10 June 2018

Is the Reasoning in “Coman” as Good as the Result?

The Court of Justice of the European Union has not always enjoyed the reputation of being particularly LGBT-friendly, but its standing among those pushing for the better protection of rights of same-sex couples is likely to have improved considerably following Coman. Deciding that the term ‘spouse’ in the Citizenship Directive includes same-sex spouses, the Court has placed all Member States under the obligation to recognise the personal status of same-sex marriage. Several commentators have rightly hailed the decision as a breakthrough for the protection of same-sex couples (here and here). I too fully welcome the inclusion of same-sex marriages in the term spouse in the Directive. Yet, while I agree with the substantive result of the decision, I am uncertain if the CJEU’s reasoning is equally convincing. My two main points of critique concern the interpretative techniques applied and the relationship between national identity and fundamental rights.

The interpretation of the term “spouse”

The stakes were high in Coman, but the questions referred to the CJEU by the Romanian Constitutional Court were rather mundane. They concerned the interpretation of the Citizenship Directive and that of the term ‘spouse’ in particular. If interpreted as including same-sex spouses, the conclusion that Member States were to grant the right to residence to same-sex spouses of EU citizens would be difficult to escape. However, as Advocate General Wathelet rightly noted, the drafting history of the Directive confirms that the definition of the term spouse was deliberately left open and any express references to spouses of the same sex were removed (para 51). In addition, the Court had previously confined marriage to unions between persons of the opposite sex in D and Sweden, while it held that the term spouse ‘refers to a person joined to another person by the bonds of marriage’ in Coman (para 34). One would expect, therefore, the CJEU to address the restrictive definition of marriage previously adopted and confront the uncertainties resulting from the drafting process.

It decided not to. The CJEU confirmed that the term spouse within the meaning of the Directive is gender-neutral and could cover the same-sex spouse of the Union citizen (para 35). In addition, it denied the Member States to define the term spouse in accordance with national law and confirmed that, whereas they are free to decide on whether to allow for marriage of same-sex couples, they must comply with the provisions on the freedom of movement of EU citizens (paras 35-38). So far so good. In what followed, however, the CJEU ostensibly confused the need for a uniform definition of the term spouse with the need for a uniform application of free movement rights. It said that allowing Member States the freedom to restrict access to their territory by third-country nationals married to EU citizens would have the effect that ‘the freedom of movement of Union citizens … would vary from one Member State to another’, which would constitute an interference with the right to free movement (paras 39-40).

I do not fault the Court’s logic here, but as a response to the Romanian Constitutional Court’s main question, concerning the interpretation of the term spouse, it is inadequate. The interpretation of that term cannot depend on the consequences for the right to move and reside freely, as the question is precisely whether the Directive intended to allow for variations in the scope of free movement for same-sex spouses, depending on the national laws of the Member States. For example, Article 2(2)(b) of the Directive makes the free movement rights of registered partners of Union citizens dependent on whether national legislation ‘treats registered partnerships as equivalent to marriage’. The variations in and restrictions to free movement, depending on the national laws in place, are a direct consequence of the Directive. Certainly, Article 2(2)(a) of the Directive on the spouse of EU citizens makes no reference to national laws, which is why the Court rightly insisted on a uniform definition of the term spouse. However, it does not follow that no variations in free movement in the different states can occur. For example, if it had interpreted the term spouse as covering solely married couples of the opposite sex, it would also have adopted a uniform definition, but one with widely varying consequences, depending on whether the national laws allow for the recognition of same-sex couples. To be clear, I do not argue for a restrictive reading of the term spouse, but the arguments the CJEU put forward do not properly support its decision.

In that respect, the Opinion by Advocate General Wathelet is more persuasive. He thought that the interpretation given to the term marriage in D and Sweden had become ‘outdated’ on the ground that national norms on marriage had changed considerably. Roughly half of the EU Member States currently allow marriages between persons of the same sex, an astonishing change when compared with the situation in 2001, when D and Sweden was decided (paras 56-58). Given the indications that the interpretation of the term spouse within the meaning of the Directive would depend on national acceptance of same-sex marriage, and subsequent developments therein (para 51), and seeing the developments that had indeed occurred, the Advocate General reasoned that there were good reasons for the term spouse to cover same-sex couples.

The relationship between fundamental rights and national identity

This is arguably but a minor issue, with little to no significance for the outcome of the case. Possibly more consequential is the response the CJEU offered to those Member States who sought to justify the restriction to the right to free movement of same-sex spouses on grounds of public interest and national identity (Article 4(2) TEU). First, it held that because the requirement to recognise same-sex marriages celebrated in other states does not entail the obligation to open up the institution of marriage to persons of the same sex, there is no threat to the public policy or national identity (paras 45-46). While I have sympathy for that position, it is debatable if not the requirement to recognise same-sex marriages challenges the constitutional identity of those states whose constitutions define marriage as unions between persons of the opposite sex to some degree. The Court could have decided instead that these interests are insufficiently weighty to restrict the rights offered to same-sex spouse by the Directive.

It may have said something to that effect when it decided that restrictions to the right to freedom of movement are justified only when consistent with the fundamental rights recognised by EU law (para 47). In other words, states cannot invoke the national identity clause successfully if the fundamental rights of Union citizens are undermined consequently. That may seem like an attractive perspective, but here the Court may have overplayed its hand somewhat. For a start, it is questionable if that reasoning is consistent with earlier case law. For example, Runevič-Vardyn permitted the non-recognition of surnames on the ground that a requirement of recognition would lead the EU to violate the national identities of the Member States. Interestingly, the CJEU recognised that names were a constituent element of a person’s identity and private life, protected by both the Charter of Fundamental Rights and the European Convention (para 66). Likewise, Coman recognised that the relationship of a homosexual couple falls within the notion of private and family life as defined by the European Court of Human Rights. The possible inconsistency between both cases consists in the relationship between fundamental rights and national identity. In Coman, the recognition that the rights at stake were recognised as fundamental rights by the Court of Human Rights meant the end of the story. In Runevič-Vardyn, to the contrary it directed the national court to strike a fair balance between ‘on the one hand, the right of the applicants in the main proceedings to respect for their private and family life and, on the other hand, the legitimate protection by the Member State concerned of its official national language and its traditions’ (para 91). The precise relationship between fundamental rights and the national identities of the Member States thus remains in doubt.

Tempting as it is to take the position that the Member States’ national identities should always give way to the fundamental rights of EU citizens, the complexity of the interrelationship between fundamental rights and national identities may warrant a more nuanced approach. Let me illustrate this by one example. The controversies surrounding recent case law on religious discrimination in the workplace (Achbita, Bougnaoui, and Egenberger) are well known and need not be rehearsed here. Imagine, however, a case concerning employment in the French public sector. Already in Bougnaoui, a private sector case nota bene, France clearly stressed its belief that its constitutional principle of laïcité cannot be undermined by the Framework Directive on Employment Discrimination (See AG Sharpston’s Opinion in that case). Few will doubt that this principle is at the heart of French national identity, but it seems equally evident that it stands in a difficult relationship with the freedom of religious expression. I am not saying that there cannot be good grounds for restricting the French principle of laïcité if the Framework Directive so requires, but the argument that fundamental rights always and immediately trump states’ national identities seems insufficiently attentive to the possible controversies that may come before the Court.


None of the above is meant to put into question the decision to include same-sex spouses within the term spouse within the meaning of the Directive and to oblige all Member States to recognise the marital status of same-sex partners. For good reasons, those benefitting most from this decision will be least concerned with those legal issues I have addressed here.

SUGGESTED CITATION  van den Brink, Martijn: Is the Reasoning in “Coman” as Good as the Result?, VerfBlog, 2018/6/10, https://verfassungsblog.de/is-the-reasoning-in-coman-as-good-as-the-result/, DOI: 10.17176/20180611-100057-0.

Leave A Comment


1. We welcome your comments but you do so as our guest. Please note that we will exercise our property rights to make sure that Verfassungsblog remains a safe and attractive place for everyone. Your comment will not appear immediately but will be moderated by us. Just as with posts, we make a choice. That means not all submitted comments will be published.

2. We expect comments to be matter-of-fact, on-topic and free of sarcasm, innuendo and ad personam arguments.

3. Racist, sexist and otherwise discriminatory comments will not be published.

4. Comments under pseudonym are allowed but a valid email address is obligatory. The use of more than one pseudonym is not allowed.

Explore posts related to this:
Coman, Discrimination, Same Sex Marriage

Other posts about this region: