Constitutional Primacy and the Limits of EU Law

https://doi.org/10.63189/ATOV2688

Constitutional courts across the EU have increasingly asserted that, in matters touching “constitutional identity”, national constitutions rank above even the case‐law of the CJEU. The Treaties contain no explicit supremacy clause. Instead, the principle that EU acts prevail is a creation of case-law (Costa/ENEL etc), balanced by treaty limits such as Article 5 TEU (conferral of powers) and Article 4(2) TEU (respect for national identity). National constitutions almost always proclaim themselves supreme (e.g. Poland’s Art. 8) and some include an identity clause. This backdrop suggests that the CJEU’s notion of primauté cannot be absolute. Indeed, a growing chorus of constitutional courts – in Poland, Germany, France and Romania – insists that where fundamental constitutional norms are at stake, the state’s basic law is the final arbiter.

Legal and Constitutional Context

The EU’s “constitutional text” lacks a supremacy provision. Instead, Article 5(2) TEU ensures the EU “shall act only within the limits of the competences conferred” by the Member States. Likewise, Article 4(2) TEU pledges respect for each State’s “national identities, inherent in their fundamental structures” and “essential State functions” (law and order, security, etc). In short, powers not granted to Brussels remain with the States. Against this, the CJEU’s doctrine (from Costa and Internationale Handelsgesellschaft) is that EU law automatically displaces inconsistent national law, including national constitutional provisions. That apparent conflict is resolved in practice by relying on national courts to “disapply” conflicting national norms when implementing EU law.

However, many constitutions declare themselves the highest law of the land (e.g. Art. 8 Polish Constitution) and explicitly subordinate international treaties to the Constitution (e.g. Art. 54 French Constitution). These texts, combined with Art. 5 TEU and Art. 4(2) TEU, imply that EU rules can never override a Member State’s core constitutional identity except by explicit amendment. Indeed, in France the Conseil constitutionnel has held that domestic law transposing EU directives “cannot go against a rule or principle inherent in the constitutional identity of France”, and absent such an identity conflict the courts must leave EU compliance to the EU judge. Similarly, Poland’s Constitution gives only ratified treaties priority over statutes – not over the Constitution itself (Art. 91) – and indeed Art. 8 designates the Constitution as “supreme law” of Poland. In Romania, Article 148 of the constitution likewise provides that EU treaties prevail over conflicting national law, but again within the framework of the Romanian Constitution (which itself may not be contravened). These legal provisions support the view that Member States have not ceded their fundamental identity or core powers to the EU; rather, EU competences remain bound by the Constitution and by the nation’s will.

National Case Law on Constitutional Supremacy^[1]

Poland (K 3/21 etc.)

In K 3/21 (7 Oct 2021), Poland’s Constitutional Tribunal reviewed several EU Treaty provisions against the Polish Constitution. It held that EU law cannot be interpreted to allow courts to disregard constitutional limits. For example, it struck down as unconstitutional any reading of Article 19(1) TEU that would entitle Polish courts to apply EU norms “in a way inconsistent with the Constitution”, or to reintroduce into domestic law any EU provision the Tribunal had struck down. In effect, K 3/21 put national constitutional review above even definitive CJEU interpretations: Polish courts were forbidden from “bypassing” the Constitution on the basis of EU law, because doing so would violate the Constitution itself. (This stance directly contradicts the CJEU’s view that ordinary courts must disapply prior constitutional rulings if they clash with EU obligations.) K 3/21 thus reaffirms that Poland’s basic law sets the ultimate boundaries of EU authority in Poland.

Germany (Weiss/PSPP)

The German Federal Constitutional Court has similarly insisted on constitutional limits. In its Weiss (PSPP) judgment of 5 May 2020 (2 BvR 859/15 et seq.), the Court held that if EU measures exceed what the Member States have conferred, they cannot bind Germany. It actually declared the ECB’s bond‑buying program (and the related CJEU ruling upholding it) ultra vires under German law, and gave the government time to renegotiate the program to comply with constitutional requirements. In doing so, the Court emphasized that EU powers derive from consent and that Germany’s “constitutional identity” – especially the eternity clause (Art. 79(3) GG) – cannot be overridden by EU law. In short, the BVerfG said it would enforce Germany’s supreme law even if doing so “disapplies” a CJEU decision.

France (Décisions 2006‑540 DC and 2021‑940 QPC)

France’s Conseil constitutionnel has likewise reserved a final say. It has repeatedly affirmed that transposition laws and implementing acts must not contradict the Constitution, and it guards core principles rooted in French constitutional tradition. In its 2006 decision (No 2006‑540 DC), for instance, it declared certain EU constitutional provisions inapplicable in France to the extent they would undermine national sovereignty (notably the special Treaty on a Constitution for Europe). More recently, in QPC No 2021‑940 (Oct. 15, 2021) the Court stressed that “the Republic participates in the EU … on the condition that [its] Constitution is respected”. Citing Art. 88‑1, it held that only laws touching “a rule or principle inherent to France’s constitutional identity” can be subject to constitutional review when implementing EU directives. The Court even identified one such identity principle (stemming from Article 12 of the 1789 Declaration: the monopoly of the state over force) to test a Schengen directive’s transposition. This approach preserves a constitutional backstop: where an implementing measure merely gives effect to clear EU obligations, French judges must apply it, unless it collides with an identity clause.

Romania (Decision No 390/2021)

Romania’s Constitutional Court has likewise rejected unqualified EU primacy. In Decision 390/2021 (8 June 2021), it held that “the primacy of EU law must not be perceived as removing or disregarding national constitutional identity.” The Court asserted that if a law is upheld as constitutional, ordinary courts may not unilaterally strike it down in favor of EU law – else legal certainty collapses. (The effect was to grant Romania’s constitution a veto over certain EU-based rules, a position since challenged by the CJEU.) In any event, these constitutional judgments underscore that Romania, too, claims the last word on conflicts affecting core constitutional norms.

Recent CJEU Jurisprudence

While constitutional courts affirm national primacy, the CJEU has reiterated EU supremacy.

A trilogy of recent cases is illustrative.

• Case C‑225/22 (AW) (Sept. 29, 2025). In this request from Poland, the CJEU was asked whether courts could ignore a higher national court’s (including a constitutional court’s) ruling if that ruling conflicts with EU law. The Court answered unequivocally: when a national decision or court ruling is irreconcilable with EU obligations, the national court must “disregard” it. In particular, an ordinary court bound by a constitutional court’s decision must still give full effect to EU rights. In AW the Court stressed that Polish judges must be free to refuse to apply any constitutional court judgment that would undermine EU law effectiveness. This means that, under EU law, a constitutional court judgment loses its binding effect whenever it conflicts with EU obligations. (The Court thus placed itself above Poland’s Tribunal, echoing its prior ruling in Križan.)

• Case C‑430/21 (RS) (Feb. 22, 2022). This Romanian reference concerned national rules that banned judges from examining the EU‐law conformity of any statute already upheld by the constitutional court. The CJEU likewise struck down this limitation. It held (1) that EU‐law primacy requires national courts to disapply any rule – even constitutional rulings – that clashes with EU law, and (2) that a disciplinary regime penalizing judges for enforcing EU law violates judicial independence. In clear terms, Article 19(1) TEU (preserving judicial independence) and Article 47(1) EUCFR mean judges must be able to vindicate EU law directly, even if that means contravening a constitutional court’s finding. In RS the Grand Chamber concluded that neither constitutional identity nor trust in the national judiciary can justify stripping courts of the power to ensure EU law’s full effect.

• Case C‑792/22 (Energotehnica) (Sept. 26, 2024). A Romanian criminal proceeding raised questions of social-protection law and national procedure. Crucially, Romanian law (as interpreted by its Constitutional Court) had given an administrative-court judgment absolute res judicata before criminal courts, even when EU rights were at issue. The CJEU struck this down. The Court held that EU law precludes any rule that forces ordinary judges to apply a constitutional court’s decision when that decision infringes EU rights[6]. In other words, no national law or practice may punish a judge (even by discipline) for setting aside a higher court’s ruling if it violates EU fundamental rights. The Court reaffirmed that national courts must have the autonomy to give EU law its full effect in each case, despite any conflicting constitutional jurisprudence.

Each of these CJEU judgments invokes the core EU principle of effectiveness and primacy: courts are required, of their own motion, to disapply any national rule (including a constitutional ruling) that conflicts with EU law. The Court even cites Costa v. ENEL and successive 1970s cases to emphasize that no later domestic law – not even a constitution – may be allowed to nullify EU law’s uniformity. In sum, the CJEU’s position is that each Member State must grant EU law unequivocal priority in all areas of EU competence.

Balancing Competing Authorities

The foregoing shows a stark clash: on one side, CJEU case-law asserts unconditional primacy; on the other side, national supreme courts and tribunals insist on constitutional finality.

Critics of the CJEU emphasize that the Union’s powers are wholly conferred – and that, because there is no “EU Constitution” overriding national constitutions, each State’s basic law must ultimately govern foundational issues (constitutional identity, structure of government, etc.). The national courts’ rulings note that Article 4(2) TEU and Article 5 TEU explicitly reserve core prerogatives to the State, suggesting that EU law’s primacy has clear textual limits. From this angle, the German, Polish or Romanian positions are not “disobedience” but a logical defense of national sovereignty. They might analogize to ordinary international law: treaties bind a State, but under Article 49 of the Vienna Convention a fundamental change of constitution or regime could (in theory) allow a State to reassess its obligations. Here, though EU membership creates a strong presumption of compliance, nothing in the text compels a State to bow to EU judgments that violate its constitutional core.

National identity clauses add weight to this view. If a constitutional court deems certain principles (e.g. democracy, rule of law, human dignity, or the very identity clause itself) beyond EU reach, it can claim a competence to shield them from EU adjudication. Indeed, Article 4(2) TEU’s reference to identity – together with Article 5’s conferral rule – arguably allows a Member State to insist that fundamental changes (e.g. altering its constitutional structure or supreme-law status) be subject to its own procedures, not the EU’s. Thus, the absence of a supremacy clause in the Treaties, in combination with these provisions, underpins the argument that constitutional courts retain ultimate authority when a EU obligation is in serious tension with core national constitutional principles.

On the other hand, the CJEU and EU advocates argue that a Member State cannot have it both ways: if it freely chose to join the Union and accept its system of remedies, it must honor the Union’s integrated legal order. From that viewpoint, national judges – ordinary and constitutional – agreed at accession to respect the Union’s primacy as a condition of membership. The Court emphasizes that allowing any Member State to disregard EU law would undermine the uniformity and effectiveness of the internal market and the rights of EU citizens everywhere. Indeed, in AW and RS it held that checks on national constitutional acts are built into EU law itself (not just the Treaties but also fundamental rights and judicial-independence guarantees).

Integrating Bóka and Krajnyák’s Central European argument^[2]

Bóka & Krajnyák’s Part IV is valuable because it connects doctrinal symptoms (identity reservations, ultra vires review, judicial “dialogue” breakdowns) to a broader institutional story about the rule of law as a driver of EU constitutionalisation.

They define their objective explicitly: to analyse trends in the supranational interpretation of the rule of law and to test the hypothesis that the normative and institutional framework of that interpretation enables EU institutions to pursue a constitutional federalisation trajectory—creating, in effect, a constitutional federation “without a constitution.” This is not merely theoretical: their argument is anchored in observed institutional developments (soft-law monitoring and evaluation, hard-law conditionality and infringement practices, and the constitutional framing of the rule of law by EU institutions).

On the normative dimension, they treat “constitutional identity” as the key category through which national constitutional courts have tried to re-specify primacy. They underline a classic but crucial clarification: primacy does not (in EU doctrine) invalidate national provisions; it requires disapplication in the individual case. This helps explain why constitutional systems can maintain a claim of domestic validity—even while accepting that EU law may prevail in application—thus sustaining pluralist “coexistence” at the level of validity but not always at the level of applicability.

On the institutional dimension, they argue that the CJEU plays a “prominent role” in EU constitutionalisation by building a constitutional-like structure and, importantly, by institutionalising hierarchy of EU law over domestic law in practice. They connect this to the Lisbon Treaty’s political context (as a compromise after the failed constitutional project) and suggest that constitutionalisation beyond Member States’ intentions triggers resistance.

Their discussion of the “rule of law toolbox” is especially relevant for understanding why supremacy disputes now concentrate on the judiciary. They point to the growing suite of soft-law instruments (e.g., scoreboards, cycles, dialogues) and their interplay with hard-law tools (e.g., conditionality, infringement), arguing that the distinction between soft and hard law is fading in rule-of-law governance, with potential consequences for democratic legitimacy. The constitutional stakes rise because these tools, in their account, can require Member States’ “national rules and structures” to reflect EU standards, implying an EU-level rule-of-law concept that develops partly independently from national constitutional interpretations.

Finally, Bóka & Krajnyák’s treatment of the preliminary ruling procedure is not only descriptive but constitutional: they highlight that Article 267 TFEU structurally creates a hierarchical relationship in which the CJEU’s conclusions bind national courts, and they warn that a forced hierarchy between constitutional courts and the CJEU—absent a legitimate transfer of power—can undermine constitutional democracy by marginalising constitutional courts’ mediating role.

Conclusion

The clash between “constitutional supremacy” and the CJEU’s primacy doctrine reflects deep choices about sovereignty and integration. On one hand, the EU’s legal order assumes unquestioned precedence; on the other, many Member States insist that their constitutions – and the identity and conferral clauses in the TEU – preserve a final border. Notably, neither the EU Treaties nor any binding EU text explicitly declares EU law supreme over constitutions. Instead, primacy comes from case-law and political acceptance. Where fundamental constitutional identity is concerned, some national courts rightly note that the Union’s competences are strictly limited by treaty.

Consequently, there remains a powerful argument that in matters of “constitutional identity,” national constitutions hold the last word. Article 4(2) TEU expressly requires mutual respect for each state’s identity, and Article 5 TEU limits the Union’s reach. These provisions, combined with entrenched supremacy clauses in national constitutions, mean that when core national principles collide with EU law, the constitutional court may be the ultimate arbiter. The CJEU may demand that EU norms be applied fully, but without a new treaty basis for absolute supremacy over constitutions, that demand faces a democratic and legal counterweight. In the end, absent a clear treaty mandate, the constitutional courts’ insistence on final authority over identity‐driven conflicts remains a valid interpretation of the existing legal framework – one that will continue to shape EU‑law enforcement in member states.

^[1] See: Tribl, N. (2024). Integration Responsibility: ‘The Relationship of National Constitutional Courts to the Court of Justice of the European Union: Hierarchy or Collegiality?’, Law, Identity and Values, 4(1), 251–269.

^[2] Bóka, J. - Krajnyák, E. (2024) ‘The Supranational Interpretation of the Rule of Law: Trends and Conclusions from a Central European Perspective’, in Bóka, János, ed. (2024) The Supranational Interpretation of the Rule of Law. Studies of the Central European Professors’ Network . Central European Academic Publishing, Miskolc - Budapest, pp. 413-478.