Biometric data and data protection law: the CJEU loses the plot

 

Steve Peers

Many people are increasingly concerned about adequate protection of their biometric data. To this end, the proposed EU data protection Regulation would classify that data as sensitive data, ensuring an extra degree of protection for it. But in the meantime, before that proposal is adopted, there are other EU measures which regulate the issue. Unfortunately, yesterday’s judgment of the CJEU in Willems and others does an inadequate job, with great respect, in applying the current EU rules to such data.

Background

The Willems judgment concerns biometric data collected for passports, as provided for in an EU Regulation of 2004, as amended in 2009. In fact, the CJEU has ruled on this Regulation several times before. In UK v Council, it (unconvincingly) ruled that the UK could not participate in the Regulation, since it was closely linked to the parts of Schengen rules (the abolition of internal border controls) in which the UK didn’t participate. In Schwarz, it ruled that the Regulation was valid from two different angles, as it was correctly adopted using the ‘legal base’ allowing the EU to adopt measures on external border control, and the interference which it entailed with the right to privacy was justified by the interest in ensuring the identity of passport holders and the validity of the passport. Finally, the Court recently ruled on the privacy aspects of displaying names in passports (as discussed here).

Building on these judgments, the national court in Willems had two questions. First of all, did the Regulation apply to some types of identity cards, given that they can in effect be used as passports for travel within the EU? Secondly, the national court asked the CJEU to interpret the data protection rules applicable to the further use of biometric data after it was collected for the purposes of passports. The latter question stemmed from the concern of the litigants in this case that their biometric data would be stored on a centralised database with inadequate security, which would be used for other purposes without a clear identification of who would have access to it.

More precisely, the national court’s second question was whether ‘Article 4(3) of [the passport Regulation, read] in light of Articles 7 and 8 of the Charter of Fundamental Rights of the [EU], Article 8(2) of the [ECHR] and Article 7(f) of [the current data protection Directive], read in conjunction with Article 6(1)(b) of that Directive’, required a guarantee that when collecting biometric data under the Regulation, Member States had to apply a ‘purpose limitation’ rule that such data  could only be used for the original purpose for which the passport was issued.

Judgment

On the first question, the CJEU looked at the wording of the Regulation, which specified that it did not apply to ‘identity cards issued to [Member States’] nationals or to temporary passports and travel documents having a validity of 12 months or less’. The Court ruled that the words ‘having a validity of 12 months or less’ only set out the scope of the Regulation as regards ‘temporary passports and travel documents’, meaning that such documents were within the scope of the Regulation if they were valid for more than 12 months. On the other hand, the words ‘having a validity of 12 months or less’ did not set out the scope of the Regulation as regards national identity cards. So no identity cards fall within the scope of the Regulation, regardless of the period of their validity.

On the second question, the CJEU ruled that the passport Regulation only governed the use of data for the purposes of that Regulation. Any further use of that data, as specified in the preamble, was regulated by national law. It followed that the Regulation did not apply a purpose limitation rule upon Member States as regards biometric passport data. Because the Regulation did not apply to such uses by Member States, the EU Charter did not apply either, although such further use of data might be restricted by national law or the ECHR. Finally, as for the data protection Directive, the CJEU stated that ‘the referring court was requesting the interpretation of [the passport Regulation] and only that Regulation’, so there was no need to examine whether the data protection Directive affected national law on the further storage and use of biometric data collected for passport purposes.

Comments

I won’t mince words: this judgment is appalling.  It’s sensible enough as regards the scope of the passports Regulation itself, which clearly wasn’t intended to apply to any national identity cards or to the creation of government databases using biometric data. But the Court’s fundamental flaw is its failure to confirm and elaborate upon the application of the Charter and the data protection Directive to such databases.

Let’s examine those two points in turn. As regards the Charter, of course it’s true, as the Court says, that it only applies when a dispute falls within the scope of EU law. But the Court made that point only as regards the scope of the passports Regulation, before (not) answering the question about the data protection Directive. Logically the Court cannot conclude that this dispute is not linked to EU law before it assesses also whether the data protection Directive applies.

Anyway, if we apply the Court’s own case law, the link to the passports Regulation alone brings this issue within the scope of the Charter. In NS, a key judgment on the scope of the Charter, the EU’s Dublin Regulation left an option to Member States to decide in their national law whether to consider asylum applications which fell within the responsibility of another Member State. But the Court ruled that the Charter applied to such national discretion. More relevantly, in a line of cases starting with Promusicae, the Court applied the Charter in detail to a national option to provide for the collection of personal data on use of the Internet set out in EU law. And in last year’s Digital Rights judgment, the Court invalidated the EU’s data retention Directive for the very reason that this Directive failed to effectively regulate the further national use of personal data collected pursuant to it.

As regards the question about the data protection Directive, the CJEU’s answer simply departs from reality. It is quite clearly not true that the national court was ‘only’ asking for an interpretation of the passport Regulation. As we can see from the text of the question excerpted above, it also asked the CJEU to interpret the data protection Directive. Admittedly, it only asked the CJEU to interpret the Directive in the context of the Regulation. But the CJEU does not make that distinction clear; and more importantly, that distinction just doesn’t matter.

Why? Because the CJEU has frequently rephrased questions by national courts in order to give a full reply to the EU law issues which they are actually having to address in the relevant litigation. The examples are legion, but the most relevant one is the judgment in Promusicae. In that case, which concerned mass interception of Internet users’ activity for the purposes of enforcing intellectual property rights, the national court only asked questions about EU intellectual property law and the e-commerce Directive. The CJEU quite rightly redrafted the questions in order to give an answer about the relevant data protection rules (in that case, the e-privacy Directive) as well. In Willems, the national court had already identified the relevance of the data protection Directive, so a comparatively minor redraft of its questions would have sufficed in order to ensure a reply that was fully relevant to the national litigation.

The Court’s ruling is also unsatisfactory in the broader context of the legislation and case law on similar issues. When it asserted that national law applied to databases of biometric data, the CJEU only selectively quoted from the preamble to the passports Regulation. Recital 4 of the preamble to the 2004 Regulation states that access to the data collected as regards biometric passports is ‘subject to any relevant provisions of [EU] law’. Moreover, the CJEU interpreted the data protection Directive as regards a comparable national database (a collection of information on foreign nationals) in the Huber judgment. I should note that the data protection Directive also applies where the passport Regulation does not: to biometric information collected as regards identity cards, and to passport biometric information collected in the Member States that are not bound by the Regulation (the UK and Ireland). Finally, the Court’s indifference to the fate of biometric data collected by Member States as regards passports seriously undercuts its own rulinge in Schwarz, when it defended the validity of the passports Regulation on the basis of the limited scope of its interference with privacy rights (proportionality), and quoted the S and Marper judgment of the European Court of Human Rights to the effect that ‘the [EU] legislature must ensure that there are specific guarantees that the processing of such data will be effectively protected from misuse and abuse’.  

At first sight, these criticisms of the ruling may seem legalistic. But my concerns are about much more than the deep flaws in the Court’s legal reasoning here. As we all know, the scope of databases and mass surveillance of individuals (‘big data’) have increased exponentially in recent years. This raises huge human rights issues and EU law has a significant role to play. Last year, in its judgments in Digital Rights and Google Spain, the CJEU genuinely tried to grapple with these issues. Many aspects of these judgments have been criticised, but the Court is at its best when it fully engages in these important legal debates. When it avoids them, with the specious legalism it spouts in Willems, it is at its worst.

 

Image credit: Dailyalternative.co.uk

Barnard & Peers: chapter 9, chapter 26

The UK's general election: a fundamental change to UK/EU relations?

Steve Peers

The result of the current British election campaign could be crucial for the future of the UK’s relations with the European Union. Every UK-wide election party which is likely to win seats in the election has now released its election manifesto, namely: the Conservatives; Labour; Liberal Democrats; UKIP; and the Greens. It’s therefore a good time to examine what the parties are saying about the EU, and what the various post-election scenarios would mean for the UK’s relations with the EU.

According to pollsters, at present the most probable outcome of the election is that no party will have an overall majority, although there is a small possibility that either the Labour party or the Conservative party will obtain enough seats for a majority. In the absence of a majority, either the Conservative party or the Labour party will try to obtain enough votes to govern from other parties, which are likely to include parties running in Wales, Scotland and Northern Ireland.  So it’s necessary to consider what these other parties’ view on the EU is, and (more indirectly) whether they are likely to support Labour or the Conservatives in office.

Of these parties, only the Welsh Plaid Cymru has released its manifesto already, but I will refer to the other parties’ positions to the extent that they have been announced to the press: the Scottish Nationalist Party (SNP) and the Democratic Unionist Party (DUP). I won’t discuss Sinn Fein, since it will not take up its seats in Parliament, or the Northern Irish Social Democratic and Labour Party, since it sits and votes with the UK-wide Labour party.

Party manifestos

There’s a lot in the manifestos that touches upon EU-related policy. For instance, the Liberal Democrats promise a ‘Digital Rights Bill’, which is closely related to EU laws on data protection and net neutrality. EU law also has a big impact on environmental law, consumer law and some other policies. But I will focus here on the key question of ‘Brexit’, ie the UK’s withdrawal from the EU.

The Conservative party re-iterates that party’s policy of renegotiating EU membership and then holding an in/out referendum by the end of 2017. The renegotiation would focus on free movement (‘immigration’) from the rest of the EU, although the manifesto also refers to changing the principle of ‘ever closer union’ of EU Member States, and protecting the interests of non-eurozone Member States.

The Labour party manifesto refers to specific EU reforms, including EU immigration issues. It promises an in-out referendum if there is a further transfer of powers from the UK to the EU  This policy pledge is a development of current legislation (the 2011 European Union Act), which requires a referendum already in the event of such transfer of powers – but not an in/out referendum. However, the party does not promise an in-out referendum relating to the renegotiation. Indeed, they have made much of their opposition to that prospect.

The Liberal Democrats also promise an in-out referendum if there is a further transfer of powers from the UK to the EU. In fact, it’s their long-standing policy. They don’t call for a referendum following renegotiation.

The UK  Independence Party reiterates its long-standing policy in favour of the UK leaving the EU. It calls for a referendum to be held ‘as soon as possible’, with the preferred question ‘Do you want Britain to be a free, independent and sovereign democracy?’. They support a negotiated withdrawal, rather than a unilateral departure.

The Green Party declares itself in favour of the EU, although supports reform of it, and favours holding an in/out referendum on EU membership. However, it seems unlikely that this party would support a Conservative-led government.

Plaid Cymru declares that it is pro-European, and makes no demand for a referendum. It also seems unlikely that this party would support a Conservative-led government. The SNP support the UK’s EU membership, but in the event of an in/out referendum, they would like the public in each region of the UK to have a veto on leaving. This party has expressly ruled out supporting a Conservative-led government. The Labour party has in turn ruled out a coalition with the SNP, although it has not ruled out less formal arrangements.  

Finally, according to press reports, the DUP supports an in-out referendum on EU membership. It’s not clear if they would support renegotiation first, or would like an immediate referendum along the lines of UKIP. The DUP has traditionally supported Conservative governments in the past, although the party has declared its willingness to negotiate with the Labour party as well. It’s not clear if they would insist upon an EU referendum as the price of their support of a Labour government (it would be superfluous to insist on one as a condition of supporting the Conservatives). It’s also not clear if, like the SNP and Plaid Cymru, they would insist that Northern Ireland would also have to vote in favour for Brexit to be valid, but I doubt that this is their view, since they traditionally seek stronger ties between Northern Ireland and the rest of the UK.

Analysing the manifestos

To make things simpler, I will assume that politicians will stick to the policies that they are promising. Of course, it’s always possible that politicians will break their promises; some might say that this is about as certain as death and taxes. But politicians do keep some promises, and in some cases there are strong pressures on them to keep those pledges. For instance, if David Cameron reneged on his renegotiation policy, he would not last 24 hours as leader of the Conservative party.

There are three different visions of the UK’s future with the EU on offer: an immediate Brexit referendum (favoured by UKIP); renegotiation followed by a referendum (supported by the Conservatives); or renegotiation without a referendum (supported by Labour and the Liberal Democrats).

The latter two parties do support a referendum in the event of a transfer of powers from the UK to the EU, but that would only happen if there were a Treaty amendment that other Member States appear to have little enthusiasm for. In the event that Eurozone Member States want new measures to ‘save the euro’, it’s possible to agree separate treaties or EU legislation among themselves (as they have before), or, if necessary, to agree on amendments to the EU Treaties which only apply to the Eurozone states, or which otherwise give the UK an opt-out. In that case, there would be no transfer of powers from the UK to the EU, and so no need for a referendum.  Simply put, an in/out referendum under the Labour or Liberal Democrat policies is very unlikely.

As for the details of the Brexit policies, I have commented already on the details of the Conservatives’ intentions to renegotiate the free movement rules, and so won’t repeat these points again.

While UKIP make much of the need for an unbiased referendum campaign, their suggested question is clearly biased. A genuine unbiased question would be: ‘Do you support the UK remaining a member of the European Union, or leaving it?’, with boxes for ‘remain’ and ‘leave’ (so that neither side has the supposed advantage of being the ‘Yes’ vote).  

The demand for regional voting on a Brexit referendum in the different parts of the UK is clearly incompatible with the constitution of the UK as it currently stands. It resembles the arrangements in federal states like Canada, Australia and Switzerland, where there are both national and regional thresholds for constitutional amendments. But the UK is not such a state – though whether it should be one is a rather broader question. In any event, it’s a moot point, since it would only arise in practice if the SNP is supporting a Conservative government – which it has said it won’t do.

Negotiations between parties

What happens if no party holds a majority? The exact composition of the next government will then depend on negotiations between the parties, which in turn depends on the exact numbers of seats each gets. It’s too early to predict that now. But some basic points can be made.

First of all, it’s simply not true, as Professor Tim Bale has pointed out, that the biggest party always gets to form the government. It’s rare for the second largest party to form one, but it’s entirely legal and consistent with constitutional convention if it does.

Secondly, the Liberal Democrats have clearly indicated via press reports their willingness to compromise on the Brexit issue, and to support the Conservative demand for a referendum subject to discussion of the details, such as the width of the voting franchise and the wording of the referendum question. A Brexit referendum is clearly a ‘red line’ (non-negotiable issue) for the Tories. So a renegotiation and referendum would be likely to take place if the Tories and Liberal Democrats hold a majority of seats between them, or if the Tories, Liberal Democrats and DUP hold a majority.

Thirdly, a coalition or other arrangement involving UKIP (Tory/UKIP; Tory/DUP/UKIP; Tory/LibDem/UKIP; Tory/LibDem/UKIP/DUP) might have difficulty agreeing on the Brexit issue, due to the different policy of UKIP. Some Conservative backbenchers would undoubtedly like to a see an immediate Brexit vote, but others in the party, and the Liberal Democrats, would not. But it would be awkward for UKIP to back down on its most important policy.

If Labour and the Liberal Democrats held a majority, or if Labour formed a government in most other scenarios (Labour/SNP; Labour/LibDem/SNP; Labour/SNP/Green/PC etc) a Brexit referendum is highly unlikely.

Finally, in the improbable event that Labour and the Conservative party do a deal, differences in the Brexit referendum policy would be one of many obstacles. But the far bigger issue would be managing the consequences of hell freezing over.

Passing EU-related legislation

Finally, would there be any difficulties legislating for Brexit? It can be assumed that a Tory-led government would immediately introduce a government bill to put into effect its planned referendum. One potential problem here is the House of Lords, which blocked a previous private member’s bill which aimed to put this policy into law. The so-called ‘Salisbury Convention’ means that the House of Lords does not block government bills which implement policies which were set out in the leading party’s manifesto. According to the House of Lords itself, the convention should only apply if all of the coalition parties have set out the same policy – but only the Tories (and likely the DUP) have set out a renegotiation and referendum plan in their manifesto. That could lead to the House of Lords blocking the referendum Bill. Although a majority in the House of Commons could use the Parliament Act to overrule the Lords, that would mean a year’s delay in the law coming into force, which might complicate renegotiation talks.

Art credit: Adam, Daily Telegraph

Barnard & Peers: chapter 2

The Commission’s power of initiative: the CJEU sets important constraints

 

Steve Peers

As every EU politics or law student learns, a key feature of the Commission’s role as the ‘motor’ of EU integration is its near-monopoly on making proposals for EU action. But does that near-monopoly – which the Commission likes to call its ‘right of initiative’ – imply a corresponding power to withdraw proposals? And if so, are there any constraints on such a power? The CJEU answered these questions in an important judgment today.

Background

The case concerned a proposal for framework legislation on ‘macro-financial assistance’ for non-EU countries. This type of assistance helps out non-EU countries which are in severe economic difficulties, for instance helping them to make an imminent loan payment. It obviously helps the economies of the countries concerned, thus indirectly helping EU companies that export to them; and it undoubtedly cements the political relationship between the EU and those countries.

Before the Treaty of Lisbon, such aid was granted on the basis of the EU’s ‘residual powers’, which are now provided for in Article 352 TFEU. However, that Treaty created a specific ‘legal base’ for the EU to adopt rules on macro-economic support for third states: Article 212 TFEU, which provides for the use of the ‘ordinary legislative procedure’ to adopt legislation on this. The Treaty of Lisbon also created an Article 213 TFEU, which allows assistance to be granted in urgent cases without going through a full legislative process.

In 2011, the Commission proposed ‘framework legislation’ on macro-financial assistance, which would have given it the power to decide on the crucial question of which third countries receive this money. The Commission’s decisions on this issue would be controlled by means of the variant of the ‘comitology’ rules known as the ‘examination procedure’, which gives Member States’ experts’ the power to block draft Commission decisions. There’s no significant role for the European Parliament (EP) in that process.

The Commission’s suggestions did not satisfy the EP or the Council. The EP suggested instead that Commission Decisions on which countries receive macro-financial help be subject to ‘delegated acts’: Commission decisions which can be blocked by either the Council or the EP. For its part, the Council position was that each decision to grant aid to a third country had to be subject to the ordinary legislative procedure. The two institutions began negotiations, and eventually agreed to use the ordinary legislative procedure to this end.

Since the Commission objected to this aspect of the EP/Council deal in principle, it withdrew its proposal before the institutions could adopt the legislation concerned. The Council responded by taking the rare step of suing the Commission. While the Council was supported by ten Member States, the EP did not intervene on either side – despite the huge stakes for that institution.

The judgment

The Court’s judgment started out by accepting that the Commission’s power to make proposals gives it a corollary power to withdraw them. However, that power did not constitute a ‘right of veto’ in the legislative process, since that would upset the principle of institutional balance and conferral of powers. In short, the Commission has the power to withdraw proposals – but that power is constrained.

So how exactly is that power constrained?  The Court ruled that the Commission had to give the EP and the Council its reasons for such withdrawals, which must include ‘cogent evidence or arguments’. Such withdrawal decisions must be subject to judicial review, in the form of actions for annulment. However, it was sufficient to give those reasons to a Council working party and an EP/Council negotiation meeting.

As for the substantive grounds for withdrawing its proposal, the CJEU ruled that the amendment which the EP and Council wanted would have changed an ‘essential element’ of the proposal, and would have been irreconcilable with the ‘objective’ of improving the efficiency of EU policy in this area. Next, the CJEU ruled that there was no infringement of the principle of democracy, since it was inherent in the Commission’s right of initiative that it could withdraw proposals as long as the Council had not yet acted. So the Commission did not infringe the principles of conferral of powers and institutional balance.

Finally, the Court considered a separate issue: whether the manner of the Commission withdrawing its proposal breached the principle of ‘sincere cooperation’ between the EU institutions. On this point, the Court ruled that the Commission could not be criticized for withdrawing its proposal at a very late stage in the Council/EP negotiations, since only then had it become clear that the co-legislators would insist that the ordinary legislative procedure had to be used for the approval of each new macro-financial assistance decision. Moreover, the Commission had attempted to reconcile the position of the other institutions, and had proposed compromises. It seems implicit from the Court’s analysis here that the EP’s position (delegated acts instead of a comitology process) would not have altered the essential elements of the proposal.

Comments

First of all, what are the immediate consequences of this judgment? The failure of the legislative process in this case didn’t stop the EU granting macro-financial assistance to third States. In practice, in the absence of framework legislation on this subject, the EU has been adopting individual legislative measures to assist each country – which is exactly what the Commission objected to in the first place.

What procedure should govern such decisions, as a matter of principle? The CJEU seems to adopt the Commission’s view that efficiency is the main criterion. There is indeed a good argument for efficiency in the specific context of macro-economic assistance – since the bills usually have to be paid imminently, and the third country concerned is often in desperate straits. But the Commission and Court don’t refer to this specific context, and don’t balance the need for efficiency against the contrary case for full democratic control.

For example, one recent major beneficiary of EU macro-financial assistance has been Ukraine, and the EU’s relations with that country have been highly controversial. Ideally, there should be a full democratic debate on whether the EU should support that country’s economy, allowing those who sympathise with Putin’s Russia or who have other reasons to criticize the Ukrainian government to debate the merits of that assistance with those who support that government. Perhaps a good compromise would have been an agreement by the Commission to make a full legislative proposal whenever a significant minority (say a quarter of the Member States and/or a quarter of the Members of the European Parliament) indicate a wish to debate the merits of giving macro-financial assistance to a particular third State.

Secondly, the broader consequences of today’s judgment concern the relations between the EU’s three political institutions. The judgment has both procedural and substantive dimensions. Procedurally, it seems very easy to justify a Commission withdrawal: it need only inform EP/Council ‘trialogue’ (negotiation) meetings, or a Council working party of its intentions. Here the Court is legitimising these informal elements of EU governance (thanks to Ron Patz for raising this point). However, with respect, its approach is not convincing. Since a legislative proposal affects the EP as well as the Council, informing a Council working party cannot be sufficient. In fact, informing a trialogue committee shouldn’t be sufficient either, since there are many MEPs and Member States not represented there (the Council Presidency negotiates on behalf of all Member States as represented in the Council).

More broadly, the Commission should have to justify its withdrawals openly to the general public. It usually does this when it withdraws proposals as part of its work programme, but the principle of openness (as set out in the Treaties) justifies it becoming a legal rule, applying to every case of withdrawal. The best practice would be to adopt a formal Commission Decision to withdraw each proposal published in the EU’s Official Journal, giving effect to the principle of active transparency.

Indeed, it’s disturbing that the Court conceives of the EU system purely in terms of the institutions’ accountability to each other, rather than to the public generally. The Court’s dismissal of the ‘democratic principles’ argument is far too brief, failing to place that principle on an equal footing (never mind a superior footing) with the institutional rules on the Treaty. The better approach would have been to interpret the lacuna in the Treaties regarding withdrawal of Commission proposals in light of this democratic principle, emphasising the key role which that principle gives to the EU institutions with a greater claim to electoral legitimacy.

On the other hand, the Court does confirm that there must be judicial control of the withdrawal of Commission proposals. The application of the annulment procedure means that not only the Council, but also the EP and individual Member States can sue the Commission for withdrawing a proposal. Other parties are likely to lack standing to do so, except for a limited number of cases where Article 263 TFEU confers it to bring challenges to non-legislative acts. The alternative possibility of bringing proceedings via national courts to challenge withdrawals appears unworkable at first sight.

What are the substantive constraints to the Commission’s withdrawal of its proposals? The CJEU hints that there might be additional constraints after the Council has ‘acted’; presumably this refers to the Council adoption of its first reading position. Only a minority of EU legislative proposals go past this first reading – and half of those have in practice been agreed between the EP and Council already at that stage. It’s not clear what additional constraints might then exist, besides those set out in the Treaty (for instance, it’s easier to amend a Commission proposal in some cases after first reading). It’s also possible that the Court would take a different approach to non-legislative measures, or those subject to a special legislative procedure.

So let’s focus on the usual case – a withdrawal of a proposal for a legislative act pursuant to the ordinary legislative procedure, before the Council adopts its first reading position. The Court did not suggest that a change in the essential elements of a Commission proposal was the only ground to withdraw a proposal. Presumably it is still open to the Commission to withdraw proposals on other salient grounds, in particular the most common grounds for withdrawal: where the proposal is obsolete, or there is no chance of its adoption.

However, the existence of judicial review must mean that the Commission’s assessment in this regard could be open to challenge. The Court might also be called upon to clarify in another case what constitutes a change in the ‘essential elements’ of a proposal, where (for instance) the EP and Council want to widen or narrow that proposal’s scope significantly. There is no reason why the ‘essential elements’ rule is restricted to cases where the objective is improving the efficiency of EU actions, since there are many other possible objectives for EU action.

What other grounds for withdrawal exist? It seems implicit in the judgment that the Commission cannot simply argue that it has changed its mind, otherwise judicial review would have no purpose. There must be substantive reasons justifying that change of mind. Would it be sufficient that there is a new Commission? This is obviously a live issue, given that the Juncker Commission recently withdrew a number of proposals (for instance, the ‘circular economy’ proposals) on the grounds that it had changed its legislative priorities, and wished to start the process from scratch (see discussion here).  At first sight, since the proposals can always be redrafted during the legislative procedure, this is not a sound enough reason to withdraw a proposal, in light of today’s judgment – and there is still time to bring an annulment action against these withdrawals.  There might conceivably be an argument that a new Commission has more flexibility to withdraw proposals – but that begs questions as to whether it has a genuine democratic mandate, in particular if the Commission President did not campaign on the basis that the proposals in question should be withdrawn.

Finally, it might be arguable that today’s judgment has implications not only for the Commission, but for the ‘sincere cooperation’ during the legislative procedure between the EP and the Council. For instance, the EP and Council have been far apart for years as regards a proposal for maternity leave. As I recently pointed out (see here), the EP has shown recent signs of willingness to negotiate, which the Council has rejected.  For its part, the Commission has done little to try and broker a compromise, but has simply resorted to threatening to withdraw the proposal if there is no deal. Can the EP sue the Council for its intransigence? Can it sue the Commission if it follows through on its threat to withdraw the proposal, having played no constructive role in the talks? If the EU institutions and Member States make frequent use of the possibilities opened up by today’s judgment, this judicial intervention into the legislative process raises many such important questions.

 

Barnard & Peers:  chapter 3, chapter 5, chapter 8

The difference between torture and other ill-treatment: Cestaro v. Italy and the “prohibited purpose” requirement

 

Christina Kosin (LL.M. Edinburgh) Ph.D. candidate and academic assistant at the German Police University

What is the central element which distinguishes torture from inhuman or degrading treatment or punishment? Some scholars claim that the “prohibited purpose” requirement of the torture definition is the most central as well as the only criterion which is decisive in differentiating between that crime and other forms of ill-treatment.. However, in my view this is incorrect, and indeed the recent judgment of the European Court of Human Rights in Cestaro v. Italy shows (once again) that the “prohibited purpose” component of torture is not the only decisive criterion in distinguishing the crime from other inhuman or degrading treatment or punishment.

The judgment

Mr Cestaro was among the protesters surrounding the G-8 summit in Genoa, Italy from 21^st to 22^nd July 2001 (para. 7). On the first day of the summit of the world’s leading industrial States around 100,000 protesters gathered and demonstrated against globalisation (para. 19). Various radical groups such as the so-called “black blocks” were amongst the peaceful anti-globalisation protesters and engaged in violent and/or criminal conduct such as the looting of the city (para. 12). During the two days of incidents, hundreds of protesters and police forces were injured and one young person died (paras. 18-21). From the 21^st to 22^nd July 2001, Mr Cestaro and other protesters were housed in a school, which was stormed by the Italian police at that time. Mr Cestaro and others were brutally ill-treated while peacefully and legally lodging in the school (para. 182). The actual aim of the police mission was to search the school for evidence that could lead to the identification and possible arrest of members of the “black blocks” (para. 182). Mr Cestaro was subjected to repeated kicks and beatings with the tonfa, which is considered a potentially lethal weapon. As a result, Mr Cestaro suffered multiple fractures and a permanent impediment in his right arm and right leg (para. 178). The ECtHR ruled that the treatment by the police amounted to torture.

Ultimately he challenged the Italian state before the European Court of Human Rights (ECtHR), which issued its judgment in the case on 7 April. He alleged that Italy had committed a material and procedural breach of Article 3 of the ECHR, which states that “[n]o one shall be subjected to torture or to inhuman or degrading treatment or punishment”. Neither torture nor the terms “inhuman or degrading treatment or punishment” are defined in the Convention. The Court’s practice has shown that it understands torture as defined in Article 1 (1) in the United Nations Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (UNCAT). According to this definition, torture needs to be inflicted on a person

“for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind”.

This is the so known “prohibited purpose” requirement. Next to this, there are four more elements of the UNCAT torture definition namely (1) the act must be inflicted intentional, (2) the act must cause severe, physical or mental, pain or suffering, (3) the treatment must be inflicted by or at the instigation of or with the consent of or acquiescence of a public official or other person acting in an official capacity and (4) pain arising out of lawful sanctions is not included.

The issue at hand is that the prohibited purpose requirement is mistaken for the only criterion which is able to differentiate between torture and less serious ill-treatment. This confusion arose because of various reasons. The European Commission on Human Rights, which used to rule on cases together with the ECtHR, stipulated in the 1969 Greek case that torture “is often used to describe inhuman treatment, which has a purpose, such as the obtaining of information or confessions, or the infliction of punishment” (p. 186). This was interpreted to mean that the purpose requirement is the distinguishing element between torture and inhuman or degrading treatment or punishment while the other elements remain static. Further, the ECtHR has never classified a treatment which lacked the purpose requirement as torture. In addition, some scholars claim that the drafting history of the UNCAT, the travaux préparatoires, indicates that this criterion is the sole distinguishing factor between the two forms of ill-treatment. Although these reasons sound compelling, the interpretation was wrong and the prohibited purpose requirement is not the sole decisive factor as is shown below.

In section “Aperçu de la jurisprudence en matière de « torture » “, the Court provided a brief overview of its jurisprudence concerning Article 3. In the subsequent section “Application en l’espèce”, the ECtHR applied the facts to the case. Both sections elucidate that the prohibited purpose criterion is a vital but not the only requirement to differentiate torture from inhuman or degrading treatment or punishment.

In the former section, the Court did not highlight the ostensible uniqueness of the prohibited purpose requirement. Rather, the Court showed that different criteria of the treatment can be decisive in classifying an act as torture. The Court highlighted that torture can be distinguished from other ill-treatment by the special stigma which is attached to the crime – torture implies very serious and cruel pain and suffering as well as the deliberate/intentional infliction of such pain. In the Court’s words, “[o]utre la gravité des traitmens, la « torture » implique une volonté délibérée.”  (para. 171).

Then, the Court referred to cases in which it found a combination of the gravity of the treatment and the intentional element such as in Aksoy v. Turkey and Yaman v. Turkey (para. 172). In this context the ECtHR noted that all these cases contained a purpose to obtain information, to punish somebody or to intimidate the person (para. 172). This is the first time the ECtHR mentioned the element of the prohibited purpose. The ECtHR further elucidated that it put a special weight on the severity of the pain in some cases such as Aslan v. Turkey or on the arbitrariness of the violence in other cases such as in Romanov v. Russia (para. 173-174). Towards the end of this section the Court explained that it did not classify some police actions as torture such as in Krastanov v. Bulgaria as the prohibited purpose element was missing and because the pain was inflicted for a short duration only (para. 175). This constitutes the second time when the Court mentioned the purpose criterion. Lastly, the Court stipulated that some cases such as Gäfgen did not constitute torture because the level of cruelty required was not reached (para. 176).

The ECtHR clearly revealed that various components can determine whether a particular treatment constitutes torture or inhuman or degrading treatment or punishment: the gravity of the treatment, the intentional infliction of the treatment, the severity of the pain, the arbitrariness of the violence, the prohibited purpose, the duration of the treatment and the level of cruelty of the act. The Court did not classify these examples as exhaustive and one can assume that other factors of the treatment as well can trigger a distinction between torture and other forms of ill-treatment. The prohibited purpose criterion of the UNCAT torture definition is not the only decisive factor.

In the section “Application en l’espèce”, the ECtHR provided reasons for classifying the treatment by the police as torture. In its reasoning, the Court did not explicitly state which element of the conduct by the police was decisive for classifying the violence as torture and not inhuman or degrading treatment. The arguments by the ECtHR suggest that a variety of characteristics of the treatment contributed to the Court’s decision.

The Court put forward that the pain was inflicted by the police with the purpose to punish and to retaliate and with the aim to cause pain and suffering (para. 177). The prohibited purpose by the police was hence to punish and retaliate. This is the first and last time that the ECtHR referred to the purpose requirement in this section. The Court noted that there is no denying that the pain and suffering was of particular seriousness and cruelty (para. 179). As mentioned above, the applicant suffered from a permanent impediment resulting out of the beatings he received. Further, it was stated that there was no link between the applicant’s conduct and the force used by the police. The Court emphasised that Mr Cestaro did not resist in any way and that the infliction of pain was thus disproportionate (para. 180).

Although the mission of the police was to search the school, the public officials immediately used force (paras. 180-182). Instead of trying to enter the building peacefully and negotiate with the protesters, the police stormed the building by breaking down the gates and promptly used violence against the protestors (para. 183). The Court then went on and stressed that the police tried to justify its actions by arguments such as the protection of the nation was at stake (para.184-185). It is indeed true that the 21^st July was marked by heavy violence through the looting and devastation of the city of Genoa (para. 20). One can imagine that the police was under paramount pressure and strain throughout the whole day. The Court, however, correctly observed that the situation in the school was entirely different as the protesters were calm and did not resist the violence of the police (para. 186). Based on these grounds the Court reached the conclusion that the treatment by the police amounted to torture as understood in Article 3 ECHR (para. 190). The ECtHR put forward several reasons for classifying the treatment by the police as torture but it did not single out one specific factor. It seems that the immediate, disproportionate and serious infliction of pain was vital for the Court to arrive at its decision. The prohibited purpose criterion was not the triggering criterion in this case.

Comments

The Cestaro v. Italy case shows that the prohibited purpose element in the UNCAT definition is not the sole criterion which distinguishes torture from other inhuman or degrading treatment or punishment. The presumption that the purpose requirement is the most important feature of torture does not automatically mean that it is the only one.

As indicated at the outset, Cestaro v. Italy is not the first case which revealed that the prohibited purpose requirement is not the sole factor able to distinguish between both crimes in the UNCAT. In the case Ireland v. the UK in 1978, the ECtHR ruled that the five techniques employed by the British forces did not constitute torture because they did not reach the level of cruelty and intensity as implied by the term (para. 167). In Selmouni v. France 1999 the Court reached a judgment on similar grounds. The ECtHR stated that the victim was subjected to torture because the pain and suffering was of particular seriousness and cruelty (para. 101-105). In Romanov v. Russia 2002, the Strasbourg Court found a multitude of reasons why the treatment by the prison wards constituted torture. However, it put a special emphasise on the arbitrariness of the beatings the victim received (para. 70).

These cases already revealed that other factors than the purpose requirement can make a difference between torture and inhuman or degrading treatment or punishment. Nevertheless, some still claim that only the purpose requirement can be decisive. They base their arguments on the travaux préparatoires of the UNCAT as well as old case law from different treaty bodies. The Cestaro v. Italy case reiterates that the prohibited purpose criterion is not the sole element which is capable of differentiating between torture and other ill-treatment.

One can now assume that the consequences flowing from this case and other similar jurisprudence of the ECtHR, will mainly concern theoretical discussions among legal scholars. The consequence could, however, have negative effects beyond that. The Strasbourg Court is not the only organ which uses the UNCAT definition of torture in its cases concerning violations of Article 3 ECHR. The Committee Against Torture (CAT) as well as the Human Rights Committee (HRC) provide decisions on the basis of Article 1 (1) UNCAT. The former body applies and interprets the UNCAT and the latter body enforces the International Covenant for Civil and Political Rights (ICCPR). If the ECtHR, the CAT and the HRC use different grounds for distinguishing torture from other ill-treatment, this could shed a negative light on the consistency and the trustworthiness of these bodies. Considering that the ban on torture is international, the jurisprudence should be consistent as well – at least with regard to the most important and international bodies which decide on torture cases, should it not? In the worst case scenario, the victims of torture could additionally suffer from legal uncertainty.

 

Barnard & Peers: chapter 9

Fundamental Rights and the European Public Prosecutor’s Office: an uncomfortable silence

 

 

Michiel Luchtman, Jannemieke Ouwerkerk, Marloes van Noorloos, Pim Geelhoed, Jorrit Rijpma and Louis Middelkoop are members of the Meijers Committee (www.commissie-meijers.nl/en).

The EU’s proposal for the establishment of a European Public Prosecutor’s Office (EPPO) has created quite a stir in the Member States. The EPPO would be competent to investigate and prosecute fraud with EU money (e.g. the misuse of EU funds), although there is already talk about extending its competence to terrorism and other serious crime. So far, political negotiations over the draft regulation have focused on the question why the EU is in in need of this new supranational body in the first place, and on the extent of the EU’s influence on national affairs, particularly in such a sensitive area as criminal justice.

Supposing that in the near future the European Public Prosecutor’s Office will indeed be established, more attention to the substance of the current proposal needs to be paid without delay, particularly to the protection of fundamental rights. The current proposal raises serious concerns on this matter, as it is unclear who will supervise the actions of the EPPO and how this may be done effectively.

Procedural Rights in EU criminal law

In the EU context, the question of who is responsible for the guaranteeing of procedural rights in transnational criminal law enforcement has already frequently been addressed by legislation. After all, Member States of the EU do cooperate intensively on a daily basis: think of the European arrest warrant mechanism, which enables the rapid surrender of suspects from one Member State to another. All such cooperation mechanisms contain provisions on legal protection.

The proposal to establish an EPPO takes criminal justice integration significantly further than any other instrument created thus far. The EPPO will be authorized to take intrusive coercive measures, such as ordering arrests, interceptions of telecommunication, or house searches, just as national prosecutors can. Yet – and unlike national prosecuting authorities – the EPPO would be competent to apply these measures in all the territories of the participating Member States, without the restriction of national borders. It might therefore be expected that the ministers of justice, the European Parliament, and the European Commission would hold extensive debates on the precise conditions for searches, telephone interceptions, arrests, and pre-trial detention in supranational investigations. Surprisingly, they have not.  

On the contrary, in their efforts to prevent a further transfer of power to ‘Brussels’, most Member States oppose any further approximation of criminal procedure. And those who do support the establishment of an EPPO hold that citizens’ rights are already sufficiently protected, referring to the fundamental rights acquis laid down in treaties and the EU Charter. They also point out that the EPPO must respect the additional procedural guarantees provided in the domestic legal order of the Member State where it is conducting its operations.

 From national to transnational criminal procedure

Such safeguards are designed for the administration of criminal justice in a domestic system and will not always fit the new transnational setting of the EPPO. The potential consequences of this systemic flaw in EPPO’s design are highly problematic. Imagine that the EPPO conducts a house search without proper judicial authorization in Amsterdam, and then that the case is subsequently brought to trial before a Milanese criminal court. Should the Italian judge apply the complicated Dutch case-law on unlawful house searches? And should the EPPO be allowed to search someone’s personal computer online, from a distance, in order to evade the stricter rules of the Member State on whose territory the computer is located? After all, it is not only the mobility of criminals that is increasing, but also of witnesses, information, and evidence in the area of freedom, security and justice to which the EU aspires. Nonetheless, the proposal remains silent on these issues, with such potential consequence as forum shopping and a race to the bottom of the lowest level of safeguards.

The envisaged framework for the EPPO leaves an important question unanswered: Do discretionary margins for choosing the applicable national law with respect to investigatory powers, procedural safeguards, judicial supervision, admissibility of evidence, etcetera permit arbitrary interference with the fundamental rights of citizens involved in criminal proceedings?

Both the European Court of Human Rights and the European Court of Justice have consistently stressed the importance of the rule of law, as a vital mechanism defending against arbitrary interference with citizens’ fundamental rights. The law provides citizens with legal certainty as to the scope of their rights and duties; it helps to effectuate their defence rights; and it enables courts to prevent executive discretion from turning into arbitrariness. Under the proposed EPPO set-up, the problem is not that national laws are absent, unclear or ignored. The problem is rather that the EPPO is given more than twenty divergent national legal system to work with. This creates the risk that a legal avenue for a particular intrusive measure might be very easy to obtain. In certain cases, the EPPO is allowed to choose the most attractive rules from multiple codes of criminal procedure – and continuously to change the rules during the game, without its adversaries having any say on the matter, and without courts being able to exercise control. This leaves citizens without an adequate indication of which set of national rules will apply in a particular case. 

As a community based on the rule of law, the European Union should protect its citizens against arbitrary action by European law enforcement authorities. In order to guarantee that the rule of law will be upheld in the European Public Prosecutor’s Office, citizens must be offered legal certainty  as to which law will be applied. Moreover, the law must be adequately accessible to all citizens: they must be permitted to know their rights and duties, and invoke them in court.

 

Barnard & Peers: chapter 9, chapter 25

Vidal-Hall v Google: Strengthening EU law remedies

 

Steve Peers

There are many laws which exist on paper, but cannot easily be enforced practically. One example is data protection law, which relies largely for its enforcement on overworked data protection authorities. Individual ‘data subjects’ whose data protection rights have been infringed can sue the infringing companies directly, but they face a number of barriers in this respect.

In particular, if they have not suffered any direct economic loss, can they sue for non-economic loss? If not, there is probably little point in bringing legal proceedings – and the infringing company faces fewer constraints upon its breach of the law concerned.

Of course, this issue is not unique to data protection law. But this was the subject-matter of the important judgment of the Court of Appeal in Vidal-Hall v Google last week, which relied upon the EU Charter of Rights to strike down UK legislation which limited the ability to sue for non-economic losses as regards EU data protection law. This ruling could have broad implications not only on the enforcement of data protection rights, but also other rights protected by EU law.

Judgment

The case concerned allegations that Google had infringed the data protection rights of users of Apple’s Safari browser, by getting around protections against tracking which were intended to prevent Google from collecting information on those users’ behaviour online. American litigation on the same issue had foundered due to the inability of plaintiffs to obtain damages for non-economic losses under American law, although Google had reached a settlement with federal and state regulators which entailed payment of a large penalty.

As regards UK/EU law, the Court of Appeal had to address four issues: (a) is there a tort of ‘misuse of private information’; (b) whether it was possible to sue for non-economic damage; (c) whether the information involved was ‘personal data’ for the purposes of data protection law; and (d) whether there was an arguable case in tort and data protection law. The Court decided all four issues in favour of the plaintiffs, but that does not mean they have won the case yet: at this stage, the Court of Appeal was only deciding whether documents could be served on Google in the first place.

Remedies for breach of EU law

This judgment raises many important questions of privacy rights and data protection law, and substantially advances the enforcement of those rights and that law in the UK. There are excellent discussions of these finer points already by Jon Baines, Christopher Knight and Alexander Hanff. My focus here is on the broader importance of this judgment for the enforcement of EU law rights.

The starting point in this case is the EU’s data protection Directive, which provides that any person who has suffered damage due to breach of the Directive can receive compensation from the data controller for that breach. The UK Data Protection Act implements that rule by providing (in s. 13) for separate claims for damages and distress, imposing conditions (a link to damages, or data processing for special purposes) upon any claim for distress. UK courts have interpreted s. 13 of the Act to mean that ‘damages’ can only refer to economic loss, with non-economic loss subject to the more restrictive rules relating to ‘distress’.

However, the Court of Appeal ruled that the rule in the EU Directive had a wider meaning, covering both economic and non-economic loss, basing itself on CJEU case law relating to the EU’s package holidays Directive. It based this argument on its interpretation of the aim of the Directive (protecting privacy rights), read alongside the right to privacy in Article 8 ECHR and the data protection rights in Article 7 of the EU Charter of Fundamental Rights.

But what were the consequences of this finding? First of all, the Court of Appeal rejected the possibility of using the CJEU’s principle of indirect effect, set out in case law since Marleasing, to strike down s. 13(2) of the UK Act.  This is undoubtedly correct: CJEU case law makes clear that the principle of indirect effect is a rule of interpretation, requiring national courts to stretch the interpretation of national law on the books as far as possible to ensure that it implements EU law correctly. That principle reaches its limits when national law cannot be interpreted consistently with EU law, as the Court of Appeal determined in this case.

So the Court of Appeal rightly relied instead on Articles 7 and 8 of the Charter (privacy and data protection rights), applying Article 47 of the Charter (the right to a fair trial and effective remedy) to strike down national law, just as it had recently done in Benkharbouche (discussed here). That case involved a claim for employment law rights against foreign embassies, and the Court struck down the relevant provisions of the State Immunity Act to allow the suit to continue (insofar as the claims were based on EU law). The Court of Appeal did limit the impact of Article 47 of the Charter by confirming that it could not be used to strike down legislation where that would involve the courts rewriting a legislative scheme and making complex choices that should be left to the legislature to make. But that was not the case here.

As in Benkharbouche, this judgment offers confirmation of the significant possibilities of using the Charter in human rights litigation. Unlike the prior judgment, it was not necessary in Vidal-Hall to distinguish between claims linked to EU law (where the Charter applies) and claims not connected to EU law (where the Charter does not apply), because all of the claims in this case are linked to EU law. The crucial relevance of that distinction is that the Charter can be used to disapply Acts of Parliament, whereas the Human Rights Act cannot. (Moreover, any UK court can disapply an act of Parliament conflicting with the Charter, whereas only the higher courts can issue declarations of incompatibility with the Human Rights Act).

It is clear from Vidal-Hall that Article 47 can be used not just to strike down Acts of Parliament that confer immunity upon defendants, but also to strike down rules that limit heads of damage that can be recovered. It follows that many other types of restrictions on remedies could be challenged: other forms of standing rules, time limits and restrictions on legal aid, for instance. It is also clear that the ‘complex legislative scheme’ exception should not be interpreted widely: plaintiffs should target their challenges to Acts of Parliament against very precise and specific limitations in order to ensure that the exception does not apply.

The existence of this exception does mean, however, that it might be significantly more difficult to use Article 47 to ask the courts to create an entirely new remedy for breach of EU law, because the counter-argument would be that the creation of new remedies is a complex issue best left to Parliament. On the other hand, plaintiffs in such cases could also argue that the common law should develop to create new forms of remedy to ensure effective protection of Charter rights, just as tort law relating to privacy rights was clarified in Vidal-Hall.  

It should be noted here that the Vidal-Hall litigation concerns the application of a Directive between private parties. The judgment thus clearly demonstrates the importance of the Charter in overcoming the traditional restriction on applying Directives against private parties, where national law is incompatible with the Directives (ie, the lack of ‘horizontal direct effect’ of Directives). The alternative option of bringing a Francovich damages action against the state for its breach of EU law is not even discussed here. Obviously it will always be simpler and cheaper for the plaintiffs to follow the direct route of disapplying the Act of Parliament in the main litigation, as compared to having to bring an action against the State instead – especially in cases like Vidal-Hall, where the merits of the case have not even been tried yet.

Furthermore, it should be emphasised that the court was applying Article 47 of the Charter to disapply UK law, not Articles 7 and 8 of the Charter, which set out the substantive rights to privacy and data protection. That is an important distinction because according to last year’s CJEU ruling in AMS (discussed here), not all Charter rights can be enforced by setting aside national law. The CJEU has yet to rule on whether Articles 7 and 8 can be enforced by this route (it dodged this bullet in Satamedia), but it will be hard to avoid it forever.

Finally, what other areas of EU law could this judgment be relevant to? A lot of EU law concerns economic damages in any event (cf the case law on private damages for breach of EU competition law). Even consumer law is largely about economic loss, too: the package holiday judgment referred to by the Court of Appeal is an exception, because in the case of holidays, consumers’ distress arises from being unable to spend their money on holiday as they had planned.

There is at least one obvious other area of EU law where this judgment may also be relevant: the free movement of EU citizens. It is possible to claim for economic loss in such cases, for instance where a person has lost his job as a result of breach of EU free movement law (see the discussion of a recent Irish judgment here). However, often the loss is purely personal: the inability to spend time with a spouse and children. There may also be some economic costs (due to the need to travel to visit a family member, or a job which is lost or cannot be applied for due to breaches of free movement law), but the inherent loss of family life is surely highly significant too. Moreover, the root human right being protected in such cases is in part the same right as that being protected in Vidal-Hall: the right to private and family life (Article 7 of the Charter, Article 8 ECHR).  The next UK government should ensure full compliance with the law on ‘Surinder Singh’ cases (as discussed here), if it does not wish to expose taxpayers to considerable liability.

Photo credit: milanox.eu 

Barnard & Peers: chapter 6, chapter 9