A. Introduction
The European Court of Human Rights (“ECtHR” or “the Court”) uses various methods to interpret and apply the rights guaranteed by the European Convention on Human Rights (“ECHR” or “the Convention”)Footnote 1 and its Protocols. The search for a possible European consensus or trend on a given issue is part of these methods. This comparative analytical approach is both well established and, at the same time, controversial and subject to various criticisms. The question of when and how a European consensus or trend contributes to shaping rights guaranteed by the Convention and its Protocols lies at the heart of the present Article.
This Article mainly aims at listing the judgments of the Grand Chamber of the Court (“the Grand Chamber”) in the last twenty-five years where a consensus or trend analysis had an impact on the ratio decidendi of the case, at measuring this impact, and at showing what shaping of rights actually means in each case. Such a triple study has never been conducted before. Based on the results obtained and of doctrinal considerations, this Article also intends, on the one hand, to address some methodological and conceptual challenges raised by the use of a European consensus or trend to shape rights and, on the other hand, to determine if some cases are more prone than others to a potentially persuasive consensus or trend analysis.
Shaping a right is an elusive notion in many ways. It relates to the delimitation of the actual protection conferred by that right. Human rights are usually formulated in a vague way. When issuing a judgment, the Court sometimes defines the scope of a right in a general way—not solely confined to the case at hand, in other words—or it indicates which points must be examined to apply a right or determine whether it has been violated. From this perspective, judicial adjudication leads to the shaping of human rights as understood in the present Article. Interestingly, shaping rights is not synonymous with defining a given right’s scope. Indeed, balancing, notably in the proportionality test, surely relates to the specific circumstances of a case, but may also reveal general considerations on the steps to be taken in the analysis.
The judgments of the Court offer an inexhaustible source of developments shaping the Convention’s rights. From a methodological standpoint, this Article examines the judgments rendered by the Grand Chamber since the entry into force of Protocol No. 11 to the Convention (November 1998–September 2024), in which the question of the existence of a European consensus or trend is addressed. This Protocol fundamentally restructured the Convention’s protection mechanism and ushered in a new era for the Court. This period of just over twenty-five years is long enough to provide an adequate perspective on the methods followed by the Court. The judgments handed down at the Grand Chamber’s level deal, as a rule, with serious questions affecting the interpretation or application of the Convention and its Protocols or with serious issues of general importanceFootnote 2 and have supposedly been subject to a particular analytical and methodological effort. The set of judgments gathered will be carefully analyzed, and the cases where rights are actually shaped by a European consensus or trend will constitute the core of this Article, as they show the Court’s approach to its final stage. It should be noted, however, that consensus or trend analyses are also performed by Chamber formations. Several Chamber judgments are quoted in the present Article.
The present Article provides information on the impact of the consensus or trend analysis performed by the Court on the outcome of its judgments. Methodologically, this impact assessment is based on the written judgments of the Grand Chamber and, more specifically, on their ratio decidendi. Of course, many considerations, written or not, affect a case’s outcome. At the end of the day, the written judgment is what counts, what transcribes the Court’s opinion and position, and what constitutes jurisprudence and, thus, builds the case law. Consensus and trend are both used in this Article to reflect the Court’s case law. The Court does not make a clear distinction in this respect, and the difference between one and the other seems gradual rather than conceptual. As such, a trend may be seen as a stage towards consensus. This Article does not address this difference as such and takes into account the judgments where these notions and similar ones are used.
This Article starts with some comments on the task of the court to interpret rights and balance these rights or related interests, which leads to shaping rights—in certain cases with the help of a consensus or trend analysis.Footnote 3 The Court quite often conducts such an analysis, but the cases where the latter actually shapes the rights at stake are rare.Footnote 4 Moreover, shaping rights through European consensus or trend is not without conceptual and methodological challengesFootnote 5 and also raises the fundamental—though unanswered—question as to whether some cases are more prone than others to a potentially persuasive consensus or trend analysis.Footnote 6 A nuanced approach appears inevitable in this complex and multifaceted context, as does the Court in most of its consensus or trend analyses.Footnote 7
B. Interpreting, Balancing, and Shaping
Interpreting rights and balancing them or interests are two different judicial tasks, at least at first glance. In various ways, however, they can be part of a continuum and help shape rights guaranteed by the Convention and its Protocols.Footnote 8 At each stage of this continuum, the European Court may search for a European consensus or trend.Footnote 9
I. The Process of Shaping Rights
Interpreting rights involves defining their scope. This aspect of rights’ interpretation can be a rather abstract exercise, even if the judges may not, depending on the circumstances, ignore or may even explicitly take into account the concrete situation with which they are confronted. As the Convention is an international treaty, its interpretation is notably governed by Articles 31 to 33 of the Vienna Convention on the Law of Treaties signed on May 23, 1969 (“the Vienna Convention”).Footnote 10 In order to identify the context in which the Convention (“ECHR”) applies, its Preamble, as amended by Protocol No. 15, refers to, among other things, “a common heritage of political traditions, ideals, freedom and the rule of law,” the principle of subsidiarity and the margin of appreciation of the “High Contracting Parties.” For its part, balancing not only involves weighing competing rights or interests against each other, but also relates to proportionalityFootnote 11 and, to some extent, reasonableness or fairness tests. It is mostly performed by the Court’s judges regarding the concrete circumstances of a case. Of course, judges may consider the expected consequences of their judgment and, in this regard, adopt a more general and abstract approach. Many rights, but not all of them, allow for some balancing in their application, if only because they contain indeterminate legal terms such as necessity or fairness.
In the rich, complex and evolutive case law of the Court, a clear line cannot always—it is a euphemism—be drawn between interpreting the scope of rights and balancing interests or rights.Footnote 12 Regrettably or not, a back and forth continuum can be observed in a significant number of the Court’s judgments between these two tasks or, in other words, between pure interpretation of a right’s scope and pure balancing. This also means that both may contribute to shaping rights, or in other words, to define their actual, precise and effective meaning, not only in the case at hand, but also in other actual or future similar cases. The last four columns of the chart in the Appendix show at what stages the process of shaping rightsFootnote 13 may take place.
In the present Article, shaping rights refers to the multidimensional judicial adjudicative work, performed by the Court, which leads to defining the scope of a right or to indicating which points must be examined to apply a right or determine whether it has been violated. Through interpretation of the scope of a given right or the Convention as a whole, and through balancing rights and interests, the Court defines the actual, precise and effective meaning of the rights guaranteed by the Convention or its Protocols. The last column of the chart in the Appendix illustrates what “shaping rights” through European consensus or trend may actually mean in each judgment listed, something that has never been done before.
It is not argued in this Article that defining the scope of a right is equal to the balancing of rights and interests. These are two different steps in the analysis that the Court is supposed to perform, even if they are not always clearly distinguished and are sometimes overlapping. However, the point is made that a consensus or trend analysis can and does actually take place at either stage, as is apparent from the judgments listed in the Appendix.
In interpreting a constitution, a court sometimes also defines unenumerated or unwritten rights. Until now, the Court has not explicitly used this technique and has always sought to link the most significant developments in case law to the rights expressly guaranteed in the Convention and its Protocols. However, some cases explore the limits of explicitly guaranteed rights and may even venture beyond. When the Court asks whether there is a general right to repatriation under Article 3 § 2 of Protocol No. 4, it—rightfully in my view—looks at international law and material, the Council of Europe’s material and European Union law or case law and also examines whether a consensus between the Council of Europe’s Member States (“the Member States”) can be established on this issue.Footnote 14 Indeed, such a consensus analysis is notably advocated for “when the Court intends to add new rights.”Footnote 15
The process of shaping rights is certainly influenced by many factors, sources and circumstances. This Article is dedicated to one of those—European consensus or trend. Furthermore, the process is evolutive and dynamic, reflecting the idea that the Convention is a “living instrument.” From this perspective, it can prove quite unpredictable. One of the main challenges facing the Court consists of framing this process and making it more predictable. In many cases, the Court has felt the need to search for an eventual European consensus or trend.
II. Use of a European Consensus or Trend by the European Court of Human Rights
A consensus or trend analysis may help circumscribe the Member States’ margin of appreciation in an area or on a specific issue and, as the case may be, define the scope, the meaningFootnote 16 or, depending on the context, the most appropriate application of a given right. From this perspective, the quest for legitimacy cannot be an independent goal, and the Court’s legitimacy “cannot be solely dependent on a comparative exercise across forty–seven states.”Footnote 17 That said, the performed analysis may help, little by little, build the Court’s legitimacy.Footnote 18 In reality though, a consensus or trend analysis can serve several functions, such that a consensus or trend “licenses and restrains, persuades and substantiates.”Footnote 19
For all that, not every European consensus or trend really shapes the right at stake, for two reasons at least. First, the Court sometimes disregards the consensus or trend for other reasons that it deems overriding, as shown at the beginning of Section III of Part C below. Second, a European consensus or trend can simply provide the Court with some general indications on the way to balance the interests at stake. In such a case, it would be a stretch to consider that the Court significantly shapes the right in question. Furthermore, a consensus can exist at the level of rules or simply at the level of principles.Footnote 20 When the Court, for instance, considers that there is a European consensus to duly consider the best interests of children when applying Article 8 of the ECHR and, more concretely, the proportionality test foreseen in the second paragraph of this provision,Footnote 21 it does not really shape the right to respect private and family life. Similarly, the observation that the “existence of a broad consensus at the international and European level” concerning the need for special protection of asylum-seekers as a particularly underprivileged and vulnerable population groupFootnote 22 has a relatively low normative density.
By contrast, when the Court recognized a right of conscientious objection for the first time, it directly shaped Article 9 of the ECHR.Footnote 23 In the Court’s case law, a proportionality analysis can also have a shaping impact on the right at hand. When, for instance, the Court tries to determine whether the right of working prisoners to unionize is covered by Article 11 of the ECHR and looks for an eventual consensus, which it denies in the proportionality analysis,Footnote 24 it addresses a question that may shape the right in question.
It follows from the foregoing that a consensus or trend analysis also relates to matters of interpretation of the Convention. Christian Djeffal refers to Articles 31 to 33 of the Vienna Convention when he considers that “[t]he use of the consensus method should be restricted to questions of balancing, no use should be made of it in the process of interpretation.”Footnote 25 However, the reference to “context” in Article 31 is flexible enoughFootnote 26 to allow for—or, at least, not to prohibit—an interpretation of human rights from a multi-level perspective. This may lead to look for an eventual European consensus or trend in the process of interpreting rights.Footnote 27
One may argue that international norms require an autonomous interpretation, so it is not obvious why reliance on the Member States’ laws and practices should be relevant to the determination of the scope of protection of international norms. While this is a valid point, human rights are, for most of them, formulated in rather vague terms and are not isolated, autopoietic creatures. Their interpretation and, thus, their actual meaning are affected by many factors, including the environment in which they are immersed. From an inter-level perspective, said laws and practices may signal how an issue is perceived and dealt with by the Member States.Footnote 28 Is this completely irrelevant on principle when the Court is faced with the very same issue? Is it out of the question for the Court to consider a consensus between the Member States in such a context—a consensus which may then influence the interpretation of the scope of the right in question? The Court’s well-established case law provides a negative answer to these two questions, as can be seen, for instance, from the judgments listed in the Appendix. Furthermore, this inter-level dynamic can relate to countless issues, as many rights are vaguely formulated and, therefore, uncertain, but potentially broad in scope.
Anyway, shaping rights, as understood here, is a broad notion that can relate to the scope of rights, the balancing of rights and interests or a blend of both.Footnote 29 It has already been used, for example, in a book that covered many aspects of the case law of the Court, in particular the determination of the scope of human rights.Footnote 30 As we shall see, the judgments rendered by the Grand Chamber during the last twenty-five years show when and how a consensus or trend analysis helped defined the contours of a right. In some cases, the Court uses such an analysis to determine the scope of the right it must interpret, while in other cases to balance the rights and interests at stake. This observationFootnote 31 is clearly supported, for instance, by the judgments listed in the Appendix.
C. Rights Shaped by European Consensus or Trend in the Last Twenty-Five Years
The Court quite often performs an analysis of the Member States’ laws and practices or of relevant international material.Footnote 32 However, the cases where rights have actually been shaped by European consensus or trend are quite rare. In our count, in the last twenty-five years, 27 out of 424 judgments on the meritsFootnote 33 by the Grand Chamber established a consensus or trend having a “shaping impact” on the protections conferred by rights guaranteed by the Convention or its Protocols, and only one advisory opinion based on Protocol No. 16 contained a section on comparative law material having such an impact.Footnote 34 No decision of the Grand Chamber on admissibility during this period offered significant developments from this perspective. In at least six other cases during the last twenty-five years, a European consensus or trend confirmed existing case law.Footnote 35 Finally, a European consensus or trend was established in a few other cases, and the Court refrained from evolving the protection conferred by the right at hand.Footnote 36
I. Consensus or Trend Having a “Shaping Impact”
Measuring the full impact of a consensus or a trend analysis on a judgment’s or an opinion’s outcome is a very difficult task, as many factors come into play—some hardly measurable. The written judgment is and should remain the main point of reference, as it reflects what the judges want to communicate to the outside as relevant. This Article is based on this source.
Methodologically, the study adopts a four-stage process. First, the selection retained the judgments using notions such as “consensus,” “trend,” “tendency,” “majority of the Contracting Parties,” “majority of the Member States,” “number of Contracting Parties,” “number of Member States,” “common European standard,” “commonly accepted standards,” “common ground,” “common approach,” or “uniform approach.”Footnote 37 The words used by the Court in the twenty-seven selected judgments is indicated in the fifth column of the chart in the Appendix. Second, the selection has been narrowed down to only those cases in which the consensus or trend analysis also appears in the reasoning of the judgment. Third, the final selection consists of keeping the judgments in which this analysis is used in the judgements’ ratio decidendi to shape the right at stake—in other words, to define the right’s scope or to indicate which points must be examined to apply the right or determine whether it has been violated. Fourth, the strength of this impact is evaluated in the last step of the process. All these steps have been double-checked. Moreover, a final cross-check has been made with all the legal literature cited in this Article to see whether their authors have selected other judgments of the Grand Chamber.
The impact can qualify as strong when the consensus or trend explicitly and prominently appears in the ratio decidendi of the judgment and contributes in a significant way in determining the outcome of the decision.Footnote 38 The impact is of medium intensity when the consensus or trend appears in the ratio decidendi of the judgment but the Court also mentions other important elements that affect the outcome.Footnote 39 The impact is weak when the Court conducts some comparative analyses or mentions some international treaties or documents but barely refers to them in the actual rationale of its decision. Similarly, the consensus or trend analysis sometimes only fulfils a supporting function, such as when it is mentioned but the rationale of the decision is already in place and the conclusion made. When international or comparative material are mentioned by the Court in the section on the relevant legal framework and practice but not at all referred to in its assessments,Footnote 40 an impact cannot be established in the ratio decidendi and those judgments will not be considered here. Of course, one may wonder why the Court would go through the trouble of including international or comparative material in the judgment if it intended this material to play no role whatsoever in the interpretation of the Convention. However, this material is logically collected before the Court carries out its assessment. Once this has been done, the Court may feel it is appropriate to leave a trace of this collection in the judgment. This is, however, only conjecture outside the scope of the present Article.
The chart in the Appendix provides an analytical overview of the twenty-seven judgments and the one advisory opinion of the Grand Chamber in which some European or international consensus or trend helped shape rights guaranteed by the Convention or its Protocols. It should be emphasized that the analysis in this chart provides an order of magnitude, as some judgments may well fall into a grey zone. In other words, choices have been made that can be debated. In particular, the difference between strong and medium impacts is not clear-cut, but rather gradual. For the sake of transparency, the points on which a classification is based are indicated in the penultimate column of the chart in the Appendix. The total number of relevant judgments or their classifications may also be discussed, but different assessments in one case or the other will not change the overall picture.
The relevant column in the chart in the Appendix evaluates the impact of the consensus or trend on the ratio decidendi of the selected judgments and shows that this impact was strongFootnote 41 or rather strongFootnote 42 in fifteen of them.Footnote 43 The consensus or trend is established based on an analysis of both international, European or other regional texts on the one hand and the Member States’ laws or practices on the other. The number of Member States considered varies widely, from around twenty up to all of them. The three judgments on Article 6 of the ECHR reflect rather incremental evolutions in the case law.Footnote 44 Furthermore, four judgments relating to Article 8 of the ECHR,Footnote 45 Article 14 in conjunction with Article 8 of the ECHRFootnote 46 or Article 3 of Protocol No. 1Footnote 47 are focused on balancing interests in rather specific situations. This means that very significant rights’ shaping prominently and, to a certain extent, predominately occurred through consensus or trend only seven times—six judgmentsFootnote 48 and two twin judgments.Footnote 49 These eight judgments relating in particular to Articles 3, 7, 8, 9, 10, and 11 of the ECHR are based on an analysis of the vast majority of the Member States’ laws and practices, as shown in the fourth column of the chart in the Appendix. The comparative analysis is, however, very succinct and indirect in at least one caseFootnote 50 and the case law on climate change of only eight Member States is mentioned in another one.Footnote 51 This small number of judgements immediately puts into perspective, and even somewhat demystifies, the real scope of the consensus or trend analysis on the evolution of the Court’s jurisprudence from the perspective of actual rights shaping.
These judgments show that the use of the consensus or trend analysis to shape a right can occur when the Court delimitates the right’s scopeFootnote 52 or when it assesses the proportionality of the challenged interference,Footnote 53 as shown in the sixth and seventh columns of the chart in the Appendix. The shaping of rights may relate to procedural or substantial aspects, as shown in the last column of the chart in the Appendix. For instance, the Court considers that a life sentence remains compatible with Article 3 of the ECHR only when there are “both a prospect of release and a possibility of review,”Footnote 54 or that “the principle of retrospectiveness of the more lenient criminal law” is implicitly guaranteed by Article 7 of the ECHR.Footnote 55 From a more substantial perspective, the Court, for example, deduced a right to obtain legal recognition of “gender re-assignment” from Article 8 of the ECHR,Footnote 56 a right for individuals to effective protection by state authorities from serious adverse effects of climate change on their life, health, well-being and quality of life from the same provision of the ConventionFootnote 57 and a right to conscientious objection from Article 9 of the ECHR.Footnote 58
According to our reading, in the twelve other judgments and in the advisory opinion listed in the chart in the Appendix, the consensus or trend established by the Grand Chamber had a mediumFootnote 59 or weakFootnote 60 impact on the shaping of the rights at stake or fulfilled a rather supportive function in this regard.Footnote 61 The consensus or trend may have contributed to the outcome of these cases, but to a lesser extent than in the other fifteen judgments. In any event, it does not form the backbone of the decision. The key findings drawn from the first fifteen judgments also apply, mutatis mutandis, to the other twelve. Some judgements relate to the scope of the right, others to the proportionality of the interference with the right at stake, some to rather procedural aspects and others to more substantial issues, as shown in the chart in the Appendix.
Under these circumstances, the consensus or trend analysis may simply show that a given solution is not unknown or is widespread among the Member States or in international material. It is not referred to in a significant way in order to shape the rights guaranteed by the Convention or its Protocols. Accordingly, the scope of the analysis—especially in terms of the Member States considered when the impact is weak—can be reduced in a more admissible way. Finally, the impact can be strong or rather strong for one aspect of a case, but medium or weak for another aspect.Footnote 62
II. Consensus or Trend Having a “Confirming Impact”
In at least six other cases during the last twenty-five years, a European consensus or trend was used by the Grand Chamber to confirm existing case law. Quite recently, it again noted the ongoing European trend—now deemed as clear—towards legal recognition and protection of same-sex couples.Footnote 63 In another case, the Court confirmed and consolidated the principles established in its case law with regard to the protection of whistleblowers, “by refining the criteria for their implementation in the light of the current European and international context.”Footnote 64
More than ten years ago, the Grand Chamber emphasized the “common ground between the Member States of the Council of Europe regarding the importance of equal treatment of children born within and children born outside marriage” and mentioned the “uniform approach today by the national legislatures on the subject” as well as the “social and legal developments definitively endorsing the objective of achieving equality between children.”Footnote 65 Referring to a previous judgment,Footnote 66 it again mentioned, more than fifteen years ago, “an emerging international consensus among the Contracting States of the Council of Europe recognizing the special needs of minorities and an obligation to protect their security, identity and lifestyle.”Footnote 67 However, the translation of this general recognition into more precise rules regarding the schooling arrangements for Roma Children was not based on a consensus or trend analysis.Footnote 68
In 2012, the Grand Chamber confirmed previous judgmentsFootnote 69 in which landowners’ obligations to tolerate hunting on their property was considered to impose a disproportionate burden on landowners who were opposed to hunting for ethical reasons, observing that various Member States had “amended their respective legislation or modified their case-law in order to comply with the principles set out in these judgments.”Footnote 70 Based on an analysis of international and European material, as well as comparative-law information, it also discerned that year some “trend, if anything—towards fewer restrictions on convicted prisoners’ voting rights.”Footnote 71
III. Consensus or Trend without Evolving Protection
In a few other cases, the Grand Chamber admitted the existence of a European consensus or trend but refrained from evolving the protection conferred by a right guaranteed by the Convention and thus from shaping it. These judgments related to the death penalty,Footnote 72 the interdiction for a lawyer to conduct her or his own defense in criminal proceedings,Footnote 73 restrictions on obtaining an abortion,Footnote 74 and the prohibition on the use of ova and sperm from donors for in vitro fertilization.Footnote 75
Finally, in about eight cases during the last twenty-five years, the Grand Chamber referred to a European consensus or a trend when the vast majority, or a large number of, Member States supported the position of a respondent state, at least in some respects, and the Court thus refrained from evolving the protection conferred by a right guaranteed by the Convention or its Protocols. The judgments concerned Swiss courts’ refusal to examine a compensation claim relating to alleged acts of torture in Tunisia,Footnote 76 the notion of “continuous criminal offence,”Footnote 77 criminal sanctions to prevent the disclosures of certain confidential items of information,Footnote 78 the absence of a special status for journalists when they fail to comply with police orders to leave the scene of a demonstration,Footnote 79 disciplinary measures applicable to members of parliament for disorderly conduct in Parliament,Footnote 80 a national legal and administrative framework aimed at achieving pluralism in audio-visual media,Footnote 81 employers’ discretion in determining sanctions that are best adapted to accusations against employees,Footnote 82 and mechanisms for transferring title in accordance with principles similar to adverse possession in common law systems.Footnote 83
While eight judgments may provide an order of magnitude, other judgments may be relevant because the Court uses very fluctuating terminologyFootnote 84 and sometimes conducts very cursory comparative analyses. Furthermore, a consensus may exist on some general principles followed by the defendant state, but not on the specific policy it has chosen.Footnote 85 All in all, it seems difficult to contend that rights guaranteed by the Convention or its Protocols have actually been shaped in these cases. Rather, a European consensus or trend shaped the area outside the rights’ scope or the area within the right’s reach but where the latter can be limited.
D. Some Conceptual and Methodological Challenges
Shaping rights through a European consensus or trend, as in the cases listed in the Appendix, raises several fundamental issues, among which some of them shall be addressed here, especially in light of the findings that can be drawn from the most significant judgments of the Grand Chamber in this field. Quite naturally, the first one which comes to mind relates to the duty or, on the contrary, the freedom to conduct a consensus or trend analysis.Footnote 86 When such an analysis is performed, the material at which the Court looks is of special interest.Footnote 87 Finally, the weight of a European consensus or trend may vary considerably depending on several factors.Footnote 88
I. Duty or Freedom to Conduct a Consensus or Trend Analysis
On the one hand, the Convention does not expressly impose any obligation on the Court to conduct a consensus or trend analysis. Moreover, there is no general principle of international law that would entail this duty. Nor does the Court’s well established and unambiguous jurisprudence suggest a systematic obligation. Indeed, in none of the judgments listed in the Appendix, nor, more broadly, in those mentioned in Parts C and E of this Article, did the Court mention any duty to conduct such an analysis, even if some of these cases are particularly sensitive because the Court shapes rights.
On the other hand, a purely discretionary power left to the Court, without criteria of exercise, may lead to arbitrariness.Footnote 89 Can the Court simply cherry pick when to conduct a consensus or trend analysis? A positive answer to this question would make the case law extremely unpredictable, including from a methodological standpoint. Moreover, this can create inequalities, not to mention discriminations, in the sense that whether or not a consensus or trend analysis is carried out possibly affects the outcome of the judgment. On the one hand, the parties in a case before the Court do not have the same resources to conduct such an analysis themselves, which could then convince the Court of the value of such an approach in casu. On the other hand, two cases raising questions of interpretation of a right or balancing of rights or interests of comparable difficulty could be treated differently by the Court without stated reasons. Any analytical inconsistency can weaken the case law and, consequently, the legitimacy of the Court itself.Footnote 90
The lack of express obligation to conduct a consensus or trend analysis and the risk of cherry–picking naturally leads to the question of a middle ground. A quasi-duty to conduct such an analysis may exist when certain criteria are met. The difficulty lies in defining these criteria, which may be apparent from judgments in which the Court feels the need to conduct a comparative study. Cases with which Part E below deals are particularly prone to a potentially persuasive consensus or trend analysis, at least when judgments are rendered by the Grand Chamber. Should the latter consider significantly evolving its case law but not conduct such an analysis in those cases, then it would show some inconsistency. Considering its own legitimacy, the Court will thus likely tend to conduct such an analysis.
II. Relevant Material
In order to establish either a consensus or a trend, the Court especially looks at the Member States’ laws and practicesFootnote 91 and at international material,Footnote 92 as shown in the third and fourth columns of the chart in the Appendix.
1. Laws and Practices of the Council of Europe’s Member States
The Member States’ laws and practices, examined in their relevant national or local context,Footnote 93 constitute material that the Court is interested in when it conducts a consensus or trend analysis, as shown in the fourth column of the chart in the Appendix. This approach raises many questions. Three fundamental ones will be addressed here.
First, consensus is not just about the vast majority of the Member States. The Court is also willing to consider trends that point in a certain direction, as seen in the fifth column of the chart in the Appendix. This seems admissible provided that the trend is robust and established in a transparent way. Generally speaking, the Court should also state why it does not take into account some Member States or why it has not been able to assess the content of other Member States’ laws and practices.Footnote 94 Full transparency in this regard is especially needed when the impact of the consensus or trend on the shaping of the right at stake is strong or rather strong. More flexibility can probably be acceptable when the impact is less important like in the cases where the impact of the consensus or trend analysis has been qualified as weak at the end of Section I of Part C. In light of the above, the idea that “in order for comparative legal research to achieve its persuasive purpose, it should take into account the law and practice of all or nearly all Contracting Parties”Footnote 95 is relevant but may have to be nuanced.
Second, a certain stability in each Member State considered is required to define its position on a given issue. Ever-changing laws and practices usually neither indicate a trend nor allow classifying the state in one category or another. The Court should also include this aspect in its analysis.Footnote 96 In the same vein, the Court may expect some settled and quite long-standing principles, rules, or practices in a very dynamic field before significantly developing its case law further.Footnote 97 Some caution may also be justified when a matter is or has been a subject of debate in a number of European statesFootnote 98 or when “fluctuating developments” in a very sensitive matter have recently occurred in the Member States’ legal systems.Footnote 99 A similar approach can make sense in a dynamic area where new developments are possible in the future.Footnote 100
Third, are any laws and practices relevant or should they rank quite high in the legal order of the Member States? In other words, does the consensus or trend also have to be about the fundamental nature of the aspect of the right in question?Footnote 101 The Grand Chamber does not appear to make such a requirement, as the comparative analysis conducted in the cases listed in Sections I and II of Part C covers various material such as constitutional and statutory provisions and case law, to name a few. Nevertheless, the eventual constitutional foundation of an issue in a state also constitutes a factor to be considered by the Court when it examines a possible European consensus or trend, especially regarding the stability and strengthFootnote 102 of one or the other.
2. International Material
The Court also refers to international treaties, customary law, case law, or published practices when it examines the existence or the lack of a consensus or trend on a given matter, as shown in the third column of the chart in the Appendix. It sometimes mentions an international consensus or trend on a given issue.Footnote 103 When an international treaty has been or is being ratified by many Member States, it may reflect some European consensus or a trend in this direction. Here also, any trend should be clearly and transparently established by the Court. In the twenty-seven judgments and the one advisory opinion where some consensus or trend contributed to shaping a right guaranteed by the Convention or its Protocol No. 1, international also meant European, as the international treaty or treaties relied upon by the Court had actually been ratified by the vast majority—if not all—of the Member States. Note that European consensus has been conceptualized as “an implicit consent which can act as a substitute to formal consent within the Strasbourg system.”Footnote 104 Caution is required here, since this consensus does not fulfil this function.Footnote 105
Documents emanating from the Council of Europe may also indicate the existence of a European consensus or trend. The same applies to EU law, practices or case law, as all EU members are Council of Europe’s Member States, but non-EU members should not be left out of the picture. As shown in the third column of the chart in the Appendix, the Court extensively refers to Council of Europe and European Union materials.
For their part, international treaties ratified by a few Member States should not suffice to build a European consensus. However, they may point towards a trend. Furthermore, the practices of international committees, such as the Human Rights Committee of the United Nations, may influence the Court, but should not in principle be considered as part of some European consensus. Yet, none of the judgments and opinion listed in the Appendix were based on international treaties, practices or case law that are not relevant for most—if not all—of the Council of Europe’s Member States.
In Verein KlimaSeniorinnen Schweiz and Others v. Switzerland, the Court based its judgment both on scientific consensus and a consensus resulting from international material.Footnote 106 The two are in fact interrelated with respect to climate change. As far as the second is concerned, the Court explicitly referred to “the international-law mechanisms to which the member States voluntarily acceded and the related requirements and commitments which they undertook to respect”Footnote 107 and to “the Contracting Parties’ accepted commitments to achieve carbon neutrality.”Footnote 108 In this context, it specifically mentioned the Paris AgreementFootnote 109 and the United Nations Framework Convention on Climate Change (“UNFCCC”).Footnote 110 It noted that all Council of Europe Member States are members of the UNFCCC systemFootnote 111 and—one may add—parties to the Paris Agreement.
Any diverging or, a fortiori, opposite consensus or trend between, on the one hand, Member States’ laws and practices and, on the other hand, international material must probably be assessed on a case-by-case basis. Depending on their respective strengths and magnitudes, one may prove preponderant or even overriding compared to the other. However, it should be observed that a conflict between the various sources of a consensus or trend analysis did not occur in any of the judgments and opinion listed in the Appendix or, more broadly, mentioned in Part C, except, to some extent, in the now-dated twin judgments Christine Goodwin v. United Kingdom and I. v. United Kingdom.Footnote 112
Finally, the reference to international material by the Court can also reflect the search for harmony between various sources of international law, including the Convention.Footnote 113 Harmonization, rather than consensus or trend, is the key word from this perspective. Granted, the difference between these notions is not clear-cut, and the Court should clarify the approach it takes when relying, in a judgment, on international material. Is it seeking to harmonize the Convention with the latter or is it trying to establish a consensus or trend? The two terms of this alternative are not dogmatically equivalent, and this double question should not be left unanswered.
III. Weight of Consensus or Trend
The weight of a consensus or trend varies from case to case. The Court has not yet developed a convincing methodology in this regard, nor has it set weighing factors. One reason for the difficulty the Court faces in this regard can be explained by the fact that this issue also amounts to one of degree—degree of consensus intensity, strength of trend, and degree of clarity of comparative analysis, among other things. Some characteristics of the impact of a possible consensus or trend on the outcome of a case nevertheless derive from the Court’s jurisprudence.
First, a European consensus or trend may influence the outcome of a case and support the conclusions reached by the Court, but it cannot be used as the sole basis for a decision.Footnote 114 None of the judgments listed in the Appendix or, more broadly, mentioned in Part C are solely based on a consensus or trend analysis. Thus, the structured step-by-step examination of the scope and restrictions of many rights guaranteed by the Convention and its Protocols is not supplanted by the consensus analysis alone. In sum, the latter fulfills at most a persuasive function to the Court’s argumentation.
Second, a tricky question is whether the existence of a European consensus or trend creates a rebuttable presumption in favor of the solution to which the analysis points. Kanstantsin Dzehtsiarou argues in this direction.Footnote 115 However, one should observe that no provision of the Convention or its Protocols foresees such a presumption. Legally and technically, there is no room for the latter. The Court remains free to decide a case without following a European consensus or trend. That said, when the Court considered that a comparative analysis was justified in a given case, but then disregards the European consensus or a “strong trend” shown by this analysis, it should give reasons for the decision in a transparent manner.Footnote 116 At this point, I agree in substance with Kanstantsin Dzehtsiarou.
Third, should the weight of a European consensus or trend be determined on a case-by-case basis, or should it be framed by certain criteria? An affirmative answer to each part of this double question seems justified. On the one hand, the Court should always assess the merit of the case at stake. On the other hand, this process should not prove arbitrary. From this perspective, at least two main criteria make sense.
The first criterion relates to the strength of the consensus or trend when it is deemed by the Court as relevant. The greater the number of Member States that have chosen, or are directly or indirectly associated with, a solution—through their laws and practices, through the international treaties they have ratified or through texts discussed and enacted within the Council of Europe, etc.—and the clearer and more stable that solution is or, when the Court observes a trend, the clearer and firmer the trend is, then the more weight the analysis is, ceteris paribus, likely to carry in a given case.Footnote 117 The same holds true when a solution clearly results from multiple international sources relevant to the Member States and thus to the Court. However, the latter remains free to judge the case otherwise, which then implies providing reasons for its decision. The analysis is both quantitative and qualitative. From a quantitative perspective, the size or number of inhabitants of a country is not considered by the Court, which provides no details in this regard in a given judgment, as shown in the fourth column of the chart in the Appendix. To some extent, all the Member States are treated equally in this context, except for very specific situations. This is debatable, but no less problematic solution comes to mind. However, in some judgments listed in Sections I and I of Part C, the Grand Chamber put some emphasis on the case law of a few—large—countries.Footnote 118 Therefore, a kind of asymmetry—not to say inequality—can still be observed. From a qualitative perspective, the careful weighing of a solution by the Member States and the thoroughness of the Court’s comparative analysis, for instance, should play a role. Thus, a consensus or trend analysis cannot be diminished to merely counting states.Footnote 119
The second criterion deals with the nature of the case and the issues it raises. Certain cases may be more prone than others to a potentially persuasive consensus or trend analysis. Still, this question remains unsolved, by the Court itself and also by legal scholars who have dealt with this type of analysis.Footnote 120 In the judgments listed in the Appendix or, more broadly, mentioned in Part C, the Grand Chamber did not develop criteria in this regard. Due to this issue’s importance, it will be addressed separately from a doctrinal and empirical perspective.
E. Cases More Prone to a Potentially Persuasive Consensus or Trend Analysis
Most of the judgments of the Court that shape rights through a European and international consensus or trend involve indeterminate legal terms.Footnote 121 The ones more prone to a potentially persuasive consensus or trend analysis would typically deal with matters of political or general policy,Footnote 122 sensitive moral or ethical issues,Footnote 123 or changes in the case law,Footnote 124 as reflected in judgments rendered in Grand Chamber and Chamber formations. The emphasis is on the former, but some of the latter are also mentioned in Part E.
I. Interpretation of Indeterminate Legal Terms
Most judgments of the Court that shape rights through a European consensus or trend involve indeterminate legal terms such as necessity, fairness,Footnote 125 law, private and family life, property, or free elections, as evidenced by the cases listed in the chart in the Appendix. These terms and the balancing they may imply create an uncertain environment. The case law reduces this uncertainty, but the Court constantly faces new issues and, more seldomly, may wish to revisit its case law. In this context, the search for a European consensus or trend may help structure the analysis of the Court, make it more predictable or even transparentFootnote 126 and acceptable from a multi–level perspective.Footnote 127
The existence of an indeterminate legal term to be interpreted and applied in a concrete situation does not suffice to justify a consensus or trend analysis. This is a first door that potentially leads to the latter. Before this, the next door usually relates to the states’ margin of appreciation which is quite often—but by no means always—inversely correlated with a European consensus on the issue in question.Footnote 128 In other words, when the Court considers reducing a supposedly wide margin of appreciation, it may wish and, actually, feel the need to conduct a consensus or trend analysis. However, some margin of appreciation may, for instance, depend on the specific local or national contextFootnote 129 and, thus, makes a consensus or trend analysis, if not fully superfluous, at least inconclusive. Furthermore, the protection of minority rights may still lead to an intervention of the Court despite the existence of various approaches to the matter at issue among the Member States.Footnote 130 The case law in this respect lacks, regrettably, sufficient coherence, and a test should be developed to identify the appropriate relationship between a European consensus or trend and minority protection.Footnote 131
Another favorable setting to a potentially persuasive consensus or trend analysis exists when the Court envisages changing its case law. True, overturning jurisprudence, especially when it is well established, is based on the merits of the case, but the Court may search for a European consensus or trend for such an important step that conflicts, to a certain extent, with the stability and predictability of the case law. From this perspective, overturning jurisprudence may be positively correlated with a European consensus or trend.
For all that, these two hypotheses—an important margin of appreciation that the Court intends to reduce or, a fortiori, suppress, on the one hand, and overturning well-established case law, on the other hand—do not exhaust the reflection. They may be the main doors to a potentially persuasive consensus or trend analysis, but they are not the only ones. Moreover, in certain situations, a relatively narrow, but possibly evolutive, concept needs to be interpreted in the absence of any governing case law. In such a situation, the Court may wish to conduct a comparative analysis as well.
II. Cases on Matters of Political or General Policy
Cases on matters of political or general policy imply balancing several interests and making institutional or policy choices. They are usually decided by recognizing a wide margin of appreciation left to the states, while, at the same time, they may raise issues whose examination can shape rights. Matters of political and general policy are indeed decided by the Member States. From this perspective, it makes sense for the Court to examine how the Member States address an issue. This comparative analysis which is focused on the laws and practices of the states—international treaties or documents possibly fulfilling a supporting function—can also help frame and structure the application of the subsidiarity principle.Footnote 132
Based on a—brief—consensus analysis, the Court has, for instance, established a European consensus on the need for strict regulation of tobacco advertising.Footnote 133 In the judgments listed in the Appendix, a consensus or trend analysis was particularly relevant and justified in cases such as Bayatyan v. Armenia Footnote 134 or Yumak and Sadak v. Turkey. The definition of the obligation to perform military service and the sanctions for refusal raise fundamental policy issues for a state. The same applies to electoral thresholds for an election. In these two judgments, the analysis conducted by the Court revealed a consensus or common practices whose impact on the evolution of the case law was strong. The Court obviously remains free to deviate from this or to consider the particular context prevailing in a state, as it did in the second judgment.
By contrast, public health and healthcare policy matters,Footnote 135 penal policy,Footnote 136 models of criminal adjudication,Footnote 137 the right to vote of citizens living abroad,Footnote 138 political advertising,Footnote 139 parliamentary proceedings,Footnote 140 planning measuresFootnote 141 or—without being exhaustive by any means—relations between States and churches, other religious communities,Footnote 142 or even religious symbols in official settingsFootnote 143 are matters where the Court has conducted an analysis of comparative law and practice or of international material and, in the absence of a consensus or trend, or for other reasons, has renounced to further shape the rights at stake. Complex social security complaints belong to the same category of cases.Footnote 144 Accordingly, a comparative analysis can also provide useful insights when the Court envisages jurisprudential developments in this field based on a dynamic and extensive interpretation of the rights guaranteed by the Convention and its Protocols. Unfortunately, the Court has not been consistent in this regard.Footnote 145
A consensus and—more precisely in this context—comparative analysis is not necessarily required in all cases on matters of general policy. In cases infringing, for instance, on human dignity, a fortiori of vulnerable persons,Footnote 146 or the prohibition of discrimination,Footnote 147 such an analysis cannot, in principle, justify any tolerance by the Court.Footnote 148 However, some flexibility may remain with respect to distinctions based, for instance, on age. Furthermore, the specific context existing in one country may be considered by the CourtFootnote 149 and, depending on the circumstances, justify a breach with an established consensus or trend.Footnote 150
Finally, the need to perform a consensus or trend analysis and the value thereof are not obvious in cases regarding specific procedural guarantees in concrete situations. For instance, the Court looked at comparative law material in a case where it found violation of Article 1 of Protocol No. 7 in view of significant limitations imposed on the right to be informed of reasons for expulsion.Footnote 151 In this case, “the applicants received only very general information about the legal characterization of the accusations against them”; “[n]or were they provided with any information about the key stages in the proceedings or about the possibility of accessing classified documents in the file through a lawyer.”Footnote 152 The value of a comparative analysis is hard to see in these circumstances. The same prevails for a judgment that may regard a matter of general policy, such as combating large-scale tax fraud, but is closely related to the circumstances of the particular case.Footnote 153 In its—short—advisory opinion on impeachment legislation,Footnote 154 the Court did not proceed to any consensus or trend analysis, except that it briefly quoted some international material.Footnote 155 This seems admissible in view of the rather banal conclusions it reached. If the court had been interventionist on that issue, it would have been well advised to conduct such an analysis.Footnote 156
III. Cases Raising Sensitive Moral or Ethical Issues
Cases may raise sensitive moral or ethical issues, for which Member States traditionally enjoy a wide margin of appreciation, notably because of the national or even local context’s relevance.Footnote 157 On some issues, a certain consensus may appear among most of them. This may signal to the Court the need to reduce the states’ margin of appreciation and to tighten the interpretation of a right or the balancing of rights or interests.Footnote 158
To some extent, the legal recognition of same–sex couples, which has been regarded as a sensitive issue in numerous jurisdictions, illustrates this approach. The debate may be somehow closed in many European countries which have made marriage available to these couples, but it is ongoing in several others. Historically, the Court has felt the need to look for evolutions at the Member States’ level, most recently in the Fedotova and Others v. Russia case mentioned in Section II of Part C.Footnote 159 This does not mean that it should wait for a consensus,Footnote 160 nor that any trend should necessarily be widespread, but the Court could hardly have acted as first mover on this issue. When a consensus or trend is established, then the states’ margin of appreciation may be reduced and the Court may shape the right at stake, for instance, in judging disproportionate, under Article 8 § 2 of the ECHR, an interdiction of home births.Footnote 161
By contrast, the Member States keep enjoying a wide margin of appreciation, for instance, with respect to the scientific and legal definition of the beginning of life,Footnote 162 assisted suicide,Footnote 163 physician-assisted dying,Footnote 164 the withdrawal of artificial life-sustaining treatment,Footnote 165 the prohibition of surrogacy arrangements,Footnote 166 or the ban of human embryos to scientific research.Footnote 167 In the area of abortion, the Court noted a European consensus in the A, B and C v. Ireland case mentioned in Section III of Part C, but then retreated behind the margin of appreciation of the statesFootnote 168 in a shallow analysis.Footnote 169 The Court could have gone further and reduced the margin of interpretation in this regard, especially when the travel to a neighboring, more liberal state is costly, entails a number of practical difficulties, or is psychologically and physically arduous.Footnote 170
The lack of consensus among the Member States does not mean, however, that they can freely address all sensitive moral or ethical issues. The Convention and its Protocols set limits in this respect. Here, again, infringements on human dignity,Footnote 171 the prohibition of discrimination,Footnote 172 notably subject to the above qualification relating to issues such as the legal recognition of same–sex couples, or the protection of a particularly important facet of an individual’s existence or identityFootnote 173 can justify a strict case law of the Court even though a European consensus or trend is lacking on the matter at hand or, as the case may be, a European consensus or trend goes in the opposite direction. In certain cases, a consensus or trend analysis may even prove superfluous. For instance, the Court has noted that “there is a clear European consensus about the recognition of individuals’ right to openly identify themselves as gay, lesbian or any other sexual minority, and to promote their own rights and freedoms.” A consensus or trend analysis was not necessary on this point.Footnote 174 In the same vein, the Court rightly considered that the right to respect for private lifeFootnote 175 of a child born abroad through a gestational surrogacy arrangement and conceived using the gametes of the intended father and a third–party donor requires that domestic law provide a possibility of recognition of a legal parent-child relationship with the intended mother, designated in the birth certificate legally established abroad as the “legal mother,”Footnote 176 despite the lack of consensus in Europe on this issue.Footnote 177
IV. Cases Where the Court Considers Changing its Jurisprudence
Stability and continuity in the case law make it predictable for individuals, non-governmental organizations, legal entities, and states. They can also contribute to the Court’s legitimacy. Evolution in society and modification of circumstances may provide the basis for a change in jurisprudence. How can the Court establish such evolution and modification? A consensus or trend analysis may prove relevant and useful in this context.Footnote 178 However, it should be done properly, and, in any event, the Court should keep in mind that a state’s laws and practices are proxies in this respect, as society may be much more diverse. As reflected in the chart in the Appendix, the Court made a quite extensive analysis in its judgment Bayatyan v. Armenia. It considered all Member States as well as various international and regional texts.Footnote 179 This approach has proven to be compelling. In other judgments listed in the Appendix where the Court changed, “reconsidered” or “clarified” its case law, the analysis was much more cursory and, to some extent, lacked sufficient transparency or predictability.Footnote 180 There is—or, at least, was—undoubtedly substantial room for improvement in this respect.
Such an analysis will be especially—but not exclusively—appropriate when the lack of consensus has been established in a previous judgment and was part of ratio decidendi. Even if no consensus analysis was conducted in an initial judgment, this does not close the door for one in later judgments. On the contrary, such an analysis can enrich the argumentative and decision-making process.
In any event, the outcome of a consensus or trend analysis can be persuasive, but the Court retains the ability to judge otherwise. Put differently, the lack of a European consensus on a given issue does not prevent the Court from changing its case law, especially if it identifies some trend pointing in this direction. It will thus have to provide adequate justification. Likewise, the existence of a European consensus or, a fortiori, trend does constrain the Court to follow it, as it can favor other, more decisive arguments and circumstances.
F. Conclusion
A European consensus or trend can shape rights guaranteed by the Convention and its Protocols. It usually fulfils a supporting function to the interpretation of rights and the balancing they may imply. Depending on the circumstances and the issues at stake, it can prove persuasive in the sense that it significantly, but not exclusively, serves to define a right, fix its scope or extent, and sets its limits.
In our count, in the last twenty–five years, 27 out of 424 judgments on the merits by the Grand Chamber established a consensus or trend having a “shaping impact” on the protections conferred by rights guaranteed by the Convention or its Protocols. Only one advisory opinion based on Protocol No. 16 contained a section on comparative law material having such an impact. No decision of the Grand Chamber on admissibility during the last twenty-five years offered significant developments from this perspective. This study shows that strong or rather strong rights’ shaping through consensus or trend only occurred seven times—six judgments and two twin judgments. This immediately puts into perspective, and even somewhat demystifies, the real scope of the consensus or trend analysis on the evolution of the Court’s jurisprudence with respect to actual rights shaping. Of course, the total number of relevant judgments or their classification may be discussed, but different assessments in one case or the other will not change the overall picture.
Cases more prone to a potentially persuasive consensus or trend analysis are usually linked with the Member States’ wide margin of appreciation. Consequently, the states’ margin of appreciation is quite often—but by no means always—inversely correlated with European consensus on the issue at stake. When consensus is deemed relevant by the Court, the greater the number of the Member States that have chosen, or are directly or indirectly associated with, a solution, and the clearer and more stable that solution is or, when the Court observes a trend, the clearer and firmer the trend is, then the more weight the European consensus or trend is, ceteris paribus, likely to carry in a given case.
In any case, the risk of incomplete or low-quality and, therefore, truncated analysis should not be underestimated. Thus, the quality, transparency, and completeness of the consensus or trend analysis—covering, in other words, all Member States or close to it when it is comparative, except for states in which the issue at stake is not relevant or for which reliable information is lacking—are parameters on which the Court should not compromise, at least when the consensus or trend is going to have a strong or medium impact on the ratio decidendi of the case. The approach followed and the analysis conducted by the Court in Bayatyan v. Armenia is compelling in this regard. More flexibility may be admissible in other cases, especially when the Court barely refers to comparative material in its assessment.
Finally, the risk of inconsistency also deserves special attention from the Court. The latter can follow a systematic approach and, for instance, decide that a consensus or trend analysis should be conducted in all cases allocated to the Grand Chamber, but only in these. Alternatively, the Court may prefer a more flexible solution by defining criteria for conducting such an analysis when it decides cases in Grand Chamber or Chamber formations. A combination of the two approaches is also perfectly feasible. Accordingly, a consensus or trend analysis would, in principle, be systematic for cases allocated to the Grand Chamber if and when relevant for the matters at hand, and possible but subject to the fulfilment of criteria for cases decided in Chambers of seven judges. The criteria highlighted in Part E of this article could guide the Court in its choice to carry out, or dispense with, a consensus or trend analysis, regardless of whether the judgment is delivered by the Grand Chamber or a Chamber. This article supports the idea that the Court cannot simply cherry–pick when to conduct such an analysis. Otherwise, the case law would become extremely unpredictable, including from a methodological standpoint. Moreover, an unprincipled practice on the part of the Court may create inequalities. Any analytical inconsistency can thereby weaken the case law and, consequently, the legitimacy of the Court itself. The status quo being unsatisfactory, the Court needs to clarify its practice in this area. On this issue, there must be a European consensus within the Court.
Acknowledgements
The author deeply thanks Dr. Daniel Rietiker, Senior Lawyer at the European Court of Human Rights and lecturer at the University of Lausanne, for his very valuable comments as well as Dr. Julia Kamhi and Stéphanie Tumini, former teaching and research assistants at the University of Lausanne, for their excellent research assistance.
Funding statement
The author declares none.
Competing interests
There is no specific funding associated with this article.
Appendix
Open access
