CONCURRING OPINION OF JUDGE PAVLI, JOINED BY JUDGE RAVARANI

1 . I have voted in favour of the unanimous finding that there has been a violation of the applicant ’ s rights under Article 6 § 1 of the Convention. I regret, however, that I cannot join the majority ’ s reasoning in arriving at this conclusion. In my view, the Court should have found a violation under the substantive test of incitement. I therefore see no need to have continued on to the procedural test in this case.

2 . I have three points that I would like to make in this connection. The first is to highlight a certain lack of clarity in the Court ’ s case-law in respect of the application of the substantive and the procedural tests in entrapment cases. The second relates to the question of the burden of proof in the context of the substantive test. The third is to explain why, given these first two points, I consider that the Court should have concluded that an application of the substantive test alone was sufficient to find a violation of the applicant ’ s rights under Article 6 § 1.

3 . As the Court notes in paragraph 46 of its judgment in the case, its examination of entrapment complaints has developed on the basis of two tests: the substantive and the procedural test of incitement. The substantive test asks whether the authorities had good reasons for mounting a covert operation. In particular, the authorities must show that they were in possession of concrete, objective and verifiable evidence indicating that initial steps had been taken to commit the acts constituting the offence, and that the criminal act was already underway at the time when the police intervened (see paragraph 48 of the judgment). The procedural test asks whether the applicant was able to effectively raise the issue of incitement or entrapment during the domestic proceedings (see paragraph 54 of the judgment).

4 . It is in the inherent logic of these two tests that they are to be applied sequentially, with the Court turning to the procedural test only if the substantive test is inconclusive. This was the approach taken, for example, in Teixeira de Castro v. Portugal (9 June 1998, Reports of Judgments and Decisions 1998-IV), in which the Court determined that the application of the substantive test indicated a violation of Article 6 § 1 and did not find it necessary to continue on to a procedural analysis. This does not exclude that, in reaching its conclusion under the substantive test, the Court may refer to evidence produced and other findings made by the national courts in the course of the domestic proceedings.

5 . The current judgment, relying on the 2010 judgment in Bannikova v. Russia (no. 18757/06, 4 November 2010), notes that “as the case-law currently stands, the Court considers the procedural aspect a necessary part of the examination of the agent provocateur complaint” (see paragraph 49 of the judgment, emphasis added). There is, however, no proper reasoning offered in Bannikova for considering such a two-tier assessment necessary, other than a circular and general reference to the case-law (see Bannikova , cited above, § 51). At the same time, that judgment provides an alternative and genuine reason for shifting to the procedural test: pointing to “cases where the lack of file disclosure or the controversy of the parties ’ interpretation of events precludes the Court from establishing with a sufficient degree of certainty whether the applicant was subjected to police incitement”; the judgment emphasises that in such cases “the procedural aspect becomes decisive” (ibid., § 52). Under the same logic, the procedural test would not be “necessary” if the outcome of the substantive test is clear and conclusive. Any other conclusion begs the question as to what would be the point of having a substantive test in the first place. Furthermore, whatever the validity of the Bannikova position in 2010, it does not appear that such an approach has been invariably followed by the case-law in the intervening years. For example, in Furcht v. Germany (no. 54648/09, § 59, 23 October 2014), the Court simply concluded that “the undercover measure at issue ... amounted to police incitement” and that the evidence thus obtained “was further used in the ensuing criminal proceedings against the applicant”, leading to a violation of Article 6 § 1.

6 . Once the Court is satisfied that there was police incitement in the case at hand, the key remaining question is whether the domestic courts excluded any evidence obtained through such incitement from the resulting criminal proceedings in order to preserve the applicant ’ s fair trial rights.

7 . Shifting now to the question of the burden of proof, the Court has repeatedly held that the authorities bear the burden under the substantive test (see, for example, Bannikova , cited above, § 48), being required to demonstrate at any stage that they had good reasons for mounting the covert operation (ibid., § 40). At the domestic level, “it falls to the prosecution to prove that there was no incitement, provided that the defendant ’ s allegations are not wholly improbable” (see, for example, Ramanauskas v. Lithuania [GC], no. 74420/01, § 70, 5 February 2008). At the Strasbourg level, when applying the criteria of the substantive test, the Court has said that it places the burden of proof on the authorities (see Furcht , cited above, § 53).

8 . It follows that, once the applicant puts forward an arguable claim of incitement, the respondent Government must provide sufficient evidence to rebut such a claim or at least to call into question a finding under the substantive test. In doing so, the Government cannot simply rely on the failure of national law enforcement to record and disclose the basics of the relevant operations, or follow any relevant procedures for its authorisation and monitoring. This would go against the very purpose of the ban on police incitement, especially in a context, such as the Russian one, where the Court has been consistently critical of the failure, at the legislative and practical levels, to subject undercover operations to prior authorisation with independent supervision (see paragraph 53 of the judgment). Such a failure tends to undermine the legitimacy of such operations “from the outset” (see Nosko and Nefedov v. Russia , no. 11789/10, § 64, 30 October 2014) and even to encourage their arbitrary use as a routine law enforcement tool.

9 . Turning to the current case, the Court concluded that the Government had not “met the requirement of the burden of proof to demonstrate that the applicant had not been subjected to police entrapment”, but opted nevertheless to “move on to the procedural test” (see paragraph 54 of the judgment). While I agree with the first conclusion, I cannot follow the second.

10 . In my view, the national authorities have failed, both at domestic level and before this Court, to put forward any concrete, objective and verifiable evidence showing that the alleged criminal act was already underway. In fact, there was prima facie evidence that the transaction in question was initiated by a police officer. Although the officer claimed that he had prior operational evidence of the applicant ’ s propensity to commit the offence in question, no court has ever seen that evidence or any details about it; and no other evidence has been provided by the Government. Under these circumstances, the Court could and should have found, under the substantive test, that the applicant was subjected to police entrapment. An arguable claim put forward by the applicant requires at least some evidence to the contrary if it is to be considered inconclusive.

11 . Finally, it is patently clear that the national courts did not exclude the evidence obtained through incitement and in fact relied on it in convicting the applicant – a conclusion that can certainly be reached without having to deploy the fully-fledged procedural test. That failure to exclude the tainted evidence is sufficient to find a violation of Article 6 § 1.