Ten things to read about today’s data retention decision

I am a fair-weather blogger, and so I cannot remember the last time I had so many visits or retweets in a day.  Piggybacking on the unexpected traffic boost, here are ten things worth reading (from various sources) about the reason for that traffic – the finding by the Court of Justice of the EU that the Data Retention Directive is, on human rights grounds, invalid.  (My own post, Data retention parrot, is here).

I had plenty to choose from in putting this list together – fortunate that the decision was published when many of us legal academics are not teaching?

  1. The decision of the Court.  The early pages are taken up with reproducing the provisions of the legislation, so if you are familiar with the Directive, those pages are most skippable.
  2. Fiona de Londras, professor at Durham Law School, writing at Human Rights in Ireland. Special mention: discussion on whether “a more tailored, narrower approach” might survive scrutiny if the Directive is to be replaced (see also her lessons for the US, posted at The Conversation).
  3. “Cybermatron”, an expert in this field, writing on her blog. Special mention: highlighting weaknesses in the decision, including where the Court may have underappreciated the significance of the legislation and of this challenge.
  4. Steve Peers, professor at the University of Essex Law School, writing on his blog EU Law Analysis. Special mention: analysis of the current status of the (invalid) Directive, and options for states and the EU from this point on.
  5. Paul Bernal, lecturer at the UEA Law School, writing on his blog. Special mention: how the decision sits within the wider debate on and advocacy for privacy.
  6. Karlin Lillington, journalist, writing in the Irish Times. Special mention: the consequences for Ireland and the EU, by someone who has been instrumental in highlighting data retention practices for over a decade.
  7. Luke Scanlon, solicitor, Pinsent Masons, writing on Out-law. Special mention: impact on other legislation, including data protection present and future.
  8. Glyn Moody, author and journalist, writing for ComputerWorld UK. Special mention: explanation, point by point, of how the court’s decision relates to specific data retention practices.
  9. Gabriele Steinhauser, journalist, writing in the Wall Street Journal. Special mention: how the decision is being reported to an international audience, including the political dimension.
  10. Press release and FAQ on the decision from the European Commission (the ‘losing’ side, not that you would know that from the statement). Special mention: reading it with a straight face.

Apologies to those omitted – additional links welcome, through the comments sections below.

The data retention parrot

One of the most-read posts on this site is a 2009 set of ten questions about data retention legislation in Ireland. It was written with a mixture of anger and detail. Today’s post contains neither. Instead, it’s relieved – but hurried.

This morning, the Court of Justice of the European Union (CJEU) ruled in a set of cases regarding the validity, from a human rights point of view, of the Data Retention Directive (which provides for the retention by service providers of phone and Internet communications data across the EU for set periods, for the purpose of subsequent access by public authorities). Here’s the decision as posted on Scribd; official link to follow. Cases C-293/12 and C-594/12.

The Advocate General had already given his Opinion in late 2013, which was in some respects very critical of the Directive, but his recommendations were also a bit limited.  Of the cases that the CJEU heard, the one I know best (unsurprisingly) is the challenge made in Ireland by Digital Rights Ireland (High Court decision of 2010). This, and other cases starting in Austria, were sent to the EU court for a ruling on points of EU law.

Here are my first-look highlights from today’s decision.

1. The Directive raises serious issues of compatibility with the fundamental rights protected under EU law (privacy and data protection) – and it is not proportionate, and therefore invalid. This was clearly flagged by the Advocate General and will be the big headline today, rightly.  I’m just going to add some more observations, but the big result shouldn’t be ignored!

2. On the other hand, the proposal of the Advocate General (that the effect of declaring it invalid be suspended to allow better legislation to be introduced; paras 154-158 of his Opinion) has been entirely ignored in the decision, and only alluded to in a footnote in the accompanying press release. If I’m reading it right, this idea has simply disappeared.  The Directive is dead and, legally speaking, should never have existed.

3. There are important warning signs to the European bodies for the (inevitable) attempt to draft a replacement. Because of the nature of the rights and the infringements, discretion of the legislative bodies “is reduced, with the result that review of that discretion should be strict” (paras 47-8). Shroud-waving should also be avoided; “the fight against serious crime, in particular against organised crime and terrorism, is indeed of the utmost importance in order to ensure public security and its effectiveness may depend to a great extent on the use of modern investigation techniques. However, such an objective of general interest, however fundamental it may be, does not, in itself, justify” a retention measure such as this one (para 51). There are a range of specific criticisms outlined from para 58 onwards that would surely be relevant, e.g. application to the whole population, temporal or geographic restrictions, lack of a definition of serious crime, inadequate limits on access/use, a retention period plucked out of the air. Export outside the EU (topical!) is also highlighted at para 68.

4. Although it wasn’t necessary to rely on it to reach today’s result (see paras 69-70) , the CJEU makes some very important comments about the relationship between surveillance and speech:

In such circumstances, even though, as is apparent from Article 1(2) and Article 5(2) of Directive 2006/24, the directive does not permit the retention of the content of the communication or of information consulted using an electronic communications network, it is not inconceivable that the retention of the data in question might have an effect on the use, by subscribers or registered users, of the means of communication covered by that directive and, consequently, on their exercise of the freedom of expression guaranteed by Article 11 of the Charter. (para 28)

(Bonus points for channelling Vizzini)

5. The Court makes significant use of the ECtHR’s decision in S & Marper v UK (about DNA databases) – three separate references, all ‘by analogy’ regarding article 8 ECHR. The significance of S was clear at the time and today’s opinion demonstrates how it valuable it is in terms of analysing questions of law and technology – especially chilling and cumulative effects.  It’s also further evidence of the way that the CJEU builds on ECtHR rulings.

6. The Court endorses the Advocate General’s point about perception. It’s not a point unknown to those in the field (especially through the jurisprudence of the German courts and others), but it’s still not fully grasped in the UK and Ireland; data retention of this nature is “likely to generate in the minds of the persons concerned the feeling that their private lives are the subject of constant surveillance” (para 37). (Which, for the record, is a bad thing).

Those are some first thoughts, and are really an extension of even earlier thoughts posted on Twitter. More later if I can!

Data Retention Ireland

The Minister for Justice in Ireland published the Communications (Retention of Data) Bill last week: it was made available on the Oireachtas website (and brought to my attention by the ever-helpful Darius Whelan), although curiously, some reputable (and normally reliable) newspapers wrote on Monday morning about the legislation being due to be published! It will presumably be debated in the Oireachtas (parliament) when its honourable members return after the summer. Data retention legislation requires service providers to keep certain types of data on the activities of their subscribers and users, and to disclose it to relevant authorities on request. I hope that this post is of interest to Irish and non-Irish audiences, though, as the issues are arising in many jurisdictions, whether through the EU’s data retention directive of 2006 or independently. I also point to this extremely helpful status report on transposition as of January 2009: it shows very clearly that many states have included both judicial authorisation and cost recovery, which are absent from the Irish proposals.

The publication of the Bill isn’t a major surprise. A draft had been leaked, and of course this is but the Irish implementation of the 2006 Directive – so we cannot blame the Irish government alone for bringing forward these proposals. The underlying Directive remains an unconvincing one. I am not opposed to all attempts to use new forms of communication in conjunction with crime prevention, detection and prosecution. Nor am I unsympathetic to the way that some in law enforcement will feel that they are falling behind those who they pursue in terms of the use of technology. But data retention carries with it a financial burden, an administrative nightmare and, most importantly, a shift in the balance between the citizen and the state that may be presumed to be irreversable: surveillance powers, once granted, are rarely rolled back. These are broad powers, requiring retention of everyone’s data even if those having data disclosed are a subset of this (rather than the alternative of notifying a service provider to retain data on a given subject for a limited, specific purpose). As is so often the case, specific information from law enforcement on the problems with existing legislation has not been forthcoming, and public statements focus on the most extreme of cases (the Irish Minister for Justice gave us international terrorism and child pornography in his public comments today). Anyway, to ten questions that occur to me after giving the Bill some consideration. (Edit: Ronan Lupton has responded to the questions with some very useful points, including his knowledge of how the Bill is less extensive than the versions previously proposed – a fair point, but read his full comments here).

(1) We are reassured that the legislation, as with the Directive, doesn’t apply to ‘content’, but getting information on who you are communicating with and (particularly in the case of mobile telephony) where you have been over the course of two years is more than trivial – it is a very intrusive way of finding out what a person (unconvicted of any crime) has been doing in their private life. How is this acceptable?

(2) The proposals follow in the disreputable tradition of sidelining the judicial branch – making the powers in essence a general authority for digital search and surveillance operations without a warrant. Nothing in EU law requires that the powers of accessing data be exercisable by senior Gardai (not to mention principal officers in the Revenue Commissioners, a new addition to the Bill that was not part of the earlier draft) – although it does appear tighter than the UK version, which appears to let anyone with a tanard or a lanyard to make a request. There are some safeguards supposedly in place (annual statistical reporting, a judge with the job of monitoring the system), but we’ve seen that they are quite weak: see for example TJ McIntyre’s recent discussion of the current judicial ‘oversight’ of phone intercept and data retention legislation. Furthermore, the officer authorising the access to data merely has to be satisfied that it is required for preventing, detecting, investigating or prosecuting a serious offence – which, for example, carries no need for reasonable suspicion of criminal behaviour on the part of the person whose data is being disclosed. It’s a dragnet-style provision that gives powers to police, Army and revenue officials and enables them to carry out large-scale investigations without any disclosure of such to the affected individuals nor any effective right of appeal or transparency. Why could this system not be restricted to cases approved by an independent judge after specific evidence of necessity is presented by the requesting officer?

(3) Data retention remains doubtful in terms of fundamental rights compliance: in the ECHR, S & Marper v UK questions mass monitoring of the unconvicted, Copland v UK reiterates that traffic data is covered by Article 8 (as I argue here); the German courts are considering various challenges (summarised by Digital Rights Ireland: 1 | 2), and DRI itself is engaged in a challenge to the Directive. The prior case brought by Ireland against the Directive related purely to legal basis and did not address fundamental rights at any stage. Does this legislation comply with the high standards of the protection of fundamental rights that Ireland aspires to meet?

(4) Under the Directive, retention is required for between six months and two years. The UK provisions (SI 2009/859) require a standard 12 month period. The Irish proposals would require it for a year for Internet and two years for telephone. Supporters of the legislation are spinning this as a reduction from the existing (and supposedly stopgap) three year period under 2005 legislation, conveniently neglecting the requirement under EU law to reduce it to a maximum of 2 years in any event. Why is a 2-year period necessary, particularly where other implementing States are able to adopt shorter periods?

(5) No information is provided in the Bill, explanatory memorandum or press release on who will bear the costs of retention. Compare this with, for example, the UK regulations which at least empower the Home Secretary to reimburse ‘any expenses incurred’ (which are well into the millions) in complying with the regulations. Bear in mind, too, that while some providers will keep billing data for obvious reasons, this is not the case for all providers. Who will pick up the bill and why has it not been ‘costed’ in a published impact assessment?

(6) The Bill applies without more to all providers of publicly available electronic communications networks and publicly available electronic communications services. These are wide (and imprecise) definitions that, given that specific statutory obligations are created (‘a service provider shall retain’), causes doubt for many (webmail? webmail-like? open wifi? voice IM?). This will cause panic and confusion across the sector and will have seriously damaging consequences for Ireland’s ability to promote itself as a destination for high-tech industries. Compare with s 10 of the UK regulations, which provide that the obligation is only activated when the Home Secretary notifies the provider (although the Secretary does have a statutory duty to notify all relevant providers!) Why does the Government wish to create new duties without precision on who the duties will affect?

(7) There is a ‘redundancy’ provision in the UK regulations (again s 10), which states that the Home Secretary doesn’t have to notify providers where the data is retained by another provider. Presumably, this protects downstream ISPs and similarly situated others. There is no such provision in the Irish legislation and the clear terms would require the same data to be collected at multiple locations. Why are the supporters of data retention so generous with the time, money and effort of others?

(8) The detailed instructions (Sch 2, Part 1, 5(d)) requires retention of the date, time and (cell ID) location of the activation of a ‘pre-paid anonymous (mobile telephony) service’. Is this the end of pay-as-you-go anonymity through the back door?

(9) The definition of ‘serious offences’ is broad (although it is an improvement on the draft, which would have allowed the powers to be used for any offence with a 12-month sentence attached to it). Any offence carrying a five-year sentence along with selected other offences (from poisoning to the false reporting of child abuse) count. How were these offences selected and what is the basis for their inclusion?

(10) The complaints procedure under s 10 of the Irish bill is bizarre – you can find out if a disclosure request has been made about you by making a request (if you believe that your data has been disclosed!!), but you will only be told if it has been made if it turns out that the rules have been contravened. Translation: meaningless. And there’s a broad barring of legal action other than the required constitutional right of action. And ‘a decision of the (referee who deals with complaints) … is final’. And evidence obtained in violation of the statute is not automatically excluded, as it should be. Given the argument that those with nothing to fear have nothing to hide, why does the Government fear challenges so much as to bar them?

Read more, from Cearta (Dr. Eoin O’Dell, who also spoke on the subject on RTE news), and of course Digital Rights Ireland. I’m sure there will be more.