The Freedom of Religion or Belief in the echr since Kokkinakis. Or “Quoting Kokkinakis

In: Religion & Human Rights
Author: Malcolm Evans1
View More View Less
  • 1 Professor of Public International Law, University of Bristol, Bristol, UK


The Kokkinakis case is probably the most widely cited judgement of the European Court of Human Rights concerning the freedom of religion or belief. Yet the constant repetition of its key passages tends to obscure the very real, and controversial, developments which have taken place in the Court’s jurisprudence. This article outlines this practice and explores its implications. It concludes that “quoting Kokkinakis” runs the risk of have a negative impact upon our understanding of Article 9 of the echr today and its famous ‘mantra’ should be replaced by a more up to date and accurate statement of the conceptual underpinnings of that article.


The Kokkinakis case is probably the most widely cited judgement of the European Court of Human Rights concerning the freedom of religion or belief. Yet the constant repetition of its key passages tends to obscure the very real, and controversial, developments which have taken place in the Court’s jurisprudence. This article outlines this practice and explores its implications. It concludes that “quoting Kokkinakis” runs the risk of have a negative impact upon our understanding of Article 9 of the echr today and its famous ‘mantra’ should be replaced by a more up to date and accurate statement of the conceptual underpinnings of that article.

1 Introduction

There are times when one wonders whether the European Court of Human Rights secretly subscribes to some doctrine of infallibility. There are relatively few examples of its simply accepting that it has got something wrong—although this does happen from time to time.1 When it does want to move on from a position previously held, it usually does so by pointing to changes in thinking or of approach which have occurred within the countries of the Council of Europe as a justification for its doing so.2 This is understandable; but even then, it tends to perpetuate an aura of infallibility by continually referring to what it first said on a question as providing the basis for its reasoning and approach; as being a conceptual template against which all else is to be explained and developed, for ever and ever (Amen). Thus, in several critical areas of its jurisprudence, it is as if time stands still, that that its first thoughts were its best thoughts, and that the current and the future must be approached and understood through that lens.

This tendency is, of course, facilitated by the relatively vacuous nature of many of these statements of grounding principle. In the context of Article 9 the oft-quoted passage from the Kokkinakis case3 setting out the essence of the Courts’ approach says that:

freedom of thought, consciences and religion is one of the foundations of a “democratic society” within the meaning of the Convention. It is, in its religious dimension, one of the most vital elements that go to make up the identity of believers and their conception of life, but it is also a precious asset for atheists, agnostics, sceptics and the unconcerned. The pluralism indissociable from a democratic society, which has been dearly won over the centuries, depends on it.

Who can disagree with such noble sentiments, stressing as they do the significance of thought, conscience and religion to those holding such beliefs and to democratic society and democratic pluralism more generally? Well, some would. Religious pluralism is increasingly seen in some quarters as being as much a threat to democratic society as it is a bulwark of it—and it is by no means obvious that democracy is necessarily a friend of religious pluralism, as recent events in some parts of Europe have made all too obvious. But even for those who endorse such noble sentiments, is there any problem with referring to Kokkinakis in this mantra-like fashion? If it does not mean very much, then surely it cannot really do any harm?

Unfortunately, it can—and it does. There are at least three reasons for this. The first is that focussing on Kokkinakis as the foundational case on the freedom of religion or belief draws a veil over both the actual origins of the jurisprudence on which the current approach to Article 9 of the echr is based, as well as over the extent to which this has changed over time. Secondly, it tends to magnify the importance of some legal claims by endowing them with some fundamental significance which, quite frankly, they do not necessarily have. Thirdly, and somewhat in contradistinction to the previous point, it tends to perpetuate the idea that the freedom of religion of belief is less of an individual right and more of a general societal question, because there is so much at stake. By masking its basic assumptions and its subsequent development, allowing the trivial to be elevated and the significant to be marginalised, ‘quoting Kokkinakis’ does little to assist the enjoyment of the freedom of religion or belief in Europe today. It has become a veil behind which almost anything can be justified, whilst seemingly maintaining a façade of continuity and consistency of approach. When it is cited, it tends have the same effect as a ‘refresh’ button—allowing thinking to start from scratch and permitting Article 9 to be develop in quite unexpected—and often quite contradictory—ways. In what follows the first of these three consequences will be explored in more detail.4

2 The Legal Dimensions of Article 9

As Carolyn Evans points out,5 the basic building blocks of Article 9 are not really explored in Kokkinakis. The Court inherited its approach from the work of the Commission in a series of decisions—many of them inadmissibility decisions—given over a twenty-year period prior to its judgment.6 Conceptually speaking, the ‘freedom of religion or belief’ before Kokkinakis is in some ways very much more interesting than afterwards. It was in these decisions that the fundamentals were determined: that there was to be a bright-line distinction between the ‘forum internum’ and the ‘forum externum’;7 that the concept of religion or belief was to be broad and inclusive;8 that the manifestation of ‘religion or belief’ meant that the manifestation of any pattern of thought or conscience, religious or otherwise (provided that it was sufficiently cogent and serious) was to be protected;9 that manifestations did not include all actions motivated by one’s religion or belief;10 that the protections of the rights and freedoms of others was to be understood in a broad fashion, not linked to a particular right or freedom or another identifiable individual, but to the general interests of society as a whole.11 In Kokkinakis the Court set out or applied these approaches but did not really reflect on them.

This is not to criticise, but to observe. However, by accepting the thrust of the Commission’s work over the years, the Court endorsed an approach which allowed a very broad range of claims to be made in terms of Article 9, but which at the same time allowed for a very restrictive approach to their enjoyment in practice. This is perhaps the logic of a jurisprudence largely grounded on inadmissibility decisions: the focus tends to be on why there is no violation, rather than on the contours of the right. It ought to be remembered that even in Kokkinakis, the violation of Article 9 was more procedural than substantive—a failure to give adequate reasons for the conviction for proselytising, rather than on the act itself.

These have all become “received wisdoms” which have become embedded in the work of the Court, and reflected in much of its output. But the Court has not stood still. Indeed, most of the fundamental issues highlighted above have been subject to re-appraisal or re-interpretation. It has also developed new strands of thinking concerning Article 9. The problem is that, due to the constant repetition of the Kokkinakis mantra, these developments tend either to get overlooked, or not incorporated into a consistent and authoritative line of reasoning. As a result, they can be “picked up” or “put down” more or less at will. The following sections will illustrate this by looking briefly at how some of these key questions have been engaged with and how other, new, approaches have been developed against the background of Kokkinakis.

2.1 The Forum Internum and the Forum Externum

Few elements of the Article 9 framework seem more settled than the distinction between the right to hold a pattern of thought, conscience or religion—the forum internum—which is subject to absolute protection, and the freedom to manifest that pattern of religion or belief—the forum externum—which is subject to limitation. Recent research in fact tends to show that the origins of this distinction are less clear and less substantial than have been assumed.12 It is certainly the case that the argument that the right to ‘have’ a religion or belief has no practical consequences in the ‘external realm’ has been called into question by decisions such as Alexandridis v. Greece,13 and Sinan Isak v. Turkey,14 in which the State was prevented from requiring a person to reveal their religious affiliation. Refusing to reveal ones religion is not aptly described as a manifestation of religion or belief: I am not aware of any religion which requires its followers to refuse to reveal their religious identity to others, but there are, of course, many situations in which religious believers might not wish to do so; some of which are more pressing than others. The only way such Court decisions can be explained convincingly is by seeing them as extending the scope of the forum internum; by accepting that the mere fact of holding a religion or belief must mean that there a range of necessary ancillary protections which flow from this.

These, then, are some examples where the Court has concluded that the state is “putting upon” a person in an improper fashion by requiring them to reveal their religious identity for some bureaucratic purpose. However, there are others, such as Wasmuth v. Germany, in which the Court thought that “putting upon” a person in a similar fashion was, in fact, acceptable.15

Does this mean that some violations of the forum internum can, in fact, be justifiable? What does seem clear is that the absolute protection of the forum internum appears to have some discretionary edges. And does this mean that there is more permeability between the internal and external spheres than was previously thought? The implications of blurring the boundaries between the two spheres are profound, but the Court has not yet explored this as a matter of principle16 and it seems unlikely that it will do so. In the meanwhile, the confusion concerning this core element of Article 9 continues.

2.2 What Beliefs Are Protected?

Another question on which some development has taken place since the Kokkinakis case and which deserves more systemic reflection from the Court than it has received concerns the beliefs which are protected by Article 9. As has been seen, Kokkinakis reflects the broad and inclusive approach that was previously found in the Commission’s case law, subject to the caveat that in order to be protected a belief must ‘attain a certain level of cogency, seriousness, cohesion and importance’.17 The Grand Chamber has recently elevated the significance of this caveat (and sent a signal that it might be taking a somewhat more robust approach) when, in S.A.S. v. France it stressed that ‘only’ such views were protected under Article 9.18 However, since denying a belief the protection of Article 9 is tantamount to saying that it is not ‘cogent, serious, cohesive or important’, it is understandable that the Court is reluctant to get drawn too far on these questions. Nevertheless, there is also a danger that by not doing so it encourages frivolous claims, or extends a mantle of legitimacy over views which are potentially problematic.

Moreover, once it has passed that threshold, it is not for the state to have a view on the legitimacy of those views: it now routinely says that ‘the state’s duty of neutrality and impartiality is incompatible with any power…. to assess the legitimacy of religious beliefs or the way in which those beliefs are expressed’.19 Or so it is said.

In fact, some views are excluded from protection by virtue of Article 17 of the Convention itself,20 although the application of this Article by the Court has not been very consistent.21 The Court, in Pretty v. the United Kingdom, has also accepted that ‘not all opinions or convictions are beliefs in the sense protected by Article 9(1) of the Convention’.22 The truth is that, despite what the Court says, its practice is less than even-handed. It may, for example, just refuse to engage with the question at all. In the Mouvement Raelian Suisse v. Switzerland case, for example, both the Chamber and the Grand Chamber found it unnecessary to determine whether the applicant movement’s claim fell under within the scope of Article 9 since it had already dismissed the substance of the claim under Article 10.23 Importantly, the Swiss Courts at all levels had considered the freedom of religion or belief to be engaged. This is significant since, in a number of cases concerning the Church of Scientology, the Court has relied on the domestic courts having considered Scientology to be a religion or belief as being a sufficient reason for it to treat the application as falling within the scope of Article 9.24 There is no consistency here.

It is not obvious why the Court takes different approaches in these cases, but it might have less difficulty if it were to quote its own jurisprudence correctly. In Eweida v. the United Kingdom the Court claims to be drawing in its seminal decision in Hasan and Chaush v. Bulgaria when explaining that it has no scope to question the legitimacy of beliefs which fall within the scope of Article 9. In fact, the paragraph it quotes says no such thing. It actually says that ‘but for very exceptional cases, the right to freedom of religion as guaranteed under the Convention excludes any discretion on the part of the State to determine whether religious beliefs or the means used to express such beliefs are legitimate.’25 In other words—the Court is entitled to take a view after all. It seems that the Court just does not want to admit it.

This helps explain the contentious decision in Leela Forderkreis ev v. Germany. In this case the Court accepted that it was acceptable for Germany to run an information campaign which highlighted the dangers to young people of becoming associated with the Osho Movement, since this was ‘consistent with the Contracting Parties’ positive obligations under Article 1 of the Convention to secure the rights and freedoms of persons within their jurisdiction’.26 Presumably, such information was also available to those who already were members of the movement—and so seems to run perilously close to seeking to persuade such members to change their religion or belief. There is no doubt that the information campaign was intended to discourage people from joining it.27 Whilst such information might be considered merely ‘informational’, not ‘coercive’—and it is of course well established that a state may not seek to coerce a person into changing their religion or belief28 —, the idea that the state cannot have views concerning various forms of belief, and act on them, is, then, simply not true. And if it is not true for the State, how is it possible for the Court to get away with hiding behind a façade of inclusivity and non-evaluation? But this is what the Kokkinakis mantra tends to encourage and facilitate.

2.3 The Nature of Manifestations

A third core ‘architectural’ element of the Article 9 framework concerns manifestations, which according to the text take the form of ‘worship, teaching, practice and observance’. This is again an area in which there has been considerable development since the Kokkinakis case. The older case law stressed that not every act ‘motivated’ by religion or belief was necessarily a manifestation of that religion or belief.29 But this has been a source of endless difficultly, sometimes leading to consideration of whether a particular practice was or was not mandated by the belief in question, an issue which courts are generally ill-suited to determine. The Court has now adopted a very different approach, and in S.A.S. v. France the Court helpfully made it clear that it is not necessary for an applicant to demonstrate that their actions are mandated by their beliefs: the Court took the view that it was sufficient that wearing a full-face veil was ‘for certain Muslim women, a form of practical observance of their religion and can be seen as a “practice”.’30

This approach builds Eweida and Others v. the United Kingdom, which has set out a fundamentally different approach to determining whether something is to be considered a ‘manifestation’ from that found in earlier cases. In Eweida, the Court drew a distinction between those forms of manifestation—such as worship or devotion—‘which form part of the practice of religion or belief in a generally recognised form’ and other forms of manifestation where, on the facts of the case, there exists ‘a sufficiently close and direct nexus between the act and the underlying belief’.31

Whether this is much of an advance is open to doubt. It certainly is confusing to say that one form of manifestation identified by the Convention—‘worship’—is in fact an example of another—‘practice’. But the gist of the distinction is that there are some things which will always be manifestations, because they reflect an established religious practice, whilst there are other acts, which do not reflect such established practices, but which may nevertheless be considered to be a manifestation on the facts of the case. What this seems to amount to saying is that if an act is not usually considered to be a manifestation of religion or belief, then it may still be treated as such if it is appropriate to do so. This is not an insubstantial or unhelpful change in approach.32 What is clear is that this approach has little to do with the Kokkinakis statement and the jurisprudential thinking on which it was based.

2.4 Positive and Negative Obligations and the Conundrum of Neutrality

The previous sections have highlighted some key developments relating to the basic building blocks of Article 9 and shown that the focus on the ‘Kokkinakis’ mantra, means that the Court fails to draw attention to how its approach has changed over time. This is further supported by highlighting two further ways in which the Court has changed its approach, but which tend to be marginalised by the constant referral back to Kokkinakis.

The first concerns the idea of positive and negative obligations. This was first developed by the Court in the context of Article 8, and has now spread out to other areas of the Convention.33 It is fast becoming a commonplace to explore Article 9 on the basis of positive and negative obligations. This is almost entirely disingenuous. Article 9 was ‘late to the party’, and it is only comparatively recently that this distinction has started being used in this context.34 Moreover, the challenge which this poses for the traditional approach to Article 9 cannot be underestimated, since in many ways it turns it on its head.

The starting point for an approach based on positive obligations is not so much the right of the individual vis-à-vis the state, but the obligation on the state to secure the right of the individual from infringement by non-state actors, infringements for which the State would not otherwise be responsible.35 In other words, it requires the state to take action to protect the right in question from breaches by others. A classic example is the Gldani Congregation case,36 where the state was required to take action to prevent attacks on worshippers by followers of another religious figure.

It is revealing that some leading works37 now trace the origins of positive obligations in the context of Article 9 to cases such as Otto-Preminger-Institut v. Austria.38 This, in fact, was a case concerning Article 10, the freedom of expression. The issue concerned the decision of the Austrian authorities to prohibit the screening of a film considered offensive to religious believers. It was, then, a case in which the protection of the sensibilities of religious believers was being used as a justification for restricting the freedom of expression under Article 10. Whilst this has a lot to do with the idea that the exercise of the freedom of expression should be ‘respectful’, it has nothing to do positive obligations under Article 9. What it does have, however, is a powerful resonance with the idea which emerged in a series of cases in the years following Kokkinakis; that the role of the state was not to take sides in matters concerning religion or belief, but that it should ensure there was a ‘level playing field’, initially in the field of proselytism.39 This idea—that it was not the role of the state to ‘take sides’ in disputes between or within religious communities—quickly developed into another judicial accretion to the corpus of doctrine surrounding Article 9, the idea of neutrality. Neutrality is not mentioned in Kokkinakis, or in the cases which came before it.40

No matter how laudable in principle, neutrality is nigh on impossible in practice. And if a state has a positive obligation to ensure that everyone can enjoy their freedom of religion or belief, it is difficult to see how that can be done without it having to make some assessments from time to time of what it is prepared to do in support of, or in opposition to, the practice of some groups of believers. This has been obvious even as far back as the Otto-Preminger-Institut case: it was only because the Austrian authorities chose to intervene by prohibiting the screening of the film that an issue arose under Article 10. It is just about inconceivable that had it not done so, Catholic believers in the region could have mounted a successful claim that their positive freedom of religion had been violated by the failure to do so. Simply saying that the state is to be the ‘independent and neutral organiser’ of religion or belief, or that it is to act ‘neutrally and impartially’, does not determine when it may be required to intervene in some way. And it goes without saying that such intervention is unlikely to be thought either impartial or neutral by those against whom it is directed.

The Court seems to have understood this when it said in the Eweida case that

[w]hilst the boundary between the State’s positive and negative obligations under the Convention does not lend itself to precise definition, the applicable principles are, nonetheless, similar. In both contexts regard must be had in particular to the fair balance that has to be struck between the competing interests of the individual and of the community as a whole, subject in any event to the margin of appreciation enjoyed by the State….41

In truth, it might as well have said that the entire distinction is little more than a heuristic device of dubious legitimacy and marginal utility.

If this were as far as it went, that might be bad enough. But there is more. The juxtaposition of an obligation of neutrality and a positive obligation to prevent breaches by non-state actors when it may be appropriate for the state to do so has at times morphed into a positive obligation to be ‘neutral and impartial’. Acting in a neutral and impartial way is one thing: being neutral and impartial is quite another. It is true that the Court has never said in categorical terms that the obligation of neutrality under Article 9 means that it is incompatible with the Convention for there to be a state church: indeed, it has accepted the legitimacy of state churches,42 provided that this does not have negative consequences for those of other religions or beliefs. Nevertheless, it is difficult to see how this is really reconcilable with much of what the Court has said concerning neutrality and impartiality.

Be that as it may, the language of neutrality has given rise to an even more contentious idea, that not only is the state itself is to be ‘neutral’ in its dealing with religions or beliefs, but that the State is itself to be ‘neutral’ in how it engages with its citizens. It is on this basis that contentious decisions have been taken concerning the wearing of religious clothing or symbols by public employees43 or in state-run facilities.44 It is on this basis that it has been claimed that religious symbols must not be present in public buildings.45 This is not the place to discuss the merits of these positions—but it is important to note that somehow, in 25 years of jurisprudence built upon the recognition of the vital importance of religious pluralism to democratic life, we have arrived at a position in which banning religious believers from merely wearing their religiously inspired clothing in public appears capable of being necessary in a democratic society.46

This is bolstered by the recognition of another right, not otherwise known from the text of convention, having come to be discerned. This is the ‘negative’ freedom. At one level, this is merely the inevitable consequence of having discovered the ‘positive’ obligation. Nevertheless, there has been some significant jurisprudential slippage. Returning to basics, it is possible to argue that even being forced to participate in ceremonies or religious practices of religions or belief systems other than one’s own does not necessarily mean one is not free to believe as one wishes.47 However, and as has been seen in connection with the discussion concerning the distinction between the forum internum and forum externum, it is increasingly recognised that the freedom of thought, conscience and religion offers protection from being forced to act in ways which are contrary to one’s beliefs, provided this does not adversely affect the rights of others.

This claim is closely related to that of conscientious objection on grounds of religion or belief, the precise contours of which are contentious and beyond the scope of this comment. But not being required to participate in an activity is not the same thing as being exposed to an activity—yet when combined with the obligation of neutrality and impartiality, it has become possible to argue that a person ought to be free in the public realm from exposure to the trappings of religion (and, presumably, of belief). The apotheosis of this argument was the case of Lautsi and Others v. Italy, where the Chamber judgment argued that, in a state school, ‘[n]egative freedom of religion is not restricted to the absence of religious services or religious education. It extends to practices and symbols expressing, in particular or in general, a belief, a religion or atheism’.48 The Grand Chamber, happily (if controversially), stepped back from this position in a judgment,49 which is notable for its lack of visibility in the reasoning of subsequent judgments by the Court on analogous issues. The problem is that it is just not possible to be ‘positive’ or ‘negative’ in a neutral fashion. The Court has never really engaged with this problem. But why do so—when it can ‘quote Kokkinakis’ and justify almost any and every outcome?

2.5 The Nature of the Right to Religion or Belief

A final point concerns the Court’s understanding of the very nature of religious belief, and can be shortly made. In Eweida and Others v. the United Kingdom it was said that

[r]eligious freedom is primarily a matter of individual thought and conscience. This aspect of the right … is absolute and unqualified. However … freedom of religion also encompasses the freedom to manifest one’s belief … in community with others … Since the manifestation by one person of his or her religious belief may have an impact on others, the drafters of the Convention qualified this aspect of freedom of religion in the manner set out in Article 9(2).’50

Six months later, in Sindicatul “Păstorul cel Bun” v. Romania51 the Grand Chamber said that

[t]he autonomous existence of religious communities is indispensable for pluralism in a democratic society and is an issue at the very heart of the protection which Article 9 affords. It directly concerns not only the organisation of these communities as such but also the effective enjoyment of the right to freedom of religion by all their active members. Were the organisational life of the community not protected by Article 9, all other aspects of the individual’s freedom of religion would become vulnerable.52

Whereas Eweida emphasises the individual nature of the right, the Sindicatul case highlights its collective nature. Both draw on the Kokkinakis statement; Eweida quoting it expressly,53 Sindicatul by using its language, although not directly acknowledging it. This tension lies at the heart of much of the controversy surrounding the practical application of Article 9: is it focussed on the individual, the religious or belief communities or the broader community as a whole? Is it about the individual versus the community? Or is it about the individual in community? And which community? Both cases raised precisely such questions. The problem with the Kokkinakis approach is that it can support outcomes based on any of these approaches—or none. But this is not a new difficulty: both of these statements are also to be found in the seminal case of Hasan and Chaush v. Bulgaria.54 The tension is longstanding and unresolved—and perhaps unresolvable: who (or what), precisely, is the freedom of religion or belief for? At the moment, it appears to be a ‘precious asset’ for everyone in general—but the thing it seems to protect in particular is the discretion of the state to do as it thinks fit. Which is a very strange thing for a human right to do.

3 Conclusion

In the 25 years since the Kokkinakis case the jurisprudence concerning Article 9 has developed prodigiously. This is only to be expected. During that time, some of the fundamental assumptions upon which the basic organisational architecture and fundamental underpinning concepts have also changed. Yet from the manner in which the Court routinely approaches Article 9—and underscores by its continual recall of its Kokkinakis mantra—one would be unware of just how significant those changes have been. Once again, the impression of serene continuity is conveyed, whilst a jurisprudential maelstrom has been taking place beneath the surface. This comment has only focussed on a very small number of issues. Moreover, the most contentious—the margin of appreciation has not been considered at all.55 Other pieces in this collection explore both these themes and many others in greater depth. But it is hoped that enough has been said to sustain the case that the time has come—after 25 years—for the Court to identify a new foundational basis for its approach to Article 9, which better reflects the manner in which its jurisprudence has evolved over the years. Of course, that might just be part of the reason why the Court continues to quote Kokkinakis: who can seriously object to what is said in the ‘Kokkinakis’ statement—and who is seriously comfortable with the Court’s current jurisprudence on Article 9?


In the context of the freedom of religion or belief a clear example is the decision by the European Commission on Human Rights in X. and Church of Scientology v. Sweden (dec.), Application No. 7805/77, Commission decision of 5 May 1979, dr 16, p. 68 at 70 in which it reversed its previous decisions that legal entities could not have rights under Article 9 and that the rights of churches were adequately protected by the rights given to their members.


Well-known examples include the suggestion, in Al-Saadoon and Mufdhi v. the United Kingdom, Application No 61498/02, 2 March 2010, para. 120, that Article 2 has been amended by state practice so that the death penalty amounts to a violation of the Convention, and—in the context of the subject matter of this contribution—that Article 9 of the echr sustains a claim to conscientious objection from military service in Bayatyan v. Armenia [gc], Application No. 23459/03, 7 July 2011.


Kokkinakis v. Greece, Application No. 14307/88, judgment of 25 May 1993, Series A, No, 260-A, para. 31.


The second and third consequences will be explored in a longer article which is to follow.


See C. Evans, ‘Pre-Kokkinakis Case-law of the echr: Foreshadowing the Future’, paper forthcoming (2018), in a book edited by J. Gunn, J. Temperman and M. Evans.


Inevitably, it is the older editions of leading textbooks and the earliest specialist works which give most attention to the contribution of the Commission to the formation of the approach to Article 9, simply as there was little else to focus on. Examples include, M. Evans, Religious Liberty and International Law in Europe (Cambridge: Cambridge University Press, 1997); and C. Evans, Freedom of Religion under the European Convention on Human Rights (Oxford: Oxford University Press, 2001). The early Commission decisions developed the conceptual foundations which have remained relatively unchallenged and still provide the basic frame of reference for an Article the meaning of which is eminently contestable. This is why these early analyses remain so widely quoted: they remain highly relevant.


This appears to have first been set out by the Commission in C. v. the United Kingdom (dec.), Application No. 10358/83, Commission decision of 15 December 1983, and is repeated in little more than half a dozen subsequent Commission decisions and about the same number of Court judgments. I am grateful to Caroline Roberts for this statistical information.


E.g. Arrowsmith v. the United Kingdom, Application No. 7050/75, Commission Report, 12 October 1978, dr 19 (pacifism). By way of relatively early examples (of which there are many) see, e.g. Knudsen v. Norway (dec.), Application No. 11045/84, Commission decision of 8 March 1985, dr 42, p. 258 (opposition to abortion); and W. v. the United Kingdom (dec.), Application No. 18187/91, Commission decision of 10 February 1993 (veganism).


Campbell and Cosans v. the United Kingdom, Application Nos. 7511/76; 7743/76, 25 February 1982, Series A no. 48 para. 36.


Arrowsmith v. the United Kingdom, Application No. 7050/75, Commission Report, 12 October 1978, dr 19.


Thus the commission rejected a number of applications from prisoners whose individual claims were rejected on the basis of the more generic need to maintain ‘prison order’. For a helpful summary see B. Rainey, W. Wicks and C. Ovey, Jacobs, White and Ovey, The European Convention on Human Rights, 6th ed. (Oxford: Oxford University Press, 2014), pp. 420–422.


P. Petkoff, ‘Forum Internum and Forum Externum in Canon Law and Public International Law with Particular Reference to the European Court of Human Rights’, 7 Religion and Human Rights (2012) 183, who explores the parallels between the uses of these terms by the Commission and within canon law. D.J. Harris, M. O’Boyle, E.P. Bates and M. Buckley, Harris, O’Boyle and Warbrick, Law of the European Convention on Human Rights, 3rd ed. (Oxford: Oxford University Press, 2014), at p. 594, also note that the distinction is not clearer defined, but observe that the case law tends towards giving the forum internum a limited meaning, and offering (as a result) little protection.


Alexandridis v. Greece, Application No. 19516/06, 21 February 2008.


Sinan Işık v. Turkey, Application No. 21924/05, 2 February 2010.


Wasmuth v. Germany, Application No. 12884/03, 17 February 2011. The recent case of Klein and Others v. Germany, Application Nos. 10138/11, 16687/11, 25359/11 and 28919/11, 6 April 2017, the Court also concluded that there had been no violation of Article 9 when a husband was required to pay church tax owed by his wife, as a result of their having opted to submit a joint tax return (and which he was entitled to seek to reclaim). The case presupposes the legitimacy of the applicant having to make known his religious, or non-religious, affiliation, but this does not appear to have been considered problematic.


There, is, however, a very interesting reflection on this relationship in the Separate Opinion of Judge Pinot de Albuquerque in Krupko and Others v. Russia, Application No. 26587/07, 26 June 2014, where he argues that: ‘The word religion derives from the Latin term ‘religare, which means bind, to bring together, to unite each woman and man with the deity (forum internum) and all women and men with one another (forum externum). The latter is no less important that the former from the believer’s point of view. For the believer, where the latter is hindered the former is per se encroached upon’.


Campbell and Cosans v. the United Kingdom, para. 36.


S.A.S. v. France [gc], Application No. 43835/11, July 2014, para. 55.


E.g. Eweida and Others v. the United Kingdom, Application Nos. 48420/10, 59842/10, 51671/10 and 36516/10, 15 January 2013, para. 81.


Article 17 provides that: ‘Nothing in this Convention may be interpreted as implying for any state, group or person any right to engage in any activity or perform any act aimed at the destruction of any of the rights and freedoms set forth herein or at their limitation to a greater extent than is provided for in the convention’. It should be noted that this refers to activities: this seems to suggest that it does not legitimate state action aimed at changing the views of those who may hold views which are antithetical to Convention values from holding them.


A conclusion reached by Jacobs, White and Ovey, supra note 11, at p. 124, following a helpful discussion, highlighting its controversial inadmissibility decision in Norwood v. the United Kingdom, Application No. 23131/03, 16 November 2004.


Pretty v. the United Kingdom, Application No. 2346/02, 29 April 2002, echr 2002-iii, para. 82.


Mouvement Raëlien Suisse v. Switzerland, Application No. 16354/06, 13 January 2011, para. 61; Mouvement Raëlien Suisse v. Switzerland [gc], Application No. 16354/06, 13 July 2012, para. 80.


Church of Scientology of Moscow v. Russia, Application No. 18147/02, 5 April 2007; Church of Scientology of St Petersburg and Others v. Russia, Application No. 47191/06, 2 October 2014.


Hasan and Chaush v. Bulgaria [gc], Application No. 30985/96, 26 October 2000, echr 2000-xi, para. 78 (emphasis added).


Leela Förderkreis E.V. and Others v. Germany, Application No. 58911/00, 6 November 2008, para. 99. Positive obligations are discussed further in section 2.4 below.


Ibid., para. 98, where the Court says that: ‘the German Government, by providing people in good time with explanations it considered useful at that time, was aiming to settle a burning public issue and attempting to warn citizens against phenomena it viewed as disturbing … The public authorities wished to enable people, if necessary, to take care of themselves and not to land themselves or others in difficulties solely on account of lack of knowledge.’


See, for example, Ivanova v. Bulgaria, Application No. 52435/99, 12 April 2007, para. 79: ‘a State cannot dictate what a person believes or take coercive steps to make him change his beliefs’.


This being derived from Arrowsmith v. the United Kingdom, Application No. 7050/75, Commission Report, 12 October 1978, dr 19.


S.A.S. v. France, para. 56.


Eweida and Others v. the United Kingdom, para. 82.


The difficulty will be knowing how to determine the basis on which such a determination will be made. It also opens up the possibility that some forms of manifestation—those associated with an established religious practice of a well-established religious tradition—may in reality receive a higher degree of protection than others. There appears to be a presumption that some forms of manifestation are to be protected, whereas in other cases a sufficiently close nexus needs to be demonstrated. This could prove problematic from a non-discrimination perspective.


Marckx v. Belgium, Application No. 6833/74, 13 June 1979, Series A no. 31. For a general account see A. Mowbrey, The Development of Positive Obligations under the European Convention on Human Rights by the European Court of Human Rights (Oxford: Hart Publishing, 2004).


Most of the earlier works on Article 9 do not identify this distinction in the work of the Court (let alone the Commission) at all.


This is acknowledged in Eweida and Others v. the United Kingdom, para. 84, where the Court accepted that since the acts in question had been carried out by private companies and so were no attributable to the state, it could only consider the matter on the basis of positive obligations.


Members of the Gldani Congregation of the Jehovah’s Witnesses and Others v. Georgia, Application No. 71156/01, 3 May 2007.


Jacobs, Ovey and White, supra note 11, p. 433.


Otto-Preminger-Institut v. Austria, Application No. 13470/87, 20 September 1994, Series A no. 295-A.


Larissis and Others v. Greece, Application No. 140/1996/759/958–960, 24 February 1998, Reports of Judgments and Decisions 1998-I. See also the contribution by B.G. Scharffs, ‘Kokkinakis and the Narratives of Proper and Improper Proselytizing’, 12:2–3 Religion and Human Rights (2017), pp. 99–111, this special issue.


The origins of neutrality lie in the cases submitted to the Court concerning the registration of religious communities in the years following the accession of former Soviet states to the Council of Europe system. It has since become a standard fixture in almost all cases concerning Article 9, although it has been articulated in many different ways.


Eweida and Others v. the United Kingdom, para. 84.


See, for example, Ásatrúarfélagið v. Iceland (dec.), Application No. 22897/08, 18 September 2012.


E.g. Dahlab v. Switzerland (dec.), Application No. 42393/98, 25 February 2001, echr 2001-v; and, recently, Ebrahimian v. France, Application No. 64846/11, 26 November 2015. See also D. McGoldrick, ‘Religious Symbols and State Regulation’, 12:2–3 Religion and Human Rights (2017), pp. 128–141, this special issue.


E.g. Köse and Others v. Turkey (dec.), Application No. 26625/02, 24 January 2006, echr 2006-ii; Aktas v. France (dec.), Application No. 43563/08, 30 June 2009.


Lautsi v. Italy, Application No. 30814/06, 3 November 2009, paras. 56–57.


Ahmet Arslan and Others v. Turkey, Application No. 41135/98, 23 February 2010; S.A.S. v. France [gc], Application No. 43835/11, July 2014.


Indeed, some old cases endorsed this position, at least as regard ceremonials. E.g., in Valsamis v. Greece, Application No. 21787/93, 18 December 1996, Reports of Judgments and Decisions 1996-vi, the Court upheld the right of a school to insist on the participation of a child in a national day parade which the child considered to be militaristic and thus not in accordance with his religiously based pacifist beliefs.


Lautsi and Others v. Italy, Application No. 30814/06, 3 November 2009, para. 55.


Lautsi and Others v. Italy [gc], Application No. 30814/06, 18 March 2011. Although the Lautsi case was determined on the basis of Article 2 of the First Protocol to the echr, this does not lessen its significance for the purposes of understanding the reach of Article 9.


Eweida and Others v. the United Kingdom, para. 80.


The origins of the Grand Chamber’s comments can be found in Hasan and Chaush v. Bulgaria [gc], Application No. 30985/96, 26 October 2000, echr 2000-xi, para. 62.


Sindicatul “Păstorul cel Bun” v. Romania [gc], Application No. 2330/09, 9 July 2013, para. 136.


Eweida and Others v. the United Kingdom, para. 79.


Hasan and Chaush v. Bulgaria [gc], paras. 60 and 62.


This is considered by S. Berry,’ Religious Freedom and the European Court of Human Rights’ Two Margins of Appreciation’, 12:2–3 Religion and Human Rights (2017), pp. 198–209, this special issue.