Consensus analysis in Vallianatos and Others v Greece

Further to my post on the ECHR Blog in respect of Vallianatos and Others v Greece, I have been reflecting more on the use of consensus analysis in the judgment.

Here is what I wrote on the ECHR Blog:

A further interesting feature of the Court’s review is its discussion of European consensus on partnership rights. The Court noted that ‘although there is no consensus among […] member States, a trend is currently emerging with regard to the introduction of forms of legal recognition of same-sex relationships’ and cited the existence of ‘seventeen member States [that] authorise some form of civil partnership for same-sex couples’. In light of this, the Court stated that ‘the trend emerging in […] member States is clear: of the nineteen States which authorise some form of registered partnership other than marriage, Lithuania and Greece are the only ones to reserve it exclusively to different-sex couples’. The Court, therefore, observes that Lithuania and Greece are out-of-step with the other seventeen states that have an alternative system to marriage for recognizing relationships and make this available to same-sex couples. However, as if to further justify the relevance of this ‘clear trend’ to its judgment, the Court states:

‘In other words, with two exceptions, Council of Europe member States, when they opt to enact legislation introducing a new system of registered partnership as an alternative to marriage for unmarried couples, include same-sex couples in its scope’.

This statement, as I read it, is factually problematic because some member states among the seventeen said to constitute the ‘clear trend’ (that is the states, excluding Greece and Lithuania, that have civil partnership legislation that extends to same-sex couples) did not enact civil partnership legislation as ‘an alternative to marriage for unmarried couples’. In the United Kingdom, for example, because opposite-sex couples are excluded from registering a civil partnership (and have no alternative to marriage) the Civil Partnership Act 2004 cannot be described as ‘a new system of registered partnership as an alternative to marriage for unmarried couples’ that was designed to ‘include same-sex couples in its scope’. The Court seems to suggest that there is a ‘trend’ in Europe for introducing legislation to make available to all unmarried couples an alternative system of registered partnership and that Greece and Lithuania are out of step with this. But the trend is not as simple as this as not all contracting states that enacted civil partnership legislation made it available to all married couples (as in the United Kingdom) and, therefore, those states are not pursuing the objective implied by the Court. In the case of the United Kingdom, for example, the ambition of the Civil Partnership Act 2004 was not to offer ‘unmarried couples’ an alternative system to marriage but to offer same-sex couples a system of partnership registration that was not marriage. In addition to this, opponents of same-sex partnership rights will point out that Greece is not out-of-step in withholding partnership rights to same-sex couples and, on the contrary, is in line with the majority of other contracting states. Therefore, the Court’s consensus analysis as a basis for supporting its judgment bears (as is often the case) some additional scrutiny. This is especially so since the consensus argument of the majority is invoked by Judges Casadevall, Ziemele, Jočienė and Sicilianos in their joint concurring opinion to justify their decision in this case as opposed to their dissenting position taken in X and Others v. Austria.

The key issue for me in respect of the Court's use of consensus analysis is its statement that 'member States, when they opt to enact legislation introducing a new system of registered partnership as an alternative to marriage for unmarried couples, include same-sex couples in its scope’.

A closer look at the Court's claim

The key question asked by the Court in Vallianatos and Others v Greece was 'whether the Greek State was entitled, from the standpoint of Articles 14 and 8 of the Convention, to enact a law introducing alongside the institution of marriage a new registered partnership scheme for unmarried couples that was limited to different-sex couples and thus excluded same-sex couples'?

To answer that question, the Court relied upon data relating to same-sex marriage and civil partnership legislation in Europe.

The most important aspect of the Court's data is as follows:

'The comparative law material available to the Court on the introduction of official forms of non-marital partnership within the legal systems of Council of Europe member States shows that [...] seventeen member States (Andorra, Austria, Belgium, the Czech Republic, Finland, France, Germany, Hungary, Iceland, Ireland, Liechtenstein, Luxembourg, the Netherlands, Slovenia Spain, Switzerland and the United Kingdom) authorise some form of civil partnership for same-sex couples'.

It is from this that the Court draws its conclusion that: 

'...the trend emerging in the legal systems of the Council of Europe member States is clear: of the nineteen States which authorise some form of registered partnership other than marriage, Lithuania and Greece are the only ones to reserve it exclusively to different-sex couples [...] In other words, with two exceptions, Council of Europe member States, when they opt to enact legislation introducing a new system of registered partnership as an alternative to marriage for unmarried couples, include same-sex couples in its scope'.

Does the data provided by the Court support its statement that there is a clear trend that states are introducing an alternative system to marriage for unmarried couples, including same-sex couples within its scope, and that Greece and Lithuania are outliers?

Of the 17 contracting states included in the Court's list, the following can be observed:

• Iceland should not be included in the list because, since 2010, it no longer provides civil partnerships for same-sex couples but enables all couples (regardless of sex) to marry. In this sense, Iceland is similar to Denmark, Norway and Sweden - they are all countries that introduced civil partnerships for same-sex couples and then replaced them with full marriage recognition.
• The only countries that have made available an alternative system to marriage for unmarried couples and included same-sex couples within its scope are: Andorra, Belgium, France, Luxemburg, the Netherlands and Spain (6).
• The countries that have made civil partnerships available exclusively for same-sex couples and offer unmarried opposite-sex couples no alternative to marriage are: Austria, the Czech Republic, Finland, Germany, Hungary, Ireland, Liechtenstein, Slovenia, Switzerland and the United Kingdom (10).

The first conclusion to be drawn from this is that only 6 contracting states can be cited in support of the claim about a trend of states enacting 'legislation introducing a new system of registered partnership as an alternative to marriage for unmarried couples [that] include[s] same-sex couples in its scope'.

The second conclusion is that a larger group of contracting states can be cited to contradict the claim about a trend because 10 states did not enact legislation to provide an alternative to marriage for unmarried couples. In limiting civil partnerships to same-sex couples to whom marriage was not available these states provided no alternative to marriage to any unmarried couple (the word 'alternative' implies choice).

The third conclusion is that Greece and Lithuania are not outliers in Europe in respect of a trend regarding partnership recognition as presented by the Court. In not making partnership registration available to same-sex couples, these two states simply conform to the majority position in Europe which is to not provide same-sex couples with any form of partnership recognition.

Ultimately, therefore, although it is correct to observe, as the Court does, that 'of the nineteen States which authorise some form of registered partnership other than marriage, Lithuania and Greece are the only ones to reserve it exclusively to different-sex couples' it is not correct to conclude that 'with two exceptions, Council of Europe member States, when they opt to enact legislation introducing a new system of registered partnership as an alternative to marriage for unmarried couples, include same-sex couples in its scope'.

Does the substance of consensus analysis matter?

In one sense, the fine detail of the Court's consensus analysis may appear unimportant - especially if one believes (as I do) that the Court ultimately reached the correct conclusion in Vallianatos and Others v Greece.

However, the methods by which the Court reaches its judgments are of crucial importance.

This is because, from a closer inspection of the data provided, it becomes clear that it would have been perfectly prossible for the Court to reach a different judgment in this case on the basis of consensus analysis. The Court could have stated, for instance, that a wide margin of appreciation was available to Greece because of an absence of consensus in Europe in respect of the arrangements relating to marriage and civil partnership. The Court could have stated that because contracting states are so divided on whether to offer  an alternative to marriage to opposite sex couples (in some states, such as Hungary, the suggestion to offer an alternative to marriage to opposite-sex couples caused a major constiutional event) that no conclusion could be drawn in respect of the situation in Greece.

In my view, consensus analysis was irrelevant to Vallianatos and Others v Greece and the Court should not have invoked it. Although, as I have argued elsewhere, the Court uses consensus analysis as a device to legitimize its reasoning, it is often highly problematic.

In my opinion, the reason the Court relied on consensus analysis in this case was because it simply asked the wrong question. Instead of asking the question it did ('whether the Greek State was entitled, from the standpoint of Articles 14 and 8 of the Convention, to enact a law introducing alongside the institution of marriage a new registered partnership scheme for unmarried couples that was limited to different-sex couples and thus excluded same-sex couples'?) the Court should have considered whether the absence of any available partnership recognition for the applicants was in principle a form of discrimination contrary to Article 14 taken in conjunction with Article 8.