Preliminary references and EU law after Brexit

Preliminary references are the institutional mechanism through which EU law becomes a judicial conversation rather than a merely diplomatic arrangement. Article 267 TFEU enables, and sometimes requires, a national court to ask the Court of Justice of the European Union for an authoritative ruling on the interpretation of EU law or the validity of acts of EU institutions, bodies, offices or agencies. The national court then applies that ruling to the facts before it. The reference procedure is not an appeal. It is a cooperative procedure designed to secure uniformity, legality and effectiveness across a legal order administered primarily by national courts.

For Durham first-year purposes, this topic consolidates the preceding seven weeks. Supremacy explains why a CJEU interpretation matters domestically. Direct effect explains why individuals can rely upon EU norms before national courts. Indirect effect explains why national law may be read through the lens of EU obligations. Francovich liability explains one consequence of serious Member State failure. Free movement law supplies the substantive setting in which preliminary references often arise: customs charges, indistinctly applicable trading rules, Union citizenship, services and establishment have all been shaped through references from ordinary domestic litigation.

Brexit has changed the United Kingdom’s position but has not made the topic historical. Before exit and during the transition period, UK courts participated fully in Article 267. After the end of the implementation period, the general power or duty to refer ceased. Section 6(1)(b) of the European Union (Withdrawal) Act 2018 provides that a UK court or tribunal cannot refer any matter to the CJEU on or after exit day. That rule is not the whole picture. The Withdrawal Agreement preserves CJEU jurisdiction for pending cases, creates a time-limited reference jurisdiction for certain citizens’ rights questions, and maintains particular CJEU roles in relation to the Northern Ireland/Windsor Framework arrangements. Further, pre-exit CJEU authority continues to matter because large areas of former EU law were carried into domestic law as retained EU law, now increasingly described as assimilated law following the Retained EU Law (Revocation and Reform) Act 2023.

In an examination, the central analytical discipline is temporal. Ask first: did the material facts, legal proceedings, or relevant domestic enactment arise before exit, during transition, or after the implementation period? Then ask whether the issue concerns ordinary domestic law, retained or assimilated EU-derived law, Withdrawal Agreement rights, or Northern Ireland/Windsor Framework obligations. Only after those questions should one discuss CJEU authority. The common weak answer states, too bluntly, that Brexit ended EU law. The stronger Durham answer explains the layered constitutional afterlife of EU law: no general references, no automatic binding force for post-exit CJEU decisions, continued significance of pre-exit case law, and exceptional continuing CJEU jurisdiction under the Withdrawal Agreement structure.

The preliminary reference procedure was part of the original constitutional architecture of the European Communities. Its importance, however, was not fully visible from the text alone. Article 267, formerly Article 177 EEC and Article 234 EC, created a channel by which national courts could ask Luxembourg for rulings on interpretation and validity. The constitutional transformation came when that procedural channel was joined to the doctrines of direct effect, supremacy and effectiveness. In Van Gend en Loos the Court of Justice described the Community as a new legal order. In Costa v ENEL it asserted the priority of Community law over inconsistent national law. Preliminary references supplied the practical means by which those doctrines were dispersed through domestic litigation.

The procedure was therefore both decentralised and centralising. It was decentralised because national courts remained the ordinary courts of EU law. An Italian magistrate, a Dutch tax court, a French administrative court, a German constitutional court or an English High Court could become the forum in which EU law was pleaded. It was centralising because the meaning and validity of EU law remained ultimately for the Court of Justice. This avoided a patchwork in which the same Treaty provision or regulation meant different things in different Member States. The reference procedure also allowed individuals and firms to act as engines of integration. Many landmark cases did not begin as grand intergovernmental disputes; they began as ordinary claims by traders, workers, consumers, employers and public authorities.

In the United Kingdom, membership from 1973 placed domestic courts within this system. The European Communities Act 1972 made directly effective EU rights enforceable domestically, and UK courts gradually internalised the logic of reference, supremacy and conformity. The most dramatic domestic symbol was Factortame, where UK courts accepted that national legislation could be disapplied where inconsistent with directly effective EU law. Yet the more routine constitutional significance lay in thousands of ordinary cases: employment protection, equal treatment, VAT, environmental law, consumer protection, intellectual property, immigration and commercial regulation.

Brexit required Parliament to dismantle and preserve the EU law relationship at the same time. Repealing the European Communities Act without more would have created vast gaps, because domestic law had been saturated with EU-derived rules. The European Union (Withdrawal) Act 2018 therefore retained much of the existing EU-derived acquis as domestic law at the point of exit, while ending the general obligation of UK courts to follow future EU law and ending the power to make preliminary references. The transition period under the Withdrawal Agreement then postponed many practical effects until 31 December 2020. The European Union (Withdrawal Agreement) Act 2020 gave domestic legal effect to the Withdrawal Agreement through section 7A of the 2018 Act.

The result is constitutionally unusual. The United Kingdom is no longer a Member State and UK courts are no longer ordinary Article 267 courts. But the CJEU remains relevant in three ways. First, pre-exit CJEU case law supplies interpretive authority for retained or assimilated EU-derived law, subject to statutory rules on departure. Secondly, post-exit CJEU decisions may be considered where relevant, though they are not generally binding. Thirdly, the Withdrawal Agreement preserves specific forms of CJEU jurisdiction, including pending cases, citizens’ rights references for a limited period, and Northern Ireland-related obligations. The topic is thus not simply a Week 8 procedural topic. It is the point at which EU constitutional method and the United Kingdom’s post-membership constitution meet.

The first principle is cooperation. Article 267 is framed around a dialogue between courts. The CJEU does not decide the dispute before the national court; it answers a question of EU law. The national court identifies the relevant factual and domestic legal setting, formulates the question, and ultimately applies the answer. This distinction matters. A reference is not an appeal from the national court to Luxembourg, nor does it make the CJEU a general constitutional court for all domestic grievances. It is a mechanism for securing authoritative interpretation and reviewing validity within the EU legal order.

The second principle is the distinction between interpretation and validity. National courts may interpret and apply EU law, but the CJEU has exclusive jurisdiction to declare acts of EU institutions invalid. Foto-Frost prevents national courts from holding EU measures invalid for themselves. They may reject arguments alleging invalidity and uphold a measure, but if they consider the measure invalid they must refer. The reason is uniformity and legal certainty. If national courts could invalidate EU acts independently, the same regulation or decision might operate in one Member State and not in another.

The third principle concerns which courts may or must refer. Any court or tribunal of a Member State may refer where a ruling is necessary to give judgment. Whether a body is a court or tribunal for Article 267 purposes is determined by EU criteria: establishment by law, permanence, compulsory jurisdiction, inter partes procedure, application of rules of law, and independence are relevant indicia. Courts against whose decisions there is no judicial remedy under national law are under a duty to refer where a question of EU law is necessary. CILFIT limits that duty through three doctrines: the question may be irrelevant; it may already have been answered by the CJEU, often called acte éclairé; or the correct application of EU law may be so obvious as to leave no scope for reasonable doubt, called acte clair. CILFIT also insists that EU law is multilingual, autonomous and purposive, so acte clair is not an easy escape route.

Van Gend en Loos gives the reference procedure its constitutional setting. The case reached the Court through a customs dispute before a Dutch tribunal. The Court’s answer did more than interpret a tariff provision. It announced that Community law could confer rights on individuals enforceable before national courts. Without Article 267, that claim might have remained nationally contained. Through the reference procedure, a private trader’s dispute became a constitutional statement about the nature of the legal order.

CILFIT is the leading case on the duty of final courts to refer. It rejects a mechanical duty to refer every arguable EU point, but it also rejects national judicial overconfidence. The final court may refrain where the point is irrelevant, already answered, or so clear that there is no reasonable doubt. Yet EU law must be read in all language versions, in its context, and with attention to its autonomous concepts and purposes. Acte clair is therefore a disciplined doctrine, not a convenience.

Foto-Frost concerns validity. It is indispensable because it explains why preliminary references are necessary for the legality of EU acts. National courts may conclude that an EU act is valid, but they may not themselves declare it invalid. That rule protects uniformity and avoids divergent national judgments on the existence and effect of EU measures. The case should be paired with Week 1 institutional material: EU legality is centralised even though enforcement is decentralised.

Cartesio protects the autonomy of lower national courts. A national rule cannot prevent a lower court from referring merely because an appellate court disagrees with the need for a reference. This reflects the cooperative structure of Article 267. It would undermine the system if domestic procedural hierarchy could silence the courts closest to the facts.

Köbler links the duty to refer with state liability. A final court’s manifest breach of EU law, including a failure to refer, may in principle found damages against the state. The case is controversial because it subjects judicial decisions to liability analysis, but it expresses the seriousness of the final court’s Article 267 obligations.

In the UK setting, post-Brexit cases such as TuneIn and Lipton show the new domestic judicial task. Courts must identify whether they are dealing with retained or assimilated EU-derived law, whether pre-exit CJEU authority binds or strongly guides them, and whether post-exit CJEU decisions are relevant persuasive material. Lipton is especially important because the Supreme Court attempted to rationalise the different categories of retained EU law and the continuing significance of CJEU case law. The case is a reminder that Brexit did not remove EU legal reasoning from domestic adjudication; it changed the source and hierarchy of that reasoning.

See practice questions section below.

See further reading section below.