterbalancing factors” is to ensure that “a fair and proper assessment of the reliability of the evidence” takes place and that the evidence “is sufficiently reliable”. 21. As rightly pointed out by the majority, the trial court approached the evidence with caution (see paragraphs 146-150 of the judgment). However, in my view, the majority attaches too little weight to the availability and strength of further incriminating evidence (see paragraph 151, with reference to paragraphs 143-144). 22. In my view, the Regional Court had before it very strong and coherent additional incriminating evidence regarding the offence of which the applicant was convicted. On the basis of the additional evidence it was possible for the trial court to perform an assessment of the reliability of the statements from the absent witnesses O. and P. In the view of the trial court “the body of evidence, taken together, gave a coherent and complete overall picture of events which supported the version provided by witnesses O. and P. and refuted the contradictory versions of events put forward by the applicant and his co-accused in the course of the trial” (see paragraph 46 of the judgment). 23. Furthermore, as rightly pointed out by the majority, the applicant had the opportunity to give his own version of the events and cross-examine the other witnesses appearing before the trial court (see paragraph 152 of the judgment). Furthermore, the applicant had the possibility to challenge the use and importance of the written statements. 24. In fact, the core argument for finding a violation of Article 6 of the Convention in the present case seems to be the fact that the domestic authorities did not make use of the possibility to appoint a lawyer for the applicant before questioning the two witnesses, which would have given the applicant the opportunity to have the witnesses questioned at the investigation stage by a lawyer appointed to represent him (see paragraphs 153-160). 25. I disagree with the majority in their assessment of the importance of the pre-trial stage for the overall fairness of the proceedings in the present case. 26. Firstly, if a lawyer had been appointed to represent the applicant in the early stages of the investigation, when the two witnesses O. and P. were questioned by the investigating judge, and if the applicant and the lawyer had been notified of the questioning of the witnesses and the accused and the lawyer had been given a chance to question the witnesses when they gave testimony, there would simply have been no complaint under the Convention. The Court’s case-law on the use of written statements from absent witnesses concerns “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” (see Al-Khawaja and Tahery , cited above, § 119). In other words, had the applicant been given an opportunity to examine and cross-examine the witnesses O. and P. when they were questioned by the investigating judge, the subsequent use of their statements would not have raised an issue concerning the fairness of the proceedings (see, for example, Sadak and Others v. Turkey (no. 1) , nos. 29900/96 , 29901/96, 29902/96 and 29903/96, § 65, ECHR 2001-VIII; Sommer v. Italy (dec.), no. 36586/08 , 23 March 2010; Chmura v. Poland , no. 18475/05 , §§ 49-59, 3 April 2012; and Aigner v. Austria , no. 28328/03 , § 41, 10 May 2012 ). 27. Secondly, the majority seems to pay little attention to the reasons given by the investigating judge for not having notified the applicant about the questioning of the two witnesses O. and P. The applicant had not been informed about the investigation “so as not to put the investigation at risk” (see paragraph 21 of the judgment). Furthermore, the investigating judge had excluded the applicant from the witness hearing in accordance with domestic law because he “was concerned that the witnesses, whom he had found to be considerably shocked and distressed by the offence, would be afraid of telling the truth in the applicant’s presence” (see paragraph 21 of the judgment). In my view, the Court should in its case-law pay equal attention to, and protect, the rights and interests of the victims of crimes; in the specific circumstances of this case, there were good reasons to protect the victims. Furthermore, the domestic courts did not consider it contrary to Article 141 § 3 of the Code of Criminal Procedure, read in conjunction with Article 140 § 1 and as interpreted by the Federal Court of Justice (see paragraphs 28-29, 57-59 and 62 above), that no defence counsel had been appointed to represent the applicant at the time of the witnesses’ hearing before the investigating judge. I take note in that context of the Government’s explanation (see paragraph 94 of the judgment) to the effect that, under Article 168c § 5 of the Code of Criminal Procedure, the trial court was authorised to dispense with giving notice of the hearing to any lawyer appointed to represent the applicant if it considered that notification would endanger the success of the investigation. 28. Thirdly, the fact that “it might not be possible to hear evidence from those witnesses at a subsequent trial against the applicant” (see paragraph 159 of the judgment) cannot, in my view, lead to the conclusion that the failure to appoint a lawyer and give the applicant an opportunity to have witnesses O. and P. questioned at the investigation stage by a lawyer appointed to represent him (see paragraph 160) rendered the subsequent trial unfair. Admittedly, there was a risk that the witnesses would not appear at the hearing, and there will always be such a risk when evidence is secured at the pre-trial stage of the proceedings. However, in the specific circumstances of the case there is insufficient basis for arguing that is was foreseeable that the witnesses O. and P. would not appear before the trial court and give testimony. The fact that it was foreseeable that the witnesses would leave Germany shortly after the hearing before the investigating judge cannot be equated to a finding that it would have been impossible to hear evidence from them in person at a subsequent trial, either in person or via a video link. In this context I would also like to refer to the applicant’s own submission to the effect that he had assumed that he would be able to cross-examine the witnesses at the trial and had not therefore had any reason to request a repetition of the witnesses’ hearing by the investigating judge (see paragraph 82 of the judgment). 29. In my view, the majority overemphasises the pre-trial stage and the decision not to appoint a lawyer and not to notify the lawyer and the applicant about the questioning of the witnesses. 30. Furthermore, the majority does not attach sufficient importance to the purpose of “counterbalancing factors”, which is to ensure that “a fair and proper assessment of the reliability of the evidence” takes place and that the evidence “is sufficiently reliable”. In a detailed and very well-reasoned judgment, the domestic courts explained why the evidence of the absent witnesses O. and P. was deemed, in the light of all the evidence, to be reliable. As already mentioned, on the basis of the additional evidence it was possible for the trial court to perform an assessment of the reliability of the statements from the absent witnesses O. and P. In the view of the trial court “the body of evidence, taken together, gave a coherent and complete overall picture of events which supported the version provided by witnesses O. and P. and refuted the contradictory versions of events put forward by the applicant and his co-accused in the course of the trial”. 31. Therefore, in my view, the use of the written statements of the absent witnesses O. and P. and the absence of an opportunity for the applicant to examine or have examined the witnesses O. and P. at any stage of the proceedings did not render the trial as a whole unfair. 32. In my view, the judgment is another example of the Court’s focus on the importance of the investigation stage for the preparation of the criminal proceedings (see Salduz v. Turkey [GC], no. 36391/02 , § 54, ECHR 2008), which means that a failure to comply with certain procedural guarantees at the pre-trial stage more or less automatically has the consequence that the evidence obtained cannot be used against the accused. 33. This is particularly regrettable in a situation where the reason for the restriction on certain procedural guarantees is the need to protect the victims of crimes and where there is corroborating evidence that make it possible for the trial court to assess the reliability of the statements given by the absent witnesses. 34. This judgment is an example of a rather formalistic approach to the importance of procedural guarantees, whereby failure to comply with or secure certain procedural guarantees at the pre-trial stage renders the evidence obtained illegal even if the use of that evidence, on the basis of an overall assessment, does not render the proceedings as a whole unfair. BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII