Cupriak-Trojan: A First Reading from a Private International Law Perspective

,

This post was written by Laima Vaigė, Associate Professor and Senior Lecturer in Private International Law at Uppsala University. It is the first contribution to the EAPIL on-line symposium on the ruling of the Court of Justice in Cupriak-Trojan.


The Court of Justice of the European Union delivered a landmark Grand Chamber judgment in Cupriak-Trojan (C-713/23) on 25 November 2025. The ruling is concerned with the obligation of a Member State to recognise a same-sex marriage lawfully concluded in another Member State when the spouses, both EU citizens, return to their State of nationality.

The dispute concerned a Polish national and a Polish-German national, who had married in Germany and requested that their marriage certificate issued in Germany be recognised and transcribed in the Polish civil register. The request was refused, as Poland does not allow same-sex couples to conclude marriages (nor registered partnerships). The case was referred to the Court of Justice by the Polish Supreme Administrative Court, and because it is about “recognition” of a marriage lawfully concluded abroad, the judgment also has implications for private international law (PIL).

Key Reasoning

The Court of Justice held that Poland’s refusal to recognize a same-sex marriage lawfully concluded in Germany, and to transcribe the foreign marriage certificate, creates serious administrative, professional, and private inconveniences (para 51), prevents the couple from continuing their family life in Poland (para 52), and exposes them to serious obstacles in everyday life (para 53). Such a refusal therefore constitutes an obstacle to Article 21 of the Treaty on the Functioning of the European Union (TFEU) free-movement rights (para 54).

Although restrictions may in principle be justified (paras 55–56), the Court found that neither national identity nor public policy can justify the refusal. Recognition of a same-sex marriage concluded abroad does not undermine Poland’s national conception of marriage and does not require the introduction of same-sex marriage domestically (para 61). Consequently, the obligation of recognition “does not undermine national identity or constitute a threat to public policy” (para 62).

Any restriction must also comply with the Charter of Fundamental Rights of the EU (the Charter) (para 63). Article 7 of the Charter mirrors Article 8 of the European Convention on Human Rights (ECHR) (para 64), and the ECtHR has repeatedly confirmed that same-sex couples fall within both private and family life (see, e.g., Orlandi and others v. Italy). With reference to a number of judgments in which the ECtHR has held that Poland has breached its duties under Article 8 of the ECHR, the Court of Justice highlighted that Poland’s failure to provide legal recognition leaves couples in a “legal vacuum”, contrary also to Article 7 (paras 65–67). While Member States may choose how to provide recognition (para 69), they must establish some effective procedure (para 68). In Poland, transcription is the only available mechanism, and without it, recognition depends on inconsistent administrative discretion (paras 71–73). Since heterosexual couples can obtain transcription but same-sex couples cannot (para 74), the situation amounts to discrimination based on sexual orientation, prohibited by Article 21(1) of the Charter (para 75).

Finally, the Court underlined that Articles 20 and 21 TFEU and Articles 7 and 21 of the Charter are directly effective. National courts must disapply conflicting national law where it cannot be interpreted in conformity with EU law (para 76).

Implications for Domestic and EU Private International Law

In PIL, substantive validity of marriage concluded abroad is a crucial question for “recognition” – different genders being one of the substantive requirements for entering into marriage in Poland and some other States. Marriages can be concluded under different substantive laws, hence choice-of-law rules are needed to determine their substantive validity.

In some States, substantive validity of marriages concluded abroad is determined by the law of the habitual residence of the spouses (e.g., Estonia) or lex loci celebrationis (law of conclusion of marriage), e.g. Lithuania. In other States, among them Poland, it is determined by the law of nationality of the spouses. The Polish Prawo prywatne międzynarodowe (Polish PIL Act) of 4 February 2011 provides that law of nationality at the time of entering into marriage determines each spouse’s substantive capacity to marry (Article 48). There is also the ordre public safeguard (Article 7) to prevent validity of certain marriages in exceptional cases. Besides PIL rules, administrative rules on transcription apply, and these also contain ordre public exception (Prawo o aktach stanu cywilnego, i.e, Law on civil status records,  of 28 November 2014, Article 107(3)).

The Court does not engage with Polish PIL in its reasoning, and only briefly mentions the ordre public safeguard in Article 7 of  the Polish PIL Act in its description of the legal context (not directly relevant, because Article 48 leads to application of the Polish law). Once the ECJ turns to the substance, its analysis is framed entirely through EU free-movement law and the Charter. When the Court assesses “public policy” (paras 57–62), it does so not as an interpretation of the Polish ordre public exception in PIL but as an EU-law review of whether national identity or public policy can justify a restriction on Article 21 TFEU. In other words, the Court interprets the EU concept of public policy, or the EU ordre public , which cannot be determined unilaterally by Member States. As a result, Polish PIL plays only a marginal role: it is effectively displaced by the primacy of EU free movement and fundamental rights.

The Court highlighted in Cupriak-Trojan that Articles 20 and 21 TFEU and Articles 7 and 21(1) of the Charter have direct effect and national courts must ensure their effectiveness by disapplying the conflicting national provisions. Various provisions of national law can be disapplied as a result. As to private international law, there is a strong case for disapplying, in EU-law situations, the PIL rule that ties marriage validity to the spouses’ nationality. Instead, a marital status concluded in another EU Member State should follow the spouses, provided that they had a residence in the Member State where marriage was concluded (see para 52). Hence, in situations where EU law is applicable, habitual residence or lex loci celebrationis appear to work better as connecting factors in determining substantive marriage validity.

Do the concepts of “habitual residence” in PIL and “residence” in Coman (C-673/16) and Cupriak-Trojan coincide? They might, but not always, because a person can obtain habitual residence very quickly in certain situations. Meanwhile, “genuine residence” (para 51 in Coman) implies EU citizen who resides in the Member State where marriage was concluded for more than three months, under certain conditions provided in secondary EU law (art 7.1. of Free Movement Directive).

It can be observed that in the previous case-law on cross-border recognition of surnames, there was no requirement of residence. For instance, in Garcia Avello case, the surname was obtained while living in the country of birth, never changing the State of residence. In Freitag, the surname was obtained in Romania, while the applicant resided in Germany. The Court found violations of Article 21 TFEU. In both cases, the applicants had dual nationality. From the reasoning in Freitag, it appears that obtaining the new civil status in the place of nationality should also be acceptable. The requirement of residence hence applies, so far, only to cross-border recognition of marriages, but not surnames. Although the requirement of residence for entering into marriage works rather well in PIL, it does not stand on a strong footing, from the perspective of EU primary law, and even less so, from the perspective of the ECHR.

Although Cupriak-Trojan is a free-movement case, it must have consequences for EU PIL. EU secondary law (inter alia, the Brussels II ter and the Succession Regulation) must apply in compliance with EU primary law, which has now been interpreted. Where these instruments rely on marriage, Member States should not exclude same-sex spouses, provided that their marriage was concluded within the EU and involved at least one EU citizen lawfully residing in the State of conclusion. Given the direct effect of the relevant Articles, national courts must disapply conflicting national PIL rules and substantive law rules where these would undermine the required recognition for the purposes of EU law.

Loopholes and Narrow Readings

One can see in this case a “free-market logic” applied to civil status. The Court treats surnames, marriages, and legal genders as if they were legal products that must move freely in the Union, to ensure free movement of persons to which they are attached.

At the same time, these cases are not really about market integration. They concern the substance of rights attached to Union citizenship and reflect the growing constitutional role of fundamental rights in the EU legal order since the Charter acquired primary-law status. Indeed, free movement cannot function if a person, after moving or returning to a Member State, is no longer sure whether they are married, what their legal name is, or which gender is recognised in law. That uncertainty creates serious practical and legal obstacles to everyday life and deters free movement.

There are certain loopholes remaining, particularly where EU law does not reach. If the EU citizens marry in a third country and then return to their State of nationality (e.g. Poland), or if it is a third country nationals marrying in the EU and then moving to Poland, they cannot be helped by neither free movement nor EU Charter rights. In these situations, the couples can still claim infringement of the ECHR (Article 8, possibly in combination with Article 14).

Another problem is that in the region where heterosexuality is the social norm, narrow readings and outright refusals of implementation (of both ECtHR’s and Court of Justice’s judgments) have been increasingly common. Despite the judgments of the Court of Justice in Coman and Pancharevo, the child of two mothers in the latter case was never recognized as the child of the Bulgarian mother, and Mr. Coman’s husband never got a residence right in Romania.

In Coman, the Court used a narrowing phrase, “for the sole purpose of” (derived residence rights) in its reasoning and conclusions. This allowed readings that “interpretation relates only to the Directive”, which means that Mr. Coman needs to get a job in Romania (or otherwise satisfy Article 7(1) requirements of the Directive), and only in that scenario, his husband might claim the derived residence right. The Romanian authorities did not even use this reasoning, but simply dismissed the case on procedural grounds; a case is pending at the ECtHR, as a result. As to the aftermath of Pancharevo, the Bulgarian authorities decided that they needed to know which one of the mothers was the birth-giving mother, and, deciding it was not a Bulgarian national, refused to recognize the filiation of the child, thus the child had no right to Bulgarian citizenship, and subsequently, authorities could not issue a Bulgarian passport to the child in question.

The narrow readings of the Coman and Pancharevo cases were still possible due to narrow formulations of the Court, despite the fact that EU primary law was also interpreted in these cases: Article 21 TFEU, Article 4.2 TEU, and Article 7 of the EU Charter, in Coman, and Articles 20 and 21 TFEU, Article 4.2 TEU, Articles 7, 24 and 45 of the EU Charter, in Pancharevo.

In contrast to previous case-law, Cupriak-Trojan contains no narrowing formulation “for the sole purpose of” (transcription). On the contrary, the judgment explicitly references the loss of concrete legal effects when recognition is withheld: access to health insurance, registration of surnames (para 50), and enforceability of obligations between spouses and third parties (para 52). These examples show that recognition must preserve the practical ability of same-sex spouses to continue living “as married persons” and to “rely on their marital status” after returning to their home state (paras 52–53).

Transcription is only one means by which Member States may implement recognition, but implementation must be effective, not formalistic or symbolic. It follows that Member States should not provide a procedure that strips the marriage of its ordinary legal consequences. That means, in my reading, full recognition of the civil status for the purposes of private international law.

Is there room under EU law for downgrading a same-sex marriage to a registered partnership? The Court in Cupriak-Trojan did not address this because neither the facts nor the legal context allowed it.

The ECtHR’s case law might suggest that downgrading is compatible with the Convention, but it concerns recognition of marriages from any country, including possible marriage tourism. The EU context differs: it concerns only marriages concluded by EU citizens in the EU while lawfully residing in another Member State. If downgrading were permitted, a same-sex marriage would not be seen as a marriage under EU secondary legislation (except the Free Movement Directive), thereby potentially eliminating or reducing the legal effects that the Court appears concerned to preserve in Cupriak-Trojan.

In some countries, even the most basic legal effects are not available to registered partners (e.g., a registered partnership in Lithuania currently cannot be dissolved and has fewer effects than cohabitation in other EU countries). Perhaps downgrading can happen only where the registered partnership confers legal effects comparable to marriage. A simpler solution is that a lawfully concluded marriage remains a marriage and cannot be downgraded in the intra-EU context. If it is a marriage under EU primary law, it should remain a marriage under EU secondary law.

Even if the room for narrow readings is limited, it remains to be seen how the Polish court ultimately decides the case on the merits and how will the Polish legislator reacts.

In contrast to the approaches taken in other jurisdictions in the field of constitutional law, the judgment of the Court of Justice is not such a gigantic step, but respects both national identity of the recognizing state, as well as the logic of private international law.

6 replies
  1. Krzysztof Pacula
    Krzysztof Pacula says:

    Thanks for your thoughtful comment. On the point of narrowing the formulation “for the purpose of,” it seems to me that Mirin already took a similar route, so the contrast with earlier case-law might not be as strong as when we compare it with Coman/Pancharevo.

    The part I find really intriguing is the reservation in Coman/Pancharevo/Wojewoda Mazowiecki about setting up or strengthening family life in another EU Member State. It could be seen as i) a secular way of reflecting the PIL idea of habitual residence within a specific mechanism resulting from EU law — aligning the solution with some PIL standards (habitual residence of spouses at the time of marriage governing its substantive validity), ii) requirement triggering the reliance on freedom of movement in a scenario of marriage and return to the Member State of origin ; or iii) as a safeguard against marriage tourism. I guess that these options do not need to be mutually exclusive.

    Reply
    • Laima Vaige
      Laima Vaige says:

      Thank you for joining the discussion and for your insightful comments.
      You are absolutely right that Mirin already moved in this direction, which makes the contrast with earlier case law slightly less stark if one compares Cupriak-Trojan with Mirin rather than Coman/Pancharevo. What I find distinctive is that the formulation “for the sole purpose of” was used in relation to same-sex marriages — by both the ECtHR and the ECJ — whereas earlier recognition cases involving different-sex marriages, surnames, or adoptions did not employ that wording. It slowed development in order to let Member States adjust more gradually, and many indeed have — but it also created uncertainty about what legal effects such fragmented recognition can produce.

      On your second point, all three readings you suggest seem plausible and not mutually exclusive. From the perspective of EU law, I read it primarily through (ii): EU law must “cover” the issue at hand in the first place, and then EU Charter rights can come in. At the same time, the logic maps onto private international law reasoning.

      Reply
  2. Laima Vaige
    Laima Vaige says:

    I would also be curious to hear your views on whether, in the longer run, nationality might operate as an alternative connecting factor. For example, imagine two Swedish nationals with residence in Poland who marry in Sweden: should their marriage not be recognised in Poland, and would refusal to recognise it not amount to an obstacle to free movement under EU law? Such reasoning would align with the ECJ’s surname cases.
    It would also resemble the logic of the 1978 Hague Convention on Celebration and Recognition of the Validity of Marriages, which employs multiple connecting criteria for celebration and lex loci celebrationis for recognition. This approach seems workable within the EU context, but I would be very interested to hear what others think.
    Human rights perspective can take it even further, of course.

    Reply
  3. Marco Pasqua
    Marco Pasqua says:

    Dear Laima, thank you very much for your post and for opening the door to such a rich reflection on Cupriak-Trojan case. I have a series of considerations and questions that touch upon different aspects of the case, which I would like to address separately in order to keep the discussion organized and to allow others to join the conversation more easily.
    The three main areas I would like to focus on are (1) the interaction between EU law and national private international law, (2) the downgrading of same-sex marriages and (3) marriages concluded in third countries.

    (1) The Interaction between EU Law and National PIL
    The effectively displacement of Polish PIL by EU law in Cupriak-Trojan raises questions about the evolving role of national PIL within the Union. Traditionally, many Member States determine the substantive validity of marriage based on the spouses’ nationality; other Member States rely on lex loci celebrationis or habitual residence. The CJEU’s emphasis on the primacy of EU free-movement rights (Articles 20–21 TFEU) and Charter protections (Articles 7 and 21) suggests that nationality-based conflict rules may increasingly collide with EU law in intra-Union contexts.
    This leads to several questions:
    -Should habitual residence or lex loci celebrationis operate as the default connecting factors for all intra-EU marriages, even where national conflict rules point to nationality?
    -Could the principles applied in Cupriak-Trojan extend beyond marriage to other areas of private international law that are not harmonized at EU conflict-of-law level and are still governed by national PIL rules, particularly where traditional rules could undermine fundamental rights?

    Reply
  4. Marco Pasqua
    Marco Pasqua says:

    (2) Downgrading and the Preservation of Marriage Status
    Your reflections on downgrading are thought-provoking. While Cupriak-Trojan does not explicitly address whether a same-sex marriage could be recognized as a registered partnership, the CJEU’s focus on preserving practical legal effects suggests that any downgrading might be incompatible with EU law. The judgment emphasizes that recognition must preserve spouses’ ability to rely on their marital status, including access to healthcare, registration of surnames, enforceability of obligations and other civil and economic rights (paras 50–53).
    This raises additional questions:
    -Could a Member State implement a “quasi-marriage” or a registered partnership that confers most, but not all, legal effects?
    -And how would courts determine whether partial recognition satisfies Article 21 TFEU and Articles 7 and 21 of the Charter?
    As you mentioned, in some countries where registered partnerships provide fewer rights than marriage this could be particularly problematic.
    The functional approach adopted by the CJEU seems to suggest that a lawfully concluded EU marriage cannot be treated as anything less than a marriage within the EU context, even if national law offers alternative forms. This also raises the question of whether downgrading might be permissible only when the alternative legal status provides rights that are genuinely equivalent in practice to those conferred by marriage, a standard that could lead to further litigation to define equivalence in legal effects.

    Reply
  5. Marco Pasqua
    Marco Pasqua says:

    (3) EU Law, Third-Country Marriages and the ECHR
    Your discussion of the limitations of EU law in protecting marriages concluded outside the Union is crucial. Cupriak-Trojan rationale applies exclusively to intra-EU marriages involving at least one EU citizen, leaving third-country marriages outside the direct reach of Articles 20–21 TFEU and the Charter. In these cases, one must rely on the European Convention on Human Rights (Articles 8 and 14) and the caser law of the ECtHR, as in Oliari and Others v. Italy or Orlandi and Others v. Italy dealing with same-sex couples within the scope of private and family life.
    An open question is whether the principles developed in Cupriak-Trojan could be extended by analogy to third-country marriages.
    Could national courts interpret domestic law in light of EU fundamental rights reflecting ECHR principles to ensure effective protection for cross-border couples?
    This raises the broader question of the interaction between EU law and the ECHR: while the Charter reach is within the Union, ECtHR’s case law has a wider geographical reach. Could we see future cases in which the CJEU takes ECHR reasoning into account to promote consistent recognition of marriages involving third-country nationals, thereby bridging potential gaps in protection?

    Reply

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Join the discussion

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Discover more from EAPIL

Subscribe now to keep reading and get access to the full archive.

Continue reading