EHRM, 15-12-2015, nr. 9154/10

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.

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.

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).

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 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:

The Government contested that argument.

Article 41 of the Convention provides:

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.

The Government did not comment on the question of just satisfaction in the proceedings before the Court.

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.

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.

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.

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).

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.

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).

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.

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.

Finally, we agree that the Court should scrutinise carefully whether the national authorities applied sufficient counterbalancing measures (see paragraphs 125 et seq.).

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.

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.

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).

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…’.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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).

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).

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).

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.

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.

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.

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.

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).

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.

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.

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).

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.

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.

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.

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.

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).

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.

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.

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.

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).

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.

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).

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’.

Therefore, in my view, the clarification does not imply a departure from the ‘sole or decisive’ test.

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).

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.

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.

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).

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).

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.

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’.

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).

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).

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.

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).

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.

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).

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.

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).

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.

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’.

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.

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.

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.

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.