May 11
Anjum Shabbir
Anjum Shabbir
29th April 2021
Institutional law

Op-Ed: “The French Data Network Judgment : A ‘Securitarian Frexit’ or Classic Conseil d’État Euroscepticism?” by Araceli Turmo

On 21 April 2021, the French Conseil d’État (Council of State) seemed to choose a pragmatic ‘third way’ in its judgment following the Court of Justice’s preliminary ruling in La Quadrature du Net. By agreeing to carry out a judicial review on the basis of the terms established by the Court,  but also refusing to commit to full compliance like the Belgian Constitutional Court, the Conseil d’État may appear to have reached a balanced solution reconciling the requirements of national security with the protection of personal data, as well as the push for a nationalist protection of the State’s interests with a willingness to respect the primacy of EU law and engage in judicial dialogue. In this piece, I will not restate the results of this approach – both the French and Belgian rulings are explained in an excellent Insight published here on 26 April. Instead, I will focus on the broader implications of the Conseil d’État’s judgment in the context of the dialogue between the constitutional courts of Member States and the Court of Justice. After all, as the Rapporteur public reminds us in the Conclusions delivered in this case, the Conseil d’État is a constitutional court.

The French Data Network ruling is a prime example of an attitude towards EU law which has become typical of the Conseil d’État in recent years: although wanting to be seen as a ‘model student’ of European integration and definitely understanding itself as such, the Conseil continues to takes liberties with core aspects of EU law and persists in asserting its own authority and a view of EU law which is peculiarly specific to French administrative lawyers. Its positions, when read from the point of view of EU law, in fact place it far behind the constructive attitude of the Italian Corte Costituzionale as evidenced in Consob, or even the ‘pragmatic resignation’ of the Spanish Tribunal Constitucional (see Daniel Sarmiento’s recent study in Weekend Edition No. 54). The contrast between the Conseil d’État’s own perception of its own case law as entirely compliant with EU requirements and what is actually the case, from the point of view of EU law, became painfully clear when the President of the Litigation Section of the Court, Jean-Denis Combrexelle, seemed both surprised and offended by the infringement finding in Case C-416/17, which was partly based on a violation of the Conseil d’État’s obligation to refer under Article 267 TFEU. In a short note published soon after the Court’s judgment (AJDA 2018, p. 1929), he explained his view of judicial dialogue and of the obligations of national supreme courts with respect to Article 267, apparently unaware that what he was presenting as the consensus on the applicable law was, in fact, clearly incompatible with the CILFIT criteria.

It is undoubtedly true that the Conseil d’État’s relationship with European law has improved greatly in recent years. In just two decades, it went from actively resisting EU law and refusing to recognise core principles such as the full scope of direct effect, to introducing the Francovich principle into French law (CE Ass., 8 February 2007, Gardedieu) and accepting the authority of the Court of Justice’s interpretation of EU law even where no preliminary reference is made (CE Ass., 11 December 2006, Sté De Groot). This welcome change was simply a (long-overdue) adjustment to the fact of membership of the EU, which has been enshrined in the French Constitution since 1992. However, rather than perceiving this case law as partially ‘catching up’ to other national supreme and constitutional courts and as an ongoing process which requires further efforts and constant attention, the Conseil d’État seems to believe that the changes made in the 2000s are enough and it is holding on to its very own understanding of EU constitutional law which, although less openly confrontational than that expressed by the BVerfG in recent years, is not necessarily less harmful.

In the French Data Network Case, the Conclusions of the Rapporteur public, M. Lallet, provide a perfect illustration of these issues as well as offering a welcome insight into what may have motivated the Conseil d’État’s ruling. First, it is indeed true that the Conseil d’État rejected the Government’s extravagant request that it hold the Court of Justice’s judgment in La Quadrature du Net to be ultra vires. However, I do not believe we should read too much into this refusal to check the compatibility of the Court of Justice’s judgment with the EU’s competences. The Conseil d’État does not provide much detail as to the motivations behind this decision, besides a willingness to preserve the binding force of the judgments of the Court of Justice. In his Conclusions, M. Lallet suggested that the Conseil d’État should not carry out this ultra vires check, not because they are ‘gentlemen of the law’  (p. 25 of the Conclusions), nor because it was not competent to do so as a ‘constitutional court’ itself, nor because it is the ‘prisoner’ of an ‘irrepressible binding force’ of the judgments of the Court of Justice, but because it did not need to. The Rapporteur writes that the Conseil d’État can determine for itself the scope of a precedent from the Court of Justice, as well as the extent of its authority on itself and on the French administrative authorities, since the Court of Justice is ‘not [their] Supreme Court’ (p. 26). In essence, his position is that judgments of the Court of Justice only have as much authority over French administrative law as the Conseil d’État is willing to give them. One cannot assume that the Conseil d’État agrees and that this was the reasoning behind its ruling, but it cannot be excluded either. If this is the position of the Conseil d’État, it is alarming. Not only does the Rapporteur refuse to engage with the concept of judge-made law and with the idea that case law of the Court of Justice is part of EU law, and has the same authority over all national courts, but he seems to be unaware that to hold such views is to actively resist the proper application of EU law.

Based on this premise, the Rapporteur public added that a BVerfG-style ultra vires reasoning would be difficult – and risky – to introduce, but suggested that it would be preferable to rely on the Conseil d’État’s previously established safeguard. The Conseil d’État followed this suggestion, relying on its famous judgment in Arcelor. However, this safeguard, based on constitutional identity and Article 4(2) TEU, had so far been understood to mean that fundamental rights which are not sufficiently protected at the EU level could justify constitutionality reviews of national rules implementing EU law by national courts such as the Conseil d’État. Under this rule, similar to those established in other Member States, the Conseil d’État and the Conseil constitutionnel will not check whether acts implementing EU law are constitutional if a similar principle exists within the EU legal order: the review will be carried out by the Court of Justice following a preliminary reference, unless the level of protection granted in EU law is not ‘equivalent’ to that which exists in French law.

Here, however, the Arcelor ‘safeguard clause’ is relied upon to achieve the exact opposite: to reduce the level of fundamental rights protection granted by the law. According to this new judgment, where the Government (and the Conseil d’État) believe that EU law provides an excessively high level of protection of fundamental rights, the Conseil must be able to check whether provisions based on EU law are compatible with national public order security requirements. Luckily, in this case, the Conseil d’État did not believe that the EU rules at issue were fundamentally incompatible with the national standards it was using, although it did make sure to check whether this was the case for each of the provisions. However, this is a very worrying precedent, as Shahin Vallée and Paul Cassia have already explained. First, the Conseil d’État’s conclusions as to the compatibility of French law with EU law are not convincing and can, in fact, be read as a further violation of EU law. The contrast with the Belgian Constitutional Court’s decision would seem to indicate that this is the case.

Second, the reasons why the Conseil d’État felt that it could carry out this constitutional review and the way in which it chose to do so are worrisome, both from a European and a national perspective. The Conseil d’État seems to have abandoned the idea that the constitutional identity safeguard should be based on particularly important or country-specific elements of constitutional law. Following the Rapporteur public (p. 31 of the Conclusions), the Conseil d’État extends this to any constitutional norm (paragraphs 6-7). Moreover, as Shahin Vallée and his co-author write, the constitutional norms invoked here are extremely broad and ill-defined. The concept of ‘objectif à valeur constitutionnelle’ (objective of a constitutional value) is highly disputed and, according to most authors, does not have the same authority as the rights and liberties enshrined in the Constitution. It has been applied to aims as varied in scope and importance as the safeguard of public order and the financial transparency of press organisations. Relying on such concepts in order to reject the primacy of EU law because it gives too much protection to fundamental rights is surprising, to say the least, and could set a dangerous precedent.

Moreover, the Conseil d’État justifies this constitutional review by the idea that the ‘fundamental interests of the Nation’ such as public order or the fight against terrorism are not sufficiently protected in EU law (paragraph 10 of the judgment). Once again, this laconic phrase can perhaps be better understood by referring to the Conclusions. The Rapporteur public is extremely dismissive of the Court of Justice’s case law on data protection. His Conclusions give the impression that the Court of Justice is populated by naive hippies, who share the views of NGOs such as La Quadrature du Net and are blissfully unaware of the dangers to public safety and the needs of the police and intelligence services. He writes that ‘for theoretical and psychological reasons’, the Court is too mistrustful of public authorities and that EU law is ‘structurally’ unable to grant public security and the fight against crime the same importance as national law (p. 16 and 45).

The Conclusions and the judgment give no regard to the work accomplished by the Court of Justice in order to find a balance between those very requirements and the fundamental right at issue. Instead, we are to believe that the true test of the compatibility of data protection with security interests is to be carried out at the national level, on the basis of vague statements such as ‘France is particularly exposed to the risk of espionage and foreign interference’ (paragraph 44). The Conseil d’État does not even try to establish its own balance between the principles at issue. The provisions are reviewed on the basis of the Court of Justice’s judgment, and the result is then reviewed on the basis of public order and the other related ‘objectives’ raised by the Government. It is as if the Conseil d’État believes that, by nature, EU institutions cannot comprehend the need for public order and security and that, for this reason, all the work done in the Court of Justice’s case law was meaningless from the point of view of the corresponding national standards. The new European Prosecutors will certainly be happy to read this.

At a time when the temptation to resist European fundamental rights standards is growing in a number of Member States, the Conseil d’État has set a dangerous precedent. Judging from its previous case law and the Conclusions delivered by the Rapporteur public, its members are probably only partly aware of the issues the judgment poses form the point of view of EU law and will probably have thought that the compromise reached in the results of the review of the provisions at issue was a sufficient sign of goodwill. However, the ratio of this judgment, especially when read in the light of the Conclusions, is worrying both in terms of the precedent it sets for other courts by stating that general national security interests justify a constitutional review of norms based on EU law, and of the persistence of the Conseil’s unwillingness to engage with EU constitutional law as it is currently understood by EU lawyers as well as by (some) other constitutional and supreme courts. This was probably not intended as a ‘securitarian Frexit’ (according to the phrase coined by La Quadrature du Net and quoted by Paul Cassia in the post quoted above), but this realisation should not be reassuring. The fact that the Conseil d’État persists in constructing its own version of EU law which gives priority to the interests of the French State, all the while believing that it is a perfect practitioner of judicial dialogue, is not a good sign for the future of its relations with the Court of Justice.


Araceli Turmo is Senior Lecturer in EU Law at the University of Nantes. Her recent publications include ‘Immutabilité des actes juridictionnels et droit de l’Union européenne: poursuite de l’exploration des mécanismes nationaux de révision et de réexamen par la Cour de justice’, RAE (2019) 3, and ‘A Dialogue of Unequals – The European Court of Justice Reasserts National Courts’ Obligations under Article 267(3) TFEU’, EuConst (2019) 340.

Araceli is also a member of EU Law Live’s Editorial Board.



Your privacy is important for us

We use cookies to improve the user experience. Please review privacy preferences.

Accept all Settings

Check our privacy policy and cookies policy.