The Identity Turn in European Citizenship Law

Table of Contents: 1. Introduction – 2. The jurisprudential context – 2.1. Citizenship from solidarity to identity – 2.2. Values and competence in EU law – 2.3. The naming cases – 3. The radicalness of the Coman line – 3.1. Old principles in a new context – 3.2. Mutual recognition – 3.3. Justification and consistency – 3.4. U-turns – 4. Conclusion.

Abstract: This article argues that the recent run of identity judgments – Coman, Pancharevo, Mirin and Wojewoda Mazowiecki, and Shipova – are doctrinally considerably more novel than they seem. They reinterpret mutual recognition, redefine the U-turn, and abandon the Court’s traditional deference to constitutional values. The result nails another coffin in the idea of a wholly internal situation, moves towards giving full effect to Article 21(1) TFEU including its right of residence anywhere in the EU, and constructs a basis for widespread harmonization of family, identity, and sexual rights in the EU. The article does not argue against the cases. However, it suggests that they need to be looked at through the lens of competence as well as the lens of rights. Therefore, it tries to unpack their legal novelty, their potential consequences, and the associated risks to legitimacy and effectiveness. Understanding these is important to deciding how far the Court’s apparent programme of value harmonization should go. For while the Court is now pushing a progressive agenda onto conservative states, the majoritarian subtext of the judgments could mean that in the future the force of EU law is the other way round. Emancipatory states might find that Article 2 TEU and the Court stand in their way.

Keywords: union citizenship – mutual recognition – derogations – u-turns – same-sex marriage – gender identity.

1.   Introduction

In parallel with political changes in society, where left-right debates about redistribution appear to have been displaced, or at the very least reframed, by debates about identity in all its forms, Union Citizenship case law has also taken an identity turn.[1] One of the most dramatic and important lines of cases to have emerged from the Court of Justice in recent years is about the rights of mobile Citizens to have their particular subjective, often sexual, identity respected and accepted by the Member State in which they live.[2] They ask a question which certainly deserves to be at the heart of any legally defined community: what rights, as a Citizen, do I have to be me?

If the older solidarity cases, which used to dominate Citizenship discussion, were all about outsiders coming in, these identity cases are really about exit, or escape, not necessarily geographically, but from domestic norms.[3] They typically involve Citizens challenging their home states, and using European Union (EU) law’s powerful bias towards individualism to insert European-majoritarian progressive social norms into corners of the EU where they do not yet prevail.

They have been well received for their extension of fundamental rights and equality, which was achieved through use of legal elements which are long established in EU law – mutual recognition, Article 2 values as expressed in the principle of equality, the Charter of Fundamental Rights, and the commitment to free movement of Citizens. However, the aim of this article is to suggest that they are more doctrinally radical than they seem. That is not to say that they are bad decisions, but to suggest that they do raise some difficult questions about the scope of EU law, and the logic that underlies them could be both disruptive and harmonizing. Indeed, the article views the cases primarily as exercises in value harmonization within the EU, something that is attractive when one agrees with the values in question, but which nevertheless invites lawyers to think about how far that harmonization should, and will, be able to go.[4]

The article first briefly sketches the jurisprudential context of the cases. This begins with the phases of development of Citizenship law and where it is now, moves on to the Court’s wider understanding of the role of values in EU law, and then considers cases on U-turns and naming which are the immediate precursors of the identity judgments. These parts broadly emphasise the conventionality of the judgments, and the sense in which they fit seamlessly with wider law.

The next section then tries to subvert this picture. It suggests that the way in which mutual recognition is understood, the majoritarian approach to national identity, and the use of U-turn rights are all in fact quite novel departures from previous law. Moreover, these novelties have consequences: extrapolating the logic of this line of cases would end any vestige of an internal situation, and amount to the harmonization of matters of family and identity, breaking the last normative grip that Member States have on their nationals. Perhaps that is desirable; it can certainly seem so in the right context. If, on the other hand, an unusually progressive Member State faces the same majoritarian, harmonizing force, it may change its mind. This point is considered more later: harmonization typically cuts more than one way.

2.   The jurisprudential context

2.1.  Citizenship from solidarity to identity

For much of its life, the central story in Union Citizenship was about solidarity. Even before Citizenship existed the question had been asked to what extent nationals of Member States could access support, assistance and benefits provided by host Member States to their populations. Answers were sketched in both case law and legislation, with a compromising position adopted that remains broadly intact today; there is a strong, if nuanced, link between economic activity and solidarity rights, reflecting a framing idea that free movement is only for those likely to contribute to their host state, rather than those likely to burden it.[5]

Yet despite this legal stability, the topic remained intensely alive in academia for several decades – from Martinez-Sala to Alimanovic, one might say, with maximum heat between Grzelczyk and Dano.[6] This academic focus arose because of a fundamental incoherence in the law: a difficulty with reconciling the exclusionary effects of an essentially budgetary approach to free movement with the universal and aspirational ideals inherent in traditional understandings of citizenship.[7]

Yet despite softening some of the hard edges of the self-sufficiency doctrine, in Grzelczyk and subsequent cases, so that the door to a more meaningfully universal Union Citizenship might seem to have been opened a crack, in Dano the Court appeared to put a firm doorstopper in place, so that if the door would remain ajar for those whose need for solidarity was temporary, marginal, and deserved, it would not open any further.[8] The issue of what Union Citizenship should be – what the EU should be – will not go away, but in its current phase the law is offering no encouragement towards the idea that a deepening of transnational economic solidarity will play a part.

In this context, the identity cases seem to be offering a new energy and purpose to Citizenship. If it cannot be a vehicle for redistribution, at least perhaps it can be a vehicle for spreading equality and identity rights. Citizenship is no longer in the doldrums. Moreover, its new energy is attuned with the times, where the politics of redistribution seems less alive than the politics of identity. These cases are not just an adjudicatory response to a dispute, but an intervention in the European debates of today, and a contribution to the construction of a certain kind of European society.[9]

2.2.  Values and competence in EU law

The Court of Justice has most often been using values, particularly as expressed in Article 2 of the Treaty on EU (TEU), as a tool to combat illiberal governments in Central Europe, some of whom are also the defendants in the identity cases.[10] The resulting judgments have culminated in Commission v Hungary, the powerful substance of which can almost be seen as a judicial manifesto for Europe.[11] Among other elements of the judgment, the Court reaffirmed that it sees commitment to EU values, as interpreted by itself, as an essential element of membership of the EU.[12] Moreover, one of those values is equality, expressed in further EU principles and laws, and entailing a rejection of discrimination on grounds of, inter alia, sex, gender, identity or sexuality.[13] A governmental or national law ‘preference’, the Court says, for some identities or sexualities over others is contrary to EU law.[14]

The power of this statement makes it particularly important to know when it applies. It is manifestly not in all circumstances, for, as will be discussed below, the Court maintains that certain matters, such as marriage, are Member State competences outside of its jurisdiction, even when Member States adopt retro-biblical views on its nature.[15] To be subject to EU norms a matter must, on the whole, be within the scope of EU law.

The Court is not entirely consistent on what this means, particularly in the situation where a prima facie national competence obstructs EU policies or rights. There it has a formula that it often uses, to the effect that although the issue in question is a competence of the Member States it must nevertheless be exercised in conformity with EU law: it is up to you to run this matter, as long as you do it in a way I approve of.[16] Quite often, it then proceeds to require the Member State to reorganize the matter in question to fit EU law. Yet not always: on the question who to marry, (rather than whether to recognize marriages), even where heteronormative national rules lead to EU rights being diminished, the Court has always backed off.[17]

In the context of Citizenship, or at least the rights of Citizens to move and reside throughout the EU, the scope of EU law is engaged where a national rule obstructs the exercise of those rights. Only then do EU principles govern the situation. The recent identity cases are an interpretation of when restrictions on marriage and identity comprise such an obstruction. They are about individual rights to move. However, they are also about delimiting the circumstances where EU principles, and the values they express, displace those found in national law.

2.3.  The naming cases

The identity cases did not come from nowhere. Their core argument, that respect for identity is necessary to allow Citizens to exercise their free movement rights, can be seen as the flowering of the cases on naming which came before the Court in the fifteen years preceding Coman – Garcia Avello, Runevič-Vardyn, Grunkin and Paul, Bogendorff and Freitag.[18]

These cases were ostensibly quite bureaucratic, about documentary issues. However, they have many similarities with the identity cases. They also involved a fight to bring home ‘identities’ – in this case names – first recognized abroad, utilizing the now well-known U-turn construction. At stake were matters such as how the surnames of mothers and fathers could be used in naming children; the status of minority and majority languages with different scripts; and the significance of names as wider identity markers, hinting at or hiding ethnicity, or implying social status. These cases, in which the claimants mostly won, thus established that collective choices about identity and its limits could be trumped by the subjective preferences of mobile individuals – the burden of adaption in identity conflicts is, or can be, on states, not free movers. Some of these were also sensitive and controversial cases, but only in specific Member States.[19] Now their logic has been extrapolated to identity issues whose salience is EU-wide, in Coman et al.

3.   The radicalness of the Coman line.

3.1.  Old principles in a new context

In Coman, and later in Wojewoda Mazowiecki, Romania, and then Poland, refused to recognize the same sex marriage concluded by one of their own nationals while living in another Member State, when that national attempted to return with their spouse. In Pancharevo Bulgaria refused to recognize a family structure lawfully constituted in Spain, consisting of two mothers and their daughter. In practice, since one mother was Bulgarian and the other was not a Union Citizen, this amounted to a refusal to recognize the non-European mother, and possibly the daughter, as family members of the returning national. In Mirin, Romania refused to recognize a gender reassignment obtained by one of its own nationals under the law of another Member State (of which the Citizen was a dual national) when that national returned. In Shipova, Bulgaria refused to issue documents reflecting the gender as which Ms Shipova lived her daily life, because it did not correspond to her biological sex.

The resistant Member States lost.[20] The Court of Justice found that EU law required them to recognize the relationships, families, and gender reassignments that nationals brought back with them from time spent living elsewhere in the Union. A failure to do so hindered their ability to move, which Article 21 of the Treaty on the Functioning of the EU (TFEU) gave them the right to do. This is hard to argue with – is not living without one’s partner or under the wrong name or gender or without a legal relationship to one’s child a disincentive to live in a state? In Shipova it found that refusing to recognize Ms Shipova’s gender stood in the way of her ability to move between Member States. It mentioned airports and hotels and border crossings as examples of where the discrepancy between her documents and her lived gender would cause embarrassment and practical problems and surely discourage her from exercising her right to move. These are just examples. One can imagine worse, nastier, situations that an apparent failure to match one’s documents might create. Once again, the Court’s Article 21 reasoning is clearly factually correct.

The cases are nevertheless novel in various ways, perhaps even radical. They rely on orthodox elements of EU law, notably mutual recognition and the U-turn construction, but do so in a way that significantly changes the meaning and scope of these elements. They also diverge from previous cases on constitutional values, replacing deference with harmonization. The following sections outline these points.

3.2.  Mutual recognition

All of the identity cases other than Shipova are about mutual recognition; the reason why the defendant Member States were expected to recognize the marriages and transitions of the claimants was because another Member State had certified them according to its own domestic law.[21] Without that, pure free movement arguments would not normally have been enough: a person may be inhibited from moving because their unmarried partner cannot come with them, or because they are hoping to undergo gender reassignment and this is difficult in the new state, but these arguments, while human and realistic, will not as such typically be enough to require the new state to adjust its laws and practices to admit the partner or ease the reassignment.[22] It is, and was in these cases, important that one Member State start the ball rolling by making a legal and documentary statement about a person and their life, which other Member States are then presumptively expected to accept.

There were initially signals that only a limited mutual recognition was involved. At least in Coman the Court emphasized that it was necessary to accept the marriage only insofar as this was required to enable the husband (Coman) to live and function in Romania.[23] This implied residence permits, the right to work and so on, but perhaps not a full recognition of marriage for the purposes of, for example, inheritance or family rights, or indeed the recognition of the marriage as such. The judgment was ambiguous, but it might have been sufficient to allow the couple to live, work, and enjoy equal treatment without recording them officially as being married.[24]

However, the Coman compromise seems to have largely unravelled in Pancharevo and Wojewoda Mazowiecki, where same sex families were also involved, and the Court required recognition of the family structure, and then some form of recognition of the marriage for official and practical purposes. The truth is that recognition ‘to the extent necessary for free movement’ may well amount to full recognition, because any limitations are likely, in the right circumstances, to cause obstacles to life – or ‘serious inconvenience’ as the Court puts it – and so in fact be prohibited.[25] The idea of a limited mutual recognition may be unrealistic.

In any case, limited or not, it is still the mechanism of mutual recognition at work – it is still the issuing Member States actions that lead to the regulatory pressure on the host Member State.

However, this is not standard mutual recognition. It differs importantly from the mutual recognition that has been a part of EU law since Cassis de Dijon.[26] The standard doctrine is underpinned by two ideas; equivalence, and adequacy.[27]

Adequacy-based mutual recognition is the well-known product of Cassis de Dijon. It applies in situations where there are certainly significant differences between the interests and values protected in Member State laws– for no-one can pretend that an absence of rules on beer composition is the same as the Rheinheitsgebot – but the Court has decided that the matters concerned in these differences do not require government intervention. They can be left to individual choice or market forces.

Adequacy-based mutual recognition is a very limited notion, because it is so sensitive to require Member States to admit the non-conforming. This shows up in the unusual situations where differences really may matter, and cannot just be left to consumer preferences, and cannot plausibly be harmonised. In these cases, the Court allows mutual recognition to be set aside. For example, significant differences in national moral perspectives can justify rejecting ‘immoral’ goods that were lawful in other Member States, genuine differences in perceptions of safety can justify rejecting foodstuffs, and differing policy attitudes towards gambling, always a difficult area, led the Court to say that mutual recognition could not be automatically and generally applied in this field.[28]

Adequacy-based mutual recognition is disruptive and deregulatory, and for this reason it is possible, and only deployed, in the internal market, where the Union has competence to harmonise.[29] Thus, if Member States truly consider that the resulting deregulation is problematic, they have the power to address this, and in practice, very often do. It has been extensively discussed how in practice Cassis de Dijon did not create a purely mutual recognition-based marketplace, but in practice led to a wave of harmonization.[30] EU legislation has largely filled the regulatory gap which Cassis de Dijon created.

By contrast, equivalence-based mutual recognition is used in situations where EU harmonization of the field is not possible, or only to a very limited extent. Given that absence of the possibility of EU corrective measures, mutual recognition can only be accepted if there is genuine equivalence between the interests and values embodied in Member State law. This is the case in the field of professional qualifications, where if facts show genuinely substantial differences between national training, then this justifies a – proportionate and often partial – rejection of the qualification.[31] It is also the case in EU criminal law, where the possibility of arrest warrants and co-operation is based on a fundamental equivalence between Member State judicial procedures. That may be contested in particular cases, as a matter of fact, but the assertion that it exists is what underpins the EU framework.[32]

Now, returning to the Coman line, the mutual recognition of Member State documents deployed here does not fit the above pattern. The differences between Member States concerning when a relationship should be legally recognized as a marriage, or when a different gender should be acknowledged by the law, reflect deeply felt, genuine, and extremely politically sensitive national choices. There is more substance and importance to these differences than is at stake in the gambling or vitamins cases or in many qualification situations, where the Court and legislation do not insist on recognition. Moreover, there is no competence to legislatively harmonise – as the Court emphasizes in Coman, this is a Member State competence – so the consequence of mutual recognition here cannot be addressed by the EU legislature. The EU (almost certainly) cannot legislate into being an EU definition of marriage or gender, even if that were politically possible or desirable.[33]

Coman et seq thus represent a deviation from existing case law. The defending Member States are compelled to recognize marriages and families and genders that are meaningfully different – in their view – from those created under their own law, but in a context where harmonization is not an option. It is Cassis for values, but without any competence to adopt legislation – pure normative neoliberalism.[34]

Mutual recognition has traditionally been used to mediate between different Member States concerns and interests. The case law sought to find ways to accommodate them both and reach a functional compromise – protecting consumers in a way that also allowed free movement.[35] Here, by contrast, it is being used to prioritise one approach over another, and to harmonise by deregulation: to achieve a regime that decentralizes as much as possible to individual choice.

How far will this go? Some regard marriage as patriarchal. Some dislike its religious roots. Perhaps some progressive Member States will (already do) recognize long term unmarried relationships as legally equivalent to marriage and also accord parental rights to long-term de facto carers. Will other Member States be required to recognize this recognition? The Citizenship Directive might have something to say on the matter, but general principles trump the details of secondary legislation.[36] What if progressive Member States recognize non-binary as a status in documents, or gender fluid, or perhaps stop certifying gender at all – will other Member States have to adapt their own registers and documents to reflect this when citizens return?[37] There are, and will continue to be, quite fierce disagreements on the age at which a child’s wish to reassign their gender should be respected. Is recognition of a child’s reassignment elsewhere also required by the principle of equality?

The ways in which people can be liberated from historical categories and structures are open-ended. It is a mistake to think that achieving equality for a few groups represents the final push towards an emancipatory EU. But for each liberatory change realized in a Member State, the free movement logic of recognition applies as impeccably as it does in the Coman line, as does the logic of equality. Nevertheless, one suspects that the Court will draw some arbitrary line beyond which it will not go – not everything has to be recognized. Not all names do, and neither will all identities.[38] Some are more equal than others.

Of course, we all have a personal morality, and most people will have a line of that form in their own minds – this to be taken seriously, and this is not. Should it be the Court who draws that line for all Europe, without any possibility of politics playing a role – because deviations from its conception of equality are outside the power of both Member States and the EU legislature?[39] Is this politically sustainable? The use of mutual recognition to deregulate identity and relationships raises many new challenges. It risks becoming over-constitutionalisation of the personal sphere.[40]

3.3.  Justification and consistency

The explanation for why the Court has extended mutual recognition is found in the way it approached justification. Obviously free movement was at stake, and so the cases turned on whether a denial of recognition could be justified. The reasons why it could not, running through all the cases, are two.

First, the rules on same-sex relationships and gender reassignment were not compatible with fundamental rights and EU values, as expressed in the principle of equality.[41] Fundamentally wrongful rules cannot justify a restriction on free movement. Second, the impact, since the applicable EU law purely concerned recognition and not domestic marriages or gender rules, would be limited and did not infringe unduly on Member State competences or interests.[42] It is inconsistent of the Court to put forward both these reasons because they cannot logically both be relevant: if denial of recognition truly is excluded as a matter of fundamental principles of EU law, then the degree of impact is legally irrelevant. They are, logically, arguments in the alternative, a tactic more commonly used by advocates than by judges in their decision. Nevertheless, they are both there in the decisions, and so deserve consideration, not least because they will certainly return in future cases. It is the first reason which is the subject of this section. The second reason is addressed in the discussion of U-turns below.

It is unusual for the Court to require a Member State to act in a way that is contrary to its proclaimed beliefs and moral choices unless there is evidence to suggest that these are not genuine.[43] The Court is typically far more deferential towards constitutional values, and policy based on national moral sensitivities, and emphasizes that these need not be shared by other Member States.[44] In the naming cases the Court overwhelmingly found in favour of the litigants – except where there was a genuine constitutional value at stake, as in Sayn Wittgenstein and Bogendorff.[45] The Belgian claim that its law on the naming of children was foundational to its national identity, in Garvia Avello, was simply implausible as a matter of fact, and was rightly dismissed, but a genuine objection to aristocratic titles, or even the hint of a title, was enough to deny Sayn Wittgenstein and Bogendorff the right to use the names they had lawfully acquired abroad. How much did the Austrian and German population really care about these laws? How politically salient were they by comparison to the question of same sex marriage in Romania? The pattern of deference in existing case law would have prima facie led to a denial of recognition in Coman et al.

Of course, the consequences for the individuals here are arguably worse. Sayn Wittgenstein could still use part of her name, and Bogendorff could still use his old one. Perhaps a failure to recognize a name is more trivial than not recognizing a marriage or gender reassignment. Yet this does not explain the different approaches, for the Court has been prepared to tolerate fairly devastating personal consequences in the service of respecting Member State values and interests. Its headscarf case law allows Member States and businesses to de facto exclude Muslim women from the workforce, in the interests of a mix of tradition, business interests and prejudice.[46] In the Zambrano follow-up KA, the interests of a Member State in immigration law – not even constitutional, just regular policy - were found to justify the expulsion of adult partners.[47] The Court was nonchalant about this – an adult is capable of living without their partner and Member States have to be able to enforce their laws. The right to reside in the EU, found in the same Article 21(1) TFEU as the Coman cases were decided upon, clearly does not entail a right to live in your own state with your partner. Expelling a partner does not prevent a Citizen from staying. Unless, following Coman, it does.

A distinctive aspect of the identity cases is that the national legal frameworks in question had all been condemned by the European Court of Human Rights (ECHR), which of course made it considerably easier to dismiss them.[48] However, the ECHR cases did not go as far as the Court of Justice; while insisting that an absolute refusal to provide any legal frame for same-sex marriage or same-sex family rights, or gender reassignment without surgery, was unacceptable, they still left space for relatively conservative rules, perhaps limited forms of registered partnership, or authority over a child, and so on.[49] They serve as background colour for the judgments, but do not provide a sufficient foundation on their own for the field of EU law that is emerging.

No, this is EU law, and an explanation for the turn away from constitutional deference must be found in EU doctrine. It is easy to find: the reason why the Court rejected constitutional justifications in the Coman line but not elsewhere was, at least partly, because it considered those justifications contrary to EU values as expressed in the EU legal principle of equality. The justifications were, essentially, inadmissible. This follows in the line of earlier cases limiting the kinds of justifications that can be put forward for derogations – reasons must not be contrary to human rights, for example, nor purely economic.[50] The Court took this approach even though in most of the cases the Member States relied on the Article 4 TEU constitutional identity clause. It was not used in Mirin, no doubt because the Romanian constitution did not say anything about gender reassignment. However, in future cases it seems very likely that there will be Member States who have put, or interpreted, the immutability of sex or the denial of gender identity into their constitutions. In Coman, Pancharevo and Wojewoda Mazowiecki however, the constitutional sanctification of heterosexual marriage and parents was put forward as an element of constitutional identity justifying a denial of recognition. It was to no effect, because the Court’s view, now repeatedly expressed, is that constitutional identities are only to be respected to the extent that they are compatible with the values, principles, and identity of the EU.[51]

The need to reach some kind of reconciliation of EU and national constitutional visions is inevitable. That Member States may have to adapt, even on matters they regard as quasi-sacred, can be seen as part of joining the EU – certainly the Court sees it thus.[52] But it is hard to read the Treaty in a way that puts all the burden of adaption on one side. After all, the entire purpose of Article 4(2) TEU is to protect constitutional and identity particularities. To limit it just to procedural or institutional matters is to make a mockery of the article. It is above all about values, and intended as a guarantee that if Member States have views on certain value issues that are in the minority in the EU, and so different from what the EU sees as its ‘own’ values, and these questions are genuinely important to that Member State or its sense of self, then they will be allowed to derogate, at least to some extent. Moreover, derogations are, by their nature, hierarchically superior to the rules to which they apply. The Court, in subordinating the derogation to the rule, reverses this and breaks from its trend of assessing constitutional derogations according to necessity and genuineness, rather than whether they fit the prevailing views elsewhere in the EU.

This is not the first case within the scope of EU law where fundamental human rights and freedoms, including the right to equality, have been restricted because of constitutional particularities or national policy preferences.[53] Indeed, every justified restriction on free movement of persons can be seen as allowing inequality between Union Citizens. Usually, however, the restrictive national – or EU – rules in question are recognized to serve some other legitimate goal, and only contingently to infringe on rights or equality. The Court has then conceded the need for a balance, or even a choice, and left Member States – or the EU legislature – quite some discretion in how to do that. Fundamental rights have never been trump cards in EU law.

The difference with the identity cases is that they involve, as the Court sees it, direct discrimination: the national rules in question serve no other purpose than to discriminate, and thus there is no legitimate national interest to balance.[54]

The categorization as direct discrimination should not be determinative. Equality is not violated when situations are treated differently but when like situations are treated differently. Most of the work is internal to the principle, in deciding what is like or unlike. There is no normative difference between balancing equality against some other constitutional value or using a constitutional value to determine what situations are alike. It is then somewhat arbitrary to say that constitutional values – or other policies – may sometimes limit equality in practice but cannot be used to determine what equality is.

In the identity cases, the constitutional or national traditions involved could be presented as concerning particular views on what differentiates men and women, to do with tradition, culture and religion. They could even be presented as aiming to protect religious values, families, and a certain conception of human dignity. I don’t endorse these views, but they are not outside the mainstream of European society, and represent the teachings of, among others, the Catholic church, to which many Europeans are still attached. Prima facie they are the kinds of values and choices that should be acknowledged and weighed, not a priori dismissed.

Moreover, in other situations where Member States care very much, the Court and the law do allow for different understandings of equality – where religion was concerned the Court has essentially washed its hands of the matter, and conceded that it was for Member States to decide whether excluding women with headscarves was an expression of, or a violation of, equal treatment.[55] Similarly, the law allows for, but does not require, positive action, so that one Member State may in good faith take action to increase the representation of people of colour, while in another Member State that same action would be seen as constitutionally problematic and precluded.[56] The Court tolerates diverse notions of what an equal society is.

But not here. There can be differing views on the proper role of religion or race in law, society, and the determination of rights, but not on the proper role of sex or sexuality. Why? The Court does not tell us. Is it merely whether the Court judges that a particular standpoint enjoys a near-consensus or strong majority in the EU? That may be politically astute, but it is constitutionally and doctrinally weak.

It is suggested here that the view that same sex relationships are relevantly different from heterosexual relationships for legal or parenting purposes, as the defending Member States in Coman and Wojewoda Mazowiecki and Pancharevo did, or that man and woman are exclusively biological concepts, could not be described as modern, progressive, empirically well-founded, or even kind. However, these views have strong roots in European traditions, religions and culture, and were almost universal just a few decades ago. They are still adhered to by significant minorities of people in many Member States. If they are the view of the majority in a certain Member States – which is far from certain but might be the case – then their understandings of marriage and the family and identity are plausible candidates for a public policy or constitutional identity derogation. They are precisely the kind of culturally rooted and deeply felt matters for which these derogations exist, and towards which the Court has traditionally shown deference. To say that as a matter of principle such identities cannot be acknowledged by EU law because they are contrary to the EU understanding of equality is to begin an open-ended judicial harmonization programme,[57] where political realities mean that inconsistencies will be inevitable, and the high moral tone of the cases will be increasingly undermined by selective blind spots.

The majoritarian nature of this programme also deserves to be noted. Right now, the Court’s value harmonization is being embraced because it has begun with sympathetic examples, where the defending states enjoy little public or governmental support: they are ruling against only the homophobes and the central Europeans. But whose choices will they block next?[58] One is reminded of Josemans, where the Court of Justice dismissed Dutch law on tolerance of cannabis because, in its view, European states should be committed to the illegality of all drugs.[59] Or Jany, where it dismissed an enlightened Dutch policy aimed to protect against trafficking in women because of relatively formal arguments to do with freedom of establishment.[60] Or its embrace of spurious and exclusionary arguments about neutrality in Achbita, before it self-corrected in LF.[61] The Court is not always woke. Coman et al raise the question whether progressive value deviations will be the ones reined in next.

3.4.  U-turns

A notable feature of the identity cases, with the partial exception of Shipova, is that they are U-turns; Citizens returning home rather than moving abroad.[62] This situation is not novel, and despite it looking a little like a judicial extension of legislation, at least in some situations, is now fairly well accepted. There was resistance from the Member States involved in these cases, but not because of the nationality of the people involved.

Yet there are particular features of these U-turns that are novel and deserve attention because of the consequences that they imply. They do not fit within the traditional U-turn reasoning developed in Surinder Singh.[63]

In that case the Court emphasized that if Member States exercised rights abroad and could not bring them home, then this would deter them from migration in the first place. That seems a counter-intuitive and over-complicated approach – surely one could just say it deters them from moving home? However, at the time of Singh Union Citizenship did not exist, and there was no obvious or clear EU law right to move to one’s own Member State as such. Thus, to protect the returnee it was necessary to make their ability to move home an element of their ability to go abroad in the first place. In the later naming cases, which mostly also involved U-turns, the Court developed this a little but did not fundamentally change it. It spoke in those cases of putting individuals at a disadvantage because they have exercised rights abroad. There is ambiguity in its formulation, but it refers to the problem of differing documents as if the restriction that they are creating is not on moving to the home state as such, but on moving around in the EU generally.[64] The home state measure is challengeable because it has, or might have, consequences abroad. Outward free movement continues to be the peg on which U-turn cases hang.

Yet this approach becomes implausible in the situations arising in the Coman line. Are we to believe that if the Citizens had known that their home state would fail to recognize their status acquired abroad then they would never have moved? On the contrary, it seems more likely that they would have stayed away. Singh logic fails. Nor is it easy to say, as in the naming cases, that the home state failure to recognize creates problems for them moving around the EU generally. With the exception of Mirin, where this argument may work, the problem is only, and exclusively, with moving back to their home state.

For this line of cases, a simpler form of reasoning is needed. The objection to the home Member State measures is simply that it made it harder for them to move home. To bring that within the law it has to be accepted that the Article 21(1) right to movement throughout the EU includes movement to the home Member State as such.[65]

This seems to be what the Court says in Coman. There it does cite Singh but does not refer to the reasoning that initial movement would be deterred by an inability to bring rights home. By contrast, it says that if Member States could refuse to recognize rights exercised abroad, then rights of Union Citizens to move and reside freely, as granted by Article 21(1), would vary from state to state (meaning, I think, according to their home Member State and its own policies), and for an individual, their own Article 21(1) right would be impeded by an inability to return. It seems to be recognizing that the right to move home is part of the Article 21(1) right to move. It lets go of the idea that to invoke their rights a Citizen always has to show a (potential) problem with moving somewhere else.

There can be no textual objection to this. Article 21(1) gives all Union Citizens the right to move and reside throughout the entire EU, without excluding the home Member State nor treating it any differently from any other Member State. On the face of it, an obstacle to movement from a host Member State to the home Member State should be treated exactly the same as an obstacle in the other direction, or one to movement between two host Member States. It is all just free movement.

However, the consequences are not trivial. For if the Article 21(1) right to reside treats the home Member State as just another Member State that the Citizen may choose to move to or live in, then it is hard to see why this should not be true of the Article 21(1) right to reside as well; Article 21(1) treats the two as parallel and apparently independent self-sufficient rights. This would amount to giving Article 21(1) full effect, by which I mean that it is understood to mean what it says – that Union Citizens must be allowed to freely move to and from and reside in all of the Member States (including their own).

But if Citizens can challenge measures which prevent them moving to their home Member State, then surely they should be able to challenge measures which prevent, or inhibit them, from living in their home Member States too, and there seems no logical reason why they should have to show prior movement to do this.[66] A stay-at-home citizen who has never moved should be able to argue: if I cannot marry my partner, or live according to the gender I identify with, I am forced to go to another Member State, and my Article 21(1) right of residence throughout the Member States are violated. My ability to live in my home Member State, following Coman, is ‘denied’. The logic follows very naturally from the textually faithful approach to Article 21 adopted in the identity cases.

This approach has echoes in Safi, where the Court found that compelling a Dutch child to move to Spain might violate its family rights and therefore be contrary to Article 20 TFEU. The reasoning was built around the best interests of the child, but adults have fundamental rights too, and one could argue that the ingredients for a right to reside in the home state, a ‘right to stay at home’, as the AG in Safi put it, is slowly emerging.[67]

This will raise doctrinal and factual questions about when such a right is violated. In KA the Court found that refusing residence to an adult partner does not force a Citizen to leave. In Coman they seem to rule the opposite. What other kinds of measures have such an impact on a person’s ability to live in dignity that they should be seen as incompatible with a right of residence? Following the AG in Safi, if you cannot marry the person of your choice, or adopt children, or combine your religion with employment, then can you be expected to live in that state? Is your right to stay at home not violated?

The obvious answer, in the light of case law, is that measures which are contrary to the Court’s interpretation of the core EU values and rights pass (or fail) this test; they are an incompatible with a right of residence, and thus contrary to Article 21. But then the combination of that article with Article 2 and the Charter becomes a far-reaching programme of value harmonization, compelling Member States to bring their laws on family, sexuality, identity, and other matters of importance to a dignified life, into line with the Court and, implicitly, the majority of other Member States.[68] The internal situation for the movement of Citizens is definitively dead, and the idea that competence to marry is outside of EU law is no longer sustainable, for restrictive rules obviously impede the right of nationals of that state to reside at home.

Many lawyers will be reluctant to go so far, despite the text of Article 21(1), and we see the Court and Advocates-General sticking for this reason to movement as an essential trigger for Article 21(1) rights. Yet this may not to be enough to contain it. In Shipova the Court found that a failure to allow gender reassignment in law is contrary to Article 21(1) because it hinders free movement, since moving with documents that do not match ones lived gender causes ‘considerable inconvenience’.[69] Factually, this is certainly true.

Ms Shipova was a Bulgarian whose case was against Bulgaria, but she lived in Italy. However, that she lived abroad is irrelevant to the reasoning, which would apply just as well if she had not yet moved but wanted to. There was no foreign documentation to be recognized in the case, and the Court’s argument does not depend on any prior movement, but on the future movement effects of limiting individual rights.[70] Could the same not be argued about marriage? Following the spirit of Shipova, do you dare to go abroad if you are not bonded by law to the people that you love, who themselves may not be EU nationals? If you cannot maintain dual nationality, then might you have to live in a Member State where you can never vote and does that not deter your movement? Might you even have to leave the Union, because your non-EU nationality is more fundamental to you – thus hindering your ability to move throughout the EU?[71] A refusal to allow same-sex marriage, same-sex parenting or dual nationality is a restriction on the Article 21(1) right to move to another Member States, just as plausibly as a failure to recognize relationships established abroad is a restriction on the right to return.

The uncomfortable truth is that Citizens are clearly not free to move and reside throughout the territory of the Member States, as Article 21(1) apparently requires, if the degree to which those Member States accept and recognize them as they are varies. Then they will always be pushed and squeezed towards some Member States and away from others.[72] This is the corollary of seeing Citizens as human beings and not simply factors of production. In viewing them in this human way, the Court in the Coman line starts a ball rolling that may be difficult to coherently stop. Yet the equally uncomfortable truth is that addressing these disparities requires the Court – since it is unlikely that the legislature will go there soon – to impose a framework to which many Member States did not think that they were agreeing, and which they may not accept, and which creates a Europeanisation of matters that are still rooted in local understandings, variable, and deeply felt.[73]

4.   Conclusion

For someone who is gay, or has gender dysphoria, and is a citizen of a conservative Member State, the motivation to move to another Member State may be high. Of course, there are all the usual considerations of language and family and friends standing in the way, but the chance to be oneself may well weigh more heavily. The Court said in Coman that it was not really impinging on Member State competences, because Member States were free to maintain their own rules – they just had to recognize rights exercised abroad. Even if that narrow reading is right – and it was argued just above that it may be hard to maintain – it seems unlikely that the U-turn will be such a marginal phenomenon. If, hypothetically, Member States were to comply with the judgments in Coman et al, it seems very likely that many of their citizens would go abroad for the purposes of establishing lives matching their sexuality and identity and then upon returning home translate their foreign-acquired rights into domestic ones. Poland and Romania might not have to perform same sex marriages, but de facto there would be same sex marriages in Poland and Romania – one would just have to have them performed abroad. In practice, the vision of a certain kind of conservative society which their governments claim to want, and which they claim that their populations support, would be unsustainable, even if it is the case that it enjoys majority democratic domestic support and reflects deeply felt aspects of religious and national identity.

If one is persuaded by the limit set by the Bundesverfassungsgericht (BvE) in Lisbon in its Treaty of Lisbon judgment, that the EU goes too far when it deprives citizens of Germany of the capacity to determine the core socio-economic conditions of their society – to shape their living space – then it would seem a fortiori the case when it is the socio-cultural conditions, for Europe would not be Europe any more if the economic was more sacred than matters of identity, culture and history.[74] Of course, this argument is troubling when the shaping involved is objectionable to most Europeans, and discriminatory according to the majority understanding of that term. However, this contribution has been written from the perspective that majority judgments about the minority culture should not be exclusively important in deciding whether it has a right to continue to exist in Europe.[75] One thing that the identity judgments show is that thanks to the broad scope of free movement, universal EU values will extend the competence of the EU and the Court considerably, and leave almost no space for normative diversity between Member States. That is a slippery slope both to conflict, and to constitutional and constitutionalized intolerance.[76] It may also lead to a middle-of-the-road majoritarian normativity imposed across the continent which is ultimately as oppressive to the avant-garde as it is to the reactionary.

The identity cases thus suggest a need for a more cautious and competence-aware approach to the law, and respect even for constitutional identities that other Member States and the Court do not like, or wish to pretend are no part of Europe. That respect need not be unbounded, but a structured, reasoned, margin of appreciation needs to enter the law, rather than the somewhat ad-hoc selections of what is intolerable to the Court that we see now. Free movement provides paradoxical support for this suggestion, for while intervention by force – and law is force – in another state’s politics may sometimes be justified by the need to protect persecuted minorities, within the EU they do at least have the option of living more freely, and on equal terms, elsewhere, and that justifies a more reserved and pluralistic approach.

The role of Article 2 TEU, and the Charter, in this pluralism could be as shared nominal values, interpreted in different ways, but still meaningful as a focus for dialogue and convergence.[77] The Court, by contrast, has a vision of harmonization. It insists that Member States should not just share the values nominally but adhere to the same interpretations of them, its interpretation. This is troubling: it is very doubtful whether the Court has the legitimacy to be the sole interpreter of family values in Europe.[78] It should also be noted that none of its many attempts at constitutional power grabs have ever fully succeeded: no Member State apex court has ever acknowledged ultimate EU law supremacy over their constitution and quite a few have shown themselves prepared to translate this resistance into action.[79] Rather than spreading EU values the Court is more likely to be stirring up conflict.[80]

Indeed, we risk a Roe vs Wade type situation in the EU, where constitutional entrenchment of a moral standpoint before it enjoys consensus support entrenches positions, denies political development, mobilises opposition, and makes discussion around the issue toxic.[81] One might ask, when each of these cases was decided, who opened the champagne? It seems clear that it was people on both sides – those supporting the claimants and what they represent, but also right-wing leaders, who were handed useful ammunition against the EU. Whatever else it was, Coman was a gift to the Orban class of European politicians.

There are of course more optimistic possibilities. It may be that the empirical question of the effects of Coman et al is more important than its doctrinal novelty. What will it actually do to societies and laws? Will backlash and non-compliance be the dominant responses, leading to an undermining of EU law without doing much in fact for individual rights? It is certainly a real risk, and it hardly needs empirical research to make the claim that saying to a person or community ‘you must do X’ is one of the least effective ways of achieving that result, and absent a very great threat of force and the ability to create fear, it is more likely to make them do ‘not-X’.

Yet the states involved are changing, and these cases might be interventions that encourage progressive and younger forces and help speed up that change.[82] They might tip the balance and be seen from the future as the beginning of the end of the social vision they attack. In Greece, same-sex marriage was legalized following Coman, and the law also provided for recognition of same sex parental rights established abroad, suggesting implementation of Pancharevo. This is, it might be said, how EU law is supposed to work. But will Greeks perceive that they were pushed into this decision by the Court of Justice? A compelled choice is far less valuable than one freely taken and often creates a ticking political time-bomb. If Coman et al nudge states towards change they may come to be seen as huge successes. If they are seen as (trying to) compel change, they may backfire, in a Union that is diverse and fragile, and where people are prepared to fight to destruction for what they perceive to be moral. Time will show us whether these cases explode.