DISSENTING OPINION OF JUDGE PAVLI, JOINED BY JUDGES RAVARANI AND ZÃœND

1. I have voted against the finding of no violation of Article 10 in this case. Apart from the difference of opinion on the outcome, my disagreement with the majority extends to certain key findings of fact as well as the methodology that has been followed to reach this conclusion, which risks muddying the waters of our developing jurisprudence in this relatively novel field.

2. First, it is necessary to start with an important factual clarification. The respondent Government have put forward the misleading argument that the applicant was “not denied access to the desired information and he did not argue that the information disclosed to him was incorrect or insufficient” (see paragraph 44 of the judgment). I am unable to agree with this submission. The applicant requested access to thousands of pages of primary source information collected by the German intelligence service in relation to the suspicious death of a senior government official several decades ago. What he received from the intelligence service, as a result of an out-of-court settlement, was merely a high-level description of the categories of information that the relevant files contained (see paragraph 16 of the judgment). While such information may be quite useful in certain circumstances, perhaps as a preliminary step in narrowing down a wide ‑ ranging request, it is simply not what the applicant asked for. It is the equivalent of requesting to read a book, and being offered its table of contents as a perfectly good alternative. The applicant has never received access to a single page from the original files. Unfortunately, the majority have chosen to gloss over this crucial distinction and to accept that the applicant’s information request was satisfied “at least in part” (see paragraph 50 of the judgment). This fallacy lays the ground for much of the majority’s reasoning.

3. A second stepping-stone to the finding of no violation of Article 10 is the majority’s reluctance to hold that the threshold conditions for the applicability of Article 10 were clearly met in this case. The question of applicability is first joined to the merits (see paragraph 37 of the judgment) and then ultimately left open, presumably on the basis that Article 10 does not necessarily grant a right of “physical access” to government files (see paragraph 51 of the judgment). This conclusion is based, in my view, on a significant misunderstanding of both our existing jurisprudence and the way virtually all access to information regimes operate in the European space.

4. To begin with, the majority’s timidity would seem to suggest that this is the first time the Court has been faced with such a question. As this is not the case, in my view, there was no need to question or cast doubt on a well ‑ established element of our Article 10 case-law. A large number of access to information cases decided by various Sections of the Court, following the Grand Chamber’s ground-breaking Magyar Helsinki judgment ( Magyar Helsinki Bizottság v. Hungary [GC], no. 18030/11, 8 November 2016), have dealt specifically with requests by applicants for access to original documents, i.e. primary-source information held by State authorities [1] . In none of these cases has it been disputed – neither by the respondent governments, nor by the Court – that Article 10 applies in principle to such requests [2] . This being a question of the Court’s ratione materiae jurisdiction, it ought to have been considered in any event of the Court’s own motion. As a result, it seems obvious to conclude that the Court has long accepted, expressly or implicitly, that a right of access to the original files or documents held by a government authority is protected in principle by Article 10. The majority have chosen to ignore this line of case-law.

5. Furthermore, the majority’s approach is at odds with a core and widely accepted conception of the right to information in European comparative law as well as under Council of Europe standard-setting instruments. The primary form of access guaranteed by these laws – including the German federal access law itself – is direct access to primary official documents and sources, irrespective of their format (whether as an original, complete and authentic paper copy, electronic copy and so on) [3] . This is also a matter of common sense: any serious journalist or researcher would want to see the original government data, not merely information about the information , the metadata or a government-prepared summary of the requested information. The fundamental guarantees of Article 10 do not rest on the assumption that the government version of events is always to be trusted. Any researcher who has spent time leafing through dusty old records in government archives would testify to that; the practice is older than the lost library of Alexandria. The Court itself has emphasised, in a national security context, that “access to original documentary sources for legitimate historical research [is] an essential element of the exercise” of the right to freedom of expression (see Kenedi v. Hungary , no. 31475/05, § 43, 26 May 2009).

6. Thirdly, the majority find that Article 10 has not been violated despite the national courts’ manifest failure to engage in any meaningful balancing of the applicant’s Article 10 interest in obtaining access to the original files (in whole or in part) against any ongoing national security interests in preserving their secrecy. The judgment does so by relying on the applicant’s supposed failure to sufficiently justify before the domestic authorities the need for physical access as such – an omission that presumably “failed to put the domestic authorities in a position to engage in the necessary balancing of the competing interests” (see paragraph 57 of the judgment). In other words, the majority (like the national courts) wonder why the applicant could not have been satisfied with having received the table of contents, rather than the whole book, while investigating a complex set of events related to a death under disputed circumstances that had attracted a great deal of public interest.

7. In fact, this core question is tied to a structural problem in the German access to information regime. The intelligence services being entirely exempted from the scope of federal access to information law, the only option left to journalists (and to journalists exclusively) for requesting access to their information is to rely on a complex mix of judge-made constitutional remedies and/or Land -based legislation. The applicant’s case suggests, however, that these remedies are imperfect and subject to a high substantive threshold and burden of proof to be met by the requester (“the right of the press to receive information may consolidate to become a right to consult files in person”, under certain undefined scenarios; see paragraph 57 of the judgment). These national thresholds are arguably stricter and therefore incompatible with the four-factor threshold set by the Court’s Grand Chamber in Magyar Helsinki (cited above) for Article 10 to become applicable. The lack of a generalised constitutional basis for the right of access to official information at the German federal level, such that it would be commensurate with the level of protection guaranteed by Article 10 of the Convention, is another source of complication. This idiosyncratic national legal framework, as applicable to the intelligence services in this particular case, is hardly compatible with our own established jurisprudence - an aspect that the judgment does not address. Any national legal framework that shields or exempts entire government agencies from the operation of the right of access to government information, or that categorically bars requesters from access to certain primary sources in the absence of any balancing of interests, is bound to run into similar Convention problems in my view.

8. Nothing in this separate opinion should be taken, of course, to suggest that people should be able to roam freely through the intelligence agencies’ archives. On the other hand, as national security data about historical events of general importance become older or liable to be declassified, the interests of historical research and the public’s right to know become stronger and may tip the balance in favour of disclosure, in whole or in part (see Kenedi , cited above, involving access to historical records of the Hungarian secret service). Established democracies have developed multiple mechanisms to do this without undermining the internal working methods or sources of the services, or other remaining national security concerns; redactions and partial disclosure being among these standard tools, in line with the principle of proportionality. Had the national authorities put forward sound substantive reasons as to why national security would have been seriously jeopardised by granting the applicant any degree of access to the physical files – and had the national courts scrutinised such arguments in line with Article 10 standards – the refusal of physical access, in whole or in part, might be considered justified. Conversely, the failure of the domestic authorities to engage in any meaningful balancing of the substantive interests at stake is sufficient, in my view, to find a violation of Article 10 in this case. The majority’s strictly procedural approach means that the Court itself has also missed an opportunity to enrich our jurisprudence on questions of historical memory in a national security context.

[1] See, among other cases, Cangi v. Turkey (no. 24973/15, 29 January 2019) ; Studio Monitori and Others v. Georgia ((dec.) nos. 44920/09 and 8942/10, 30 January 2020); Centre for Democracy and the Rule of Law v. Ukraine ((dec.) no. 75865/11, 3 March 2020); Centre for Democracy and the Rule of Law v. Ukraine (no. 10090/16, 26 March 2020); and Leshchenko v. Ukraine (nos. 14220/13 and 72601/13, 21 January 2021).

[2] The fact that the relevant Article 10 complaint may have been found inadmissible on other grounds, or with respect to the specific nature of the information being requested (see e.g. Studio Monitori , cited above), does not change this general conclusion.

[3] See, in particular, the Committee of Ministers’ Recommendation (2002)2 on Access to Official Documents; and its Explanatory Memorandum, paragraphs 5-7 and 34-35 (including with reference to “confidential, secret or top secret” documents).