THIRD SECTION

CASE OF XENOFONTOS AND OTHERS v. CYPRUS

(Applications nos. 68725/16, 74339/16 and 74359/16)

JUDGMENT

Art 6 § 1 (criminal) • Fair hearing • Applicants’ murder conviction decisively based on confession by accomplice, who was not prompted by any deal, but was later placed in a witness protection programme and spared prosecution, not rendering trial unfair • Length of proceedings, five years and nine months, justified by complexity of case

STRASBOURG

25 October 2022

Request for referral to the Grand Chamber pending

This judgment will become final in the circumstances set out in Article 44 § 2 of the Convention. It may be subject to editorial revision.

In the case of Xenofontos and Others v. Cyprus,

The European Court of Human Rights (Third Section), sitting as a Chamber composed of:

Georges Ravarani , President,

Georgios A. Serghides ,

Darian Pavli ,

Anja Seibert-Fohr ,

Peeter Roosma ,

Andreas Zünd ,

Frédéric Krenc , judges,

and Milan Blaško, Section Registrar,

Having regard to:

the applications (nos. 68725/16, 74339/16 and 74359/16) against the Republic of Cyprus lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by three Cypriot nationals, Mr Grigoris Xenofontos (“the first applicant”), Mr Anastasis Krasopoulis (“the second applicant”) and the latter’s sister, Ms Elli Krasopouli Skordelli (“the third applicant” – collectively “the applicants”), on 21 November and 1 December 2016;

the decision to give notice to the Cypriot Government (“the Government”) of the complaints concerning the use of accomplice evidence at the applicants’ trial and the length of that trial, and to declare the remainder of the applications inadmissible;

the parties’ observations;

Having deliberated in private on 27 September 2022,

Delivers the following judgment, which was adopted on that date:

INTRODUCTION

1. The case mainly concerns the question whether a criminal trial can be considered to be “fair” within the meaning of Article 6 of the Convention if it led to a conviction based to a decisive extent on accomplice testimony.

THE FACTS

2. The applicants were born in 1981, 1973 and 1968 respectively and are currently serving life sentences in the Central Prison of Nicosia. They were represented before the Court by Mr R. Vrahimis and Mr C. Paraskeva, lawyers practising in Nicosia.

3. The Government were represented by their Agent, Mr G. Savvides, Attorney-General of the Republic of Cyprus.

4. The facts of the case, as submitted by the parties, may be summarised as follows.

5. In December 2009 the police received a tip-off about a murder-for-hire plot against an executive of a private television channel. The informant said that the murder would be carried out by two men, one of them being F.H., who was known to the police. The police alerted A.H., the channel’s CEO, but he saw no reason for concern.

6. On 11 January 2010 A.H. was shot dead near his home as he was leaving his car after a day in the office.

7. Later the same evening the police received another tip-off suggesting that the shots had been fired by F.H.

8. On 12 January 2010 the police questioned F.H. on two occasions about the murder. He denied any involvement, claiming that he had been elsewhere at the time.

9. More evidence against F.H. was gathered, and on 14 January 2010 the police arrested him and took him into custody at Pera Chorio Police Station.

10. On 15 January 2010 the police received information about the possible involvement of the first applicant in the murder. On the same day, the first applicant took a flight to Moldova and an arrest warrant was issued in respect of him.

11. On 17 January 2010 at 5 p.m., two investigators questioned F.H. as a suspect. They told him that they knew of his possible involvement in the murder but that they had no strong leads. They also told him that they could not see any personal motive for F.H. to kill the victim because, unlike F.H. himself, the victim had no connection with the criminal underworld or illegal activities and thus was unlikely to have had any differences with F.H. The investigators observed that F.H. became agitated each time they mentioned the victim’s good character. They also sensed that F.H. was regretting being caught up in a situation that could affect his family and himself.

12. Convinced that F.H. knew details of the crime but was afraid to speak, the investigators offered to protect him as a witness. F.H. hesitated over the offer because he mistrusted the investigators and would have preferred to talk to someone whose word carried more weight. His lawyer warned him of the implications of entering the witness protection programme.

13 . The investigators reported their impressions of the suspect to their superiors and suggested that a meeting between F.H and the Attorney ‑ General be arranged. The Chief of Police spoke with the Attorney ‑ General, who agreed to the meeting. At about 11 p.m. the investigators took F.H. to the office of the Chief of Police in Nicosia. While the men were waiting for the Attorney-General, one of the investigators encouraged F.H. to take the opportunity to ask the Attorney-General whatever questions he had. At some point, the Attorney-General walked in and said to F.H.: “Tell the truth, son, and you won’t lose anything. Whoever tells the truth never loses.” In less than a minute he walked out, followed by the Chief of Police. F.H. did not report his fears to the Attorney-General or make any request. The investigators drove F.H. back to his cell.

14. Despite having met with the Attorney-General, F.H. still had reservations about testifying. He asked whether he would be expected to give a deposition. The investigators replied that he could simply start talking and if his first statements proved true, the investigators would formally ask the Attorney-General to protect him as a witness.

15. On 18 January 2010 the investigators secured evidence disproving F.H.’s alibi. They confronted him with that evidence, but he still refused to break his silence. The investigators sensed that he was struggling with himself.

16. On 21 January 2010 F.H. told the investigators that he could no longer endure the pangs of conscience which he was experiencing and would testify. He did not ask for anything in return. He told the investigators how he had become involved in the murder and discussed with them how his forthcoming revelations and protection as a witness would change his life and that of his family. F.H. announced his decision to his lawyer and refused his further services.

17. F.H. dictated his confession to two investigators. He said that the murder had been masterminded by the second and third applicants, brokered by A.G. and carried out by the first applicant and himself; the first applicant had shot A.H. and F.H. had driven the getaway motorcycle.

18. F.H. asked what would happen to him and his family now that he had confessed. The investigators replied that they could only comment on the safety of his family, and that the decision concerning F.H. himself would be taken by the Attorney-General. They did, however, explain that, depending on the value of his testimony, the Attorney-General could request that F.H. be given a non-custodial sentence or a prison sentence with possible parole, or could even refrain from prosecuting him altogether.

19. After checking the confession against other evidence at their disposal and consulting their superiors, the investigators suggested that instead of prosecuting F.H., the Attorney-General could use his testimony as evidence for the prosecution. They argued that F.H. had spoken freely, was ready to repeat his statements at the trial and had not tried to diminish his role or that of his accomplices, and that pursuing the organisers of the crime would better serve society. This suggestion was recorded in a police activity log ( ημερολόγιο ενεργείας ) and later discussed in person with the Attorney ‑ General in his office. The investigators confirmed to the Attorney ‑ General their impression that F.H. was telling the truth and that he feared for his safety, having received a threat while in detention.

20. On 28 January 2010 F.H. was placed in a witness protection programme.

21 . On 5 February 2010 the Attorney-General decided not to prosecute him.

22. On 19 February 2010 the prosecuting authorities brought a case against the three suspects whom they had been able to arrest: the second and third applicants and A.G. The case against the first applicant was suspended pending his extradition from Moldova.

23. On 14 June 2010 the first applicant was arrested in Moldova.

24. On 18 August 2010 he was extradited to Cyprus and placed in detention.

25. On 25 August 2010 a new case was brought, this time against all four suspects, and was referred to the Nicosia Assize Court.

26. On 25 October 2010 the four co-defendants were committed for trial (case no. 19325/2010) before the three-member Nicosia Assize Court.

27. The first applicant and A.G. each had personal counsel and the second and third applicants shared a group of four counsel.

28. The prosecution relied on F.H.’s testimony. The defendants pleaded not guilty and alleged that F.H. and the police had fabricated his testimony to falsely implicate them in the crime. They asserted that F.H. had been the shooter, not the getaway driver, and that he was seeking to shift the blame onto them in exchange for witness protection and being spared prosecution. They also argued that F.H. was seeking to exonerate his two friends and his girlfriend, who were also implicated in the murder, and that the actions of the police, in their turn, had been based on “extraneous motives” ( αλλότριων στοχεύσεων ).

29. On 9 November 2010 the hearing stage began.

30 . Throughout the trial, the court held 280 hearings (some of them after working hours and during summer holidays), issued twenty-three interim decisions, inspected the scene of the crime, heard eighty-nine prosecution and sixty-one defence witnesses, and examined 905 exhibits, including audio ‑ and video-recordings and forensic evidence. The full record of the trial exceeded 11,300 pages.

31. F.H. was cross-examined by all the defence counsel at twenty-one hearings amounting to a total duration of almost one month.

32 . The proceedings were adjourned for forty-one days because one of the judges was on a diplomatic mission abroad. They were also adjourned on account of absences of defence counsel.

33. On 13 May 2012 the court closed the hearing stage.

34. On 13 June 2013 the Nicosia Assize Court delivered a judgment in which it found the defendants guilty of murder and sentenced them to life imprisonment.

35. The court found that F.H. was trustworthy, devoted to him fifty of the almost 400 pages of the judgment, and relied on his testimony to a “most decisive” ( καθοριστικότερο ) extent. The court stated:

“F.H. answered with exemplary stability and detail on every aspect he was asked about, in a manner that clearly showed a person who was telling the truth. He remained literally unshaken in his very long, very demanding and exhaustive cross-examination. He was disarmingly genuine, vivid and figurative in his descriptions, which he gave in a simple and unpretentious manner ... He referred to facts and details that only someone who had actually experienced them could relate, thereby excluding any possibility that they might have been a figment of the imagination or, as was alleged many times and at every opportunity, the product of collaboration or direction, either by the investigators or those identified as the real moral perpetrators, whoever they were. His testimony was also characterised by spontaneity and readiness ... [This court has] dealt with all the examples that the defence submitted without identifying anything that points to false or extraneous motives on the part of the witness or, incidentally, to any divergence from accepted police practice and the realities for the effective and speedy resolution of a crime – [the police] always showed respect for the rights of the suspects and the defendants ...”

36. The court disagreed with the defence that F.H.’s trustworthiness was undermined by his history of unlawful behaviour (misconduct during military service and convictions for illegal possession and use of firearms and assault).

37. Regarding the circumstances of his confession, the court found:

“Nothing that was said to [F.H.] by the Attorney-General, the Chief of Police or the investigators could ... be interpreted as encouragement – through promises, exchanges or otherwise, in any unlawful sense – to tell anything but the truth or to falsely testify for any other purpose ...

[The statement that] the Attorney-General ... made to [F.H.] was a simple yet meaningful piece of empirical wisdom ...

[T]here is nothing special, from a legal point of view, in the Attorney-General’s choice not to prosecute [F.H.] for the offences for which he prosecuted [the applicants] ...”

38. The court explained that F.H.’s denial of his involvement in the murder at the initial police interviews had been a “perfectly normal, human reaction” to being accused of a crime.

39. The court explained F.H.’s decision to testify by

“... his determination to tell the truth to the investigators, having realised the great evil he had committed and after having weighed and calculated all the possibilities and options that were available to him ...”

40. The court found that, among other things, its use of the accomplice testimony was justified by the Court’s case-law (referring to X v. the United Kingdom , no. 7306/75, Commission decision of 6 October 1976, Decisions and Reports (DR) 7, p. 115).

41. Because F.H.’s testimony was not corroborated by any other evidence, the court issued a “self-warning” ( αυτοπροειδοποίηση ), stating that it was approaching that evidence with “caution, suspicion, consideration and attention of the highest degree”. The court found it possible to rely on that testimony “with absolute safety”, stating:

“[The testimony is] one solid whole without cracks or contradictions, and juxtaposes details such that only someone who had actually experienced all [the events] could describe them in the way and to the extent that [F.H.] did ...”

42 . In addition to F.H.’s testimony, the court relied on the testimony of three “very important” ( πολύ σημαντικούς ) witnesses: the defendants’ friends who testified, in sum, about having been approached with the proposal to carry out the murder, about having lent a motorcycle helmet to F.H., about the defendants’ movements and emotional state. The court also relied on material evidence and evidence of the applicants’ motives for killing A.H. (personal gain, revenge and a desire to take control of the television channel concerned).

43. On 14 June 2013 the defendants lodged appeals with the Supreme Court.

44. On 23 July 2013 the court received the trial record.

45 . Between October 2014 and February 2016, the court held six directions hearings at which it allowed counsel for the second and third applicants to amend the grounds of their appeals, granted counsel for the first applicant extra time to submit his grounds of appeal, set and postponed deadlines for the filing of skeleton arguments ( διάγραμμα αγόρευσης ) by the parties and granted the applicants’ counsel extra time to correct page-number references in their submissions.

46. There were seventy-two final grounds of appeal, the main one being the perceived unfairness of the use of the accomplice testimony.

47. On 16 and 17 March 2016 the Supreme Court heard the appeals and reserved its judgment.

48. On 6 June 2016 the Supreme Court delivered its judgment, dismissing the appeals (nos. 103/2013, 102/2013 and 101/2013) and upholding the sentence.

49 . The court found that the trial court’s assessment of F.H.’s testimony had not been arbitrary or unreasonable and that it had been lawful and justified not to prosecute F.H.

50. The court also found that the use of his testimony had not rendered the trial unfair. Domestic law permitted the trial court to rely on F.H.’s confession alone. But the trial court had gone further and tested that confession in multiple rounds of “tug of war” ( διελκυστίνδες ) with other evidence, such as that concerning the pre-operational surveillance of the victim, the perpetrators’ escape route, modification of the motorcycle used by them and dozens of other details.

51 . The court found that F.H. had confessed of his own volition and that the Attorney-General’s words to him did not suggest any improper dealings.

RELEVANT DOMESTIC LAW

52. Article 113 § 2 of the Constitution provides:

“The Attorney-General of the Republic shall have power, exercisable at his discretion in the public interest, to institute, conduct, take over and continue or discontinue any proceedings for an offence against any person in the Republic.”

THE LAW

53. Having regard to the similar subject matter of the applications, the Court finds it appropriate to examine them jointly.

54. The applicants complained that their trial had been unfair, as the only evidence against them had come from an accomplice who had been spared prosecution. They relied on Article 6 § 1 of the Convention, which, in so far as relevant, reads:

“In the determination of ... any criminal charge against him, everyone is entitled to a fair ... hearing ... by [a] ... tribunal ...”

55. The Government argued that the complaint was manifestly ill ‑ founded.

56. They disagreed that F.H.’s confession had been the only evidence against the applicants. Just because the trial court had found no corroborating evidence did not mean that it had used no other evidence. Indeed, the domestic law distinguished between corroborating ( ενισχυτική ) and supporting ( υποστηρικτική ) evidence. Corroborating evidence proved that a crime had been committed and implicated the defendant, whereas supporting evidence reinforced witness testimony concerning the issues at stake. In addition to evidence given by F.H., there had been supporting evidence against the applicants.

57. However, even if F.H.’s testimony had been the only evidence, under domestic law its use would still have been lawful if the court had made clear, as the Assize Court had, its awareness of the inherent unreliability of such evidence. A trial where such testimony was used would also be “fair” from the standpoint of the Convention (the Government cited Cornelis v. the Netherlands (dec.), no. 994/03, ECHR 2004-V; Habran and Dalem v. Belgium , nos. 43000/11 and 49380/11, 17 January 2017; and, by contrast, Adamčo v. Slovakia , no. 45084/14, 12 November 2019).

58. The Government submitted that F.H. and the prosecution had not struck a deal. They contended that F.H. had confessed out of remorse, without any inducement by means of promises of favours and after having been cautioned. The Attorney-General had done no more than make a hard choice in the public interest and refrain from exercising his constitutional prerogative to prosecute.

59. According to the Government, questions of the admissibility and evaluation of evidence were generally beyond the Court’s competence. The findings of the domestic courts had not been arbitrary and should not be altered.

60. They further submitted that the rights of the defence had been respected. The defence had known that F.H. had confessed, what he had confessed to, and that he would not be prosecuted (his name had not been listed on the charge sheet). F.H. had testified under his own name and had been cross-examined extensively. His written confession had been tested at the trial as well. The investigators who had worked with him had also been cross ‑ examined. If the applicants had also wished to cross-examine the Attorney-General, they could have asked the court to summon him.

61. Even after his confession, F.H. had been kept in custody for the time needed to verify his statements. Those statements had been verified at the trial as well.

62. Lastly, although domestic law did not provide for a separate judicial review of the Attorney-General’s decision not to prosecute, the Convention did not oblige the State to make such provision, and any possible improper dealings between the witness and the prosecution could be revealed at the trial.

(a) The first applicant

63. The first applicant maintained his complaint.

64. He submitted that he had not had any motive to shoot the victim and that the shooter had probably been F.H.

65. F.H.’s confession – the only evidence against the first applicant – had been unreliable. F.H. was a person of bad character, his testimony had been inconsistent and the courts had failed to resolve those inconsistencies.

66. F.H.’s confession had been prompted not by remorse, but by the impunity promised by the Attorney-General. The two of them must have agreed a deal that was suspicious and unprecedented in the country’s legal history. That deal had not been put in writing and its details had remained unknown to the courts.

67. By deciding not to prosecute F.H., the Attorney-General had abused his discretion. His decision had served no public interest and had been neither regulated nor overseen by the courts. By letting a murder suspect walk free, the State had breached its obligations under Articles 2 and 3 of the Convention.

(b) The second and third applicants

68. The second and third applicants also maintained their complaint.

69. They submitted that the Government had misinterpreted the domestic judgments and had not understood the true evidentiary value of F.H.’s testimony and the other evidence. According to the applicants, there had been nothing but F.H.’s testimony against them. No amount of “self-warnings” issued by the trial court could alter the fact that the court had relied on that testimony.

70. Before confessing, F.H. had denied his guilt four times. The domestic courts had failed to account for that inconsistency, as required by Article 6 (the applicants cited Zhang v. Ukraine , no. 6970/15, 13 November 2018).

71. Cypriot law did not provide for a separate judicial review of the Attorney-General’s decision not to prosecute. Nor had the courts properly reviewed the Attorney-General’s decision during the applicants’ trial.

72. There had been no written record of the deal between F.H. and the prosecution.

73. The applicants argued that the Court’s case-law on accomplice testimony supported their position. There were important factual differences between their case and the cases in which no violation had been found (they referred to Cornelis and Habran and Dalem , both cited above) and similarities with a case in which a violation had been found (referring to Adamčo , cited above).

74. The Court notes that this complaint is neither manifestly ill-founded nor inadmissible on any other grounds listed in Article 35 of the Convention. It must therefore be declared admissible.

75. The parties’ submissions are summarised above.

76. The Court reiterates that, as the Government correctly pointed out, Article 6 § 1 of the Convention does not set out any rules on how evidence should be assessed. The Court may interfere in this field only if a domestic court assesses evidence arbitrarily or manifestly unreasonably (see Bochan v. Ukraine (no. 2) [GC], no. 22251/08, § 61, ECHR 2015).

77. The Court also reiterates that the Convention does not prohibit a domestic court from relying on incriminating testimony given by an accomplice (see Wilhelm Tatzel v. Austria , no. 1599/62, Commission decision of 16 January 1963, Yearbook 6, p. 348), even if that witness has been known to move in criminal circles (see Habran and Dalem , cited above, § 111).

78. However, the testimony of an accomplice, given in exchange for immunity from prosecution, may render a trial unfair (see Erdem v. Germany (dec.), no. 38321/97, 9 December 1999). This is because such testimony by its very nature is open to manipulation and may be given purely to obtain those advantages or for personal revenge (see Adamčo , cited above, § 59).

79. In the past the Court made the conclusion about the effect of such testimony on the fairness of a trial taking into account, inter alia :

– the defence knew the witness’s identity (see Habran and Dalem , cited above, § 104);

– the defence knew about the existence of an arrangement with the prosecution (see X v. the United Kingdom , cited above);

– a domestic court reviewed the arrangement (see Adamčo , cited above, § 70);

– the domestic court paid attention to all possible advantages received by the witness (ibid., §§ 65-66);

– the arrangement was discussed at the trial (see X v. the United Kingdom , cited above);

– the defence had the opportunity to test the witness (see Cornelis , cited above);

– the defence had the opportunity to test the members of the prosecution team involved (see Verhoek v. the Netherlands (dec.), no. 54445/00, 27 January 2004);

– the domestic court was aware of the pitfalls of relying on accomplice testimony (see Cornelis , cited above);

– the domestic court approached the testimony cautiously (see Verhoek , cited above);

– the domestic court explained in detail why it believed the witness (ibid.);

– untainted corroborating evidence existed (see Habran and Dalem , cited above, § 105, and Salmon Meneses , cited above);

– an appeal court reviewed the trial court’s findings in respect of the witness (see X v. the United Kingdom , cited above); and

– the question was addressed by all the courts dealing with the various appeals (see Adamčo , cited above, § 63).

80. First and foremost, as was established by the domestic courts in the present case, there was no deal between F.H. and the prosecution. The witness confessed and turned in the applicants out of remorse, without having been promised anything in return. The Attorney ‑ General’s decisions to grant him protection and not to charge him involved the exercise of the official’s discretion, rather than a follow-up on a promise he had given.

81. All the applicants, and especially the first applicant, alleged that there was more than met the eye in the interaction between the prosecution and the witness. At the trial, the applicants also asserted that the police had acted on the basis of “extraneous motives”. The applicants failed, however, to specify what those motives were. They also failed to provide any evidence of a deal between F.H. and the prosecution. This being so, the Court must accept the circumstances of F.H.’s confession as established by the domestic courts and as described above.

82. The Court further notes that the applicants knew of F.H.’s identity, the contents of his confession and the fact that he would not be prosecuted.

83. At the trial, the applicants were able to examine at will both F.H. and the police officers who had questioned him. As to the Attorney-General, the applicants did not ask that he be examined.

84. The trial court was fully aware of the dangers inherent in using accomplice evidence and took pains to explain in detail why it believed F.H.

85. As to the parties’ disagreement about the existence of other incriminating evidence against the applicants, the Court considers that its role in that field is limited. It therefore accepts the Supreme Court’s relevant findings and concludes, as argued by the Government, that the conviction did not rely solely on F.H.’s evidence.

86. The trial court’s assessment of F.H. was reviewed by the Supreme Court (see paragraphs 49-51 above), the only appellate court available.

87. Lastly, unlike in Zhang , which was cited by the second and third applicants, in the present case the trial court did explain why F.H. had changed his mind.

88. In such circumstances, it cannot be said that the Nicosia Assize Court’s reliance on the accomplice testimony rendered the trial unfair.

89. There has accordingly been no violation of Article 6 § 1 of the Convention.

90. The first applicant also complained that his trial had lasted an unreasonably long time. He relied on Article 6 § 1 of the Convention, which, in so far as relevant, reads:

“In the determination of ... any criminal charge against him, everyone is entitled to a ... hearing within a reasonable time by [a] ... tribunal ...”

91. The Government argued that this complaint was manifestly ill ‑ founded.

92. The applicant’s trial had lasted thirty-four months: from 25 August 2010 (when the case against the first applicant had been brought) to 13 June 2013 (when the Nicosia Assize Court had delivered its judgment).

93. That period had been reasonable in the circumstances.

94. The case had been complex (see paragraph 30 above) and the trial had been, as the Assize Court stated, “arduous” ( επίπονος ). It had concerned a murder for hire with many competing versions of the events. There had been four defendants, represented by different counsel pursuing different defence strategies.

95. Numerous procedural objections had led to several interim decisions. The defence counsel had pleaded their cases and cross-examined prosecution witnesses slowly and repetitively. They had examined the same witness in turn.

96. To compensate for the delays caused by the adjournments, the trial court had held some hearings after working hours and during summer holidays.

97. The duration of the appeal phase – three years – had also been reasonable because the appeals were complex. The Supreme Court had had to study thousands of pages of the trial record and to examine multiple grounds of appeal.

98. The first applicant maintained his complaint.

99. He submitted that the proceedings had lasted from 14 June 2010 (when he had been arrested in Moldova) to 6 June 2016 (when the Supreme Court had delivered its judgment).

100. According to the first applicant, the case was simple: he had faced only two charges, the only evidence had come from the accomplice, only the Cypriot police had been involved in the investigation and all the witnesses had lived in Cyprus.

101. The first applicant asserted that he had not resorted to delaying tactics. The extensive cross-examination of witnesses by each counsel had been explained by the need to fill in gaps left by the police investigation. By contrast, the trial court had postponed hearings on account of its other engagements, the judges’ absences or other reasons. Some postponements had remained unexplained. The domestic courts were known to be experiencing a backlog.

102. The first applicant argued that the Supreme Court should have given its judgment sooner than it had since it had not been required to hear witnesses and had only had to decide on questions of law.

103. The Court notes that this complaint is neither manifestly ill-founded nor inadmissible on any other grounds listed in Article 35 of the Convention. It must therefore be declared admissible.

104. The parties’ submissions are summarised above.

105. The Court considers that the period to be examined lasted five years, nine months and nineteen days: from 18 August 2010 (when the first applicant was arrested in Cyprus upon his extradition from Moldova) to 6 June 2016 (when the Supreme Court delivered its judgment).

106. The Court cannot agree with the Government that the period started as late as 25 August 2010 (when the case was brought against the first applicant), because by then he had already been “substantially affected” by his arrest (see Eckle v. Germany , no. 8130/75, § 73, 15 July 1982, and Panayiotou v. Cyprus , no. 20009/06, § 35, 20 January 2011).

107. Furthermore, the Court cannot agree with the first applicant that the period started as early as 14 June 2010 (when he was arrested in Moldova), because the duration of the extradition proceedings cannot be attributed to the Cypriot authorities.

108. The Court reiterates that the reasonableness of the length of proceedings is to be assessed in the light of the particular circumstances of the case, regard being had to the complexity of the case, the applicant’s conduct and the conduct of the competent authorities (see Pélissier and Sassi v. France [GC], no. 25444/94, § 67, ECHR 1999 ‑ II).

109. The Court agrees with the Government that the case was complex. Contract murders are generally hard to investigate and as many as four co ‑ defendants were on trial. The Assize Court was faced with voluminous evidence (see paragraph 30 above). In its turn, the Supreme Court had to study thousands of pages of trial records and to respond to seventy-two grounds of appeal submitted by four lawyers.

110. As to the behaviour of the first applicant and the authorities, the Court cannot conclude that either of them caused more delay than the other. Both parties caused adjournments of the trial (see paragraph 32 above) and the rather long preparatory phase of the appeal proceedings (see paragraph 45 above).

111. The Court has previously found a violation in a comparable case against Cyprus (see Panayiotou , cited above). That case concerned a sex ‑ offence trial before an assize court and an appeal before the Supreme Court, which together lasted five years and three months. However, despite that outward similarity, the present case is markedly different. Unlike in Panayiotou (cited above, §§ 40-41), in the present case the Nicosia Assize Court did not allow consecutive adjournments and its trial record reached the Supreme Court only one month after the judgment, as opposed to one year.

112. In sum, the Court considers that the complexity of the case justified the somewhat lengthy duration of the proceedings.

113. There has accordingly been no violation of Article 6 § 1 of the Convention.

FOR THESE REASONS, THE COURT

Done in English, and notified in writing on 25 October 2022, pursuant to Rule 77 §§ 2 and 3 of the Rules of Court.

Milan Blaško Georges Ravarani Registrar President

In accordance with Article 45 § 2 of the Convention and Rule 74 § 2 of the Rules of Court, the separate opinion of Judge Serghides is annexed to this judgment.

G.R. M.B.

Partly DISSENTING OPINION OF JUDGE SERGHIDES

I. Introduction

1. My disagreement with the judgment concerns only point 3 of the operative provisions, namely, the finding that there has been no violation of Article 6 § 1 of the Convention on account of the use of immunised accomplice testimony.

2. The present case concerns the applicants’ complaint that their trial had been unfair, as the only evidence against them, which led to their conviction for murder and a sentence of life imprisonment, had been given by F.H., an accomplice, who had been spared prosecution.

3. I will explain my disagreement with the majority on point 3 of the operative provisions, by adopting the following methodology. I will first examine the relevant case-law of the Court on the admissibility and assessment of evidence. I will subsequently explain that, if that relevant case ‑ law were followed, the sole or decisive reliance on the testimony of F.H., the immunised accomplice, would entail a violation of Article 6 because there were no sufficient safeguards to counterbalance this fundamental flaw. For the sake of a more comprehensive discussion, however, I will also elaborate on the view that the trial would necessarily be unfair where the sole or decisive testimony was that of an immunised witness, irrespective of any alleged “safeguards”. I will conclude that, on the basis of the relevant case ‑ law, there would be a violation of Article 6 in any event.

II. Relevant case-law of the Court on the admissibility and assessment of evidence in criminal proceedings

4. Article 6 does not lay down any rules on the admissibility of evidence as such, as this is primarily a matter for regulation under national law (see SA-Capital Oy v. Finland , no. 5556/10, § 73, 14 February 2019) [1] . The Court has consistently held that, as a general rule, it is a matter for the domestic courts to assess the evidence before them. Consequently, and as the judgment also states (see paragraph 76), the Court will not, in principle, intervene in issues concerning the assessment of evidence and the establishment of the facts, or in the interpretation of domestic law, unless the decisions reached by the domestic courts appear arbitrary or manifestly unreasonable and provided that the proceedings as a whole were fair as required by Article 6 § 1 (see, for instance, Ajdarić v. Croatia , no. 20883/09, § 32, 13 December 2011; SA ‑ Capital Oy , cited above, § 73; and Bochan v. Ukraine (no. 2) [GC], no. 22251/08, § 61, ECHR 2015). Article 6 § 1 of the Convention does not set out any rules on how evidence should be assessed. There is a distinction between the admissibility of evidence, that is, the question of which items of evidence may be submitted to the competent court for its consideration, and the rights of the defence in respect of evidence which has in fact been submitted before the court (see SA-Capital Oy , cited above, § 74). An assessment of the fairness of the proceedings may thus depend, inter alia , on whether the defendant was given an opportunity to challenge the authenticity of the evidence and to oppose its use (ibid., § 75).

5. The Court’s primary concern under Article 6 § 1 is to evaluate the overall fairness of the criminal proceedings. Its task under Article 19 of the Convention is to ensure the observance of the obligations undertaken by the States Parties to the Convention. In making its assessment, the Court will look at the proceedings as a whole, having regard to the rights of the defence but also to the interests of the public and the victims in having crime properly prosecuted and, where necessary, to the rights of witnesses (see Adamčo v. Slovakia , no. 45084/14, § 56, 12 November 2019).

6. In determining whether the proceedings as a whole were fair, the quality of the evidence must be taken into consideration, including whether the circumstances in which it was obtained cast doubt on its reliability or accuracy (see Jalloh v. Germany [GC], no. 54810/00, § 96, ECHR 2006 ‑ IX). The use of “intrinsically unreliable” evidence may render a trial unfair (see Söylemez v. Turkey , no. 46661/99, § 122, 21 September 2006).

III. Relevant case-law of the Court concerning sole or decisive reliance on the testimony of an immunised accomplice

7. The use at trial of evidence obtained from an accomplice, by granting him or her immunity from prosecution [2] , may call into question the fairness of the hearing granted to an accused person (see Erdem v. Germany (dec.), no. 38321/97, 9 December 1999, and X v. the United Kingdom , no. 7306/75, Commission decision of 6 October 1976, Decisions and Reports 7, p. 115). This is because such testimony, by its very nature, is open to manipulation and may be given purely to obtain advantages or for personal revenge (see Adamčo , cited above, § 59; Habran and Dalem v. Belgium, nos. 43000/11 and 49380/11, § 100, 17 January 2017; and Armold G. Cornelis v. the Netherlands (dec.) , no. 994/03, 25 May 2014). A clear and foreseeable procedure for authorising investigative measures as well as their proper supervision should be put in place in order to secure the authorities’ good faith and compliance with the proper law-enforcement objectives (see Khudobin v. Russia, no. 59696/00, § 135, 26 October 2006; Habran and Dalem, cited above, § 101; and Ramanauskas v. Lithuania [GC], no. 74420/01, § 53, 5 February 2008).

8. The intensity of the requisite scrutiny with regard to evidence from an accomplice has a correlation with the importance of the advantage that the accomplice obtains in return for the evidence he or she gives ( Adamčo , cited above, § 69).

9. In Adamčo , as in the present case, the advantage obtained by the accomplice went beyond a reduction of sentence or financial benefit, but practically meant impunity for an offence of unlawful killing (ibid.). As the Court held in that case, in view of the importance of the evidence from the accomplice, the use of that evidence at the applicant’s trial, on the specific facts of the case, had not been accompanied by appropriate safeguards so as to ensure the overall fairness of the proceedings (ibid., § 71). It concluded that the applicant’s trial had fallen short of the guarantee of fairness under Article 6 and that therefore there had been a violation of that provision (ibid., § 72).

IV. Application of the relevant case-law to the facts and circumstances of the present case

10. In the present case, the applicants’ conviction and sentence of life imprisonment were based solely, or at least to a decisive extent, on the evidence given by a single witness, F.H., who had been an accomplice to the offence and had turned informer, as a result of which he was spared (“immunised” from) prosecution for his direct participation in the premediated murder of A.H. It is clear from the decision of the Assize Court that, in the absence of any corroboration by other independent or relevant evidence, F.H.’s evidence was decisive for the conviction of the applicants to the extent that, in the absence of F.H.’s testimony, no other evidence before the court could have led to their conviction.

11. The evidence underpinning the applicants’ conviction was intrinsically unreliable. Indeed, F.H. had a significant personal interest in giving evidence against the applicants and amending his evidence to suit the prosecution’s case, because, although he had been an accomplice to A.H.’s murder and had confessed to his involvement in the crime, the Attorney ‑ General decided not to prosecute him, thus giving him the significant advantage of not being convicted or sentenced to life imprisonment, as was the case for the applicants. It is not so relevant whether there was actually an agreement between the prosecution and F.H. regarding his immunity from being charged in the case, so long as F.H. reasonably hoped not to be prosecuted – a hope that ultimately materialised.

12. I consider that F.H.’s personal interest did not bestow on him the objective guarantees of trustworthiness that a witness should have. Contrary to the findings of the domestic courts, F.H. could not be considered a trustworthy witness, in disregard of the fact that he had an extremely significant personal interest in giving evidence against the applicants and in the light of the lack of adequate and sufficient safeguards.

13. The evidence given by F.H. was intrinsically problematic. In my view, the use in evidence of his statements rendered the trial as a whole unfair on account of: (a) the sole or decisive reliance on the evidence of F.H., as an accomplice who was spared prosecution altogether, and (b) the lack of adequate and sufficient safeguards to ensure the overall fairness of the proceedings.

14. I humbly submit that the alleged genuineness of F.H.’s remorse, the trial court’s assurance that F.H.’s testimony was solid and tested by other evidence, and the court’s self-cautioning, although these are relevant factors, were not sufficient safeguards in the circumstances of the case to counterbalance the admission of F.H.’s evidence, especially in the absence of any corroborating evidence and there being no certainty as to the circumstances under which F.H. was eventually granted immunity by the prosecution for the serious offence of unlawful killing. Nor was there any element of judicial control over the decision not to prosecute. In this connection, it is to be noted that in Adamčo (cited above, §§ 70-71), the Court held:

“ ... Moreover, it is noted that all the decisions concerning the prosecution of M. were taken under the sole responsibility of the prosecution service with no element of any judicial control.

Accordingly, in view of the importance of the evidence from M. in the applicant’s trial, the Court finds that, on the specific facts of the present case, its use at the trial was not accompanied by appropriate safeguards so as to ensure the overall fairness of the proceedings ...”

15. Although the decision of the Attorney-General to prosecute or not to prosecute a person is at his or her absolute discretion under Article 113 of the Cypriot Constitution, and it being evident that the testimony of some witnesses may be an important tool for the domestic authorities’ fight against serious crime, had the Attorney-General decided to prosecute F.H. while at the same time admitting him, as he did, into a witness protection programme [3] , the fairness of the applicants’ fair trial would not have been in question. If F.H.’s alleged remorse was truly genuine, as the domestic courts accepted, and if his confession that he had contributed to the crime was not dependent on any understanding or hope that he would not be prosecuted, it is difficult to understand why F.H. was not ultimately prosecuted and instead was allowed to go free and remain unpunished, given that his testimony alone had led to the conviction of four persons for murder, three of whom were the applicants, and to their being sentenced to life imprisonment. The nature and scope of the powers of the prosecution authorities of a member State according to its domestic law do not absolve it from its positive obligations under Article 1 and Article 2 of the Convention or indeed under Article 6 of the Convention to secure to every individual his or her right to have a fair criminal trial.

16. Regarding the point in question, the Assize Court’s judgment referred only to X v. the United Kingdom (cited above), and to two Canadian cases [4] , to refute the applicants’ lawyers’ allegation that the exercise of the discretionary power of the Attorney-General not to prosecute F.H. had been a “world first” (“παγκόσμια πρωτοτυπία”).

17. Here, my concern is, however, the compatibility of the issue raised in the present case with the guarantees of Article 6, so I will confine myself to the reference to X v. the United Kingdom. It should be made clear that, as regards its authoritative or persuasive force, that case was a Commission decision and not a judgment of the Court. Furthermore, that case should be distinguished from the present one: firstly, because, unlike in the present case, in X v. the United Kingdom “[b]efore the trial, prosecuting counsel [had] disclosed tο applicant’s counsel the agreement with the Director of Public Prosecutions and the jury were told [about] the agreement and everything that was known of S.” (i.e., the immunised accomplice witness); secondly, because in X v. the United Kingdom the deal (see clause no. 1 of the agreement) with witness S. explicitly excluded immunity from a criminal charge in respect of homicide; thirdly, the benefit for S. in X v. the United Kingdom was less than that granted to F.H. in the present case since the criminal charge there was for armed robbery and not murder as in the present case; fourthly, because, unlike in the present case, in X v. the United Kingdom, the “[a]pplicant’s counsel did not object to S. giving evidence”; and, lastly, because , unlike in the present case, in X v. the United Kingdom witness S. was arraigned at the Central Criminal Court on an indictment and the “prosecution offered no evidence and the judge ordered a verdict of not guilty to be entered and S. was acquitted”. There is no reference in the Supreme Court’s judgment, acting as an appeal court from the decision of the Assize Court, to X v. the United Kingdom or any judgment of this Court. What the Supreme Court did was to agree with the Assize Court’s decision and hold that the domestic court had the option, if it properly cautioned itself, which it had done, of relying on the uncorroborated testimony of an accomplice.

18. I have extensively submitted elsewhere (in other separate opinions and in academic literature) that the principle of effectiveness, or otherwise the principle of effective protection of human rights, which is the overarching principle of the Convention, underlying all Convention provisions safeguarding human rights, is not only a method or tool of interpretation, but also a norm of international law embodied in each of those provisions. According to the said principle as a method of interpretation, the right must be interpreted broadly, any interference to it must be interpreted narrowly and restrictively, and in case of doubt in the fair balance test between the right and the interference, the right should prevail over the interference: in dubio in favore pro jure/libertate/persona.

19. In Delcourt v. Belgium (17 January 1970, § 25, Series A no. 11) the Court held that “[i]n a democratic society within the meaning of the Convention, the right to a fair administration of justice [held] such a prominent place that a restrictive interpretation of Article 6 para. 1 would not correspond to the aim and the purpose of that provision”. Similarly, in Moreira de Azevedo v. Portugal (no. 11296/84, § 66, 23 October 1990), the Court held that “the right to a fair trial [held] so prominent a place in a democratic society that there [could] be no justification for interpreting Article 6 para. 1 of the Convention restrictively”. Paul Lemmens, dealing with the broad interpretation of Article 6 § 1, emphasised “the need to avoid the guarantees of Article 6(1) becoming merely theoretical or illusory and to ensure that they [remained] practical and effective” [5] . In addition, the International Criminal Court Appeals Chamber has adopted a broad approach to the concept of fairness underlying the fact that it is aimed at the “judicial process in its entirety” [6] .

20. Regarding the principle of effectiveness as a norm of international law, it is my further submission that this norm is enshrined in Article 6 in its totality (since this provision is a coherent unit) and is an element or component of the notion of fairness of Article 6 § 1 of the Convention, which not only requires and demands that the exercise of the right to a fair trial be practical and effective, but which also protects the fairness of the trial from different fundamental flaws and defects.

21. Finding a person guilty of an offence on the basis solely or to a decisive extent on the testimony of an accomplice who, though confessing his or her participation in the crime, is spared prosecution, and without the existence of adequate and sufficient safeguards, would lower or reduce the level of protection afforded to accused persons by Article 6 and their effective participation in the trial, but, most importantly, it would render, in my view, the defence rights theoretical and illusory, not practical and effective, unless there are adequate and sufficient safeguards. This is so, because the very nature of the testimony by an immunised accomplice and sole witness is so flawed that it cannot lead to a fair trial without adequate and sufficient safeguards, which, in line with the principle of effectiveness, like any interference with a right, must be applied with strictness and meticulous and most searching scrutiny.

22. Procedural fairness may operate as a shield for substantive fairness [7] , and thus the Article 6 right to a fair trial may be a shield for protecting substantive justice, but it cannot serve this aim if the “safeguards” accepted by the Court to counterbalance the flaw in question are not of a nature or a capacity that could really ensure the overall fairness of the trial – a fundamental procedural flaw closes the door to substantive justice, without the existence of adequate and sufficient safeguards.

23. Before concluding, I also wish to make some comments regarding the reasoning of the present judgment. In its paragraph 79, the judgment states that in the past the Court has reached its conclusion about the effect of testimony by a sole immunised witness on the fairness of a trial by taking into account several factors, which it then mentions in a non-exhaustive list.

24. One of these factors which is mentioned in the list is the existence of “untainted corroborating evidence”. However, it is clear from the Assize Court’s decision that F.H.’s testimony was not corroborated by any other evidence and that is why the court issued a “self-warning”, stating that it was approaching that evidence with “caution, suspicion, consideration and attention of the highest degree”. This is also clear from the decision of the Supreme Court. That there had been no corroborating evidence is also noted in the present judgment in paragraphs 41 and 56. It is to be clarified, however, that the testimony of the other three witnesses mentioned in paragraph 42 of the judgment was not evidence which corroborated F.H.’s accusation against the applicants. Those witnesses did no more than give “general credence” to F.H.’s testimony in the view of the domestic courts, and this in no way could have led to the applicants’ conviction without the testimony of F.H.

25. The above-mentioned list also includes the following three factors: (a) the fact that the defence knew about the existence of an arrangement with the prosecution, (b) the fact that a domestic court reviewed the arrangement, and (c) the fact that the arrangement was discussed at the trial. However, as is clear from paragraphs 80 and 81 of the judgment, these factors were not applicable in the present case, since the domestic courts established that there had been no arrangement between F.H. and the prosecution.

26. Finally, there is neither a clear assessment in the judgment of the remaining factors on the Court’s list, nor a clear discussion or analysis as to whether those factors were sufficient safeguards in the circumstances of the present case, capable of counterbalancing the flaw of F.H.’s testimony.

27. In the light of the above, I would conclude that there has been a violation of Article 6 § 1 of the Convention.

V. The different meanings of “the overall fairness of trial”: can “safeguards” be enough?

28. Having concluded as above, on the basis of the case-law of the Court, requiring the counterbalancing of the fundamental flaw in question by adequate and sufficient safeguards, I wish now for the sake of a more comprehensive discussion on the topic to present a view which I would have followed, had I not felt somehow obliged to follow the case-law of the Court and had I been a member of a Grand Chamber composition, before which the development of the case-law and any departure from it would have been more apposite.

29. I will start by proposing that the prohibition or preclusion from a criminal trial of fundamental flaws, which inherently taint the whole trial, is an implied specific guarantee of a fair trial which springs from the general right to a fair trial under Article 6 § 1 of the Convention. To put it otherwise, there is a sub-right or an implied specific guarantee in Article 6 § 1 that the right to a fair criminal trial of a person should be free from any fundamental flaws which contaminate the procedure as a whole. An instance of such a fundamental flaw is when a person is convicted on the sole or decisive testimony of an immunised accomplice. Hence the prohibition on convicting a person on such evidence is an aspect or sub-guarantee of the implied specific guarantee or sub-right in question. Article 6 contains not only specific guarantees expressly contained in Article 6 §§ 2 and 3 but also specific implicit guarantees, falling under the umbrella of the general right to a fair trial, which can be considered as specific aspects of this general right to a fair trial, and one of these is the guarantee in question.

30. In my submission, the most compatible approach to the concept of a fair trial under Article 6 would be that a criminal trial under no circumstances, and, irrespective of the existence of any “safeguards”, can be fair if a person is convicted on the basis of sole or decisive reliance on the testimony of an immunised accomplice. That would be so, because the flaw is intrinsic and so fundamental to the procedure that the whole trial and any balancing test within it would be tainted completely by the flaw, and no “safeguards” would be able to counterbalance it. On the contrary, it would be futile and pointless to counterbalance any factor or “safeguard” against a fundamental flaw which would pervade the whole trial to such an extent so as to contaminate it completely. As a result of the flaw in question, not only did the whole trial become unfair for the applicants, it also ran counter to a common sense of justice and the public interest, which does not tolerate that an accomplice should be spared prosecution and face no consequences for participation in a murder, while at the same by his testimony rendering the criminal trial of four persons wholly unfair, with the consequence that they received a life sentence. It is not permitted by the interests of justice for an accused to have an unfair trial and consequently an improper conviction.

31. The view under discussion is, in my view, compatible with the literal and purposive meaning of the concept of “the overall fairness of trial” under Article 6, and, of course, is compatible with the principle of effectiveness both as a norm of international law and as a method of interpretation. In order to serve the true meaning of the right to a fair trial, the concept of the overall fairness of a trial, which provides for an all-embracing and “unifying standard” [8] , must be used so that: (a) every part or component of the trial and all its proceedings is fair, and (b) all the implicit guarantees of the general right to a fair trial under Article 6 § 1 and the explicit guarantees of the specific aspects of fair trial under Article 6 §§ 2 and 3, are satisfied.

32. A consequence of this would be that any shortfall in an Article 6 guarantee would automatically entail a violation of Article 6. In other words, according to this view, each guarantee of Article 6 is an indispensable element or component of the concept of a fair trial and incorporates the norm of effectiveness. The lack of any such guarantee may implode the fairness of a trial. Put in negative terms, a trial cannot be fair unless it is fair regarding all its procedural aspects. As Ioannis Sarmas eloquently argued “the right to a fair hearing is indeed the right to a fair trial , a right in constant normative extension, including all aspects of procedural justice” [9] . In the same vein, Georghios M. Pikis insightfully argues [10] :

“Assurance of a fair trial is a fundamental human right, the emblem of restorative justice. A fair trial constitutes a fundamental right of man and correspondingly, a principal duty of the State to secure [ sic ] in all circumstances. No deviation or shortfall of a fair trial should be countenanced. The norms of fair trial are fashioned to the needs of justice. It is the inexhaustible duty of the Court to do justice according to the norms of a fair trial. In its absence, freedom of man is muted and man’s hypostasis undermined. A fair trial can appropriately be labelled as the bedrock of human rights.”

33. Furthermore, the preferred view is more compatible with assigning a “moral reading” to human rights provisions [11] . The moral character of human rights is perhaps, more than in any other provisions of the Convention, deeply rooted in Article 6, the right to a fair trial, and is reflected in the adjective “fair”, referring to the fair hearing within the meaning of Article 6 § 1, and, of course to the heading of the Article, namely, “Right to a fair trial”. The concept of fairness is frequently used as a standard of just treatment [12] . The principle of fairness has its roots in the teachings of the ancient Greek philosopher Aristotle, according to which, justice, as a moral value, consists of both lawfulness and fairness, “‘[t]he just’ therefore means that which is lawful and that which is equal or fair and ‘the unjust’ means that which is illegal and that which is unequal or unfair” [13] .

34. It is my submission that in all the Article 6 guarantees, in which the norm of effectiveness is incorporated, a moral element is also incorporated. In my view, both the norm of effectiveness and the moral element make the provisions of this Article a more solid and coherent unit in the sense that every guarantee should be respected for its necessary and indispensable contribution to the overall fairness of the trial.

35. This view is in line with Dworkin’s constructive interpretation, the foundation of which is the conception of “law as integrity”, namely, the moral reading of the law as a single, coherent and internally consistent whole [14] , which renders practice in its best light, facilitating consistency and unity [15] . Article 6 is considered as a “framework right” covering all aspects, guarantees and principles of fair trial [16] .

36. Regrettably, however, the Court treats the concept of overall fairness as the result of a balancing exercise, by weighing up a shortfall in an Article 6 guarantee against certain interests or factors or alleged safeguards [17] . Andreas Samartzis argues “that the overall fairness assessment can be viewed as a pathology of the Court’s proportionality analysis, an instance of spillover into Article 6 cases of the Court’s tendency to determine its conclusions on the basis of ad hoc balancing” [18] . He also observes that “[t]he novelty of the overall fairness may override the literal meaning of Article 6, not only to expand, but surprisingly also to negate the minimum fair trial guarantees of Article 6(3)” [19] . Paul Lemmens remarks that “[t]he Court seems to have become more ‘result-orientated’ which allows it to leave room for domestic policy considerations when it comes to creating the framework for judicial proceedings” [20] .

37. Unfortunately, this concept of overall fairness is misleading, because even though an important guarantee of Article 6 may be missing or violated, the trial may still be considered by the Court to be fair overall. As Ryan Goss, argues on the same lines, “[f]requently the European Court purports to be applying the ‘proceedings as a whole’ test but plainly considers something less than, or different to, the proceedings as a whole” [21] . This understanding and use of the concept, apart from being an oxymoron, also undermine the rule of law and negate the minimum fair trial rights of Article 6 § 3 and other guarantees of Article 6, as they sacrifice the indispensable guarantees of Article 6 for the sake of other considerations. It would be contrary to the meaning of the holistic approach to interpreting and applying Article 6, in the context of the overall examination of the fairness of a trial, if one of the guarantees of this Article, which is in fact a fundamental sub-right, were devoid of essence.

38. It can be argued that the balancing test employed in order to assess the overall fairness of a trial cannot operate as if it were a magic formula to excuse or pardon any violations of Article 6. Regrettably, what happened in the present case is not only that F.H. was granted immunity by the Attorney-General, but also that the fundamental flaw of his testimony was, metaphorically speaking, “immunised” by “the overall fairness of the trial test”, as employed by the domestic courts.

39. Though the relevant issue in the present case does not come under Article 6 § 3, but under Article 6 § 1, reference must be made to all the provisions of Article 6 in order to show that this – in my opinion – erroneous concept of the overall fairness of the trial pervades Article 6 in its totality. To give some examples: the overall fairness test is used by the case-law when examining: (a) the minimum right to have a lawyer of one’s own choosing under Article 6 § 3 (c) (see Beuze v. Belgium [GC], no. 71409/10, §§ 120, 144-150, 165, 171, 9 November 2018) [22] ; (b) an issue regarding conviction based solely or to a decisive degree on statements that have been made by a person whom the accused has had no opportunity to examine or to have examined – the so called “sole or decisive rule” (see Al-Khawaja and Tahery v. the United Kingdom [GC], nos. 26766/05 and 22228/06, §§ 119, 126-65, 15 December 2011 [23] , and Gani v. Spain, no. 61800/08, § 42, 19 February 2013) [24] , and an issue concerning the use of unlawfully obtained evidence (see Allan v. the United Kingdom, no. 48539/99, § 42, 5 November 2002). Perhaps the only exception where, according to the recent case-law of the Court, the overall fairness of the trial is not examined is where evidence is obtained through torture (see Jalloh v. Germany [GC], no. 54810/00, § 105, 1 June 2010) [25] .

40. The Court applies a balancing test and procedural safeguards not only in relation to Article 6 but also when interpreting and applying some other provisions of the Convention, such as Article 1 § 1 of Protocol No. 7 to the Convention. Regarding the latter, the Court in Muhammad and Muhammad v. Romania ([GC], no. 80982/12, 15 October 2020), albeit erroneously, in my view, considered an absolute procedural right concerning expulsion of aliens to be a qualified one, placing on it implied limitations and then applying counterbalancing factors which are not mentioned in the said provision [26] . In a different context, but with a similar effect, the Court in interpreting Article 18 of the Convention (“limitation on use of restrictions on rights”), in Merabishvili v. Georgia ([GC], no. 72508/13, § 305, 28 November 2017) and in Navalnyy v. Russia ([GC], nos. 29580/12 and 4 others, § 165, 15 November 2018), regrettably [27] applied the “dominant-purpose test”, which tolerates the presence of illegitimate purposes which are not predominant. Of course, here, I am not dealing with an Article 18 issue nor do I discuss whether this provision applies also for the purposes of Article 6, but the comparison is born out of the case-law of the Court that unfortunately interprets and applies Convention provisions so that certain fundamental flaws or defects are not given their appropriated consideration by following the principle of the effective protection of human rights.

41. I believe that the erroneous meaning of the overall fairness of the trial under Article 6 is related to the misconception in the case-law that the right under this Article is a qualified right in all its aspects rather than an absolute one with only limited exceptions. That may explain why the case-law gives only a relative meaning to Article 6 guarantees, putting them “in a wider perspective and assessing them in the light of the general right to a fair trial” [28] .

42. In this connection, the following comparison may be useful. Regarding qualified rights, such as those guaranteed by Articles 8-11 of the Convention, the balancing test is made between the right concerned and a lawful and legitimate interference, while in the case of Article 6, the balancing is between a shortfall in an express or implied sub-right or guarantee and implied interests or factors or alleged safeguards. The difference between these two situations is significant in that in the first situation what is compared is a right, while in the second situation what is compared is not a right, but a shortfall in a guarantee of a right.

43. Accordingly, it is not correct to accept a shortfall in a guarantee of a right, or a fundamental flaw in the procedure, as something which can ultimately be overlooked, by comparing it with implied interests, factors and alleged safeguards. To do so would risk allowing deviations from Article 6 guarantees to persist, with the result of undermining or breaching the rule of law which pervades the provisions of Article 6 arguably more than any other Article of the Convention. The variables compared should be legal or legitimate ones, like rights and lawful and legitimate restrictions and within the legal balancing scale; there should not be anything other than such variables or anything which may infringe or be incompatible or irreconcilable with the concept of a fair trial and, as inferred therefrom, the requirement of the proper administration of justice (on the latter, see Ramanauskas v. Lithuania, cited above, § 53).

44. Consequently, where there is a balancing test there should be no room in it to place fundamental defects, flaws and shortfalls in guarantees. According to the view under discussion, however, there should be no balancing of any interests under Article 6, save in exceptional cases, such as where the interests of children or vulnerable victims so require.

45. It is to be observed that, in fact, the balancing test made in the present case by the domestic courts and the Court involved weighing up a fundamental flaw – a potential violation – against certain alleged counterbalancing factors mainly concerning the merits of the case, treating them as “safeguards”, for example by finding that the testimony of F.H. was solid and that he was a trustworthy witness. Lastly, it must be emphasised that fairness for the purposes of Article 6 tolerates neither disrespect for the rule of law and the values of civilised societies founded upon the rule of law, nor the non-preservation of the integrity of the judicial process, considered as an organic whole. A fundamental flaw, such as the one under discussion, does not amount merely to breaking a link in the chain of evidence but to not having any chain of evidence tο begin with and ultimately to the absence of evidence against the accused. This amounts not to a mere handicap in the procedure but to the amputation of the whole procedure.

46. This hybrid defect-curing balancing test, however, is a paradox, because a court should enter into the merits of the case only if it has first established that the procedure was fair.

47. In my view, a court should not examine the case on the merits in order to pardon a fundamental procedural flaw under Article 6 § 1. If it does so, one can rightly argue that the aim of this Article is defeated and the very existence of this provision “holding so prominent a place in a democratic society” (see paragraph 19 above) is overlooked.

48. I will finish with some concluding comments on the view under discussion. Convicting a person solely or decisively on the evidence of an immunised accomplice is a fundamental flaw of a trial which by its very nature is bound to make the whole trial unfair, without the need for any balancing test and without any counter-balancing factors being sufficient. Not only does such a flaw mean that there is no fair trial, but it is the very enemy of a fair trial. The inherent fundamental flaw of the testimony of F.H. was not only a source of legal uncertainty for the applicants, while giving the prosecution a clear advantage in the handling of their case – possibly breaching the principle of equality of arms –, but it also deprived the applicants of their right to a fair criminal hearing in its essence. Stated otherwise, this fundamental flaw, in my view, deprived the applicants of their liberty, without due process (also safeguarded by the 5th Amendment of the U.S. Constitution), thus without ensuring them a fair trial.

49. I wish to make it absolutely clear that in no way has the view I have presented under this head influenced my judgment that there were no sufficient safeguards in the present case to counterbalance the flaw in question. My present opinion is based exclusively on the existing case-law regardless of whether, by following the view I have elaborated under this head, the result would be the same, thus a finding of a violation.

VI. Conclusion

50. In conclusion, by following the case-law of the Court, I would find that there has been a violation of Article 6 § 1 of the Convention and I would award the applicants an amount in respect of non-pecuniary damage, costs and expenses; an amount which, however, it is not necessary to determine since I am in the minority.

APPENDIX

​

No.

Name

Born in

Mr Grigoris XENOFONTOS

(“the first applicant”)

1981Mr Anastasis KRASOPOULIS

(“the second applicant”)

1973Ms Elli SKORDELLI KRASOPOULI

(“the third applicant”)

1968[1] See further comment on this in Harris, O’Boyle and Warbrick, Law of the European Convention on Human Rights, 4th edition (OUP, 2018), pp. 420-422; see also Pieter Van Dijk, Fried Van Hoof, Arjen Van Rijn and Leo Zwaak (eds), Theory and Practice of the European Convention on Human Rights, 5th edition (Intersentia, 2018), pp. 643, 650.

[2] The terminology varies for an accomplice turned informer, as reflected in the English expression “to turn Queen’s/King’s evidence”. The term “crown witness” is used, for example, by Pieter Van Dijk, Fried Van Hoof, Arjen Van Rijn and Leo Zwaak (eds), cited above, pp. 649-650. Other languages have specific terms such as repenti (French) or pentito (Italian) for offenders who cooperate with the authorities.

[3] See the Witnesses Protection Law of 2001 (Law 95(I)/2001), as amended. It is to be noted that this Law includes neither a provision for granting immunity to an accomplice who becomes a prosecution witness, as a measure for his or her protection, nor a provision that an accomplice must first be convicted (after his or her confession or a trial) before becoming a prosecution witness giving evidence against a defendant. Therefore F.H. could still have been in the witness protection programme without being granted immunity.

[4] See also the reference to these two Canadian cases as well as the reference to two other similar cases, one of the Supreme Court of California and the other of the Supreme Court of Texas, in Takis Eliades and Nicolas G. Santis, with the collaboration of Georgios Stavrinakis, The Law of Evidence: Procedural and Substantive Aspects (Hippasus Publishing, 2014), pp. 659-660 (in Greek).

[5] See Paul Lemmens, “The Right to a Fair Trial and its Multiple Manifestations”, in Eva Brems and Janneke Gerards (eds.), Shaping Rights in the ECHR – The Role of the European Court of Human Rights in Determining the Scope of Human Rights (Cambridge University Press, 2013), p. 314.

[6] See Situation in the Democratic Republic of Congo in the case of the Prosecutor v. Thomas Lubanga Dyilo, Judgment on the Appeal of Thomas Lubanga Dyilo against the Decision on the Defence Challenge to the Jurisdiction of the Court, ICC-01/04-01/06-772 (OA4) (2006) at paragraph 37. See also Nicolas A. J. Croquet, “The International Criminal Court and the Treatment of Defence Rights: A Mirror of the European Court of Human Rights’ Jurisprudence?”, in Human Rights Law Review 11:1 (2011), p. 100.

[7] I elaborated on this point further in my Partly Dissenting Opinion in Regner v. the Czech Republic [GC], no. 35289/11, 19 September 2017, especially in paragraph 37.

[8] A term used by Andreas Samartzis, “Weighing Overall Fairness: A Critique of Balancing under the Criminal Limb of Article 6 of the European Convention on Human Rights”, in Human Rights Law Review, 2021, 21, p. 410.

[9] See Ioannis Sarmas, “Fair Trial and Search for Truth in the Case Law of the European Court of Human Rights”, in R. Spano, I. Motoc, B. Lubarda, P. Pinto de Albuquerque, M. Tsirli (eds), Fair Trial: Regional and International Perspectives , Liber Amicorum Linos-Alexandre Sicilianos (Anthemis, 2020), p. 500.

[10] See Georghios M. Pikis, Justice and the Judiciary (Martinus Nijhoff Publishers, 2012), at § 145 (p. 63).

[11] To use the term of Ronald Dworkin. See Ronald Dworkin, “Law’s Ambitions for Itself” (1985), 71(2) Virginia Law Review , pp. 173, 176, 178, 181-182 and 185; Ronald Dworkin, Law’s Empire (Bloomsbury, 1986, Hart Publishing, 2021), p. 411; and Ronald Dworkin, Freedom of Law: The Moral Reading of the American Constitution (Harvard University Press, 1997) .

[12] See Francis A.R. Bennion (ed.), Bennion on Statutory Interpretation: A Code , 5th edition (LexisNexis, 2008), section 265, p. 797.

[13] See Aristotle, The Nicomachean Ethics, Book V. i, 8 (Loeb Classical Library XIX, 1990), translated into English by H. Rackham.

[14] See on Dworkin’s law as integrity: Ronald Dworkin, Law’s Empire , cited above, pp. 176-224, 225-275, 410-411; Ronald Dworkin, “Law’s Ambitions for Itself”, cited above, pp. 184-186; and M. D. A. Freeman (ed.), Lloyd’s Introduction to Jurisprudence , 9th edition (Sweet & Maxwell, 2014, 2016 2nd impression), pp. 599-602 .

[15] See Ronald Dworkin, Law’s Empire , cited above, pp. 410-411.

[16] See Nicolas A. J. Croquet, cited above, pp. 100, 128; Van Dijk, “Universal Legal Principles of Fair Trial in Criminal Proceedings”, in Rosas and Helgesen (eds), Human Rights in a Changing East-West Perspective (Pinter Publishers, 1990), 89, p. 112.

[17] See Paul Lemmens, who characterises the approach of the Court, which involves weighing in the balance the competing interests under Article 6, as “the new approach – or better, the return to the original approach” (Paul Lemmens, cited above, p. 313, see also ibid., pp. 311-312).

[18] See Andreas Samartzis, cited above, p. 421.

[19] Ibid ., p. 416

[20] See Paul Lemmens, cited above, p. 313. Lemmens considers that this development “seems to fit with an increased emphasis on the ‘subsidiarity’ character of the Convention protection system”.

[21] See Ryan Goss, Criminal Fair Trial Rights – Article 6 of the European Convention on Human Rights (Bloomsbury, 2016), p. 125.

[22] In that case, as in Simeonovi v. Bulgaria [GC], no. 21980/04, 12 May 2017, the overall fairness of the trial had been examined even though there were no compelling reasons for the restriction of the right to have a lawyer. Compare, however, these two cases with Salduz v. Turkey [GC], no. 36392/02, § 58, ECHR 2008, which is more in line with the preferred view, where it was held that the right to have a lawyer should be available unless there are compelling reasons to refuse, but even where such reasons exceptionally justify a denial of access, such restriction should not unduly prejudice a defendant’s rights. The approach in Salduz was followed by my Partly Dissenting Opinion in Simeonovi, cited above, and my Joint Dissenting Opinion with Pinto de Albuquerque in Farrugia v. Malta , no. 63041/13, 4 June 2019. In John Murray v. the United Kingdom [GC], no. 18731/91, § 59, 8 February 1996, it was held that “the decision to deny [the applicant] access to a solicitor unfairly prejudiced the rights of the defence and rendered the proceedings against him unfair contrary to Article 6 paras. 1 and 3 (c) of the Convention”. This view is even more in line with the preferred view because there the Court found a violation of Article 6, without examining whether there had been compelling reasons justifying the restriction and, of course, without referring to or examining the overall fairness of the trial.

[23] In paragraph 146 of its judgment in Al-Khawaja and Tahery v. the United Kingdom , cited above, the Court held “it would not be correct, when reviewing questions of fairness, to apply the sole or decisive rule in an inflexible manner”, and “nor would it be correct for the Court to ignore entirely the specificities of the particular legal system concerned and, in particular its rules of evidence, notwithstanding judicial dicta that may have suggested otherwise”. In its view, “[t]o do so would transform the rule into a blunt and indiscriminate instrument that runs counter to the traditional way in which the Court approaches the issue of the overall fairness of the proceedings, namely, to weigh in the balance the competing interests of the defence, the victim, and witnesses, and the public interest in the effective administration of justice.”

[24] On the “sole or decisive rule”, see also William A. Schabas, The European Convention on Human Rights – A Commentary (OUP, 2015), p. 313; Ryan Goss, cited above, pp. 170 ‑ 176; Elmar Widder, A Fair Trial at the International Criminal Court? Human Rights Standards and Legitimacy – Procedural Fairness in the Context of Disclosure of Evidence and the Right to Have Witnesses Examined (PL Academic Research, 2016), pp. 74-82.

[25] See also on this, Paul Lemmens, cited above, p. 309; Harris, O’Boyle and Warbrick, cited above, p. 420; and William A. Schabas, cited above, pp. 320-321.

[26] Of relevance is paragraph 5 of my Concurring Opinion in the cited case of Muhammad and Muhammad .

[27] For criticism of the dominant-purpose test, of relevance is my concurring opinion in Merabishvili, cited above.

[28] See Paul Lemmens, cited above, p. 309.