Related Articles/Videos

Piet Eeckhout on EU human rights law in 2012

9th March 2012

Picture by Djtm from

Professor Eeckhout is Director of the Centre of European Law, at King’s College London. He is co-editor of the Yearbook of European Law, and also teaches at the College of Europe, Bruges. He is an associate academic member of Matrix Chambers, London.

20 years ago, the title of this post would have raised many eyebrows. There was of course, even then, some case law on human, or fundamental rights, as they are officially termed. But the EU was certainly not seen as a human rights organization. At most the Maastricht Treaty indicated a somewhat more important role for human rights. Today, it must be said, there is still a debate on whether it is proper to call the EU a human rights organization, but I am myself firmly on the side of those arguing that it is. Of course, in contrast with for example the European Convention system, the EU is clearly more than just a human rights organization - but it is clearly also that. The EU now has its own Charter of Fundamental Rights, which is becoming ever more significant in the case law. That is not all, however, we also have a Commissioner for fundamental rights, there is a Fundamental Rights Agency, and the EU is, notwithstanding Treaty language suggesting that it has no competence to legislate as regards human rights, doing exactly that. It has enacted impressive legislation in the field of anti-discrimination law; its legislative efforts to build a common asylum system are impregnated with mechanisms for human-rights protection; and there are a number of initiatives in the field of criminal law which are human-rights oriented. For the last five years or so the EU Courts have been at the forefront of the struggle to ensure that human rights are protected in the sensitive area of counter-terrorism policies.

Let us start by looking a little bit at the broader constitutional landscape, because there is a lot to see there. The Charter of Fundamental Rights has become binding, and has acquired Treaty status on 1 Dec 2009. In the last one-and-a half year or so, the EU Courts have been starting to apply it properly. I have not counted the cases, but there are very many, and it is therefore already clear that the Charter plays and will play a significant role in EU case-law.   In addition, the Treaty provides that the EU shall accede to the ECHR. The process of negotiation is under way, but I understand it is not running entirely smoothly, and actual accession is probably still some way off. Even so, the relationship between the Charter and the Convention is already a live question, as the Charter incorporates Convention rights, but also others. Will the Charter end up being an instrument emphasizing more of an EU version of fundamental-rights protection, perhaps in some areas in tension, or even conflict, with the Convention? On top of this duality, the Lisbon Treaty drafters have, in their wisdom, also maintained "general principles of EU law" as a separate category. The EU constitutional system of human rights protection therefore resembles a 3D movie, the three dimensions being the three sources of fundamental rights: the Charter, the Convention and General principles of EU law. I am afraid though that, unless you have the right spectacles, the movie does not make too much sense.  And as this is a kind of badly organized movie theatre, no one can tell you which are the right spectacles. 

The case law of the Court of Justice does not really tell us much yet about which spectacles are the right ones. But in some of the Opinions of Advocates General there are already some indications that there is scope for some tension between the Charter, the EU's homegrown bill of rights, and the Convention and its case law. There is for example the Opinion of AG Cruz Villalon in a case called Scarlet Extended v Sabam (Case C-70/10), regarding restrictions on the business activities of internet service providers, restrictions aimed at combating illegal music downloading. One of the issues he addressed was the scope for "limitations" to the Charter rights. The AG accepted that there could be divergences, in this regard, between the Charter and the Convention, in which case the Charter would require an autonomous interpretation. The word autonomous clearly has strong connotations in that it is employed by the Court in cases where there is a need to protect the EU legal order from unacceptable outside interference. The AG also indicates that the Court of Justice must "take into account" the ECHR case law. I note here that this language is fairly weak, and can perhaps not be equated with the concept of consistent interpretation. The German Constitutional Court has been criticized for adopting that same principle, of requiring German courts to "take account" of the Strasburg case law, and no more.
But let us now leave the broader constitutional landscape, and look at a couple of more specific questions and issues. I start with what is one of the most vital, if not the most vital, question of EU human rights law. Clearly, that law binds the EU institutions, but when does it also bind the Member States? In an article I wrote ten years ago (“The EU Charter of Fundamental Rights and the Federal Question”, CMLRev 2002) I called this, somewhat provocatively, the federal question. In that article I anticipated that that question might particularly arise in the Court's citizenship case law. Clearly, a citizen who moves to another Member State is protected by EU law, and such protection must extend to his or her fundamental rights. But what about the non-moving EU citizen? Well I am happy to see that I rightly anticipated that this was going to be one of the most sensitive questions, but very saddened to see what the Court has done with it, in particular in Ruiz Zambrano (Case C-34/09), McCarthy (Case C-434/09) and Dereci (Case C-256/11). As all these cases concerned the rights of non-EU family members of EU citizens, there is a clear connection with the fundamental right to private and family life. In Zambrano this was noted by AG Sharpston, who made a very radical proposal as regards the answer to the federal question. She suggested that EU human rights law should be binding on the Member States in all areas of EU competence, independently of whether those competences were in any way exercised or not. If accepted, that would mean an enormous expansion of the reach of EU human rights law, and in particular the Charter. At present EU human rights law binds the Member State, generally only where they are implementing EU law (for exampling implementing a directive or applying a regulation) and in a free movement context, when questions of derogations arise.
Of course in Zambrano the Court did not take up the AG's suggestion, as it said hardly anything at all. The fundamental rights of the Zambrano family were not mentioned. Nor did the Court in McCarthy address the question whether Mrs McCarthy could rely on those rights. AG Kokott made it clear, though, that, in her view, the case did not come within the scope of EU law, meaning, in the classical conception, that EU human rights law does not apply. However, in Dereci - clearly the case with the strongest "wholly internal situation" dimension, as it concerned Austrian citizens who had never moved and did not have dual nationality - the Court did consider the fundamental rights dimension. After having decided that the citizenship provisions do not assist the various families involved, the Court turned to the question whether the litigants could rely on the right to private and family life, as protected under the Charter, and indeed the Convention. The reply, I have to say, is mind-boggling. The Court states (§ 72):
“Thus, in the present case, if the referring court considers, in the light of the circumstances of the disputes in the main proceedings, that the situation of the applicants in the main proceedings is covered by European Union law, it must examine whether the refusal of their right of residence undermines the right to respect for private and family life provided for in Article 7 of the Charter. On the other hand, if it takes the view that that situation is not covered by European Union law, it must undertake that examination in the light of Article 8(1) of the ECHR.”
That is literally all the Court says. After having found that the citizenship provisions do not apply, the Court leaves it entirely to the national court to decide whether there may be some other connection with EU law which justifies applying the Charter, and in particular the right to private and family life. If ever there was an unhelpful preliminary ruling, this is the one. How is the national court supposed to decide this? Although I am no expert in this field of family re-unification, there must be hundreds if not thousands of EU citizens, residing in their own Member State, but with a spouse (or other family member) who does not have residence rights in that State. In those circumstances, all the Court is able to say is that national courts must decide whether any of these cases have a sufficient link with EU law, without giving any indication itself? And it does not stop there. The icing on the cake is that the Court points out that the national court may want to think about applying the Convention - actually, the language is stronger: if it takes the view that the situation is not covered by EU law, it must undertake that examination in the light of Article 8(1) of the ECHR. What business is this of the EUCJ? How can EU law instruct the Member States to apply the Convention? Do they have to give direct effect and primacy to the Convention? Not every Member State does, of course.
We all know that Europe is in crisis. To me, Dereci is a decision at the height of the constitutional crisis engulfing the Court of Justice. It is becoming increasingly difficult to accept reverse discrimination of EU citizens who do not move, as well as to define the required movement for the purpose of applying the citizenship - and indeed the internal market provisions. There are therefore those who argue for weakening or even severing the link with free movement. But then the reach of the Charter would also become much broader. There are, on the other hand, those, who emphasise the principle of conferred powers, and insist that there must be limitations to the scope of EU law.
That is a fundamental constitutional debate, and the incapacity of the Court to take a clear position in the Zambrano, McCarthy and Dereci case law shows that there is an internal constitutional crisis.
There is no space here to explore this much further, but I would nevertheless like to say something about when, in my view, EU human rights law and the Charter ought to bind the Member States. It seems to me that that question has always been approached from the wrong angle. Perhaps the focus should not be on the Member States: whether they are acting within the scope of EU law. The foundational principle, it seems to me, is that, in all that it undertakes, the EU must respect the fundamental rights of the individual. Every individual - not just the EU citizen - is entitled to respect for their fundamental rights when exposed to some kind of EU action. The basic question may therefore not be whether a Member State is acting within the scope of EU law, but rather whether the particular case of the individual is sufficiently connected with EU law to trigger the application of EU human rights law. That in itself does not resolve the Zambrano - McCarthy - Dereci conundrum, but it could be a basis for resolving it. In Dereci, what this would mean is that, since the Court decided that these families cannot rely on their citizenship rights, their particular case is insufficiently connected with EU law to trigger the application of their right to private and family life.
This is the critical part of my post. Fortunately, there is quite a bit of good case law, too. An example is the famous N.S. case (Joined Cases C-411/10 and C-493/10). Again this is a judgment which illustrates in what way the Member States may be bound by EU human rights law, when implementing EU legislation. The case is a fall-out from another crisis, one in Greece, and no, it is not Greece's debt crisis (though there may be a link), but rather Greece's asylum crisis. Notwithstanding the fairly comprehensive EU legislation on the common European asylum system, there are clearly great deficiencies in Greece's treatment of asylum seekers - of which there are very many, since the Greek external borders are an ideal entry point into the EU. Under the EU system, this makes Greece the Member State responsible for dealing with these asylum seekers, but the Greek system simply cannot cope, and the large majority of these people are left on the streets, without food or shelter. In January of last year, the ECtHR established that these deficiencies were such as to violate a number of Convention provisions, and found against Belgium in a case where that country wanted to return an asylum seeker to Greece.   But this raised very significant questions about the future of the European Asylum System, which is, like any other EU law, based on a principle of mutual confidence. Member States cannot, as a rule, judge whether other Member States are complying with their obligations.
The NS judgment follows references from the UK and Ireland, and concerned the question as to what the UK and Irish authorities ought to do, when in implementation of the relevant EU regulation, they are minded to return an asylum seeker to Greece. The judgment is very well reasoned. The main point is that the Court accepts that it has been established that the Greek system does not work, and violates fundamental rights, which are also protected under EU law. Member States have to respect those fundamental rights. Clearly, this is again a case where a Member State is bound by the Charter, because it is implementing EU law. Note further the extent to which the Court of Justice bases its finding of systemic deficiencies in Greece on the ECtHR judgment - see § 88.
Fortunately, the regulation also allows any Member State to exercise its discretion and to accept to examine an asylum claim, even if that Member State is not, under the normal criteria, the responsible Member State. This is what the UK and Ireland now need to do, to avoid being in breach of the Charter.
Is that it, as regards this case? Those with long memories may recall that Tony Blair left a European summit, on the Lisbon Treaty, claiming that he had obtained an opt-out from the EU Charter. There is a Treaty Protocol on that. Does that Protocol not mean that the UK is not bound by the Charter? It does not mean that, and the Court has hardly any difficulty at all to reach that conclusion. All the Protocol does, in terms, is to clarify the position of the Charter in UK law.
Let me now turn, lastly, to judicial review of EU acts. These last couple of years we have seen a remarkable willingness of the Court of Justice to review, and indeed annul, EU acts on the basis of violations of fundamental rights. There was of course Kadi I in 2008 and there are now a couple of judgments in which the Court annuls provisions in secondary legislation. The most remarkable case, and one which has been much commented upon, is Test-Achats (Case C-236/09). The Court decided, in Test-Achats, that once the EU legislature had decided to ban sex discrimination in the field of insurance services (among others), it could not provide for a derogation, potentially unlimited in time, which would enable the MS to allow differences in premiums and benefits based on sex, and resulting from the use of sex as a determining factor in the assessments of risks based on relevant and accurate actuarial and statistical data. It may well be statistically true that women cause fewer car accidents, as the principle of equal treatment aims to prohibit different treatment based on sex, one simply cannot take the sex of the person into account when calculating premiums and benefits. Although the judgment has been much criticized, I think this is the right approach, and it is easier to see that when one thinks about other forms of prohibited discrimination, such as race. We would not accept that sickness insurance premiums are different between races, on the ground of the prevalence of certain diseases within a race.
These are just some examples, taken from recent case-law, showing the extent to which the EU is becoming a human rights organization. This is certainly an area which merits further study, as the EU has its own characteristics as a human rights organization, which are in many ways very different from other international organizations concerned with the protection of human rights.

Recommended Readings