Guest post: Data Retention: I can’t believe it’s not lawful, can you? A response to Anthony Speaight QC

Guest post by Matthew White

Introduction:

Ladies and gentlemen, Bagginses and Boffins. Tooks and Brandybucks. Grubbs! Chubbs! Hornblowers! Bolgers! Bracegirdles! Proudfoots. Put your butter away for I am about to respond, rebut, rebuke and more to a recent blog post for Judicial Power Project, by Anthony Speaight QC on data retention.

Blanket data retention is unlawful, please deal with it:

Speaight starts off by referring to the recent Court of Appeal (CoA) judgment in  Tom Watson and Others v Secretary of State for the Home Department [2018] EWCA Civ 70 and how the Court of Justice of the European Union (CJEU) has created problems and uncertainties with regards to data retention. As David Allen Green would say, ‘Well…’ Well, just to be clear, the position of the CJEU on blanket indiscriminate data retention is crystal clear. It . Is . Unlawful . It just happens that the CoA took the position of sticking their fingers in their ears and pretending that the CJEU’s ruling doesn’t apply to UK law, because its somehow (it’s not) different.

Just billing data is retained? Oh really?

Next, Speaight recaps the data retention saga so far, in that telecommunications companies have always recorded who uses their services, when and where, often for billing purposes. A long time ago, in a galaxy far, far away (a few years ago, and anywhere with an internet connection) this position was a robust one. But the European Commission (Commission) in 2011 highlighted that:

[T]rends in business models and service offerings, such as the growth in flat rate tariffs, pre-paid and free electronic communications services, meant that operators gradually stopped storing traffic and location data for billing purposes thus reducing the availability of such data for criminal justice and law enforcement purposes.

So, in a nutshell, data for billing purposes are on the decrease. This would explain why the Data Retention Directive (DRD) (discussed more below) affected:

[P]roviders of electronic communication services by requiring such providers to retain large amounts of traffic and location data, instead of retaining only data necessary for billing purposes; this shift in priority results in an increase in costs to retain and secure the data.

So, it’s simply untrue to refer to just billing data when talking about data retention, because this isn’t the only data that is or has ever been sought.

It’s the Islamists fault why we have data retention:

Speaight next points out that it was the advent of Islamist international terrorism that made it advantageous to place data retention obligations on companies. Oh really? Are we going down this route? Well….. demands for data retention can be traced back to the ‘International Law Enforcement and Telecommunications Seminars’ (ILETS) (6) and in its 1999 report, it was realised that Directive 97/66/EC (the old ePrivacy Directive) which made retention of communications data possible only for billing purposes was a problem. The report sought to ‘consider options for improving the retention of data by Communication Service Providers.’ Improve? Ha. Notice how 1999 was before 9/11? Funny that.

It doesn’t stop there though. A year later (still before 9/11), the UK’s National Crime and Intelligence Service (NCIS) made a submission (on behalf of the Mi5/6, GCHQ etc) to the Home Office on data retention laws. They ironically argued that a targeted approach would be a greater infringement on personal privacy (para 3.1.5). Of course, they didn’t say how or why this was the case, because, reasons. Charles Clarke, the then junior Home Office Minister, and Patricia Hewitt, an ‘E-Minister’ both made the claim such proposals would never happen (Judith Rauhofer, ‘Just Because You’re Paranoid, Doesn’t Mean They’re Not After You: Legislative Developments in Relation to the Retention of Communications Data’ (2006) SCRIPTed 3, 228; Patricia Hewitt and Charles Clarke, Joint letter to Independent on Sunday, 28 Jan 2000) and should not be implemented (Trade and Industry Committee, UK Online Reviewed: the First Annual Report of the E-Minister and E-Envoy Report (HC 66 1999-2000), Q93).

Guess what? A year later Part 11 of the Anti-terrorism, Crime and Security Act 2001 (ATCSA 2001) came into force three months after 9/11 (Judith Rauhofer, 331). The Earl of Northesk, however, pointed out that ‘there is no evidence whatever that a lack of data retained has proved an impediment to the investigation of the atrocities’ on 9/11 (HL Deb 4 Dec vol 629 col. 808-9). What this demonstrates is that data retention was always on the cards, even when its utility wasn’t proven, where the then Prime Minister Tony Blair, noted that ‘all the surveillance in the world’ could not have prevented the 7/7 bombings. It’s just that as Roger Clarke succinctly puts it:

“[M]ost critical driver of change, however, has been the dominance of national security extremism since the 2001 terrorist attacks in the USA, and the preparedness of parliaments in many countries to grant law enforcement agencies any request that they can somehow link to the idea of counter-terrorism.” (Roger Clarke, ‘Data retention as mass surveillance: the need for an evaluative framework’ (2015) International Data Privacy Law 5:2 121, 122).

Islamic terrorism was just fresh justification (7,9) for something that ‘the EU governments always intended to introduce an EC law to bind all member states to adopt data retention.’ Mandatory data retention was championed by the UK during its Presidency of the European Council (Council) (9) (and yes, that includes the ‘no data retention from us’ Charles Clarke (who was accused of threatening the European Parliament to agree to data retention (9))) and described as a master class in diplomacy and political manoeuvring (Judith Rauhofer, 341) (and they say it’s the EU that tells us what to do!!). Politicians goin’ politicate. Yes, the DRD makes reference to the Madrid bombings, but the DRD was not limited to combating terrorism (6), just as the reasons for accessing communications data in UK law under s.22 of the Regulation of Investigatory Powers Act 2000 (RIPA 2000) were not solely based on fighting terrorism. There is nothing wrong with saying that data retention (yeah, but not blanket, of course) and access to said data can be important in the fight against Islamist terrorism, but would you please stop pretending that was the basis on which data retention was sought?

Data retention was smooth like rocks:

Next, Speaight points to the ‘smooth operation’ of the data retention system. Smooth how and in what ways? Harder to answer that is, yess! Well….. in 2010, the Article 29 Working Party (WP29) pointed out that ‘the lack of available sensible statistics hinders the assessment of whether the [data retention] directive has achieved its objectives.’ The WP29 went further pointing out that there was a lack of harmonisation in national implementation of the DRD (2). This was, the purpose of the DRD (harmonising data retention across the EU), and it didn’t even achieve what it set out.

What about its true purpose? You know, spying on every EU citizen? Well the European Data Protection Supervisor (EDPS) responded to the Commission’s evaluation of the DRD. WARNING: EDPS pulls no punches. First, the EDPS reiterated that the DRD was based upon the assumption of necessity (para 38). Secondly, the EDPS criticised the Commission’s assertion that most Member States considered data retention a necessary tool when conclusions were based on just over a third (that’s less than half, right?) of them (para 40). Thirdly, these conclusions were in fact, only statements (para 41). Fourthly, the EDPS highlighted there should be sufficient quantitative and qualitative information to assess whether the DRD is actually working and whether less privacy intrusive measures could achieve the same result, information should show the relationship between use and result (43).

Surprise, surprise, the EDPS didn’t find sufficient evidence to demonstrate the necessity of the DRD and that further investigations into alternatives should commence (para 44). Fifthly, the EDPS pretty much savaged the quantitative and qualitative information available (para 45-52). A few years later, the CJEU asked for proof of the necessity of the DRD. There was a lack of statistical evidence from EU Member States, the Commission, the Council and European Parliament, and despite that, they had the cheek to ask the CJEU to reject the complaints made by Digital Rights Ireland and others anyway (ibid). Only the Austrian government were able to provide statistical evidence on the use (not retention) of communications data which didn’t involve any cases of terrorism (ibid). The UK’s representatives admitted (come again? The UK admits something?) there was no ‘scientific data’ to underpin the need of data retention (ibid), so the question begs, wtaf had the DRD been based upon? Was it the assumption of necessity the EDPS referred to? Draw your own conclusions. The moral of the story is that the DRD did not operate smoothly.

Ruling against data retention was a surprise?

Speaight then moves onto the judgment that started it all, Joined Cases C‑293/12 and C‑594/12, Digital Rights Ireland in which the CJEU invalidated the DRD across the EU. According to Speaight, this came as a ‘surprise.’

I felt a great disturbance in the Law, as if thousands of spies, police, other public authorities, politicians and lawyers suddenly cried out in terror, as the State were suddenly unable to spy anymore. I fear something terrible has happened.

So, who was surprised? Was it the European Parliament who had initially opposed this form of data retention as they urged its use must be entirely exceptional, based on specific comprehensible law, authorised by judicial or other competent authorities for individual cases and be consistent with the European Convention on Human Rights (ECHR)? Was it a surprise to them when they also noted that that ‘a general data retention principle must be forbidden’ and that ‘any general obligation concerning data retention’ is contrary to the proportionality principle’ (Abu Bakar Munir and Siti Hajar Mohd Yasin, ‘Retention of communications data: A bumpy road ahead’ (2004) The John Marshall Journal of Computer & Information Law 22:4 731, 734; Clive Walker and Yaman Akdeniz, ‘Anti-Terrorism Laws and Data Retention: War is over?’ (2003) Northern Ireland Legal Quarterly 54:2 159, 167)?

Was it a surprise to Patrick Breyer who argued that data retention was incompatible with Articles 8 and 10 of the ECHR back in 2005 (372, 374, 375)? Was it a surprise to Mariuca Morariu who argued that the DRD had failed to demonstrate its necessity (Mariuca Morariu, ‘How Secure is to Remain Private? On the Controversies of the European Data Retention Directive’ Amsterdam Social Science 1:2 46, 54-9)? Was it a surprise to Privacy International (PI), the European Digital Rights Initiative (EDRi), 90 NGOs and 80 telecommunications service providers (9) who were against the DRD? Was it a surprise to the 40 civil liberties organisations who urged the European Parliament to vote against the retention of communications data?

Was it a surprise to the WP29, the European Data Protection Commissioners, the International Chamber of Commerce (ICC), European Internet Services Providers Association (EuroISPA), the US Internet Service Provider Association (USISPA), the All Party Internet Group (APIG) (Abu Bakar Munir and Siti Hajar Mohd Yasin, 746-749) and those at the G8 Tokyo Conference? Hell, even our own assistant Information Commissioner, Jonathan Bamford, back in 2001 wouldn’t be surprised because he said ‘Part 11 isn’t necessary, and if it is necessary it should be made clear why’ (HL Deb 27 Nov 2001 vol 629 cc183-290, 252). Was it a surprise when prior to Digital Rights Ireland:

Bulgaria’s Supreme Administrative Court, the Romanian, German Federal, Czech Republic Constitutional Courts and the Supreme Court of Cyprus all [declared] national implementation of the DRD either invalid or unconstitutional (in some or all regards) and incompatible with Article 8 ECHR?

Was Jules Winnfield surprised?

The point I’m trying to hammer home is that (you’ve guessed it), the CJEU’s ruling in Digital Rights Ireland should come as no surprise. Still on the issue of surprise, for Speaight it was because it departed from decisions of the European Court of Human Rights (ECtHR) and the CJEU itself. Ok, let’s look at these ECtHR cases Speaight refers to. The first is Weber and Saravia v Germany, a case on ‘strategic monitoring.’ This is a whole different kettle of fish when compared to the DRD as this concerned the surveillance of 10% (I’m not saying this is cool either btw) [30, 110] of German telecommunications, not the surveillance of ‘practically the entire European population’ [56]. Ok, that may have been an exaggeration by the CJEU as there are only 28 (we’re not so sure about one though) EU Member States, but the point is, the powers in question are not comparable. The DRD was confined to serious crime, without even defining it [61]. Whereas German law in Weber concerned six defined purposes for strategic monitoring, [27] and could only be triggered through catch words [32]. In Digital Rights Ireland, authorisation for access to communications data in the DRD was not dependent upon ‘prior review carried out by a court or by an independent administrative body’ [62] where in Weber this was the case [21, 25]. Apples and oranges.

The second ECtHR case was Kennedy v UK, and it’s funny that this case is brought up. The ECtHR in this case referred to a previous case, Liberty v UK in which the virtually unfettered power of capturing external communications [64] violated Article 8 of the ECHR [70]. The ECtHR in Kennedy referred to this as an indiscriminate power [160, 162] (bit like data retention huh?), and the UK only succeeded in Kennedy because the ECtHR were acting upon the assumption that interception warrants only related to one person [160, 162]. Of course, the ECtHR didn’t know that ‘person’ for the purposes of RIPA 2000 meant ‘any organisation and any association or combination of persons,’ so you know, not one person literally.

And this was, of course, prior to Edward Snowden’s bombshell of surveillance revelations, which triggered further proceedings by Big Brother Watch. A couple of years ago, in Roman Zakharov v Russia, the ECtHR’s Grand Chamber (GC) ruled that surveillance measures that are ‘ordered haphazardly, irregularly or without due and proper consideration’ [267] violates Article 8 [305]. That is because the automatic storage of clearly irrelevant data would contravene Article 8 [255]. This coincides with Advocate General (AG) Saugmandsgaard Øe’s opinion that the ‘disadvantages of general data retention obligations arise from the fact that the vast majority of the data retained will relate to persons who will never be connected in any way with serious crime’ [252]. That’s a lot of irrelevant data if you ask me. Judge Pinto de Albuquerque, in his concurring opinion in Szabo and Vissy v Hungary regards Zakharov as a rebuke of the ‘widespread, non-(reasonable) suspicion-based, “strategic surveillance” for the purposes of national security’ [35]. So, I’d say that even Weber v Saravia is put into doubt. And so, even if the CJEU rules that data retention in the national security context is outside its competence, there is enough ECtHR case law to bite the UK on its arse.

Probably the most important ECtHR case not mentioned by Speaight (why is that?) is that of S and Marper v UK, this is the data retention case. Although this concerned DNA data retention, the ECtHR’s concerns ‘have clear applications to the detailed information revealed about individuals’ private lives by communications data.’ What did the GC rule in S and Marper? Oh, was it that blanket indiscriminate data retention ‘even on a specific group of individuals (suspects and convicts) violated Article 8’? Yes, they did and it was S and Marper to which the CJEU referred to on three separate occasions in Digital Rights Ireland [47, 54-5]. Tele 2 and Watson (where the CJEU reconfirmed that blanket indiscriminate data retention is prohibited under EU law) is just the next logical step with regards to communications data. And so far from being surprising, the CJEU in Digital Rights Ireland and Tele2 and Watson are acting in a manner that is consistent with the case law of the ECtHR.

The CJEU case law that Speaight refers to is Ireland v Parliament and Council which was a challenge to the DRD’s legal basis, not whether it was compatible with the Charter of Fundamental Rights, so I’m not entirely sure what Speaight is trying to get at. All in all, Speaight hasn’t shown anything to demonstrate that Digital Rights Ireland has departed from ECtHR or CJEU case law.

You forgot to say the UK extended data retention laws:

Speaight then rightly acknowledges how the UK government replaced UK law implementing the DRD with the Data Retention and Investigatory Powers Act 2014 (DRIPA 2014) in lightspeed fashion. What Speaight omits, however, is that DRIPA 2014 extended retention obligations from telephone companies and Internet Service Providers (ISPs) to Over-The-Top (OTT) services such as Skype, Twitter, Google, Facebook etc. James Brokenshire MP attested that DRIPA 2014 was introduced to clarify what was always covered by the definition of telecommunications services (HC Deb 14 July, vol 584, 786). This, of course, was total bullshit (5), but like I said, politicians goin’ politicate.

Claimants don’t ask questions, courts do:

Speaight moves onto the challenges to DRIPA 2014, we know the story already, the High Court (HC) said it was inconsistent with Digital Rights Ireland, whereas the CoA disagreed, blah, blah. Speaight points out that the claimants had no issue with data retention in principle, which is true, but so what? Speaight also points out that the CJEU went further than what the claimants asked by ruling that blanket indiscriminate data retention was not permissible under EU law. Wait, what the fark? It’s not the bloody claimants’ that ask the CJEU a question on the interpretation of EU law as I’m pretty sure it was the Swedish referring court (via Article 267 of the Treaty on the Functioning of the EU, you know, a preliminary reference) that asked the CJEU:

Is a general obligation to retain traffic data covering all persons, all means of electronic communication and all traffic data without any distinctions, limitations or exceptions for the purpose of combating crime (as described [below under points 1-6]) compatible with Article 15(1) of Directive 2002/58/EC, 1 taking account of Articles 7, 8 and 15(1) of the Charter?

And the CJEU said no. End of discussion.

The ends don’t always justify the means and for clarity, the CJEU didn’t reject shit:

Speaight also says that the CJEU in Tele2 and Watson rejected AG Saugmandsgaard Øe’s advice that the French governments found access to communications data useful in its investigations into terrorist attacks in 2015. Such a position however, falls victim to several questions, such as under what circumstances was the data sought? Was it accessed as a consequence of the legal obligation to retain? Or was it already retained for business purposes? What were the results of the use of that data? Could the same results have been achieved using less intrusive means? Saying it is useful tells us nothing as the ECtHR has plainly said necessity (in a democratic society) is not as flexible as expressions such as ‘useful’ [48], and as the CJEU rightly noted, a measure in and of itself, even in the general interest cannot justify general indiscriminate data retention [103]. This demonstrates that the CJEU didn’t reject anything, they didn’t even refer to the French government’s evidence, they just said as fundamental as fighting serious crime may be, and the measures employed, cannot by themselves justify such a fundamental departure from the protection of human rights. Just because you can, doesn’t mean you should. A certain ECtHR said something similar in Klass v Germany in that States ‘may not, in the name of the struggle against espionage and terrorism, adopt whatever measures they deem appropriate’ [49].

The CJEU doesn’t have to answer what it wasn’t asked:

Speaight then whines about the CJEU not addressing the issue of national security, well they weren’t asked about national security in Tele2 and Watson, were they? Like I said, even if the CJEU doesn’t have competence to rule on national security based data retention, Roman Zakharov is watching you from Strasbourg (he’s not actually in Strasbourg, I don’t think, but you dig).

What’s your problem with notification?

Speaight also bemoans the obligation to notify saying this requirement could damage investigations and surveillance and went beyond what the claimants had asked. Well, again, the claimants weren’t asking the questions, ffs, and the CJEU made this point by referring to previous case law, notably, Schrems [95]. The CJEU made very clear that notification should be done ‘as soon as that notification is no longer liable to jeopardise the investigations being undertaken by those authorities’ [121]. This is consistent with the ECtHR’s stance. Both courts are aware that notification can defeat the purpose of the investigation, and sometimes even after it has concluded, notification may still not be appropriate. But Speaight seems to omit this crucial detail.

Lawyers getting mad:

Speaight notes that criticism of Tele2 is not confined to Eurosceptics. Sure, but you don’t have to be a Europhile to defend it either. He also noted that it was roundly condemned by all the participants at a meeting of the Society of Conservative Lawyers. Well, no shit to my Sherlock, the name kinda gave it away. He also notes that the former Independent Reviewer of Terror law, David Anderson QC, said it was the worst judgment he knew of. Wait til Anderson reads the ECtHR’s case law on this matter then, which if anything, on proper reading goes further than Tele2. Speaight also points out that Demonic Grieve QC MP was pissed and that a well distinguished member of the French Bar, Francois-Henri Briard basically saying we need more conservative judges to trample on fundamental rights. If a judgment that protects the fundamental rights of all EU citizens pisses off a few lawyers, so be it.

Conclusions:

I’ve spent way too much time on Speaight’s post, and the really sad thing is, I’ve enjoyed it. It’s hard to have a conversation about data retention when you first have to sift through a load of bollocks, and there was plenty of bollocks, just to make your point. And by the time you’ve cleared through all the falsities and misleading or exaggerated points, you run close to 4k words without actually saying what your position is. So, my position for this blog post is, we should always shoot down rubbish when it shows its ugly face or else it festers. Actually, the point is, I can believe that blanket indiscriminate data retention is unlawful.