PARTLY CONCURRING, PARTLY DISSENTING OPINION OF JUDGE KOSKELO, JOINED BY JUDGE TURKOVIĆ

1. I have voted, and agree, with the majority as regards points 1 to 3 of the operative provisions of the judgment, which concern the admissibility of the complaints. I have also joined the majority in finding a violation of Article 8 in respect of both the section 8(4) regime and the Chapter II regime. As regards the section 8(4) regime, however, I am not able in all respects to subscribe to the reasons given by the majority. As far as the intelligence sharing regime is concerned, unlike the majority, I have voted for finding a violation of Article 8.

I. The RIPA section 8(4) regime

2. The present case concerns legislation providing for secret surveillance, by means of bulk interception, of electronic communications which qualify as “external” (for an understanding of the concept of “external” communications see paragraphs 69-71 of the judgment). It is important to note that this type of secret surveillance of communications is not limited to certain already known or identified targets but is aimed at the discovery of threats and hitherto unknown or unidentified targets which might be responsible for threats (see paragraph 284 of the judgment). The relevant threats are broadly framed and comprise threats to national security or to the economic well-being of the country as well as threats arising from serious crime (see §§ 57-59).

3. It is obvious that such an activity – an untargeted surveillance of external communications with a view to discovering and exploring a wide range of threats – by its very nature takes on a potentially vast scope, and involves enormous risks of abuse. The safeguards against those risks, and the standards which under the Convention should apply in this regard, therefore raise questions of the highest importance. I am not convinced, in the light of present-day circumstances, that reliance on the Court’s existing case-law provides an adequate approach to the kind of surveillance regimes like the one we are dealing with here. A more thorough reconsideration would be called for. I acknowledge that this would be a task for the Court’s Grand Chamber. I will only raise some concerns which, in my view, require attention in this regard.

(i) The context of earlier case-law

4. Apart from the recent Chamber judgment in Centrum för Rättvisa v. Sweden (no. 35252/08, 19 June 2018), which is not yet final, the Court’s case-law has not dealt with the present kind of surveillance but with regimes which, as a matter of either law or fact, have been narrower in scope. Furthermore, in the light of current developments, I consider that reliance on the line of existing case-law is no longer an adequate basis for assessing the standards which under the Convention should govern this particular domain.

5. The Court’s case-law on secret surveillance of communications essentially dates back to Klass and Others v. Germany (cited in the judgment) which was decided by the Plenary Court four decades ago, and the admissibility decision in Weber and Saravia v. Germany (also cited in the judgment), which concerned an amended version of the same German legislation and was decided twelve years ago, in response to a complaint lodged in the year 2000.

6. As the Court noted in Klass and Others , the German legislation then at issue (the G 10) laid down a series of limitative conditions which had to be satisfied before a surveillance measure could be imposed. Thus, the permissible restrictive measures were confined to cases in which there were factual indications for suspecting a person of planning, committing or having committed certain serious criminal acts ; measures could only be ordered if the establishment of the facts by another method was without any prospect of success or considerably more difficult; even then, the surveillance could cover only the specific suspect or his presumed “contact-persons”. Thus, the Court observed, “ so-called exploratory or general surveillance [was] not permitted by the contested legislation ” (see Klass and Others , § 51).

7. In this regard, the RIPA section 8(4) regime which is at issue in the present case is different from that in Klass and Others in that the section 8(4) regime does encompass what the Court then referred to as “exploratory” surveillance and which in fact constitutes an essential and critical feature of this particular regime. Consequently, the scope and purpose of the surveillance regime now at issue is wider than that addressed in Klass and Others .

8. In Weber and Saravia , the complaint concerned a revised version, adopted in 1994, of the German G 10, whereby the scope of permissible surveillance was extended to cover the monitoring of international wireless telecommunications (see Weber and Saravia , § 88) in order to allow a “strategic surveillance” of such communications by means of catchwords. According to the Government’s submissions in that case, at the relevant time merely some ten per cent of all telecommunications were conducted by wireless means, and thus potentially subject to monitoring. In practice, monitoring was restricted to a limited number of foreign countries. The telephone connections of the State’s own (i.e. German) nationals living abroad could not be monitored directly. The identity of persons telecommunicating could only be uncovered in rare cases in which a catchword had been used (ibid., § 110).

9. The surveillance regime at issue in Weber and Saravia covered international wireless communications traffic, i.e. traffic transmitted via microwave or satellite, the latter operating through a survey of the downlink to Germany. Line-bound international communications were not subject to monitoring except where the risk of a war of aggression was concerned.

10. It is noteworthy that at the time of the surveillance regime which gave rise to the complaint in Weber and Saravia , strategic monitoring was mainly carried out on telephone, telex and fax communications. In those days, surveillance did not extend to email communications (see the judgment of the Federal Constitutional Court of 14 July 1999, 1BvR 2226/94, 1 BvR 2420/95, 1 BvR 2437/95, Rn 230, according to which, at the time of the hearing of the case in 1999, an expansion of strategic monitoring to email communications was only being planned for the future). One significant feature of communications by email, apart from the fact that nowadays they are so common, is that the identity of both the sender and recipient is usually directly available. Furthermore, many currently used means of communication or access to information through the Internet were only at embryonic stages at the time of the domestic complaint in Weber and Saravia .

(ii) The context of the present case

11. My point with the remarks above is to draw attention to the factual environment against the background of which those earlier cases were adjudicated, and the dramatic changes that have occurred since. The applicants have indeed referred to the technological “sea change” which has taken place.

12. What is important to note in this regard is that the technological “sea change” has had a twofold impact. On the one hand, technological developments have advanced the means by which surveillance of communications can be carried out. On the other hand, new technologies have revolutionised the ways in which people communicate, access, use and share information. That change is deeper than just a matter of volume. The digital age has in some respects transformed people’s lifestyles.

13. As a result of these changes, the potential exposure nowadays of a vast range of communications and other online activities to secret surveillance is far greater than before. In the wake of such developments, the potential risks of abuse arising from such surveillance have increased as well. Thus, the factual context in which “exploratory” or “strategic” secret surveillance operates is dramatically different from the circumstances that still prevailed a couple of decades ago, when the Weber and Saravia application was lodged, let alone four decades ago, when Klass and Others was decided. In the light of such changes, it is problematic and troubling to approach the question of the necessary safeguards against abuse simply by applying standards that were considered sufficient under significantly or even essentially different factual circumstances.

14. Furthermore, the “sea change” in terms of technologies and digitalised lifestyles is not the only development to be taken into consideration. The threats on account of which surveillance of communications is considered necessary have also changed. In this regard, too, the picture is twofold. One the one hand, for instance, there have been real and well-known aggravations in the risks of international terrorism. On the other, there is also increasing evidence of how various threats can be invoked, rightly or wrongly, in order to justify measures that entail restrictions on individual rights and freedoms. The notion of terrorism, for instance, may sometimes be used quite loosely and opportunistically in a desire to legitimise interferences with such rights and freedoms. Especially where secret surveillance is conducted in order to discover and explore broadly formulated threats such as those to national security or the nation’s economic well-being, the need for real safeguards through independent control and review is obvious.

15. There is yet another “sea change” calling for heightened attention in the assessment of the necessary standards in the context of secret surveillance of communications. It is the degradation of respect for democratic standards and the rule of law of which there is increasing evidence in a number of States. While I am not suggesting that the present respondent State is a case in point in this regard, the Convention standards must nevertheless be considered in the light of the fact that such developments testify to the actual or potential fragility of safeguards, institutional arrangements and the underlying assumptions that in ideal circumstances might appear adequate in order to minimise the risks of abuse. In fact, the same threats that are invoked to justify secret surveillance may also serve to reinforce tendencies toward a weakening of the checks and balances which underpin adherence to the rule of law and democratic governance.

(iii) Concerns

16. In line with the majority, I agree that the Contracting States must enjoy a wide margin of appreciation in determining whether the protection of national security requires the kind of surveillance of communications which is at issue in the present case (paragraph 314 of the present judgment). However, given the high risks of abuse, which at worst may undermine not only individual rights and freedoms but democracy and the rule of law more generally, the margin must be narrow when it comes to the necessary safeguards against abuse.

17. Under the impugned legislation, one of the striking features is that all of the supervisory powers entrusted to authorities with independence from the executive are of an ex post nature. Another striking feature is that not only are the general protective aims of the legislation very broadly framed, but also the specific authorisations (warrants and certificates) issued by the Secretary of State appear to be formulated in very broad and general terms (see paragraphs 156 and 342). Furthermore, the concrete search and selection criteria which are applied to filter intercepted communications for reading of their content are determined by the analysts conducting the surveillance (see paragraphs 157, 340 and 345-46 of the present judgment). As indicated by the domestic findings, the latter are not even subject to any meaningful subsequent oversight by independent bodies (see paragraphs 157 and 340).

18. Ever since Klass and Others , the Court has indeed held that in view of the risk that a system of secret surveillance for the protection of national security may undermine or even destroy democracy under the cloak of defending it, the Court must be satisfied that there exist adequate and effective guarantees against abuse (see Klass and Others , §§ 49-50). This assessment depends on all the circumstances of the case, such as the nature, scope and duration of the possible measures, the grounds required for ordering them, the authorities competent to authorise, carry out and supervise them, and the kind of remedy provided by the national law (ibid., § 50).

19. As discussed above, in the light of the changes in both the nature and scope of surveillance and in the prevailing factual realities, the circumstances have indeed evolved in such a way and to such an extent that I find it difficult to accept that the adequacy of safeguards should nevertheless be assessed simply by relying on the case-law that has arisen under different legal and factual framework conditions.

20. In particular, given the present overall context, I question the approach according to which prior independent control by a judicial authority should not be a necessary requirement in the system of safeguards.

21. Already in Klass and Others , when considering the initial stage of control, the Court stated that, in a field where abuse was potentially so easy in individual cases and could have such harmful consequences for democratic society as a whole, it was in principle desirable to entrust supervisory control to a judge (see Klass and Others , § 56). Under the G 10 legislation, judicial control was replaced by an initial control effected by an official qualified for judicial office and by the control provided by the Parliamentary Board and the G 10 Commission. In that case the Court concluded that, having regard to the nature of the supervisory and other safeguards provided for by the G 10, the exclusion of judicial control did not exceed the limits of what might be deemed necessary in a democratic society. The Court noted that the Parliamentary Board and the G 10 Commission were independent of the authorities carrying out the surveillance and vested with sufficient powers and competence to exercise an effective and continuous control. Furthermore, the democratic character was reflected in the balanced membership of the Parliamentary Board, on which the opposition was represented and was thus able to participate in the control of the measures ordered by the competent Minister, who was accountable to the Bundestag. The Court found that the two supervisory bodies could, in the circumstances of the case, be regarded as enjoying sufficient independence to give an objective ruling (ibid.).

22. As indicated above, in my view the legal and factual circumstances of that case, which go back four decades, cannot be considered comparable to the situation now under consideration. It is somewhat striking that in Weber , despite the important changes in the legislative and factual framework, the Court succinctly stated that it saw no reason to reconsider the conclusion in Klass and Others (see Weber and Saravia , § 117). In any event, in the light of the circumstances prevailing at the present time, such reconsideration seems to me to be indispensable.

23. Where, as in the present case, the interception (as a matter of technical necessity) encompasses vast volumes of communications traffic in an indiscriminate manner, without being linked to any kind of prior elements of suspicion related to the threats by reason of which the surveillance is conducted, everything in terms of the protection of individuals and their rights depends on whether and how the subsequent stages of the treatment of the intercepted communications provide effective and reliable safeguards for those rights, and against any abuse of the surveillance. Under such circumstances, given the potential intrusiveness of the surveillance and the abundant risks of abuse, I consider that it cannot be appropriate that all the ex ante safeguards remain in the hands of the executive. I think the applicants are right to argue that there is a need for an “updating” of the standards as regards prior independent judicial authorisation. It seems to me to be important that the authorities of the executive branch should be required to explain and justify before an independent judicial authority the grounds on which a particular surveillance should be authorised, and to account for the search criteria on the basis of which the intercepted communications will be filtered and selected for a review of their content.

24. In this respect, I am not convinced by the arguments advanced by the majority in support of the position that prior judicial control is unnecessary (paragraphs 318-20). The majority acknowledge that judicial authorisation is not inherently incompatible with the effective functioning of bulk interception (paragraph 318). Indeed, the recent case of Centrum för Rättvisa v. Sweden (cited above) offers an illustration, as it deals with Swedish legislation under which prior judicial authorisation is required.

25. The main argument against imposing such a requirement appears to be that it would not entail a sufficient safeguard, and that even in the absence of prior judicial authorisation the existence of independent oversight by the IPT and the Interception of Communications Commissioner provide adequate safeguards against abuse. In my view, it is obvious that prior judicial authorisation cannot in itself be sufficient and that further, robust safeguards such as those in place in the UK are indeed required. However, the fact that a given safeguard would not be sufficient is not enough to support a conclusion that it should not be considered necessary. In my opinion, it is quite essential to have in place an adequate system of safeguards, including controls exercised by independent bodies, both ex ante and ex post .

26. While the safeguards ex post that are provided for in the UK legislation and practice appear to set a good model in this domain, this does not in my view suffice to remedy the fact that the authorisation and implementation of the surveillance are wholly in the hands of the executive authorities, without any independent control ex ante. In this respect, the system of safeguards is even weaker than that considered by the Court in both Klass and Others and Weber and Saravia, in that under the German G 10 regime, although the surveillance was not subject to prior authorisation by a court, it had to be authorised by the G 10 Commission (see Weber and Saravia , cited above, § 115), which was not an executive branch body (ibid., § 25). Moreover, according to the judgment of the Federal Constitutional Court of 14 July 1999 (cited above, Rn 87), a list of search concepts was part of each restriction order, whereas in the present case it has transpired that the search and selection criteria are determined by the analysts operating the surveillance and are not subject to any prior supervision, nor any meaningful subsequent oversight (see paragraphs 157, 340 and 345-46 of the present judgment).

27. In sum, what we have before us now is a regime of secret surveillance, the reach of which under the prevailing factual circumstances is unprecedented, and under which a very wide operational latitude is left to the services operating the surveillance, without any independent ex ante control or constraint, and under which the search and selection criteria are not even ex post subject to any robust independent control. I find such a situation highly problematic. An independent ex ante control is all the more important because of the secret nature of the surveillance, which in practice reduces the possibility that individuals will have recourse to the safeguards available ex post .

28. I also consider that the remarks made by the majority in paragraph 319 of the judgment are not capable of supporting a conclusion according to which prior independent judicial authorisation should not be required. Rather, the argument that even judicial scrutiny may fail its function serves to underline the crucial importance which attaches to the requirement that such control must have effective guarantees of independence, in order to meet the proper standards of the necessary safeguards.

29. In short, while I agree with the conclusions set out in paragraph 387 of the judgment, I do not consider those shortcomings to be the only ones that justify a finding of a violation of Article 8 in the present case. In particular, taking into account the present legal and factual context, I do not believe that the necessary safeguards in the circumstances of surveillance based on the bulk interception of communications can be sufficient without including an independent ex ante judicial control. The position according to which prior judicial control of authorisations for secret surveillance of communications was a desirable but not a necessary safeguard stems from Klass and Others which, firstly, concerned a more limited surveillance regime than the one now at issue and did not permit “exploratory surveillance” at all, and which, secondly, was decided four decades ago against the backdrop of factual circumstances that in many relevant respects were different from those prevailing today. That position was later, in Weber and Saravia , carried over to a surveillance regime which did have more similarities with the RIPA section 8(4) regime but nevertheless operated in conditions very different from those prevailing in the modern digitalised societies. For the reasons outlined above, that position should, in my view, no longer be maintained by the Court.

II. The intelligence-sharing regime

30. It is easy to agree with the principle that any arrangement under which intelligence from intercepted communications is obtained via foreign intelligence services, whether on the basis of requests to carry out such interception or to convey its results, should not be allowed to entail a circumvention of the safeguards which must be in place for any surveillance by domestic authorities (see paragraphs 216, 423 and 447). Indeed, any other approach would be implausible.

31. On this basis I consider, in sum, that the shortcomings referred to above in the context of the section 8(4) regime also attach to the intelligence-sharing regime (see paragraphs 109 and 428-29). I therefore conclude that the safeguards have not been adequate and that there has been a violation of Article 8 in respect of this regime also.