In its recent Opinion 2/13 the Luxembourg Court found that plans for the EU to accede to the ECHR are not compatible with Union law as it currently stands. This ruling has been critically received, including on this blog (see here (Lock), here (Besselink), here (Michl), here (Douglas-Scott), here (Peers) and here (O’Neill)). The immediate focus has been on how the Opinion should be evaluated as a matter of Union law and followed up inside the EU. To that effect it has been suggested that the draft accession treaty would need renegotiation, or that a text with Treaty status (a Protocol) should be added to the existing EU Treaty texts which would themselves be left intact. Are other options available too in Brussels, for example freezing accession ambitions for a while or changing existing Treaty texts?
Clearly, no matter what solution is eventually found, it will (once again) take years. This raises another important prior question: what is the Strasbourg Court likely to do until an EU accession solution 2.0, or any EU-internal alternative is in place? In particular, will Opinion 2/13, and its revealing reasoning for how the Luxembourg Court currently views the place of human rights protection in Union law and the leeway that Member States have in diverting from Union law if their ECHR obligations so require, have implications for the Strasbourg Court’s “EU approach” (see here and here for its own factsheets regarding its general approach and that in Dublin cases)? This contribution offers some first reflections on moving on in Strasbourg and Brussels.
The Opinion: first a step back
It is quite understandable that the Luxembourg Court’s ruling has met with considerable disappointment. It has been suggested that it has prioritised the protection of its own position over EU human rights protection. But, taking a step back, perhaps this time there was actually (also) a veritable case for “blaming Brussels” too. For were the instructions laid out in the Treaty and the Protocols sufficiently clear to begin with? Is it at heart at all possible to establish independent external judicial review, apparently for reason that human rights protection was felt not to be properly safeguarded in the existing set-up (article 6 TEU), without redistributing competences to the disadvantage of the Luxembourg Court (Protocol 8)? In other words, if the whole point of accession actually was to change something in the institutional design of the EU, the make-up of legal remedies and even the way in which Union law had been interpreted so far by the Luxembourg Court, why not state that more clearly from the outset? From that perspective the Opinion by the Court may be a reflection of the convoluted drafting of the EU accession instructions by the Herren der Verträge.
Looking at the state of affairs from another perspective, perhaps the fact that it is now “back to the drawing board” is also a unique second (or third) chance to ask the basic prior question of the Lisbon Treaty text: quite how can the application of an EU internal human rights document (the Charter) that the Luxembourg Court is under an obligation to apply, be combined with external judicial review by the Strasbourg Court of the EU’s (including the Luxembourg Court’s) performance with regard to the ECHR, if that ECHR and the way in which it is interpreted are themselves part (but only part) of the normative content of the Charter? This is not an easy one. A binding Charter and EU accession, it should be remembered in this context, were initially alternative solutions to “fill the EU human rights gap”. Only later did they become cumulative elements in the EU treaties, as a “solution” without a prior problem analysis justifying this double-headed approach. Yet, curiously, given the great stress accorded to EU accession so far their development has somehow remained unconnected. Additional instructions on how to dovetail their two separate logics may be unavoidable.
Then Opinion 2/13 itself. In fact, by the standards of any Court ruling, it offers a surprisingly candid, concise and (mostly) clear analysis. Some of us may not like what we read, and some cross-translation from Union law to human rights law expertise may be required to clarify its full significance, but it is extremely helpful for considering future directions in Strasbourg and Brussels. In particular the reasoning under the headings “preliminary considerations” (par. 153-177) and “the specific characteristics and the autonomy of EU law” (par. 179-200) is revealing in a number of different respects, including with regard to
- the Court’s extension of its Melloni-reasoning to ECHR Member States’ freedom to go beyond what is required by ECHR minimum norms, as well as further (unstated) implications of this reasoning with regard to interpreting the Charter and the ECHR side-by-side, and
- the Full Court’s approach to mutual trust in the EU in the light of EU Member States’ parallel ECHR commitments, and its reference to the 2011 Luxembourg Grand Chamber ruling in NS.
These two elements will be briefly highlighted below. I agree with Scheinin that in thinking about responses to the Court’s Opinion it seems more fruitful to consider these (and many other relevant) elements of the analysis rationally, in the light of the broader questions of post-Lisbon Union human rights protection architecture, rather than being stuck in disappointment for too long.
Opinion 2/13: two important elements
I. Melloni-extension and further implications for ECHR-interpretation within the field of application of Union law
In its section about the specific characteristics and the autonomy of EU law (specifically par. 187-190), the Court points to its previous 2013 ruling in Melloni. In that case the Court established that it follows from article 53 Charter that the application of national human rights standards must not compromise the level of protection provided for by the Charter or the primacy, unity and effectiveness of EU law. The problem that the Luxembourg Court sees (already currently existing, but formalised in case of an accession) is that the Strasbourg Court, on the basis of article 53 ECHR, could allow an ECHR Member State that is also an EU Member State to go beyond the ECHR minimum level of protection required, while in doing so it could suggest that there is more “policy space” for that Member State than the Luxembourg Court would itself allow under article 53 Charter.
As I have pointed out earlier on this blog (see here), Melloni can be viewed as problematic (or at least under-explained) from the viewpoint of higher national constitutional human rights standards. It may make sense from the viewpoint of protecting primacy and unity of Union law categorically to state that any national norm standing in its way should be disapplied. But the CJEU has, in the rather specific case of a national constitutional standard offering higher human rights protection, simply asserted this without giving a substantive reason for why this is a justifiable approach. Differences in human rights standards that go beyond the minimum level (and provide additional protection to the citizens of that Member State) are an essential expression of differences in priorities of Member States. It cannot be simply presumed that all of these priorities were left to one side when policy areas were transferred to the EU, or that in negotiations in Brussels they can be easily “sacrificed” without a proper discussion. For this assumes at the same time that this aspect was ever seriously discussed in national or EU fora. The Opinion 2/13 application of the Melloni-reasoning to the (international law of the) ECHR is a sweeping extension of this logic, effectively making the complete field of application of ECHR norms prisoner to some prior check of whether primacy, unity and effectiveness of EU law could be encroached. The effect is that large, because it is never quite clear in advance precisely where the normative protection offered by the ECHR ends and where national policy space then begins. The extension suffers from the same shortcoming in substantive justification.
Even more significant is the build-up to the Melloni-element of the reasoning. It reveals, although curiously implicitly, a crucial element of how the Luxembourg Court looks upon possibilities for coordination and cooperation in interpreting the ECHR in the field of application of Union law. The relevant paragraphs are 183-186, and read as follows:
- … the Court of Justice has … declared that an international agreement may affect its own powers only if the indispensable conditions for safeguarding the essential character of those powers are satisfied and, consequently, there is no adverse effect on the autonomy of the EU legal order.
- In particular, any action by [the ECtHR] must not have the effect of binding the EU and its institutions, in the exercise of their internal powers, to a particular interpretation of the rules of EU law.
- It is admittedly inherent in the very concept of external control that, on the one hand, the interpretation of the ECHR provided by the ECtHR would … be binding on the EU and its institutions, including the Court of Justice, and that, on the other, the interpretation by the Court of a right recognised by the ECtHR would not be binding on the … ECtHR.
- The same would not apply, however, with regard to the interpretation by the Court of Justice of EU law, including the Charter. In particular, it should not be possible for the ECtHR to call into question the Court’s findings in relation to the scope ratione materiae of EU law, for the purposes, in particular, of determining whether a Member State is bound by fundamental rights of the EU.
The key issue here is this: the Charter, primary Union law, has integrated and subsumed ECHR norms, but goes beyond it in some important respects. In that frame, the ECHR norms and their contents have themselves been made “rules of EU law” in the sense of par. 184. There is no doubt that the Luxembourg Court claims sole and ultimate authority to interpret any rule of Union law (par. 186), including in this case the ECHR norms being constitutive elements of the primary Union law of the Charter. Clearly, in this logic, for the Luxembourg Court the only cooperation in interpreting the ECHR and the Charter that would be at all compatible with Union law would be for the Strasbourg Court to follow the way in which the Luxembourg Court interprets the ECHR norms (as constitutive elements of the Charter), and not the other way around. All other solutions would go against the Luxembourg Court’s current understanding of the specific characteristics and the autonomy of EU law.
That appears to mean – and, if indeed the case, this is absolutely crucial – that in the view of the Luxembourg Court, it is required by Union law for it to become somehow involved in defining the content of ECHR-norms when they come into play in the field of application of Union law as they are part of interpreting Union law. On a more minimalistic, Strasbourg Court-friendly reading of these paragraphs of the Opinion it could perhaps be argued that each time it would need at least to be clearly distinguished somehow what the ECHR-element and what the Charter-add-on element is, keeping the first under the authority of the Strasbourg Court and the latter under the authority of Luxembourg. And that to establish this shared/common/joint definition, there should be some direct form of coordinating between the ECHR and Charter, and possibly between those interpreting them.
The handling of the normative overlap between the ECHR and the Charter has not been addressed in much greater detail in the Charter and its Explanations than a general indication that the Charter rights in important part originate in and reaffirm the ECHR and Strasbourg case law, and that these Charter-rights are intended to have the same meaning and (in some cases) scope than their ECHR-equivalents (Charter preamble, par. 5 and article 52(3) Charter). It is clear that, if indeed this also acquires an institutional dimension (i.e. Luxembourg claiming, on the basis of Union law, some prior involvement in interpreting ECHR-law as constitutive element of “its own” Charter), it will cause a colossal tension with the Strasbourg Court’s usual interpretative approach, e.g. considering the ECHR as a “living instrument”. For if the ECHR is a living instrument, so are the ECHR-elements of Charter-rights (for which the Luxembourg Court sees itself as the primary interpreter). This can cause various complications. For example, if an add-on aspect of the Charter has been given an interpretation by the Luxembourg Court, and the Strasbourg Court ratchets up its interpretation of an underlying and partly corresponding ECHR-norm, is the Strasbourg Court bound to follow the Luxembourg take on things with regard to the add-on? In such cases the question becomes: who is to follow whose interpretation?
In the view of the Court, this – as well as the Melloni-point – needs to be addressed. How that can be done is not answered by it, but needs to be considered in Strasbourg and Brussels.
II. Mutual trust and ECHR obligations of EU Member States
In par. 191-195 of the Opinion the Court provides its views on the importance of mutual trust at the heart of many rules of Union law in relation to ECHR obligations held separately by Member States. EU Member States may be required to presume that other EU Member States comply with fundamental rights, and are then legally prevented from checking whether the partner EU Member State has observed them (par. 192). Given that through EU membership these Member States have accepted a transfer of powers to the EU, they must also accept that in these areas their affairs “are governed by EU law to the exclusion, if EU law so requires, of any other law” (par. 193) (including, clearly, ECHR law). If the ECHR would require any different approach from Member States, e.g. on occasion checking whether another Member State really has observed fundamental rights to give effect to its positive obligations under the ECHR, such a requirement would be “liable to upset the underlying balance of the EU and undermine the autonomy of EU law” (par. 194).
Two points are particularly interesting here.
First, one would want carefully to compare the reasoning of the Full Court in this Opinion and that of the Grand Chamber in NS – both from the viewpoint of what is now the state of Union law and from the viewpoint of the Strasbourg Court by way of analysis of how its Grand Chamber MSS– and Tarakhel-rulings are being followed up in the context of Union law. The Luxembourg Grand Chamber had stressed in December 2011 that the Dublin Regulation contained a “rebuttable” presumption of compliance with human rights (par. 106), because “to require a conclusive presumption of compliance, [the Regulation at issue] could itself be regarded as undermining the safeguards which are intended to ensure compliance with fundamental rights by the EU and its Member States” (par. 100). The presumption would be rebutted if, like in NS itself, one Member State whose control individuals are under at a certain moment and who has to decide whether to comply with EU law requirements to transfer these individuals cannot be unaware of systemic failures/deficiencies in the fundamental rights protection provided by the other EU Member State (par. 86 and 89). The Full Court never uses the notion of “rebuttable presumption” or “systemic flaws/deficiencies”, but rather states twice (par. 191 and 192 – and notably no longer in the concluding paragraph 194) that EU Member States are required to consider other Member States to be in compliance with EU law and (EU) human rights, “save in exceptional circumstances”. Is this approaching the same issue from a different side (mutual trust), using different wording? Or is this an evolution in the thinking of the Luxembourg Court in the direction of giving more weight to mutual recognition/mutual trust as compared to human rights in Union law? In this context it also becomes relevant that, as observed by Scheinin, Judge Rosas, the rapporteur in NS, apparently opted to stay outside of the Full Court that delivered Opinion 2/13 even after participating in the hearing, possibly – Scheinin speculates – for lack of agreement with the approach taken.
Second, the view taken with regard to the relative importance of mutual trust here must be understood in the light of the more general exposé given by the Court about the constitutional order of the Union in the “preliminary considerations”. In par. 168 it states that the notion of mutual trust follows from the premise that the EU is based on shared values (article 2 TEU), which “implies and justifies” the existence of mutual trust in legal (constitutional?) terms. The Charter is “also” at the heart of the constitutional structure (par. 169). The interpretation of the rights therein needs to be ensured “within the framework of the structure and objectives of the EU” (par. 170). These objectives, aimed at achieving the implementation of the integration process that is the raison d’être of the EU itself, concern those providing for free movement of goods, services, capital and persons, Union citizenship, the area of freedom, security and justice and competition policy (par. 172). The Charter rights “must therefore be interpreted and applied within the EU in accordance with [this] constitutional framework” (par. 177). In other words, the Court remains somewhat vague about the relative importance of the various different constitutional aims, but appears to suggest that they all have equal value. The fact that the Charter has been given primary Union law status does not appear to mean for the Court that something has been changed in this respect.
This outcome may have been perfectly predictable for Union law experts. But some of the disappointment with the outcome in the Opinion may lie in the fact that human rights/ECHR experts tend to operate from a rather different notion, i.e. that human rights protection actually functions to correct and shape other constitutional principles in whatever way they are given effect. For example, if there is a tension between human rights and a rule that gives effect to the constitutional principle of mutual trust, the view of human rights experts will likely be that the application of that latter rule will have to be brought in line with human rights because these form minimum rules underlying any form of cooperation. In other words: human rights trump other constitutional values in so far as their application comes into tension with them. It appears that the Luxembourg Court simply sees this in a fundamentally different way.
Opinion 2/13, the ECHR and the Strasbourg Court
Opinion 2/13 gives a fairly unambiguous summary of how the Luxembourg Court views the relevance of human rights protection in Union law generally, of the place to be awarded to in it by the ECHR and its interpretation by the Strasbourg Court, and the role that EU Member States’ ECHR obligations can play in implementing rules of Union law that are structured on the basis of mutual recognition. This is hugely beneficial. From all of these perspectives the Opinion must also make for a very interesting read in Strasbourg. At this stage the question has become whether Opinion 2/13, or the Luxembourg Court’s future human rights interpretation based on it, will amount to a reason for the Strasbourg Court
- to reconsider what has been de facto marginal supervision with regard to EU Member States if they implement EU law in cases where they do not have discretion (known as the “presumption of equivalent protection” developed in Bosphorus), and
- to reconsider or refine its approach with regard to EU Member States in other situations, such as where the underlying Union law rules are premised on mutual recognition (including in “Dublin cases”)
These are of course large and fundamental questions. The current contribution will provide but some thoughts on how to go about considering this issue and a few initial observations.
Re-reading Bosphorus at this time is interesting, particularly the section in which the Strasbourg Court set out its general approach (par. 149-158). It held that, on the one hand, the ECHR does not prohibit Member States from transferring sovereign powers to a supranational organisation (par. 152). On the other, Member States remain responsible for all acts and omissions of organs of such an organisation regardless of whether that act or omission results from domestic law or the necessity of complying with international legal obligations (such as Union law) – the jurisdiction to scrutinise any act under the ECHR remains in place regardless of the origin of the legal obligations on the Members (par. 153). Therefore absolving Members completely from their ECHR obligations in areas transferred to a supranational organisation would be incompatible with the object and purpose of the ECHR, as otherwise commitments under the ECHR could be limited or excluded at will at the cost of the ECHR’s peremptory character (par. 154). More particularly, Members can only transfer powers in line with ECHR obligations as long as that organisation is considered, as regards both substantive guarantees offered and mechanisms controlling their observance, in a manner which can be considered at least equivalent to that for which the ECHR provides. With regard to the substantive guarantees the analysis only pointed out how the ECHR was given special significance in the Luxembourg Court’s case-law, and how the ECHR was established as the minimum level of protection vis-à-vis many Charter rights based on it (par. 159). With regard to protection mechanisms the Strasbourg Court summarised the Union law system of legal remedies (par. 161-164). Yet, any such finding of equivalence would be relevant only in cases where Members would not be left any discretion by the commitments coming from the international organisation (par. 157) and would remain susceptible to review in the light of any relevant change in fundamental rights protection (par. 156). In that way, the presumption of equivalent protection would remain rebuttable, so that if and when the protection of ECHR rights in the context of the supranational organisation became manifestly deficient the ECHR’s role as constitutional instrument of European public order would regain precedence (par. 157).
So how does Opinion 2/13 look through the lens of the Bosphorus-reasoning? As was pointed out earlier, it appears that the Melloni-reasoning of the Luxembourg Court, and particularly some implicit elements at the heart of that position which were also inferred from the specific characteristics and autonomy of Union law, could have far-reaching consequences for interpretation of ECHR norms by the Strasbourg Court in the field of application of Union law. Already at first sight this would appear to go to both the substantive and procedural leg of the equivalence-test. Substantively, has the fact that the Charter has subsumed ECHR norms (something that was applauded still in Bosphorus, par. 159) and has made their contents de facto Union law (something that the Strasbourg Court clearly did not take into account) changed something in the set-up on which the Strasbourg Court based its reasoning in 2005? Is it a requirement of ECHR law that ECHR norms are interpreted recognisably and as ECHR norms by its Members and organs of supranational organisations that these Members have decided to set up? Procedurally, does it matter that the Luxembourg Court appears to claim sole interpretative authority for interpreting these ECHR-norms now encapsulated by Union law, or (less plausibly from the viewpoint of the Luxembourg Court’s reasoning) at least some co-interpretative role without accepting the Strasbourg Court’s word as binding as a matter of Union law? Does that stance of the Luxembourg Court have implications for the Strasbourg Court’s interpretative approach with regard to the ECHR as a living instrument, and could that have implications for its self-perception as the sole and final authority as ECHR interpreter? These issues are highly complex, and deserve thorough scrutiny from the viewpoint of protecting the ECHR.
The reasoning in Opinion 2/13 about mutual trust probably puts rather a different perspective for the Strasbourg Court on how it has so far dealt with “Dublin cases”. Was it aware that the Luxembourg Court had such a strong opinion on the sole relevance of Union law in these cases, effectively requiring them to act in a way that is reflective of the fact that these issues “are governed by EU law to the exclusion, if EU law so requires, of any other law” (including, therefore, ECHR law) (Opinion, par. 193)? Should this have implications for the way in which the Strasbourg Court formulates findings of possible violations by Member States, perhaps also addressing these more clearly to EU institutions such as the EU legislator and the Luxembourg Court itself? How, in other words, will the Strasbourg Court go about stressing its Bosphorus-viewpoint that ECHR obligations remain relevant for EU Members at any stage, including in cases where they have discretion in implementing Union law? Again, these are highly explosive issues for which there is not a beginning of an answer. But it seems clear that the Strasbourg Court may have to act on this.
If the conclusion in Strasbourg were to be that the substantive and/or procedural leg of Union-based ECHR-protection no longer suffice, it would not seem unreasonable for the Strasbourg Court to describe its “equivalent protection”-line as something outdated, (strategically) put into place at the time as a temporary and transitional measure awaiting effective EU accession to the ECHR. Similarly, it is not unimaginable for it to take Opinion 2/13 as a basis for changing the way that it approaches EU Member States in Dublin cases. For in both cases there appears to be nothing inherent in the set-up of the EU that would justify a conclusion that, in contrast to other settings, policy developed in it would somehow generally be less likely to come into tension with the ECHR. Perhaps Opinion 2/13 could even lead the Strasbourg Court to revisit its position according to which it has “recognised the growing importance of international cooperation and of the consequent need to secure the proper functioning of international organisations” as “such considerations are critical for a supranational organization such as” the EU (Bosphorus, par. 150) – and more precisely lay out the conditions under which such international/supranational cooperation is acceptable from the viewpoint of ECHR protection for which it carries final responsibility.
Some options in Brussels not previously mentioned
Back to Brussels. Lock and Besselink have both proposed concrete measures to be taken by the Brussels institutions. Lock argues that the draft accession treaty would need to be amended. He appears to assume that no EU Treaty-change is needed for that, but acknowledges that non-EU Council of Europe Member States may be less than enthusiastic about this. Besselink proposes a Protocol along the lines that the EU should accede to the ECHR notwithstanding the Court’s objections. This would effectively add an extra layer to the instructions by the Masters of the Treaties to clarify the contradictory instructions given by them earlier (without, however, removing the underlying tension). In my view these options are probably not mutually exclusive, as Besselink appears to suggest, but may have to be developed at different points in time. It is not impossible that if indeed it was decided to add a Protocol along the lines suggested, there would still be a need felt to adapt the draft accession treaty, perhaps at the request of non-EU States. Importantly, both these suggestions do not provide an answer to the prior question mentioned above about how their solutions would fit in the post-Lisbon human rights architecture that has incorporated both the Charter and accession route that were originally developed as alternatives. Moreover, both proposals are premised on the notion that efforts to accede to the ECHR should be continued immediately. The implications of their proposals could, in my view, therefore only be well appreciated once it is explained how they would fit in this “bigger post-Lisbon picture”.
An equally plausible line, that would at least be worth exploring seriously, could be actually first to focus on creating more elaborate EU-internal policy with regard to protecting the EU Charter, and only then see in what form and for what (remaining) reason(s) EU accession could still add value to EU human rights protection. I have pointed out elsewhere the anomaly of focusing on designing external supervision before setting up, developing and somehow evaluating an internal (in the EU-case, Charter-based) system of human rights protection (see my contribution “Kissing awake a sleeping beauty? The Charter of Fundamental Rights in EU and Member States’ policy practice” in this recent book). This may in fact have been precisely the Luxembourg Court’s unstated primary aim with its Opinion: pointing out that there may be a contradiction in the Lisbon architecture and buying time to establish a clear(er) line with regard to how it applies “its own” Charter. Importantly, this option to prioritise the Charter for now may also allow the Strasbourg Court, if it so sees fit, to re-position itself with regard to the Luxembourg Court and the EU more generally (see above).
Another alternative, or cumulative, avenue would be to amend the existing Treaty text or that of the Charter. Such change would be aimed not so much at facilitating EU accession at any cost, but at clarifying the desired status of the ECHR as a matter of Union law. Given the stress placed by the Luxembourg Court on the fact that the ECHR is not currently formally directly binding on it as a matter of Union law (par. 179/180), a possible alternative to formal accession (that would also be very helpful in clarifying how the Charter should be interpreted) could be to insert an explicit instruction (including for the Court) in article 6 TEU and/or the Charter and/or the Charter’s Explanations to the effect that the Charter and Union law general principles can only be given meaning by explicitly referring to (and taking on board the substantive content of) the ECHR and the Strasbourg Court’s interpretation of it. It would be a “legal upgrade” of some language in the Charter’s preamble and in its article 52(3), clarifying that, as a matter of Union law, the ECHR and its interpretation by the Strasbourg Court should be leading. In that way the underlying rationale of EU accession to the ECHR, which was to ensure that the combination of the ECHR and its interpretation by the Strasbourg Court would also apply to the EU, could be achieved indirectly in a manner that is binding in terms of Union law. It would protect and empower the Strasbourg Court, subject the Luxembourg Court to it de facto but not formally (and avoid putting in place all sorts of complex and time-consuming coordination mechanisms).
John Morijn contributes in a personal capacity; the opinions expressed cannot in any way be attributed to the Dutch government.