This contribution looks at the German Federal Constitutional Court’s PSPP judgment and its scrutiny of the methods of interpretation of EU law by the Court of Justice of the EU. Notably, the GFCC introduced a new Solange: its intention to accept the CJEU’s methodology as long as the latter does not arbitrarily disregard the principles common to national constitutional traditions. In discussing this “methodological Solange”, I will first show why the German court taking ever greater ownership in the interpretation of EU law, without engaging the CJEU and while circumscribing the preliminary rulings avenue, is the emerging “new normal”. Then I will describe how in several previous occasions the national courts’ concerns regarding the methodology employed by the Court of Justice have led the latter to gradually improve its reasoning, and how the PSPP piece might fit into this puzzle.
- WHAT ABOUT METHOD?
The story is already well-known: the GFCC in PSPP took issue with the CJEU’s methodology in Weiss. More specifically, the flaws in the proportionality assessment performed in the latter made it ultra vires in Germany. The GFCC thus repeatedly treated the CJEU’s method with words like “objectively arbitrary” (para. 118), “untenable” (para. 119), “meaningless” (para. 127), and “incomprehensible” (para. 153). This was so due to a lax scrutiny with which the CJEU approached the review of the European Central Bank’s decision, which in turn, the GFCC added, was plagued by insufficiencies in its assessment of economic and monetary policy consequences of the bonds purchases program.
However, in reaching that decision, the Karlsruhe judges were accused (illustratively, here and here) of seriously misreading the EU law against which they assessed the ECB’s decision and the Luxembourg court’s judgment reviewing it. Notably, they failed to interpret appropriately Articles 5 and 19 of the TEU. Here, two key points emerge.
First, the ultra vires review was not entirely based in the German Basic Law. One would expect this given the earlier jurisprudence of the GFCC. Rather, the ultra vires review was in big part based in EU law – as interpreted by the GFCC itself. This was routinely emphasized by the commentators. So, as Gareth Davies notes, it is
“less an attempt to keep the EU out, than to shape it in a certain image. That may be why it is so controversial; in a club of many members, it is more offensive for one to tell the others how it should be run, than for that member to simply turn their back”.
But, can a national court ever legitimately do that: lecture not only its peers but the chairman of the club too on how the club rules should be understood and followed?
Second, the CJEU’s judgment was deemed ultra vires due to its methodological shortcomings. But, can a national court ever legitimately police a Court of Justice’s judgment in that manner?
- WHO CAN SAY WHAT?
The GFCC in PSPP opens with the acknowledgment that the Article 19 TEU mandate – to ensure that in the interpretation and application of EU law the law is observed – primarily refers to the CJEU (para. 112). Primarily, but does that mean exclusively?
The GFCC clearly thinks that it does not. Albeit this reading of Article 19 TEU would go beyond the text of that provision, it would be in line with the CJEU’s pronouncement in ASJP (para. 33), which emphasizes that the EU court and national courts have a joint duty in carrying out this mandate. Pushing this dicta to its limit, and arguably against any conceivable intention of the CJEU to allow national courts to go against its holdings, the GFCC has recently made a couple of unguided excursions into the field of authoritative interpretation of EU law. Two examples are indicative of it taking ever-greater ownership in the interpretation of EU law (as it is applied in Germany, obviously), to the detriment of the CJEU’s institutional position under Article 267 TFEU and its division of labor with national courts under the preliminary rulings procedure.
In 2015, the GFCC issued an order on the European Arrest Warrant, in which it declared its intention to review the application of EU law in Germany for its compliance with human dignity under Article 1 of the Basic Law. At the same time, however, the GFCC interpreted away the conflict between EU law and human dignity guarantees under the German constitution. In doing so, it tried to present the matters of EU law to be “so obvious as to leave no scope for any reasonable doubt” (para. 125) – in other words, “acte clair” – in order to exonerate itself from the obligation to refer preliminary question to Luxembourg. How “clair” things indeed were remained highly doubtful.
In a recent case from 2019 – Right to be forgotten II – the GFCC dealt with a constitutional complaint by applying, for the first time ever, the EU Charter of Fundamental Rights as the relevant standard of review of domestic application of harmonized EU law. Asserting a novel jurisdiction in this way, the GFCC promoted itself to a role of “co-curator of the EU Charter, alongside the CJEU”. In a way, it arrogates the competence to authoritatively interpret the Charter rights in domestic situations where the CJEU, due to procedural reasons remains uninvolved (e.g. where lower courts do not refer preliminary questions). And albeit expressing its fidelity to a close cooperation with Luxembourg via Article 267 TFEU in such circumstances, the Karlsruhe judges here again dared to conclude that the matter of EU law is “clair”; yet again, it remained unclear whether the CILFIT criteria on what makes an issue “clair” were properly interpreted.
Now, with the bold move that followed in PSPP, the GFCC is inevitably contesting the CJEU’s exclusive jurisdiction under Article 19 TEU. As such, this judgment considerably differs from earlier decisions of high national court in cases like Ajos or Landtovà. Therefore, by “stepping into the shoes of the Luxembourg judge”, the Karlsruhe court is straightforwardly “undermining the Court of Justice’s role as the supreme interpreter of EU law”.
- METHODOLOGICAL SOLANGE
The GFCC in PSPP also acknowledges that the Article 19 TEU mandate covers “the methodological standards for the judicial development of the law” (para. 112). So, there is an EU method of interpretation that is by and large constructed in Luxembourg. But this method is constructed with an eye on the common constitutional traditions of the Member States and jurisprudence of the highest European courts, similarly to other general principles of EU law. The GFCC agrees that that method does not have to correspond to national jurisprudential traditions in a 1-to-1 manner. After all, “the particularities of EU law give rise to considerable differences with regard to the importance and weight accorded to the various means of interpretation” (para. 112). For this reason, national courts ought not to substitute the CJEU’s interpretations of substantive EU law with their own, when a given interpretation stays within the boundaries of accepted methodology. When the law is indeterminate and open for several reasonable interpretations, the GFCC does not reject the CJEU’s interpretation simply because it favors a different one.
Yet the situation is quite different when the CJEU’s interpretation (in the sense of the outcome of the interpretive process) does not square with a reasonable interpretation (in the sense of the interpretive process itself). Granted, the EU method of interpretation is constructed independently. But in doing so, the CJEU cannot “simply disregard” the national jurisprudential traditions. Surely it can make a mistake, and the German court grants it “a certain margin of error”. Because judges are humans, and even Luxembourg judges are humans, and humans make mistakes. And a fair share of ill-reasoned judgments did come out of the Luxembourg benches. However, it cannot construct the meaning of an EU norm out of thin air or on a whim. In that sense, the Article 19 TEU mandate “is exceeded where the traditional European methods of interpretation or, more broadly, the general legal principles that are common to the laws of Member States are manifestly disregarded” (para. 112). And then the GFCC concludes with the methodological Solange:
“as long as the CJEU applies recognized methodological principles and the decision it renders is not objectively arbitrary from an objective perspective, the Federal Constitutional Court must respect the decision of the CJEU even when it adopts a view against which weighty arguments could be made”.
What all this means is that what the CJEU got wrong in Weiss is not that it failed to honor German constitutional standards of what makes an appropriate method of interpretation. Rather, it failed its own standards. And this is the “manifest error” in method that made it ultra vires. It failed the EU standards – as the GFCC understands them.
With this, the GFCC takes even greater ownership in the interpretation of EU law than suggested above. This time this is not at the level of substantive rules. Rather, the German court takes ownership over (in Herbert Hart’s parlance) the secondary rules, which are “the rules about rules”; here, the rules of interpretation. For this, the challenge in PSPP is much greater than in Right to be forgotten II. It not only contests the “judicial autonomy” of the Luxembourg court in the EU constitutional realm – that is, “the sole power to state the right answer to a specific case”. Now it contests “methodological autonomy” of the CJEU, which ensures that “the means to arrive to such answer cannot be contested”.
- IS YOU IS OR IS YOU AIN’T?
Discussions about legal methodology led by national courts and the CJEU need some context.
First, it is difficult to deny that methods of interpretation, and techniques of legal argumentation more broadly, employed by the CJEU are essentially the same ones that national courts know and use. True, different courts may differ in their preferred interpretive approaches, like the pragmatic, functional interpretation the CJEU champions, with the effectiveness of EU rules at the center stage, as opposed to dogmatic (in a non-pejorative sense) interpretation often followed by the GFCC. This may cause serious frictions between the opposing approaches. And no court is completely unbiased and impartial as to its preference, which may lead them to distrust each other. After all, no judicial method is ever apolitical. But at least here the interlocutors speak the same language. The situation is different when one side engages in whimsical and out-of-nowhere interpretive moves. Then the language is not shared anymore. And for this the CJEU has long been accused. Remember Herzog and Gerken’s “Stop the European Court of Justice” and their (in)famous criticism of the Luxembourg court’s flawed methodology in cases like Mangold?
In PSPP, the Karlsruhe court similarly claims not to understand the CJEU anymore. It expressed concerns already in Gauweiler, albeit it stopped short of pulling the trigger; and the entire strand of the CJEU’s case law regarding the euro-crisis has been considered as over-stretching the recognized interpretive methods and being unprincipled on the matters of principle. This led to reaction in PSPP, after the preliminary reference in Weiss showed that the CJEU intends no change.
The GFCC suggested further that the judicial “language” is mutually constructed by the courts in the EU. If the rules of interpretation reflect the shared constitutional traditions, national courts have their share in constructing them. For these unexpressed “meta-rules”, in this view, the Article 19 TEU mandate is a shared task. A similar idea – that in a discursive development of the unexpressed general principles of EU law – both structural, like proportionality, and substantive, like fundamental rights – the task is shared between the EU court and national courts – appeared long ago but somehow got lost. After PSPP, it might be picked up again.
Second, national courts contesting the methodology and reasoning of the Luxembourg court’s judgments is nothing new or unique, I believe. This seems to be the way EU law has always been developed. Textbook examples of classical doctrines on general principles of EU law may illustrate the point. Recall the structural principle of (vertical) direct effect of directives, where the CJEU initially offered only the argument of effectiveness (effet utile) (van Duyn, para. 12), after which, faced with unimpressed national courts, it subsequently introduced another argument, the estoppel principle (Ratti, para. 22). Or, the substantive principle of fundamental rights protection as developed originally in the 1970s: in the earliest judgments, the CJEU stuck to the “aprioristic” approach of merely “discovering” pre‐existing fundamental-rights-as-principles. However, soon facing challenges from national constitutional courts, especially the GFCC in Solange I, it changed its approach of interpreting EU fundamental rights to a more “positivist” one. Therefore, it gradually introduced and refined the sources out of which EU rights have been constructed: first “common constitutional traditions”, and then “international human rights treaties”, in particular the ECHR. It was primarily after these methodological developments that the GFCC stepped back in Solange II with a familiar outro: as long as the EU generally retains an effective protection of human rights, substantively similar to German constitutional standards – “effective” and “similar” as interpreted and understood by the GFCC itself – we yield.
The same would go under the methodological Solange in PSPP: as long as the CJEU applies appropriate interpretive methods – “appropriate” as interpreted and understood by the GFCC itself – we yield. Perhaps the German court will in the future be celebrated for this, as it deservedly was for its Solange I & II push against the unbridled supremacy of EU law and for an increase in the standard of fundamental rights protection in the Union, as Bobić and Dawson remarked.
This leads to the third and final point. If the CJEU depends on national courts in order to uphold a workable and efficient system, and it itself emphasizes this regularly, then national courts accepting its interpretations of what the EU law means is essential. To have them accept those interpretations, in big part, hinges on the quality and persuasiveness of the reasoning behind those interpretations. On this depends the very authority of the EU apex court. Unfortunately, a discontent with poor reasoning of the Luxembourg court has been boiling for a while now. Simple, self-referential, and wannabe-authoritative “because I say so” reasoning is not (and probably never was) enough.
The institutional development of national court’s increased ownership in the matters of interpretation of EU law seems inevitable as the process of EU integration matures. This might be a hard version of the long-advocated “interpretive pluralism”. Still, this may sound unthinkable for the many; even blasphemous, having the uniform application of EU law in mind. But, for others, “less uniformity does not necessarily entail complete disintegration”. Situations of legal divergence like these are merely a mark of common law heritage of the EU legal order; something not to lament but celebrate.
The intervention of national courts on the level of rules of interpretation of EU law, demanding more from those in the position of authority and calling them to order when defaulting, is likewise not so outrageous. The rule of law is hardly breached when a check on arbitrary judicial decision-making is introduced by a greater demand for justification. This is what the rule of law – in contrast with the rule of courts or men – means in a pluralist, democratic, constitutional community. Why not read the GFCC’s PSPP judgment in this spirit, as “a desperate cry for more methodological integrity” in the EU judicial exchanges? If the reading of that spirit is the right one, it seems right. Like many times before, it might lead to a better and more consistent reasoning of the CJEU, to everyone’s satisfaction.