Skip to main content
European Commission logo
English English
CORDIS - EU research results
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary
Content archived on 2024-06-18

Religious Diversity and Secular Models in Europe – Innovative Approaches to Law and Policy

Final Report Summary - RELIGARE (Religious Diversity and Secular Models in Europe – Innovative Approaches to Law and Policy)


Executive Summary:

In an increasingly globalised world, the EU today to an unprecedented degree has come to be composed of people with different ethnic, religious and national backgrounds. Legislators and policymakers at various levels of the decision making process are left with the question: How to respond effectively and appropriately to increasing social, cultural, religious and philosophical diversity in a democratic context. The RELIGARE project assesses in particular the question how to strike a balance between the application of non-discrimination norms (and their further expansion) and the protection of the right to freedom of religion and belief. The search for the right balance requires that controversies be addressed on both empirical and normative grounds: (1) discrimination against non-believers, non-confessional groups and religious organisations other than the ones already regulated or admitted by law (thus including non-established churches); (2) the contours of the right to freedom of belief, both individual and collective, in both the forum internum (conscience) and the forum externum (expression, practices); (3) unequal distribution of State support;. The RELIGARE research covers 10 countries (Belgium, Bulgaria, Denmark, Germany, Great Britain, France, Italy, the Netherlands, Spain and Turkey, and focuses on four areas of participation in social life: (1) employment, (2) family life, (3) access to and the use of public space, and (4) State supported activities.

The project’s approach has been comparative and interdisciplinary, in combining legal analysis with sociological data and insights. It yielded three types of research instruments:

(1) a database of case law for the 10 countries involved in the research;
(2) a series of thematic templates that summarize the relevant legislation, court cases and controversies in the various countries (these reflect, in a synthetic fashion, the arguments made both by the legislative branch and by the courts and tribunals, in order to address a number of particular situations); and
(3) sociological reports, drawn up for the six countries where fieldwork was conducted (Bulgaria, Denmark, France, the Netherlands, Turkey and the UK).

Based on the main findings, the RELIGARE project advances a number of recommendations that are addressed both to the domestic authorities (Member States) and, in particular, to the EU Institutions. The recommendations call for a more direct and active role for the EU Institutions in developing a coherent policy framework that would strengthen the combat against discrimination on the grounds of religion or belief that is compatible with a democratic understanding of the functioning of pluralist democracies and can therefore help overcome divisions and segregations.

Project Context and Objectives:

1.1 European secular models and their interplay with Europe’s religious heterogeneity

The RELIGARE research project takes a thorough look at the complex interactions between two principles enshrined both in EU law and in the national constitutions of the Member States and aims at critically examining, for the 10 aforementioned countries, the extent to which these principles are in tension with one another: on the one hand, freedom of religion and belief and on the other non-discrimination. Freedom of religion and belief is to be understood in the generic sense: it refers not only to the protection, in law, of religion but also extends, in line with the generally accepted meaning, to non-religious individuals and groups.

With a view to ordering the relationships between State institutions and religious groups and communities of conviction, the majority of national legislators in Europe have made the – historic – decision to secularise State law. The hypothesis was that detaching the State and its institutions from religion would ultimately make society in democratic States more inclusive, providing a unitary legal and political framework within which religious and non-religious people can live together. In reality, however, the ‘secular model’ followed in various European States is both complex and varied. While most States officially commit themselves in one way or another to secularity, often they also maintain support for one or more religious organisations or communities of conviction. Furthermore, in a rapidly globalising world, one can observe in Europe both a de-traditionalisation in the form of the rise of the ‘unchurched’ as well as an emergence of new religious movements and alternative, non-institutionalised spiritualties. New religious (or confessional) pluralism is a fact, as is the dispersal of governance functions among various levels of decision-making bodies in matters that touch upon the relationships between the State and religions and beliefs (supra- and sub-national in addition to the national). In the face of these facts, what remains of the constitutional principle of the autonomy of the State from (organised) religion and vice versa? Is the gap between an increasingly unchurched majority sentiment and the legal regulation of religion still in force evidence of hypocrisy, ethnocentrism, or perhaps just bona fide neglect of a developing trend? And what are (still) legitimate policy aims in this changing societal context?

The RELIGARE project distinguishes four different models in place in European States.

Models Examples

States which proclaim a strict separation between State and religious groups France; Turkey

States with established majority Churches England and Scotland; Denmark

States that have signed concordats/agreements with Churches and religious communities Italy; Spain; Germany

States with ‘conditional pluralism’ Belgium, Bulgaria and the Netherlands

The first involves countries where there is – in principle – a strict separation between religious bodies and the State. In such systems, such as that in France (laïcité model), State policy is required to be conducted by means of secular laws alone and is therefore kept strictly separate from religious beliefs. States following the French example relegate religion, in theory at least, to the private sphere and oppose an inclusion of religion in State law and administration. A second model, by contrast, accepts institutional links between State and religion, such as the arrangement between the Church of England and the State in England. The third model is the concordat model in place in Italy, Germany and Spain. In these countries, the State has signed an agreement with some religions, but this does not mean that public authorities ignore other religions. A fourth model combines non-establishment with conditional (limited) legal, administrative and/or political pluralism. A State may support religions, but should it decide to do so, it must beware of favouring one religion over another. Countries like Belgium and the Netherlands fall in this last category. It is important to note that non-institutional links and engagements with religious communities (funding; support; dialogue) are the rule rather than the exception in most States. In the Netherlands, the monarchy has been tied to the Protestant Church, though there is no State church (since 1815).

This complex feature of the ‘secular models’ throughout Europe gives rise to controversies on both empirical and normative grounds: (1) unequal distribution of State support among various religious bodies (churches, denominations, and religious communities); (2) discrimination against non-believers and religious organisations other than the ones already regulated or admitted by law, (the latter group including ‘new’ religious minorities); (3) disagreement on the contours of the right to freedom of belief.

1.2 Between EU competency and EU relevancy

Before sketching the main research findings of the RELIGARE project, it may be helpful to bear in mind the legislative framework put in place at EU level, as well as a number of recent initiatives that were (or are being) taken by the EU authorities with regard to combating discrimination on the ground of religion or belief, since they are closely connected with the topics of research addressed in the RELIGARE project.

In at least three ways, the EU is addressing problems linked with discrimination on the ground of religion and belief:

First, there is the legislative framework put in place at EU level to combat discrimination on the ground of religion or belief: Article 10 of the Treaty on the Functioning of the European Union (TFEU) requires that the EU shall “aim to combat discrimination” on the ground of religion or belief (among other grounds), “in defining and implementing its policies and activities”. Article 19 TFEU enables the Union to take appropriate action to combat discrimination based on religion or belief, among other grounds. One such action has been to get the Framework (Employment) Directive 2000/87/EC, which prohibits discrimination on the ground of religion or belief in employment, occupation and vocational training, approved and effectively applied. Another legislative action has been the - still pending - Directive proposed by the European Commission (COM (2008) 425, Brussels, July 2, 2008) that aims to extend the scope of anti-discrimination to fields beyond the labour market. Following calls by the European Parliament and various civil society actors for the expansion of EU law protecting against discrimination, and after an elaborate consultation process, the European Commission submitted a Proposal for a Council Directive on implementing equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation on 2 June 2008. Keeping the political reality in mind, and the fact that the proposed ‘horizontal directive’ may not – at least in the foreseeable future – be adopted, the RELIGARE research project has nonetheless taken a particular look at the text of the directive, in particular at the provisions regarding protected grounds which may affect the rights and duties of religious or philosophically-motivated individuals and groups (e.g. prohibition of discrimination on the basis of sexual orientation). We will come back to this below, in the section on the potential impact of the RELIGARE project.

Second, among the action(s) and programmes to combat discrimination on the ground of religion or belief, the following publications have proven to be of particular relevance for the purposes of the RELIGARE research project:

(1) In 2011, Equinet, the European network of equality bodies, published “Equality Law in Practice: A Question of Faith – Religion and Belief in Europe’. This examines cases of religious discrimination and freedom of religion under Article 9 ECHR;
(2) The overview prepared by the European Network of Legal Experts and published in 2011 in the European Anti-Discrimination Law Review. The article reviews ten years of case law on discrimination on the ground of religion or belief and other grounds;
(3) Also in 2011, the European Network Against Racism (ENAR) published a report entitled “Reasonable accommodation of cultural diversity in the workplace”. The document defines reasonable accommodation as “an adjustment made in a system to accommodate or make fair the same system for an individual, based on a proven need”;
(4) The Network of Socio-Economic Experts in the anti-discrimination field, funded by the European Commission, is currently preparing reports (at Member State and European level) on discrimination on the ground of religion or belief. These reports will have a particular focus on employment and education;
(5) Recently, the Directorate-general for Internal Policies (Department C: Citizens’ rights and constitutional affairs) issued a study entitled “Religious Practice and Observance in the EU Member States”. The study highlights “present sources of conflict, (…) best practices and put(s) forward recommendations to promote both religious practice and observance and the respect of human rights”.

This is just a selection, but what all these publications demonstrate is the growing interest on the part of various actors, including the EU, in combating discrimination on the ground of religion and belief.

Third, one must not lose sight of Article 17 TFEU, which states: “1. The Union respects and does not prejudice the status under national law of churches and religious associations or communities in the Member States; 2. The Union equally respects the status under national law of philosophical and non-confessional organisations; 3. Recognising their identity and their specific contribution, the Union shall maintain an open, transparent and regular dialogue with these churches and organisations”. While it appears that implementation of the first two subsections is left to the EU and ultimately to the European Court of Justice (ECJ), to respect the third subsection requires specific action by the Union. In practice, this action is now in the hands of the President of the Commission, specifically his Bureau of European Policy Advisers (BEPA), which acts under the President’s authority. The RELIGARE findings suggest that a number of the issues investigated may impact on both the first two subsections and subsection (3) of Article 17 TFEU: religion and belief indeed involve, if not directly then indirectly, many fields of EU competence. They are listed below, in the section on the potential impact of the RELIGARE research project. Since the Commission can be expected to act as a monitor of the implementation of EU law, nothing prevents it from taking a more forward thinking approach in its next Action Programme.

1.3 Normative framework: ‘Inclusive State neutrality’ and ‘justice as even-handedness’

Two core principles are enshrined in the Constitutions of Europe’s States as well as in international human rights law: (1) freedom of thought, conscience and religion and (2) equality or the right to non-discrimination, including on grounds of religion or belief. The ideal that the EU itself stands for in this regard is formulated as follows: “The Union is founded on the values of respect for human dignity, freedom, democracy, equality, the rule of law and respect for human rights, including the rights of persons belonging to minorities. These values are common to the Member States in a society in which pluralism, non-discrimination, tolerance, justice, solidarity and equality between women and men prevail.” (Art. 2 TFEU). “In defining and implementing its policies and activities, the Union shall aim to combat discrimination based on sex, racial or ethnic origin, religion or belief, disability, age or sexual orientation.” (Art. 10 (ex Art. 13) TFEU).

In assessing the various ways in which the 10 countries under scrutiny handle issues that relate to freedom of religion and belief and non-discrimination, the RELIGARE research project proposes two yardsticks: ‘inclusive State neutrality’ and ‘justice as even-handedness’.

‘Inclusive State neutrality’ rejects a strict separation and strict neutrality between the State and any religions or secular beliefs insofar as those positions could be interpreted as public hostility towards churches/religious communities and/or secular values. Experience shows that it is an illusion to hold rigidly on to separation and strict neutrality. It also shows that separation can be used to exclude, willingly or unintentionally, vulnerable minorities who are seen as ‘outsiders’. Strict neutrality, when it takes on an ideological dimension, risks entering into competition with other ideas or systems – including religions - with claims on the public space. Inclusive State neutrality, on the other hand, offers an overarching framework for ‘managing’ competing claims on the public space, including those of religions and beliefs, in a way which both recognises their roles, and their right to make claims, while also refusing to favour one over another. The notion of inclusivity therefore serves as a guide to an equal opportunities policy, which within the framework of contemporary constitutional democracies includes equality of treatment of all individuals based on the principle of equal respect of human rights, civil liberties and social and economic rights. Inclusivity differs from a strict equality and other strict standards that are in practice unachievable in the current state of affairs and raise incommensurable questions (what is strict equal treatment of two religious communities, one with millions of followers and a well-established presence and the second counting mere hundreds of individuals, often newcomers in a country?).

Justice understood as ‘even-handedness’ further builds on this concept and recognises the legitimacy of active policies towards religion and belief, as long as they are applied in an inclusive way. The related idea of equidistance to be respected by the State towards different religions and beliefs under its jurisdiction connoted a certain hands-off attitude. From the perspective of inclusive even-handedness, diversity is an integral part of people’s right to participation in society. Beliefs and practices are thus not seen as an obstacle, and neither are they seen as static. In fact, they may change over time depending on the circumstances (e.g. mainstream or minority practices; conversion; revival of some traditions and interpretations, etc.). ‘Inclusive State neutrality’ and ‘justice as even-handedness’, taken together, mean that the State’s policy should be fair to all in granting recognition to religions, beliefs and practices; the principle is respect, with any limits placed only by way of an exception, e.g. for the protection of the fundamental rights of others in accordance with the case law of the European Court of Human Rights (ECtHR).

The data collected within the framework of the RELIGARE research project offer various concrete illustrations of what these two yardsticks, in combination, can mean in practice: the purpose in using them is to develop a more substantive understanding of equality, one which improves inclusivity in practice. Ultimately, ‘inclusive State neutrality’ and ‘justice as even-handedness’ seek to render more transparent and clarify the grounds for the protection of the freedom of religion and belief, taken in the broadest possible sense (i.e. including secular beliefs) and to help fight discrimination and exclusion on these grounds as consistently as possible.

Project Results:

In this section the main research findings that have emerged from the RELIGARE research are briefly sketched. The presentation of each of the four areas follows the same structure: after a summary of the key research findings, there follows a series of recommendations. The key insights and recommendations differ in various ways. These differences may be attributed to various reasons, such as the specificity of the challenges addressed in each of the four areas, the existence of a (EU) legal framework and the prominence of the relevant challenges in the (155) sociological interviews conducted by the project team members. The search for a balance between the protection of the freedom of religion and belief, on the one hand, and the principle of equal treatment and non-discrimination on the other, combined with an attempt to engage with the two yardsticks of ‘inclusive State neutrality’ and ‘justice as even-handedness’, is however, common to the different parts.

The presentation starts with employment and goes on to look at family law issues, public space and, finally, State support.

2.1 Employment: Protecting the freedom of religion and right to equality of employees and employers in the workplace

2.1.1 Religious diversity and the workplace – two clusters

The RELIGARE findings regarding the workplace were divided into two clusters: the individual and the collective cluster. The individual cluster looks at a number of mechanisms that have been implemented in the workplace to protect and to various degrees accommodate (or not) religious or non-confessional beliefs, practices and identity. These mechanisms are mainly grounded in human rights reasoning and anti-discrimination legislation. They also include ‘voluntary’ good practices. Furthermore, the analysis makes a distinction between front-office and back-office positions when it comes to the issue of visibility of religious symbols and dress in all sorts of workplaces.


Under the individual cluster, the distinction was made between (rare but present) discrimination cases sensu stricto (dismissal because of religious affiliation ), and various accommodation cases involving private employees and civil servants:

(a) conflicts or requests relating to religious dress (including wearing turbans, not cutting hair, covering hair, etc.) in both front-office and back-office positions;
(b) requests motivated by a need to reconcile conflicting religious time-working time obligations;
(c) requests for exemptions or alterations of particular job duties or circumstances, including socialising customs; and
(d) requests to use certain (often already existing) facilities or space, typically for prayer or meditation.


Only some of these cases involve conflicts with other fundamental rights and can thus be regarded as ‘hard cases’. ‘Soft cases’, where accommodations harm no compelling interests and are (nearly) cost-free, have also been identified. Contrary to what one may think, even these ‘soft cases’ have not been unequivocally successful in court.

• With regard to religious dress and symbols, the main issue relates to the Islamic headscarf. Noticeably, most instances identified from the data relate to front-office workers (e.g. cashiers, bus drivers, and also – in the public sector – teachers); this points to the fact that the visibility of distinctive religious dress is regarded as problematic by many private sector employers (the same can be said for public sectors that regulate the dress of public teachers). The case law, combined with the sociological country reports, confirm a pattern whereby religiously distinct workers are routed to the back-office, a tendency that may be symptomatic of a European discomfort with the existence or visibility of heterogeneity.
• In conflicts between religious duties or observances and working time, objections to accommodation here often relate to formal equal treatment. However, the organisation of working time in Europe adheres to a calendar strongly inspired by the dominant religions, with only a few instances of structural accommodation for minority groups. In this context, maintaining a regime of formal (or strict) equal treatment under a non-neutral standard (in the sense of being inspired by a dominant religious tradition which it privileges) may turn out to be problematic and in the long run indefensible.
• The RELIGARE findings further point to a wide variety of issues such as workers refusing to perform certain job functions (registering same-sex partnerships; handling alcohol or pork), refusing to adhere to dominant socialising customs (e.g. shaking hands with people of the opposite sex), workers challenging the position of women in managerial positions in the workplace, workers seeking to be excused from work birthday parties, amongst others. In the RELIGARE research the focus has been on ‘hard cases’ where religious freedom appears to conflict with gender equality, equal treatment of sexual minorities, etc. In the sociological interviews conducted, the overwhelming consensus amongst respondents was that in the workplace employers should try to accommodate the religious practices of employees (notably, also of secular respondents), but they do not all agree that a duty to do so should be enshrined in law.

Under the collective cluster, the following was addressed: (a) hiring (and dismissal) practices involving the ‘religious ethos’ exception (Article 4.2 Directive 2000/78/EC); (b) balancing job requirements with employees’ privacy and family rights.

With regard to faith-based employers, there are a number of issues to be mentioned here.

• First, there are tensions between the claims of collective autonomy and the rights of individual workers. In this regard, ‘more protection’ for one translates in more restrictive rights for the other. In order to determine the appropriate response, one must therefore look at broader societal goals, e.g. a wide leniency in the area of autonomy rights will limit access and inclusion.
• Second, the current legal framework with a limited exemption for religious ethos companies has created misunderstandings and complexities of its own. This is an area where the legal framework could be fine-tuned and/or good practices in the national approaches could be helpful to other Member States struggling with similar issues.

2.1.2 Two relevant legal frameworks

Two legal frameworks are relevant when it comes to the protection of religion and belief in the workplace: the human rights framework and the non-discrimination framework. The first has been the purview of the European Convention on Human Rights and the Court in Strasbourg, while the EU has taken the lead in formulating non-discrimination standards under a series of Directives, in particular EU Directive 2000/78/EC, in the area of employment, occupation and vocational training. The interaction between these two frameworks is reciprocal and complex: sometimes complementary in their operation, sometimes in clear tension.

The RELIGARE data show that there are gaps in the existing legal protection that make it difficult adequately to address (even modest) claims for religious accommodation in Europe and, in particular, that the tool of human rights “allows for diversity only under certain limitations, thereby conditioning and at times even distorting these religious experiences based on interpretations that form, what some do not hesitate to call, the core of a (contentious) ‘Eurocentric approach’. Freedom of religion derives its language and working categories from historical and dominant values.” This is one of the reasons why the RELIGARE project makes a case for using the important additional tool of legally enshrined reasonable accommodation on the basis of religion and belief in the workplace.

With regard to the workplace, the RELIGARE data reveal various conceptions of ‘equality’, and the tensions between these conceptions and freedom and other values. The idea of substantial equality may conflict with formal equal treatment: while under formal equality, the focus is on identical or similar treatment irrespective of particular personal characteristics, including religion, under substantive equality, context-relevant religious or belief-based identities, beliefs and practices of employees sometimes are to be taken into account. This is a multifaceted exercise and entails: inclusiveness (full-enjoyment of citizenship rights); redistribution (problematic socio-economic status of some ethno-religious communities); recognition (signal function); symmetrical protection (procedural even-handedness applies to everyone, and is not conceptually limited to particular minorities). Within this conception of equality there is a clear role for reasonable accommodation, which questions the existing anti-discrimination paradigm. The idea behind reasonable accommodation is that “by failing to accommodate, the characteristic can result in denying an individual equal employment opportunities.” Therefore: “Instead of requiring … people to conform to existing norms, the aim is to develop a concept of equality which requires (mutual) adaptation and change.” In light of these normative choices, the mode of handling issues/conflicts (judicial, legislative or regulatory means, or simply through the use of certain voluntary (good) practices) is not irrelevant. Soft measures should be complementary to more ambitious proposals for legislative change in a volatile environment.

2.1.3 Redefining the reasonable accommodation debate

To the extent that the debate on reasonable limits to accommodation was centred on the issue of ‘cost’ (economic or otherwise), the RELIGARE project argues that the question needs to be redefined: one must also take account of the cost to minorities and non-conformists and to the social/welfare system: from the question “How much can it cost? / When is it (dis)proportionate for the distributive agent (employer) to carry the cost?” to “Who should carry the burden? / What is a fair distribution of burden?” This provides an alternative way to consider the occurrence of religious ‘claims’; in other words, we argue that the process of accommodation is always a two-way street and oftentimes the degree of (self)accommodation and adjustment by the employees in question is duly recognised.

There should be a genuine and meaningful right to freedom of religion, including for employees in the workplace. It can further be argued that the collective aspect of freedom of religion requires a level of autonomy granted to churches and religiously-oriented organisations, possibly justifying exemptions to general non-discriminatory hiring and dismissal practices. Certain national (and European) approaches to freedom of religion in the workplace arguably do not meet minimum standards, in the sense that religion is relegated to the private domain or that it is held that an employee can always resign from a ‘voluntary position’ in the event of a conflict between professional and religious commitments. Yet, beyond minimum standards there are various approaches, which give adequate consideration to freedom of religion, including in cases where it conflicts with other rights, freedoms and values. Other values, such as social cohesion and pluralism, remain of course important but are at best derivative of (the product of) meaningful freedom of religion and (substantive) equality. They should not/cannot be ‘forged’ on their own.

One particular challenge when it comes to the management of diversity in the workplace regards ‘avoiding conflict’ behaviour. Avoidance is not necessarily good practice, as it may mean denying certain rights. The discomfort with issues of religion, and especially minority religions and Islam, could lead to an avoidance strategy that involves shunning confrontations by excluding the workplace as a site for the enjoyment of the freedom of conviction and belief, or by diluting the ban on discrimination on the basis of religion or belief to theoretical/rhetorical minima. Pursuing certain mechanisms of protection (of human rights) often exposes contentious issues and can lead to even more conflicts, at least in the short and medium terms. The wish to avoid problems might then be a reason for upholding the status quo. The RELIGARE project argues that the idea of inclusiveness should not be seen as a way to ‘contain’ minorities and those who do not conform to the prevailing workplace standards. The idea is, rather, to give individuals adequate room to reconcile their various identities and commitments; certainly a degree of adaptation to the workplace setting is necessary, but the integration of a widely diverse workforce in Europe should not require an assimilative approach to religion in the workplace, where ‘passing’ and coping (or even hiding) are routine requirements for employment participation. Sometimes confrontations are not only inevitable but also necessary on the way to achieving sustainable long-term justice.

Similarly, there is a real danger that initial ‘good practices’ of accommodating religious diversity in the longer term may trigger restrictive responses in society and in the case law, and may even lead to restrictive legislation. To give an example, in Denmark, a Muslim trainee was turned down because she did not comply with the dress code prevailing in the Danish department store Magasin. The Eastern High Court of Denmark in 2000 found that the negative treatment of the trainee based on her religious garb constituted indirect discrimination. In particular, the court noted that Magasin did not have a clear dress code. The ruling in this case initially led many companies to change their employee clothing policies to accommodate religious dress. However, the first case prompted a chain of Danish supermarkets including a group called Føtex to adopt a restrictive dress code and strictly enforce it. When a case against Føtex came before the Danish Supreme Court in 2005, involving the claim of a Muslim employee who had been fired for wearing a headscarf on the job, her claim was rejected. The reasons were formalistic: unlike the Magasin trainee, this employee had been aware of company policy regarding clothing and had signed a document saying that she would adhere to these rules. The Føtex case thus established the right of employers to set requirements for employee clothing, without a need to make room for dress mandated by employees’ religions, as long as these applied to all employees in the same position, thus as long as there was formal equality or strictly equal treatment.

There are both economic arguments (cost-benefit) and rights-based perspectives; at the EU level both arguments are relevant. In light of the need for more substantive equality and a more positively conceived freedom of religion, more needs to be done to allow workers of minority religions, here especially women, to participate meaningfully in the workplace. There are individual-level arguments: freedom and accommodation would allow the self-realisation by these women living in a society based on traditional Christian values where they are already disadvantaged in various aspects, but it also allows for more (macro)economic arguments for making the ‘business case for diversity’ (the fact that the European ageing workforce cannot afford to place certain groups and individuals on stand-by but must give them meaningful opportunities to engage, build skills and succeed).

2.1.4 Policy recommendations

Three main topics have been identified for concrete policy recommendations, in particular at the EU level, as relevant for future policy initiatives:

(a) Reasonable accommodation for religious observances in the workplace;
(b) Exemptions for religious employers/organisations from anti-discrimination laws and labour laws;
(c) Enforcement of human rights and equality legislation, in particular as it relates to religion (i.e. through Human Rights Commissions/Equality Bodies for religious inequalities).

Recommendations addressed to EU policy-makers

(1) Amend the Employment Equality Directive to add a right for public and private employees to request reasonable accommodation on the basis of religious beliefs, practices and observances, unless such accommodation would result in a demonstrated disproportionate burden or cost for employers;
(2) Track the legislative and judicial implementation of the Employment Equality Directive, in particular through available national case law applying the prohibition against discrimination on the basis of religion or belief. Develop a coherent and clear policy to address national decisions or developments that (often inadvertently) undermine the purposes of an anti-discrimination policy;
(3) Include the aspect of religious and philosophical diversity more clearly in the EU’s discourse on the value of diversity;
(4) Adopt explicit strategies accommodating EU employees from diverse religious backgrounds, promoting such strategies in other public and private workplaces through example;
(5) Create or encourage platforms for exchange of ‘good business’ practices in respect of accommodating requests for religious accommodation, and promote these national or cross-national practices across EU Member States.

Recommendations addressed to national governments

(1) Adopt national legislation that gives employees a legal right to request reasonable accommodation for religious beliefs, practices and observances in public and private workplaces, unless doing so would impose a demonstrated disproportionate burden or cost on employers;
(2) Establish an advisory service (‘helpline’) for employers and other decision-makers with questions about religious accommodations in various work settings, possibly operated by an equality body;
(3) Encourage the inclusion of workers from diverse religious backgrounds in the workplace through a variety of policy tools: e.g. encourage and incentivise company diversity plans, which also touch on religious diversity, address legitimacy of company ‘neutrality policies’;

Recommendations addressed to local governments, trade unions, equality bodies and NGOs

(1) Demystify the concept of indirect discrimination using information campaigns and informational resources directed at employers and employees;
(2) Voluntary accommodations that help employees reconcile professional and religious duties and do not cause any organisational hardship should be privileged and proffered as good examples;
(3) Educate labour market stakeholders and the general public on common religious practices and observances of religious minorities in the country;
(4) Emphasise the unacceptability of disadvantaging, negative treatment or discrimination against employees on the basis of their religion or belief as well as their religious or philosophical practices and observances.

2.2 Family law: Religiously inspired family laws and Private International Law systems within the territory of the EU

There is no direct EU competence for regulating substantive family law, and hence this area of the law continues to be a competence of the Member States. While the legal background of most States under study reveals the long-standing influence of religious laws based on Christianity (the exception being Turkey), secularisation has been a dominant trend across all the countries studied. Nevertheless all countries also display considerable cultural and religious specificities in their legal ordering of the family. In recent years, family law issues have become - from various points of view - increasingly relevant from an EU perspective. Not the least because the differential treatment among Member States of identical family situations - which touches on the most intimate sphere of life - certainly has considerable impact on the effective implementation of the fundamental right of free movement within the EU (see Article 45 of the EU Charter of Fundamental Rights and the Directive 2004/38/EC). A major issue in this context are the significant differences that continue to exist among Member States with regard to same-sex marriages and non-registered partnerships. Under EU law, recourse to arguments which prevent recognition within the internal legal order of situations created abroad, and which thus put an obstacle to the free movement of people, can only be acceptable if they are justified, proportionate and if they are applied in a non-discriminatory way. Decisions of the ECJ, so far, have demonstrated that even comparatively less sensitive issues – with no links to religion or belief – such as the family name, fall under EU-granted freedoms and basic rights (e.g. the ECJ judgment of 14 October 2008 in Grunkin and Paul v. Standesamt Niebüll, Case C-356/06).

As things stand, it is unclear how the impact of cross-border family law restrictions will be judged by EU bodies in the foreseeable future, not to mention Article 9 of the EU Charter of Fundamental Rights, the implications of the EU’s membership in the ECHR (particularly regarding Articles 8 (right to family and private life) and 12 (right to marry and found a family)) or the possible implementation of the proposed Equal Treatment Directive (COM (2008) 425, Brussels, July 2, 2008). It should be also noted that in application of the new Rome III Regulation on divorce and legal separation (Regulation No 1259/2010) and other initiatives to unify conflict of law norms relating to the family (e.g. Regulation No 650/2012 on matters of succession), family law will gain ever more relevance at EU level.

2.2.1 Private International Law (PIL): The bulk of the case law on religiously inspired family laws in Europe

A large number of court decisions collected within the framework of the RELIGARE project reveals that the main legal conflicts revolving around issues of religious diversity and the family that have arisen in the last decades relate to private international law (PIL) issues. The RELIGARE project’s research on family issues has thus been devoted to a large extent to disputes that fall within the domain of private international law as traditionally understood. In practice, interactions with religious(ly inspired) family laws, and hence the accompanying questions of private international law, are of particular concern to people who have settled in Europe on a more or less permanent basis but who nevertheless retain connections with their country of origin. They thus look for solutions that would allow them to live a harmonious family life, with international status, i.e. enjoying equal recognition under civil (in Europe) and possibly religiously inspired law (in the country of origin). Their situation is ‘transnational’. The difficulty in their case lies in articulating their (personal) status in such a way that it can be effectively received in both legal orders involved: on the one hand, that of the country of emigration (i.e. of origin), and on the other hand, that of their habitual residence. Today it seems that it is primarily Muslims who – in various parts of Europe – are faced with the difficulty of combining respect for the laws enacted by the (secular) State in their country of residence with the laws that apply in their country of origin. The values of the latter are often endowed, in their view, with particular legitimacy. Among the issues most frequently submitted to the courts are: the prerequisites for legally valid marriages and their recognition if concluded under religious (and state) laws abroad, some specific contents like dower payment obligations and divorce issues, in particular the recognition of foreign divorces performed according to religious laws and the limits of the applicability of such laws under PIL provisions. In the latter case, the invocation of public policy appears to be the most recurrent argument: religious laws are criticised for treating sexes or religions/beliefs unequally. The case law significantly differs, however, from country to country.

2.2.2 Beyond Private International Law

Yet, the research done under the RELIGARE project on issues relating to family life also takes a more forward-looking approach, going beyond the boundaries of private international law. This approach is of a more exploratory nature: what other potential approaches might there be when it comes to the treatment, in secular civil law, of personal statuses that are religious in nature? The RELIGARE research looked at the following three alternative (possible) solutions: (1) the incorporation into domestic (civil) law of religious rules, taking the example of the marriage contract, in particular; (2) the development of alternative mechanisms of dispute resolution (ADR) for certain types of family disputes; and finally, (3) recourse to Community intervention (EU law) in (international) family matters.

The reason for supplementing the repertory of private international law conflicts with a more future-oriented approach that explores the three above-mentioned alternative approaches is the following: over the years, the demographic profile of European communities of (im)migrant origin has undergone profound change. The presence of these groups and communities is not only linked to (im)migration, but now represents a permanent settlement in the countries of habitual residence in Europe. In many cases, these residents hold the nationality of the country of residence, while also retaining the nationality of their country of origin. These developments are changing the status quo: matters of personal status become, in such cases, first and foremost matters governed by domestic substantive law. Account is generally taken of religion when some form of incorporation of religious law(s) in the domestic State legal order is allowed, such as the recognition granted to marriages contracted (and registered) before religious authorities. One possible alternative would be to encourage spouses to enter into (secular) contractual arrangements under civil law, thus to allow such arrangements to accommodate religious requirements. Certainly the debate is a sensitive one, in particular when it comes to alternative solutions that raise the prospect of calling into question arrangements that have emerged from historical developments such as the monopoly of State law in family affairs. The use of prenuptial arrangements to govern relations between spouses are a well-established practice in the United States and seem perfectly reconcilable with the requirements of both religious (in particular, Islamic) and civil law. There is no reason to fear for the introduction of similar practices in Europe, they will not put at risk the exclusivity of State law in family matters.

The second avenue explored by the RELIGARE project is to accept that religious minorities be entitled to a certain extent to handle their family disputes in the informal sphere before bodies of their own choice (see section below on alternative dispute settlement bodies), and in their own way. Yet, on condition that they are very carefully elaborated and create neither new forms of discrimination nor unjustified privileges for some groups only, the work of these bodies may offer some helpful prospects. This may be the case, in particular, with individuals who, on the one hand, have become full-fledged citizens of their country of residence and are thus bound by its legal system, while, on the other hand, keep placing that system alongside religious norms that are reflected in practices and traditions. Admittedly, there is much contestation about the issue of religious ADR in particular. However, it may be worth the effort to further assess the pros and cons of ADR in family matters, and while so doing to give particular attention to the experience of countries where ADR is increasingly accepted in religious contexts also.

The third line of thought, which calls for EU action, is connected to the diverse forms of treatment that religious personal statuses receive in the different national private international law systems of the Member States. The research conducted in the course of the RELIGARE project reveals important differences in the way traditions such as arranged marriages, polygamy, or dissolution of marriage through repudiation are treated in the national legal orders of the Member States. Cultural traditions of Member States themselves play an important role here; the guiding principles regulating family relationships vary. Institutions or traditions that run counter to individual autonomy and gender equality are in most cases refused any form of recognition, while in some cases mitigating circumstances may permit finding a viable solution. Yet the disparate approaches to these issues constitute a true challenge from the point of view of EU law, since in practice it can give rise to ‘limping situations’ – whereby a situation lawful in one country is not recognised in another – that present a serious hindrance to the free movement of people within the Community. The above-mentioned decision of the ECJ in the Grunkin and Paul case may serve as an illustration here.

2.2.3 Policy recommendations with regard to family issues

Three observations form the basis for the recommendations below: First: the RELIGARE data show a widespread lack of sound legal information on the meaning and the role of religious family laws among minorities in Europe; Second: the starting point for elaborating new approaches to religious family laws should preferably be the existing legal frameworks, and their main characteristics; Third: there is no specific EU competence for regulating family law, thus the RELIGARE recommendations below include, to a great extent, informal initiatives and recommendations.

Recommendations with regard to PIL family issues

(1) The Commission should assess the extent to which the intersection between religion and family law constitute, in practice, barriers in the implementation of the Citizens’ Directive 2004/38, with a focus on the implications of these restrictions for the principle of non-discrimination, sufficient legal certainty, and proportionality.
(2) Courts and administrations often lack the means to thoroughly interpret foreign family laws and the way these laws are concretely applied/interpreted in the country of origin. Databases of case law and of foreign family laws (and their concrete application) and institutions for professional education on an EU level should be established, and better coordination of the different approaches is necessary.
(3) The EU should encourage and develop further efforts in fostering mutual recognition of family law decisions by administrations and courts in other Member States. This can be achieved by either unifying PIL norms according to the Rome III model (e.g. implementing the Rome IV regulation on matrimonial property and encouraging member states to join the existing regulations), or by developing common standards regarding the scope and limits of the application of public policy against foreign norms. Unless further measures are taken at EU level, even if only on an informal level - information, cooperation or education - it is likely that the low level of legal coherence within the EU today will continue to prevent an increasing number of EU citizens from making use of their basic right to move freely within the EU. This situation clashes with the international reputation of the EU with respect to the implementation of human rights standards in an equally efficient and neutral way. In its negotiations with third countries, the EU’s credibility when it comes to defending and advertising such standards is permanently at stake.
(4) EU-wide uniform standards should be developed regarding the application of foreign religiously inspired norms or decisions delivered abroad under such norms. The traditional approach of examining only the result of the application of foreign norms – instead of (re)examining them thoroughly – grants more stability (and also international harmony) in family relations and legal security for the parties irrespective of their actual residence. A careful contextual appreciation, of each case in particular, involves three steps: First, the foreign institution has to be evaluated based on reliable information. Second, it has to be compared to other institutions of the law of the land in its relevant elements, distributing rights and duties between the parties involved. Third, it has to be clarified whether and to what extent the parties are entitled to waive certain rights voluntarily (by using dispositive law) in the event of differences between the requirements of the legal systems involved, provided there is no mandatory State protection governed by mandatory legal rules, including ordre public, at stake.
(5) Awareness of the fact that European family law systems derive from a Christian background, although it has been translated into secular forms, should be increased. Thus, specific needs of newly installed religious minorities should be appropriately addressed. Inclusive even-handedness can help maintain the balance: whereas the application of the public policy exception in PIL is meant to meet the requirements of human rights and their high standards of protection, irrespective of the concrete situation of the parties involved, sustainable solutions - including for families with a religious minority background - should be given priority under the applicable law. This may mean, under certain conditions, recognition of decisions delivered in the countries of origin of couples/litigants of migrant/religious minority background. In particular, initiatives to further coordinate the recognition of divorces under foreign (religious) laws are recommended.
(6) Initiatives to increase mutual cross-border recognition of State court (or administrative) decisions should be fostered, enabling parties to regulate their family relations in a harmonious way, including across borders if they wish to do so. In particular, the potential benefits to be drawn from optio iuris (choice of State law determined by contract) in family relations, i.e. by encouraging parties to make use of their autonomy, in particular with regard to marriage contracts, should be further assessed.
(7) Legal practitioners (judges, State officials and lawyers in general) should be offered professional training in techniques of PIL and accommodation available within existing law, as well as on the content(s) of religiously inspired foreign legal institutions. Application of foreign religious laws should not be rejected automatically, based on the perception that “secular courts are not entitled to enter into religion” ). Foreign family laws, provided they are the formal/official law of a country, regardless of whether they are secular or religious, should indeed be treated according to the common principles of PIL.
(8) Databases of case law and good practices for professional education on EU level should be established; they should include information on foreign, religiously inspired laws, in particular those laws that conflict with domestic law, and on the way they are applied in the country of origin.
(9) The use of legal experts in courts (e.g. arranging settlements) should be extended where necessary; experts should be given instructions as to what the court requires of them.
(10) Customers/litigants/disputants: individuals - in particular of migrant background - should be appropriately informed of the existing legal procedures as well as of the contents/orientation of the law of their habitual residence.
(11) Cooperation with religious institutions: joint initiatives should be taken by State authorities and religious organisations/persons with a view to taking – to the extent possible – an unequivocal stand on the fact that some relevant aspects of religious (family) law(s) are in conflict with European family law, making it clear that in such cases the law of the country of residence has precedence in principle. Such a stand would necessitate greater outreach on the part of the State to these institutions without necessarily seeking to control them; ‘dialogue’ is the keyword here

Recommendations with regard to domestic legal institutions and ADR

According to contemporary European legal understanding, family law issues are perceived as private matters but the State is entitled and even obliged to intervene in case of violation of the weaker party’s interests (not restricted to domestic violence). Therefore:

(1) ADR mechanisms should meet the standards of impartiality and professionalism. Access to State courts should remain open, at any time. This condition is equally true for religious ADR mechanisms and institutions. All kinds of groups - including non-faith-based groups – should be entitled to set up their own ADR mechanisms within the limits set by the procedural laws of the State.
(2) The main goal of allowing ADR mechanisms to operate in family matters should be to strengthen/improve acceptance of the principles of the legal order and access to its mechanisms. Free choice to regulate one’s private matters has to be respected, and the State should not intervene there, as long as there is mutual respect for the rights of the persons involved and decisions are made professionally and free from undue pressure.
(3) There are profoundly different convictions as to the acceptance of marriages registered by religious institutions but having civil legal effects. Some Member States maintain a well-established system of such registrations (e.g. Denmark, Spain or the UK), whereas others, like France or Germany, would reject these for legal-historical reasons. Thus, any recommendations must take such differences into account. As already mentioned above, in all Member States, public information about the prerequisites for legally valid marriages should be increased. Norms simply prohibiting exclusively religious marriages do not appear to solve the whole problem of legally invalid religious marriages.
(4) The legal prerequisites for entering into a legally valid marriage - whether civil or religious - should be much clearer. As a matter of fact, religious marriages as such have no legal effects under the family law rules of most Member States. There are countries (Italy and Spain for example) where religious marriages can have civil effects provided they respect all the conditions required for performing a civil marriage. Yet where religious organisations are authorised to register marriages under the law of the State, as is the case for England and Wales, they however do not always meet the necessary requirements, thus producing marriages that are invalid under civil law. Parties who trust in the validity of such marriages may find themselves in the position of not being able to claim their rights in State courts and before the public authorities.. It is thus strongly recommended that the legal prerequisites for a valid marriage (civil and religious) be made clear in all EU Member States.
(5) Legislators in the Member States should be encouraged to draw clear legal lines between desirable and non-desirable forms of ADR. Insofar as Member States wish to keep the monopoly for family law issues with State courts and institutions, which is also justified, they should take the appropriate measures to guarantee access to these institutions to everyone efficiently. This relates to:

(i) Information about access to State institutions and their functions, including information about the compatibility of State law and religious convictions, e.g. in marriage issues;
(ii) Providing effective legal aid for those in need;
(iii) Increasing cultural sensitivity on the part of State institutions concerning the religious and cultural needs of minorities, which are less well known among the religious or cultural majority, trying to find ways for reasonable accommodation within the framework of the existing law and its institutions ;
(iv) Information should be made available about the formal and material legal prerequisites to enter into a legally valid marriage and about the consequences of (religious) marriages that are considered legally invalid since not contracted in the proper (secular) forms;
(v) In countries where the registration of legally valid marriages by religious institutions is permitted, public authorities should make sure that religious institutions (and the public) are correctly informed of the legal prerequisites of such marriages. Cooperation with the religious organisations can considerably increase the efficiency of such efforts.

2.3 Access to and use of the public space

2.3.1 Two thematic foci

The RELIGARE research on issues relating to the (neutrality of the) public space focused on two cases in particular: religious dress codes and places of worship. The empirical findings of the RELIGARE research with regard to public space issues show that the debates in the countries under scrutiny tend to revolve particularly around these two cases. These issues are prominently present in the media and also in the parliamentary debates, but it also appears that in the past few years a significant number of disputes have been settled through the courts. National case law has been collected by the different RELIGARE national teams and analysed comparatively.

2.3.1.1 Religious dress codes

The RELIGARE data show that each of the 10 countries under scrutiny has to date followed a different route to regulating the question of religious dress codes.

Full-face veil

Regarding the full-face veil, France and Belgium have adopted laws that prohibit appearing in public with one’s face covered. In both cases the prohibition applies to the entire public space. At the opposite end of the spectrum is the UK, where the minister for immigration in 2010 qualified any provision aimed at banning the burqa or niqab as “unBritish”. There are no legislative or administrative provisions forbidding the burqa or the niqab at the national or local level, and neither are there any directives by professional organisations on this matter. A third route hinges on local law: the State abstains from outlawing the use of the full-face veil throughout its territory, but the ban may be introduced by mayors or other local authorities by means of administrative provisions. This is what happened in Spain and, for a certain period, also in Italy. A different approach is followed in Denmark, where, apart from a single piece of legislation prohibiting the wearing of religious or political symbols in court, there is no law forbidding the wearing of the full-face veil, nor are there any local administrative provisions outlawing its use. Instead, there is some case law and also a series of documents issued by professional bodies and government directives, which supply guidelines for dealing with the most controversial cases. By these means, for example, it has been established that women wearing the burqa or niqab can ride public transport but train cards require the identification of the bearer; that schools and universities may (but are not obliged to) prohibit the full-face veil where it hinders non-verbal communication between students and teachers. In these and other cases a pragmatic and functional approach has been adopted. The full-face veil as such is not perceived as a garment that violates fundamental human rights, and thus the general attempt has been to find a solution that guarantees, as far as possible, women’s right to wear the burqa or the niqab; only for functional reasons has the wearing of this garment been forbidden.

RELIGARE data indicate, however, that current debates on the full-face veil are frequently driven by an ideological approach that focuses on values and principles, claiming their general and undifferentiated application, but they fail to examine, case by case and within their own context, those situations where the full-face veil ban might clash with the principle of the non-discriminatory access and use of the common public space. The debate tends to be monopolised on the one hand by those who believe that the full-face veil – always and everywhere – offends women’s dignity, public order, security, and gender equality, and on the other by those who consider it a manifestation of faith, personal self-determination, freedom of expression and practice of religion that must be guaranteed without exception.

Based on the RELIGARE findings, which include the sociological reports, it appears that the two yardsticks of ‘inclusive State neutrality’ and justice as even-handedness’ would permit a less radical or confrontational and a more balanced way to tackle this issue, namely by carefully assessing the different problems that the full-face veil can present to an orderly enjoyment of the common space. A general ban on wearing a full-face veil affects the accessibility of the common space, for women in particular. Yet it is a space where the right to move freely within the territory of a State is, in principle, granted to any person and the scope for limitations is thus restricted (see, in this sense, the ECtHR decision in the Ahmet Arslan v. Turkey case ). Evidently, there are situations where seeing the face of a person is necessary (when checking identity documents, for example), and there are also certain activities that can be hindered by a veil covering the whole face, such as when driving a car, for example, and occasions when appearing with one’s face covered can give rise to social alarm. In these and (a limited number of) other instances it appears to be legitimate to prohibit the wearing of the full-face veil in common areas, based on a pragmatic rather than an ideological approach, that is, on the basis of a concrete examination of the situation that takes account of the rights of others to the enjoyment of the (same) common public space, and their right to free movement as well. In these cases it is not the symbolic meaning of the full-face veil (which varies from person to person and can never be ascertained with certainty) that is at stake, but rather the social difficulties it may cause. This (functional) approach allows for measures to be adopted that are proportionate to the practical problems raised by the wearing of the full-face veil and that also take account, as far as possible, of the individual’s freedom of religion and expression. If there is evidence that wearing the full-face veil is not a matter of free choice but the consequence of an imposition (by others), for example, public authorities may be expected to prioritise the protection of individual freedom. Once this freedom is safeguarded, however, wearing clothes that are part of a religious practice or manifest a religious and cultural conviction cannot be limited in the public space as long as there is no evidence of infringement upon the right of others to freely move within that same space.

Religious symbols worn by teachers at school

As far as religious symbols worn by teachers and students at school are concerned, a general prohibition is in force in French public schools. In Belgium the trend is to follow the same approach. Some German Länder prohibit teachers from wearing religious symbols, but the prohibition does not apply to Christian symbols.

Here again, an inclusive and even-handed approach would offer a more nuanced solution. A first step in the reasoning is to take for granted that fostering tolerance and respect for pluralism is (or should be) one of the most important goals of the educational process in a context of pluralistic democracy. This goal can be reached only if teachers provide an education based on an impartial approach to the religious and philosophical convictions of the students and their parents (see Article 2 of the First Protocol to ECHR ), in order to prepare students to recognise the role that religions and other beliefs and worldviews play in individual and social life. At the same time, students should be shielded from any bias or false comparisons as well as from any ostentatious and/or imposed manifestation of religion or belief on the part of the teachers. Inclusive even-handedness also takes account of the fact that in performing their work, teachers enjoy the right to freedom of thought, conscience and religion and cannot be prevented from manifesting their religious or philosophical convictions except if these were to conflict with the school’s secular character. However, these manifestations must respect the right of students and (depending on the students’ age) of their parents not to be exposed to forms of indoctrination that violate their freedom of thought, conscience and religion.

Teachers can therefore be prohibited from wearing religious symbols and clothing that, in a given context, might have an ‘indoctrinating’ effect on students, provided that this prohibition is proportionate to the specific situation and is applied in a non-discriminatory manner. By way of a provisional conclusion and considering that each educational context has its own specificities, it appears sensible to adopt a case-by-case approach that refrains from applying the same rule to different situations: an even-handed approach offers more possibilities for taking into account the distinctive circumstances that impact on the delicate balancing of teachers’ and students’ rights. At the same time, such an approach requires some guidance, to ensure that it offers the necessary legal certainty.

Religious symbols worn by students at school

In the case of religious symbols worn by students at school, the reasoning is slightly different: manifesting one’s religious or philosophical conviction through the wearing of symbols and religious items of clothing by students at school is an expression of the right to freedom of thought, conscience and religion granted by Article 9 ECHR. According to the ECtHR case law, such manifestation can be subjected only to the restrictions “necessary in a democratic society in the interests of public safety, for the protection of public order, health or morals, or for the protection of the rights and freedoms of others” (Article 9 ECHR). As repeatedly affirmed by that Court, there is less potential for students to indoctrinate others than for teachers: therefore the right to manifest one’s religion or philosophical convictions through symbols and clothing should be limited only when it represents a concrete danger to the principles of tolerance and mutual respect that are essential for any educational environment. Once again, each case has its own specificities that need careful (even-handed) consideration. Moreover, a general ban on religious symbols worn by students, though accepted by the ECtHR when the State’s policy is to safeguard the secular character of its schools, may be problematic in light of another fundamental principle of the educational process, namely, pluralism. Justice understood as even-handedness requires that a general prohibition should be the last resort, to be used only in cases where other less restrictive means do not allow for an adequate solution. The arguments in favour of an inclusive case-by-case approach (as well as those that require recourse to even-handed means), already mentioned in reference to teachers, apply a fortiori in the case of students.

2.3.1.2 Places of worship

Places of worship play a key role in the expression of any religion or belief. A comparative survey carried out by the RELIGARE research project indicates that although, in principle, all religious communities can expect to be treated equally (this applies explicitly to Belgium, Bulgaria, Italy, the Netherlands and Spain), in practice this is clearly not the case. In Turkey for example non-Sunni groups are not granted the treatment enjoyed by Sunni Islam; in Italy, although the law is the same for all and does not distinguish between religions, in practice it is extremely difficult to open/build any type of religious monument that is not a Catholic church. In some cases formal procedural reasons are invoked to withhold administrative permission: protest by neighbours, sometimes accompanied by threats of violence (in Bulgaria, England and Wales); excessive noise at night (Germany); and traffic congestion combined with a lack of parking space (France).

There are also important differences between national policies as to the financing of places of worship. The Danish State supports the building of places of worship belonging to the (Lutheran) state church. In Belgium, the government does not generally give financial support to the building of places of worship, but it does offer, to recognised religions, the right to such support for maintenance and renovations of their buildings. Bulgarian law allows the government to support the building of a place of worship, but when and how this is done is at the discretion of the local authorities. In France, the government may not in principle fund places of worship, but recent legislation allows local authorities to issue leases for places of worship, especially where the building in question is also used for other purposes (see also below).

The RELIGARE research addressed three sub- issues in particular: the construction and the opening of places of worship, the reassignment of former places of worship for other purposes and the location as well as regulations/rules that apply in the case of (public and private) burial grounds or cemeteries. For each of these sub-issues, the analysis proceeds in an identical way: after a comparative examination of the way these issues are handled in the present situation in all (10) countries under scrutiny the report sketches what would be the solution if an inclusive even-handed approach were to be taken. In order to save space here, we refer to the RELIGARE website, where more detailed analyses are made available (www.religarepoject.eu). They address the abovementioned three issues, as well as the controversial issue in several countries of the calls to worship and the ringing of the church bells.

2.3.2 A key conceptual proposal: not one, but three kinds of public spaces

The data collected within the framework of the RELIGARE project all clearly indicate the intricacy of the question of the neutrality of the public space: can the public space ever be neutral (from a religious or philosophical point of view)?

The answer to this question depends on what one understands by ‘neutral’: if a neutral public space ought to be one that has no distinctive quality or characteristics, then the answer cannot but be negative. Public space, in all its various manifestations, has always had distinctive characteristics that derive from the history, culture and belief of the people who live in it. As religion is one of these characteristics, a public space cannot be expected to be entirely religiously neutral: even the most anti-religious States (e.g. Enver Hoxha’s Albania) could not, in practice, build a public space totally devoid of religious characteristics. If, however, the term ‘neutral’ is used in the sense of ‘having no preference’, being even-handed, fair and impartial, the conclusion cannot be so clear-cut, and one needs to distinguish between the various possible meanings underlying the expression ‘public space’; in combination they allow for a more accurate analysis of the comparative data.

These three spaces need not necessarily be seen as physically or temporally separate; rather, they coexist and overlap to a considerable extent. A square, for example, is typically a public space that can be common, political and institutional at the same time, depending on the use made of it. The same applies to the school: it is a space that is common, political and institutional at the same time. With this consideration in mind, it is possible to differentiate some characteristics of these three dimensions of the public space:

a) Under this classification, the common space is understood as the physical space that people have to enter in order to meet their basic needs: in this sense, it is inescapable. This space is not accessed with the intention to participate in a political debate, but simply to get to work, for example, or to buy what is needed for daily life. The communication process that takes place in this space is not primarily political. Of course, wearing a cross, a kippa or a turban may convey a message concerning the religious belief of an individual, even when that individual is doing his or her shopping. But the same kind of message can also be conveyed by his/her haircut, earring or tattoo and there is no reason why religious symbols should be regulated more restrictively in such situations than other symbols. From a normative point of view the common space must be kept as accessible as possible to avoid segregating in their homes people who feel bound to manifest their religion or belief in some way when entering the public space.
b) The political space in this schema can be understood as the space of debate and discussion where the public discourse takes shape. It should not be regarded only as a space of intellectual “argumentation about the truth value of propositions”, but more broadly as “a realm of creativity and social imaginaries in which citizens give shared form to their lives together, a realm of exploration, experiment, and partial agreement”: it is the space where new “normative worlds”, as Robert Cover calls it, take shape. It is frequently also a metaphorical space, although it can take the form of physical locations (Speaker’s Corner in London, or a political rally in a square, for example). In order to perform its function of providing room for open and democratic debate, the political space should be free and pluralistic: the visible presence of different religions and beliefs in this area is indispensable to the pluralism on which a democratic society is based (as affirmed by the ECtHR’s decision in Metropolitan Church of Bessarabia v. Moldova, 2001).
c) Finally, the institutional public space can be understood as the place where binding deliberations, which are compulsory for all, are conducted (parliament, the law courts, public administration, etc.). It is not (only) the space of debate and discussion, but the place of decisions that, once taken, have to be respected by everybody. A law court is not a TV talk show: a court delivers an enforceable judgment that determines who is right and who is wrong. In order to gain the general respect and recognition that is required for the enforcement of such binding decisions, this space must not only be, it must also be seen to be, fair and impartial.

In the three public spaces identified above, individuals act in a different capacity according to the role they take or the task they are expected to perform in that space. A public school, for example, is attended by students and teachers. Once they have entered the school doors, students retain their private status, while teachers acquire a more public status. Consequently when a student wears a headscarf, she is manifesting her personal conviction/belonging; when the scarf is worn by a teacher, this private dimension cannot be dissociated from the public one. This role differentiation may help explain why the case law in most countries investigated within the framework of the RELIGARE project indicates greater willingness to accept the students’ scarf than the scarf worn by a teacher. One speaks of the same space (the school), but seen through the lens of the various roles performed by the actors, one distinguishes two different groups of people who take different roles within that (same) space. Even-handedness implies that one takes a different position depending on the role and responsibility of the person(s) involved.

One can also look at the public space issue from an angle that is neither spatial nor personal, but functional. From this perspective, the notion of public service comes to the fore: the school or the hospital provides a public service, independently of whether it is a public or a private school or hospital. A public service can be provided both by a public and by a private institution. At present the tide seems to be turning towards the expansion of public services rendered by private institutions (including in fields that were traditionally reserved to the State, like security, management of prisons, etc.). From this functional point of view, new questions arise: are all public services, without distinction, to be provided in a ‘neutral’ way? When a religiously or belief-oriented private institution provides a public service, is it obliged to give up its religious/belief characteristics?

The distinction between the three spaces, each with its own characteristics and uses, allows for the application of differentiated solutions to different situations. The typology also yields two yardsticks of equal treatment, through the combined test of ‘inclusive State neutrality’ and ‘justice as even-handedness’, to play the role assigned to them: by emphasising the diversity of spaces, the inclusivity test focuses not so much on the substance of religions or beliefs, but distinguishes instead between modalities of management of the public space, and it makes this management depend on the functions one particular space has within a particular context. By focusing on the various functions a particular space may have in a concrete situation, no actor in the public space is confined to a single identity. The classification system offers new possibilities for finding an equilibrium among at least three functions of the public space: it allows for a (functional) approach that is specific to the two proposed yardsticks of ‘inclusivity’ and ‘even-handedness’: the aim is inclusion of all parties (individuals) involved, through a careful examination of the characteristics of each situation put in its own context.

It is against this backdrop that a number of specific recommendations have emerged from the RELIGARE project. These are listed in the section that follows.

2.3.3 Recommendations with regard to the public space

2.3.3.1 Recommendations with regard to religious dress codes

(1) Combating discrimination. Both the EU Commission and EU Member States should assess whether the existing rules concerning the wearing of religious symbols or clothing by students and teachers in public schools – insofar as public schools that provide vocational guidance, vocational training, advanced vocational training and retraining are concerned –respect the national provisions adopted pursuant to Council Directive 2000/78/EC. While pursuing the discussions on the pending proposal for a Council Directive on implementing the principle of equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation, one relevant question to assess is whether the prohibition on wearing the full-face veil in public spaces and the provisions concerning the wearing of religious symbols by students in public schools, which are in force in France and Belgium, constitute a form of discrimination in light of Art. 2(2) of the proposed Council directive.
(2) Providing guidance. Drawing inspiration from best practices that have proven successful at national and local levels, the Commission should help disseminate information that can aid the public authorities of the EU Member States in addressing the issues of the full-face veil and of religious symbols worn by students and teachers at school, in a way that is in compliance with the ECHR norms and the EU laws. EU Member States should also consider the possibility of establishing advisory bodies at the local level, with the participation of different stakeholders, which would provide guidelines on the issue of religious symbols/clothing worn by students and teachers at school in a way that respects the specificities of each situation.
(3) Protecting women’s freedom of choice. EU Member States should assess the effectiveness of the existing legal frameworks aimed at protecting women against being forced to wear religious or special clothing and, where there may be such compulsion, increase the level of protection.
(4) Avoiding unnecessary prohibitions. Legislators should refrain from enacting a general prohibition preventing teachers from wearing religious symbols and clothing at school, provided it is possible to address the issue on a case-by-case (school-by-school) basis through the intervention of the school authorities. Authorities should limit the prohibition for students to wear symbols and clothing expressing their religious or philosophical convictions to situations where such restriction is strictly required to guarantee the carrying out of school activities and no reasonable accommodation of the religious needs of the student seems negotiable. In both cases, if the competent authorities deem that a general ban on religious symbols and clothing worn by teachers or students at school is required, this should be formulated in terms that are non-discriminatory and that place as few limits as possible on the freedom of thought, conscience and religion of students and teachers alike. Finally, EU Member States should refrain from introducing a general ban on the full-face veil in all public spaces, in particular by recourse to criminal law provisions, and should assess the suitableness of limiting the wearing of the full-face veil only in those places and situations where seeing the face of an individual is required for security purposes or where public or professional requirements demand that civil servants wear religiously neutral attire or that their face can be seen.

2.3.3.2 Recommendations with regard to places of worship

(1) Government authorities should effectively apply the principle of equal treatment to the building and opening of places of worship. It cannot be accepted that the discretionary powers of local authorities allow them to treat different religious groups differently when it comes to the acquisition or lease of a place of worship.
(2) Religious communities are entitled to be treated even-handedly with respect to bell-ringing or calls to worship. Even-handedness also means that government authorities take into account all contextual factors, looking for a fair balance between what is acceptable - preferably, through negotiation - to impose on a neighbourhood and the requirements of the protection of religious freedom.
(3) If a place of worship is no longer used for its initial purpose, it is recommended that the owner (most often the religious community concerned) be allowed to link the selling of the building to conditions governing its future use and that are legally enforceable on the buyer and his successors.
(4) Finally, it is recommended that legal steps be taken to ensure equal treatment of all religious groups as regards burial or cremation. Restrictions based on public health considerations, imposing serious constraints on the wishes of various religious communities concerning burial or cremation, should be critically assessed: are they still justified? An inclusive, even-handed approach would require that these restrictions be critically examined in light of contemporary means of protecting public health.

2.4 State support mechanisms

The RELIGARE project also investigated the issue of State support to religious and other communities of conviction, in particular the funding of clergy salaries and the financing of a number of core religious activities and institutions. Special attention was paid to the funding of places of worship, the access to and funding of religious and life-stance broadcasting on public media channels, as well as the funding of training for religious leaders. Subsidies for charitable, educational or social activities under the guardianship or responsibility of religious or philosophical organisations were not taken into account. Where religious institutions receive public subsidies for these activities, it is generally conceived of as remuneration for a public mission. Such “secular” activities carried out by institutions or individuals sharing a common religious conviction therefore fall outside the scope of the RELIGARE research strictly speaking.

At first sight, the issue of public financing of religious and philosophical organisations seems remote from the competencies of the EU. Article 17(1) and (2) of the TFEU confirm this. Sociological and demographic changes in European societies show, however, that the financing of religions and philosophical organisations poses a key challenge to constitutional arrangements and regulations of State support as they have been elaborated in the past in the internal legal order of the Member States. In order to assess the challenges that State support to religious groups (and occasionally also to secular organisations, as is the case for the humanist movement in Belgium and the Netherlands) actually poses to public policies, closer attention needs to be paid to the impact, in practice, of State support: who are these groups and is State support even-handedly distributed among them? Two recent developments that are characteristic of European societies appear of particular relevance here; they challenge policy- and decision-makers ever more pressingly in an increasingly diverse societal context: individualisation of life styles and (personal) convictions and the steady increase of secularisation. They are reflected in policies of deregulation and privatisation which, in turn, result in the financing of religious groups being left to market forces and the private sphere.

The RELIGARE project proceeded to a comparative examination of the various regulations that apply in the countries under scrutiny. The method adopted has been an analysis in two steps. First, it distinguishes five models of support by the State to various religious groups, paying special attention to certain injustices that are the consequence of a lack of adaptation (of the historical legislative framework in place) to recent socio-demographic developments in contemporary European societies. Secondly, the analysis looks toward the future: what models offer better alternatives, and is there a role to play for the EU in promoting such alternatives? In the section below the focus is on the latter question: what alternatives to the existing models of State support could be developed and what role could the EU play in this? The comparative analysis of sociological data collected by the RELIGARE project researchers shows increasing criticism of the historical models of State support; the critique however targets not so much the principle underlying such support but rather the way the existing mechanisms are being implemented: discriminatory biases and the lack of equivalent public support for sometimes long-established communities and newly arrived religions or beliefs. Adaptations have been made in some countries, but for the most part they are still based on the existing design, often with no assessment of the place of and the role played by religious and, more broadly, belief groups in State and society. It is not enough to extend support to other groups beyond the ones that have hitherto been the beneficiaries, since even then there will still be groups excluded, or even discriminated against. Moreover, it does not suffice to draw inspiration from (experiences and models from) abroad given how deeply each model is rooted in the history of a particular country and in certain equilibria that have not been easy to achieve.

2.4.1 A forward-looking approach: pragmatic steps in search of a more ‘inclusive’ and even-handed way to justify State support

The comparative investigation of different models in force reveals that mechanisms of State support for religion, which were mostly designed during an era when traditional religions still formed the majority (and were hence much more important than they are in Europe today), and, moreover, could in some cases lay claim to compensation for the confiscation of their property, are now the subject of increasing criticism. The main criticism is that they fail to meet the conditions of equal treatment of religious and belief groups both old and new, with the numbers of the latter growing just about everywhere in Europe. These groups are not only linked with post-war migrations, but also with the free movement of persons within the territory of the EU and with the phenomenon of new individual religious choices. Moreover, non-religious people ask for respect for the freedom not to have a religion.

In the section below, a number of policies (possible alternatives to the present models) are sketched out. They draw on what has already been put in place in some countries, or adopt a more future-oriented view. At this stage, it may be useful to recall the position of the ECtHR on the principle of State support to religions. An examination of the case law of the ECtHR reveals that the Court does not exclude differential treatment, but it requires a balanced and explicit justification in an objective and reasonable manner. However, one significant trend in the ECtHR’s interpretation of the Convention is the progressive strengthening of the requirements of non-discrimination. It entails that national legislators are required, whenever necessary, to adjust existing mechanisms of support with a view to offering equivalent benefits to all, including non-majority and/or non-traditional groups. The requirement comes very close to the standards developed by the RELIGARE project: ‘inclusive State neutrality’ and ‘justice as even-handedness’.

2.4.1.1 The fiscal route

One way to support religions is indirect financial support of religious denominations (and other non-religious organisations) by means of tax benefits for their charitable, educational or social activities. This arrangement exists in all 10 countries analysed by the RELIGARE project. In some countries, this indirect form of support is also made available for religious activities. In practice, tax exemptions and tax deductibility arrangements for donations and bequests allocated to religious communities along with other non-profit organisations are a way to support funding of religions by members and followers: they can benefit from an extensive range of tax exemptions and deductions that are in some cases linked to activities and services (VAT exemption or VAT reduction for the supply of goods and services, exemption or reduced rate of business and corporation tax). Other types of exemption may also apply, such as exemption from inheritance tax, exemption from tax on donations and bequests, tax deductibility for donors, exemption from property tax for buildings mainly used for public worship, etc.

If some activities of religious organisations are to be considered as serving economic purposes, they are generally subject to taxation rules that apply equally to (other) economic activities as well. In certain countries religious organisations are exempted from value-added tax (VAT) on the acquisition and construction of real estate or on services and supplies of goods, provided their services are open to the public (Belgium: VAT Code, Art. 44, § 2, 11°; France: General Tax Code, Art. 261). In most cases, a distinction is made between commercial and non-profit activities, as well as between religious or non-religious purposes. In the UK, for example, construction or acquisition of a building is zero-rated for VAT if it is wholly used for charitable purposes (HM Revenue & Customs 2008). In practice, a de minimis non-charitable use of 5% of the building’s floor area is permitted. The rule sounds simple, but is not always easy to implement since it may not always be possible to distinguish between commercial and non-profit or religious activities carried out within the same building.

Similar exemptions apply to property tax, provided the buildings are used for public worship (Belgium: Income Tax Code, Art. 12; England and Wales: Places of Worship Registration Act 1855; France: General Tax Code, Art. 1382-4e and Art. 1407). In case of mixed commercial and non-profit use of one and the same building, the exemption applies on the basis of evidence that the space is used mainly for public worship (at least 70% in the Netherlands: Hoge Raad [Supreme Court of the Netherlands]), 4 December 1991); in other cases the exemption is strictly limited to the percentage of the property in which non-commercial activities take place (IMU, Imposta municipale unica in Italy). In Spain, tax benefits are granted to the main denomination and to historical religious denominations (recognised, registered or having an agreement with the State). Therefore, tax exemption is available to the Catholic Church but also, albeit to a lesser extent, to religious denominations that have cooperation agreements with the State. Respondents interviewed on the issue were asking for an extension of the tax exemption to all registered religions (Registro de entidades religiosas), arguing that, if benefits are granted on the grounds of religious aims and public benefit, it does not make sense to limit them to denominations with cooperation agreements with the State. Alignment of religious organisations with other types of organisations should make it possible to extend the tax benefits to other denominations as well, provided they comply with minimum standards that apply equally to all.

The RELIGARE data regarding allocation of tax incentives show a certain convergence across the different national legal systems; this applies to the requirement that the activity be non-profit in nature, as well as to more specific admissibility conditions such as numeric size of the beneficiary groups, reference to promotion of human rights, presence or absence of an explicit reference to religious activities, financial transparency, etc. A presumption of public benefit for religions is included in some domestic legislation, by explicitly including certain religious organisations among tax-exempt Public Benefit Organisations (PBOs) (Bulgaria, Denmark, France, Italy, Spain) or treating “advancement of religion” as a charitable purpose (Germany, UK).

A significant - more recent - trend throughout Europe is the demand that minority groups be better supported financially. A recurrent claim is that tax benefits for groups or organisations should not depend on their legal status, but should instead be allocated on the basis of their public benefit purpose, a non-discriminatory criterion ensuring (inclusive) State neutrality, equality and transparency. Equal treatment should apply to groups and communities, irrespective of their doctrine, historical role or legal status. Public authorities (tax administration, state body, independent commission, etc.) should not be allowed to require more than the fulfilment of minimum conditions, in particular by checking an entity’s compliance with public order, a criterion that applies indistinctively to every type of organisation.

Tax exemptions may also serve another aim, namely to encourage taxpayers to make donations to groups and communities of their own choice. This type of funding therefore seems to be consistent with the principle of freedom of religion, as well as with the freedom not to identify with any religion: it is based on social consent and to a certain extent also on social utility.

At present, Member States retain a considerable margin of appreciation in granting tax exemptions. For the sake of ‘inclusive State neutrality’ and also of ‘even-handed’ distribution of the benefits, it should be a requirement that public authorities in each country disclose clear and reliable information on their assessment criteria when granting exemptions to groups and/or associations. The EU plays only the subsidiary role of tax coordination (and harmonisation for VAT). The European Commission checks that taxation (or non–taxation) does not create an obstacle to the free movement of goods (Art. 28 TFEU), free movement of capital (Art. 63 TFEU) and the free supply of services (Art. 56 TFEU) within the internal market. If the Commission detects a failure to comply with EU law, it may initiate an infringement procedure.

In future, it would no doubt benefit minority groups in particular if the EU could provide the Member States – proactively – with a (common) set of assessment criteria that could ensure a better coordination of the conditions placed on tax benefits and, by so doing, promote non-discriminatory tax treatment across all Member States.

2.4.1.2 Other methods: the funding of the training of community/group leaders; the protection of cultural heritage

The RELIGARE project also shows evidence of alternative models of public support to religions and beliefs. One notices in particular the growing importance of the technique of project financing. The funding of the training of religious and philosophical leaders, and programmes for the protection of cultural heritage are two such forms of project financing.

The training of religious leaders

Religious leaders can be a driving force in helping, in particular minority groups or communities, to strengthen their position in society and to express their claims whenever they feel it necessary. As spokespersons for the community, they are usually also in contact with State officials and/or members of other communities and intermediary groups. However, empirical data show that leaders are often not (yet) sufficiently acquainted with the institutional landscape of the country, let alone with its legislation. It often happens that, for lack of adequate training, they cannot offer the appropriate support to their group or community.

In most European countries the training of religious leaders has established itself as a historical feature of the status of religious denominations and, consequently, of their funding. In practice, ministers of religion from “established” or “recognised” religions (or mainstream, as is the case for Sunni Islam in Turkey) are either trained at public universities (Germany, Denmark, United Kingdom) and university seminaries (Turkey, Bulgaria) or at private universities and university seminaries (France, Belgium, Spain, Italy, Netherlands). Faculties of theology in public universities, for example, are funded at least partly by public authorities. Religious groups more recently established in Europe cannot offer equivalent training facilities for their religious leaders, with the exception of Germany, where Islamic theology has now become part of the university system and is funded by the State authorities. In the Netherlands a similar development appears to be in process.

Support from the Member States for setting up a suitable system of training for religious leaders, ensuring that minority groups are offered equivalent facilities to those offered to other religious and philosophical groups, would constitute a form of support that could be perfectly in harmony with the principles of ‘inclusive State neutrality’ and ‘justice as even-handedness’.

State support for the protection of cultural and religious heritage

An important form of project financing is the protection of religious and cultural heritage. Although to a varying degree, in all 10 countries under scrutiny, public interest in heritage and its maintenance is considerable, and it has grown over the years. With some noticeable but very limited exceptions, one can observe many common points in the way European countries manage and preserve religious heritage. Among these commonalities figures the financial support given to religious groups and communities for the maintenance of historic buildings. This support is quite systematically and narrowly linked to the cost of such maintenance. The logic applies particularly to religious buildings, especially places of worship that are seen as counting among the most valuable parts of common cultural heritage. Whether located in the countryside or major cities, religious buildings play a central part in the debate of how best to preserve heritage.

A challenge that should not be underestimated, however, has been the steady growth of mass tourism. This has given rise to new situations: religious heritage buildings have in a growing number of cases are considered to be first and foremost a cultural heritage, separated from their traditional function as places of worship. This gradual, but nonetheless radical, change in both the perceptions and the actual use of religious heritage buildings changes the situation: does the subsidy given for the upkeep of a building or place of worship benefit primarily the religious group that claims it, or the visitors who are interested mainly in the cultural significance of a place whether or not this is linked to religion?

The present situation is generally speaking the following: religious heritage can benefit from direct State support, but only as a part of cultural heritage and not on religious grounds.. The main criterion for allotting funds is the condition that the property be an open place of worship. It is the result of religious heritage now being conceived of as part of national history (Belgium, France, the Netherlands, some buildings in Germany, where churches are state-owned, and Italy – Fondo edifici di culto). In the Netherlands, the legislation accepts that although religious buildings are privately owned by religious groups, some of these buildings, as listed monuments, are entitled to benefit from regular direct public grants. The same is true in Bulgaria and Spain. In the case of Bulgaria, additional funds can be provided by the Council of Ministers of Religious Denominations Directorate

2.4.1.3 State support to religion and belief in the media

Another particular form of State support that deserves to be mentioned here is State support to religion and belief in the media. The State has a specific responsibility in plural(ising) societies. Its task “is not to remove the cause of tension by eliminating pluralism, but to ensure that the competing groups tolerate each other”, where the position of minorities needs special attention. Public media are an important instrument for the State to protect, advance or even guarantee this pluralism.

Generally speaking, the national broadcasting regimes in place in several Member States offer airtime also to religious groups, in one way or another. This tradition goes back to the origins of the broadcasting regimes. Situations vary across Europe, but in general access to public broadcast/television is made possible for a plurality of religions in the majority of the 10 countries researched. Such arrangements do not exist, however, in Bulgaria and Turkey. The outcome for the Muslim communities is very different in the two countries.

Table: see attached pdf

Guaranteeing access to radio and television channels does not however necessarily go hand in hand with full State funding of the programmes broadcast. As things stand, public channels have very limited resources for offering airtime to religious groups. In Belgium for example, the national legislation provides for extremely poor airtime and budgets. The Netherlands will soon change its policy on religious broadcasting, by ending access and funding in 2016. This radical change has its origins in current policies of financial restraints.

Strikingly, State support to religious broadcasting does not match with the traditional systems of State support to religion: countries without regular church financing such as France and the Netherlands for example, have nevertheless set up a generous policy of airtime for religious broadcasting. Possibly, this may have to do with the fact that broadcasting is a technology that developed well after the establishment of the historically grown State-religion(s) regimes.

The growing variety of groups that claim access to airtime raises questions of selection of the groups eligible for a broadcasting licence. In their policies of distribution of airtime (i.e. setting the criteria for access to and the different allocations of airtime), the State neutrality principle requires that authorities refrain from being partial to certain religions. The criteria developed for the allocation of airtime have raised questions on State neutrality in different countries (Belgium, the Netherlands, Germany ), as a consequence of the fact that the State tends to show more “interventionist” when the number of competing candidates for airtime is growing.

The growing number of State-funded religious broadcasters also gives rise to problems of accountability and transparency (the Netherlands ). As a result of their internal diversity (and disunity), Dutch Muslims for example, lost their share of Dutch radio and TV airtime in 2011, and only got it back in 2013. In Bulgaria, the Turkish Muslim minority watches the Turkish media given the absence of regular Muslim religious broadcasting on Bulgarian public television channels. Non-religious communities (especially the humanist movement) have entered the radio and television scene in Belgium (since 1932 (radio) and 1951 (TV)) and the Netherlands (since 1926 (radio) and 1971 (TV)). In other countries (France, Germany, United Kingdom ), non-religious organisations have not yet succeeded in gaining access to broadcasting time on a comparable basis on nationwide channels.

The available data on the broadcasting regimes in the 10 countries researched for RELIGARE show that, over time, the evolution of State policies follows the same pattern, starting with regimes that only accept the presence of a traditional majority (State) religion and ending with a more inclusive broadcasting regime, mirroring in some way the new religious and philosophical diversity in society. The research data for the “type A” countries show that minority religions attach considerable importance to being present in the mass media, not represented by others, but gaining access to public channels themselves (Bulgaria and Turkey ). The media regulator in Bulgaria received complaints from non-registered religious minority communities with reference to the presentation of their activities in broadcasts by TV operators. The research data for “type D” categories point to the growing organisational difficulties arising from an increasing religious ‘external pluralism’ on the public channels (the Netherlands).

2.4.2 Policy recommendations for State support to religion and belief

In assessing techniques for the public funding of religious (and exceptionally also secular groups, as is the case in Belgium and in the Netherlands), the RELIGARE project focused on activities that are specifically religious or philosophical in nature; it left aside the funding of more peripheral forms of social action. One major finding resulting from this assessment is the increased importance given to (the requirement of) greater transparency and objectivity of public funding techniques, rather than calling into question the principle of such funding itself. Applying inclusive even-handedness in a procedural approach, would mean that all stakeholders take part in the discussion on whether and how State support should be provided and which participatory techniques are best adapted for facilitating the distribution of this support.

Given that the EU authorities lack competence in this area, over the short term substantial differences among the policies of the Member States will necessarily subsist. A lack of competence at EU level does not mean, however, that EU authorities, who are themselves bound to respect the European Charter on Fundamental Rights, do not bear responsibility for the possible effects of EU policies on the status of religious and philosophical groups (or their individual followers) in Member States. The EU authorities are not empty-handed: the Commission provided itself with an instrument that enables the EU policy-makers to evaluate the human rights aspects of new proposed legislation at an early stage: the technique of Impact Assessment (see below, IV.: The potential impact of the RELIGARE research project). These Impact Assessments, which also take into account the possible impact of policy proposals on human rights (the catalogue, formulated in the Charter), are relevant for the position of the freedom of religion or belief in its collective dimension too.

(1) The EU Institutions should include religion not only in its individual but also in its collective dimension (the human rights position of religious and non-religious communities of conviction) systematically in the Integrated Impact Assessments as a structural component in the policy-making process.
(2) Regarding tax policies: any policy that is being put in place should promote the alignment of churches, religions and belief groups with organisations acting for public benefit (PBOs), irrespective of their legal status within the State (recognised, registered, etc.)
(3) Tax deductions for donations, bequests and inheritance in favour of public benefit organisations (PBO), including religious organisations, may be a means to encourage the funding of religions by their members and followers: (a) by increasing the rate of deductibility; (b) by subtracting deductions from taxable income, rather than from actual tax due; (c) by raising the limit of the deduction and d) by lowering the threshold for deductibility, in order to open up the tax facilities to lower income groups as well.
(4) The EU promotes projects that benefit society at large. With regard to the status of PBOs and the European tax framework for PBOs the EU could play a more pro-active role in determining a common concept of public benefit with a view to achieve a better coordination of the national systems in place.
(5) Within its competence of tax coordination, the EU should further promote transparency and accountability of the groups receiving support, as a way of combating any indirect discrimination of further developing a common tax policy.
(6) In the area of religious / cultural heritage, the EU could play a constructive role by supporting programmes that make it possible to critically assess the difficulties that come with the interminglement of religious and cultural heritage, and in particular in the case of reassigning religious heritage buildings for other uses.
(7) More specific means that are to be taken into consideration if a system of state supported religious broadcasting is maintained (external pluralism):

- When public radio or television channels are operated by (or under the responsibility of) the State, an equitable and transparent procedure should be allowed for providing various religious or belief groups with access to these public channels
- The position of minorities should not be neglected, nor should the organisational specificities of some minority religions in particular (as is the case for Islam).
- Even where religious communities or groups have no access to licences for their own radio or television on public channels, an access to other radio and television programmes (for instance news programmes and documentaries) should be made possible in order to reflect diversity of religions and beliefs in public dialogue, on various subjects that matter in society, including ‘religion’ itself.
- Religious and belief groups should not be particularly disadvantaged from owning or operating radio or television facilities (on national or local level). States upholding a ban on religious channels should reconsider their position [reassess the necessity and effectiveness of a ban] in the light of the growing religious diversity.

Potential Impact:

Introduction

The RELIGARE project focused on tensions between the application of non-discrimination norms and the fundamental right to freedom of religion and belief. It shows that tensions arise in various areas of social life which may in the end escalate into intricate legal conflicts. Some of these clashes have gained considerable public attention, such as the debate about the wearing of the Islamic headscarf in French public schools and in private workplaces, the discussions about the display of crucifixes in public schools in Italy and, more recently, the ritual circumcision debate in Germany. In cases of ‘clashing rights’, balances must inevitably be struck, however imperfect these may be, and any constraints must be justified. What role(s) should the EU institutions play in this area? In what matters can they be expected to take the initiative in the years to come?

This report concludes with the question: what impact could the RELIGARE report be expected to have? With that aim, this section advances a number of concrete recommendations to the European Institutions for developing a coherent legal and policy framework that addresses issues of freedom of religion and belief. It identifies no less than three types of actions that would enable the EU to become more deeply engaged in the combat of religious discrimination and in the guarantee of religious freedoms broadly understood. The proposals are based mainly on the findings of the RELIGARE research project, but do not of course exclude the possibility that in the years to come, the EU would take yet other initiatives or become more closely involved in the struggle for equal treatment of all persons in Europe, taking account of religion or belief.

The first type of action consists of legislative instruments: the EU has already developed several such instruments - some more constraining than others -, which are intended to render more effective the fight against certain forms of exclusion or discrimination on grounds relating to religion and/or belief. To date Directive 2000/78/EC is one of the principal such instruments adopted by the European Union to combat discrimination in the employment sector. A new directive, known as the “horizontal directive”, has since been proposed. A second type of initiative involves the various participatory procedures (encounters, dialogues and reflection on the future) – whether already in place or yet to be set up – among the European Institutions and various groups and/or persons concerned directly or indirectly by discrimination on grounds of religion and/or belief and, more generally, by diversity policies. Article 17(3) TFEU reflects this sort of deliberative participation. Finally, a third type of action or initiative is the involvement of experts, either by commissioning reports on specific topics, as was the case with the RELIGARE project, or in the form of experts groups, with a consultative role, who are available to the EU institutions, when needed, to examine specific questions in greater depth.

From a policy perspective, the three types of initiatives are clearly different in nature – it is only those legislative instruments with direct effect that are binding on Member States. But taken together they would enable the EU to make a greater investment in elaborating a policy that on the one hand takes into account the sovereignty of States in managing religious and philosophical diversity in the internal legal order (the principle of subsidiarity), while on the other hand ensuring that this management is respectful of minimum standards of protection (fundamental principles) that are applicable across the entire territory of the EU. All three types of actions are necessary for setting up a policy that is inspired by the ideal of the greatest possible inclusivity, that is to say, where individuals are respected in their diversity, and seeks to benefit from the active and committed participation of all its members, taking account of their religion or belief, if that is their wish. The fight against discrimination is not sufficient in and of itself; it must be accompanied by participatory procedures that are both open and democratic.

The comparative examination undertaken by the RELIGARE project for 10 countries under scrutiny reveals that a European policy that is committed resolutely to an effective protection of religious freedom, in the broadest possible sense, both within the internal legal order of the Member States and at EU level, is not only desirable but necessary in view of the objective of conducting a European policy of the greatest possible inclusivity of all persons that takes into consideration the question of full respect for, and the effective protection of, the fundamental rights of individuals regardless of religion, faith or non-religious beliefs and ensures, whenever necessary, that the religious beliefs and traditions of individuals and communities are respected, while guaranteeing that a due balance is struck with the rights of others in accordance with the case law of the ECtHR.

3.1 Three types of action at EU level

3.1.1 The legislative route: On the proposed horizontal directive

Building on the EU’s existing framework of rights, and more particularly on Directive 2000/78/EC, the EU could decide to extend its non-discrimination (and integration) arsenal to issues pertaining to religion and belief outside the scope of employment by giving further support to the legislative procedure conducting to the adoption of the proposed Council Directive on Implementing Equal Treatment Between Persons Irrespective of Religion or Belief, Disability, Age or Sexual Orientation.

On the one hand, such an extension would align EU law with what has already been achieved in several Member States which, in their domestic anti-discrimination legislation, go further than Directive 2000/78/EC, albeit with some modifications. On the other hand, one must remain realistic: the text of the so-called horizontal directive has for several years now been the object of considerable political reticence; its adoption would not be a smooth path. Moreover, it is essential to study attentively the reasons given by those Member States which oppose the directive, in order to be able to take these into account without necessarily giving up the aim of having the debates reach a successful end and then seeing the adoption of a European legislative instrument that would considerably extend the scope of the struggle against unjustified exclusion. As stated above, in its Green Paper accompanying the proposed horizontal directive, the European Commission has expressed its wish to ‘complete’ the European non-discrimination framework.

Considering the above, and without in any way detracting from the advantages one might expect from an additional European legislative instrument that enlarges the scope of the fight against discrimination to include the entire field of services, it can be said that from the perspective of the principles of religious freedom and non-discrimination, the proposed horizontal directive is a missed opportunity. One may, for example, criticise the current version of the text for limiting the EU’s reservation of reasonable accommodation to one particular vulnerable group (individuals with disabilities) and not extending it to reasons of religion or belief. This may block the avenue of reasonable accommodation for reasons of religion or belief in the foreseeable future, by prioritising the protection of persons with disabilities. One wonders whether this normative choice is consistent with the EU’s ‘Europe 2020’ strategy for ‘smart, sustainable and inclusive growth’: good policies can add to sustainable and inclusive growth, and work towards an inclusive labour market that fully takes advantage of the skills and talents of its (ethno-religiously) diverse population, with due regard for fundamental rights in the workplace.

There are two possible attitudes in this regard: either one takes a pragmatic attitude that consists of maintaining the text in its present version, without adding to the protections it currently contains, in the hope of thereby maximising the chances of success of the difficult negotiations among Member States concerning an instrument that does not enjoy unanimous support. Or one tries to be more ambitious and takes advantage of the fact that the directive has been in the pipeline for a long time - the fate of the proposed horizontal directive at this time still remains unclear – to suggest amendments to the text that has been put forward.

Another possible critique, based on a detailed analysis of the text of the proposed directive, is that religious-ethos organisations may feel their freedom and autonomy is restricted by standards of non-discrimination, while individual members of religious minorities facing hostility will obtain additional protection against discrimination in important public spheres. The proposed horizontal directive, by expanding the right to non-discrimination on the basis of sexual orientation and of religion or belief (two highly relevant grounds) to the broad societal areas of education and goods and services, would indeed affect the position of commercial operators with strongly held beliefs, religious schools and religious-ethos companies offering goods and services available to the public. The proposed directive would not only give certain rights and remedies to individuals, but also implies obligations on the part of faith-based organisations and religious individuals in certain circumstances (e.g. hotel owners). Tensions could arise where religious groups, particularly religious minority groups, were no longer allowed to define autonomously the specific services they (want to) offer to particular individuals or groups. There may indeed be justification for allowing direct distinctions to be made by religious-ethos organisations or individuals offering services from the perspective of a consistent religious ideology, in light of the legitimate aim of freedom of religion. Such an arrangement may be one way to ensure improved consideration of the fundamental right to freedom of religion in the balancing of rights and interests that is inherent in formulating norms regarding non-discrimination.

These few criticisms in no way detract, as pointed out above, from the merit of a European legislative instrument that would extend the fight against unjustified exclusions - including discrimination on grounds of religion and/or belief - to the vast area of services. But one should not miss out on the opportunity also - from the outset - to examine in greater detail the weaknesses of the existing text, or at least the undesirable and/or unforeseen effects that risk emerging, as well as the new tensions to which it might give rise. This is all the more urgent since the directive has not yet been given final approval.

One of the highest priority recommendations put forward by the RELIGARE research partners is therefore the suggestion to undertake further study - on a comparative basis, taking into account the specificity of national situations (i.e. in domestic law) - of the various potential effects of the entry into force of the horizontal directive, including any deleterious (i.e. discriminatory) and/or unpredictable effects of such a legislative instrument.

Among the questions to be looked at in greater depth, there is the issue of what to do about ambivalent situations: what would be the implications of such an EU directive for cases that are on the borderline between religious and secular? If the horizontal directive comes into force, there will be stiff competition between autonomy of religion and the obligations that fall under the directive. Where will the boundary be drawn between the autonomy of faith-based organisations and the expectations of public/state support to combating discrimination based on sex and gender? Will it be as broadly drawn as in the US (US Supreme Court case law)? Should the directive also apply to religious education, or only to all the other aspects of schooling? Religious discrimination within the context of religious education offered by confessional schools is probably difficult to address, but as far as sex discrimination is concerned there should be no exemption. If the directive is approved, faith-based organisations providing a public service, such as adoption agencies, would also be under pressure to comply with non-discrimination rules, even if they are not State-funded; only in the realm of their own, directly religious/spiritual services could they continue to choose freely whom they appoint and to whom they offer services. And if they want to have charitable status, they will also have to comply with non-discrimination laws. All these issues, and so many more, would need to be further investigated.

3.1.2 The consultative process

A second type of action that would enable the EU institutions to give more weight to a resolutely inclusive European policy is that of consultation. Three forms of this process can be identified: the participatory process as enshrined in Article 17(3) TFEU; internal consultation among EU services and administrations; and, finally, the obligation to undertake an ‘impact assessment’ before introducing any new legislative or judicial measure that directly or indirectly affects religion in the broadest sense. The impact assessment indeed involves a form of consultative procedure. As indicated above, participatory procedures are all the more important given that they draw inspiration from a more constructive logic than anti-discriminatory measures. The struggle against discrimination does not suffice in and of itself - it must be accompanied by participatory procedures that are open and democratic.

3.1.2.1 On Article 17 of the Treaty on the Functioning of the European Union (TFEU)

According to Article 17(3) TFEU, the EU is bound to maintain an open, transparent and regular dialogue with churches, religious associations or communities and philosophical and non-confessional organisations that are recognised as such at national level in the Member States and adhere to European values. In principle, Article 17 is incumbent upon all EU institutions; in practice however, it is the European Commission, and more particularly the Bureau of European Policy Advisers (BEPA) that is being entrusted with maintaining such dialogue with interlocutors at various levels, whether in the form of written exchanges, meetings or of specific events. In a recent opinion the European Ombudsman, in response to a complaint in which mainly the Commission was targeted, attempted to clarify the meaning of the three adjectives ‘open’, transparent’ and ‘regular’. Which kind of dialogue should the EU institutions have with religious and civil society groups? The EU institutions have a certain margin of discretion in framing the dialogue with view to the implementation of Article 17, in principle however they are bound to invite to the dialogue all churches, religious associations or communities as well as philosophical and non-confessional organisations that are recognised or registered as such at national level and adhere to European values. One of the principal aims of the consultation made obligatory by the procedural provision of Article 17 is to guarantee regular and direct contact - at various levels and in different ways - between the highest European political institutions and representatives of religious communities and various communities of conviction - also non-confessional - throughout Europe. In the long run, this makes Article 17(3) TFEU a potential tool of a truly participatory democracy at a very high level. In its own way, it reiterates the principle of participatory democracy also contained in Article 11 (2) TUE.

The future will determine what will be impact of this principle, but its importance for issues that are in particular related to religion and belief should not be underestimated. On the one hand it enables EU institutions - at this stage BEPA more specifically - not only to be kept informed of the situation and aspirations of various interlocutors on matters that are of particular relevance to them and exchange views with them. On the other, and this what the authors of this RELIGARE project report would very much support, it enables EU institutions and the Commission in particular to take certain initiatives which, without infringing on the provisions of paragraphs 1 and 2 of the said Article 17 TFEU (which confirm the principle of sovereignty of the Member States in all areas concerning their relations with churches and non-confessional organisations under their jurisdiction), would guarantee a more consistent implementation of EU law in areas that relate to the protection of the freedom of religion and belief taken in the broad sense. In his decision the European Ombudsman rightly reminded that the main objective of Article 17(3) TFEU is to allow the discussion of concrete legal questions and existing legislative provisions and that doing so is perfectly compatible with the provisions of paragraphs (1) and (2) of the same article: “Conducting a dialogue on an issue dealt with in existing legislation cannot but constitute constructive action” (Decision 2097/2011/RA against the European Commission; 48). The European Commission embodies and upholds the general interests of the Union and should be the driving force in the Union's institutional system, also in matters that relate to religion and belief whenever relevant within the framework of administering and implementing Community policies and of enforcing Community law (jointly with the Court of Justice).

The RELIGARE project fully supports the Ombudsman’s suggestion, articulated in the above-mentioned decision, in which he “(…) urges the Commission (…) to develop the understanding of how Article 17 TFEU should be interpreted and applied (…) and, if necessary, to draw up guidelines in terms of how exactly it plans to implement Article 17 TFEU” (ibid., 63).

3.1.2.2 Consultation among various services (Directorates-General) and administrations

Without waiting for the Commission to align itself with the position of the European Ombudsman in its dialogue with civil society on questions relating to freedom of religion and belief, it may be useful here to stress the urgency of another type of consultative procedure that could be set up at European level. The RELIGARE research partners have observed that there is a lack of coordination internally among the different services that, within the European Institutions, are called upon to consider questions of religious discrimination. A suggestion that thus emerges from the RELIGARE project after three years of research on EU policy in ensuring respect for freedom of religion and belief is that a permanent consultative mechanism be put in place among the services, in order to enable the teams to listen to each other’s views on certain priorities and, as far as possible, to undertake joint actions. One possibility would be to activate an inter-service group on religion and belied that would facilitate the systematic and regular exchange of information among Directorates-general at the Commission’s level and coordinate the Commission services working in the area. This would enable the putting into place of a policy of internal consultation on religious questions and matters relating to the diversity of beliefs any time these require action on the part of the EU. We give a few examples of such situations below, based on the findings of the RELIGARE research, which in our view would merit being discussed internally and in a transversal fashion within the EU institutions so as to facilitate better coordinated action.

The RELIGARE findings suggest that there are a number of issues which require forward-thinking in terms of the current (and future) BEPA Action Programme(s):

With regard to the workplace: the EU has competence to intervene in the way Member States, within their internal legal order, balance claims pertaining to the protection of the freedom of religion and belief and the requirement of non-discrimination, including the public acceptance of the expression of certain identities. A potential expansion of the equality principles contained in EU Directive 2000/78/EC (Employment Equality Directive) – i.e. combating discrimination on the basis of religion or belief – to other fields of social life as well, e.g. the provision of services (the so-called ‘horizontal directive’ project), would necessarily enhance the scope for such intervention and therefore create the need for optimal coordination among various Directorates-General.

With regard to family life: both historically and in the present, majority Christian traditions have had a considerable impact on family arrangements. The variations among the different legal practices of the EU Member States can adversely affect the freedom of movement and residence within the Union. Further coordination among the Member States is necessary to take account of this, but doing so is likely to put some pressure on interpretations of Article 17 (1) and (2) TFEU and so should be anticipated by the Commission.

With regard to the public space: two aspects in particular have a significant impact on areas of EU competence, directly or indirectly (through the application of the European Convention on Human Rights in EU law). First, recent (national) court decisions have driven Member States further apart in their views of the place of religious expression in the public space; this applies particularly to the bans on certain kinds of religiously motivated dress. Second, in a number of Member States the application of equal treatment requirements is being extended to faith-based organisations. These are two sources of evidence of the fact that, within the EU, the concept of ‘public order’ is given divergent meanings in the different Member States. Here again, expectations of equality and freedom of movement potentially clash with the protection of the freedom of religion and belief enjoyed by individuals and communities within Member States, and therefore put Article 17 (1) and (2) TFEU under pressure.

In sum, the RELIGARE project has identified a significant and growing tension between basic human rights, especially respect for private and family life (Article 8 ECHR; Article 7 EU Charter), freedom of religion and belief (Article 9 ECHR; Article 10 EU Charter), and prohibition against discrimination (Article 14 ECHR; Articles 20 & 21 EU Charter). These tensions intersect with tensions between the programs of various EU Commission directorates, especially Education, Employment, and Home Affairs and with the principles of reasonable accommodation in connection with religion.

A mechanism of continuous and transversal reflection and mutual consultation at the level of the Directorates-General would probably be the most suitable place for coordination to take place, in terms of the internal and external networking on which the current (and future) Action Program focuses. This may require that in the longer term questions of the interaction between religion and belief, the wider secularity of the European space, and the programmes of the various relevant Commission Directorates-General be addressed more firmly after consultation between these programmes and in line with the future work of the Bureau of European Policy Advisors (BEPA). Moreover, a more active role could be envisaged for the Fundamental Rights Agency regarding religious affairs.

3.1.2.3 Impact assessment and the compatibility with fundamental rights

The third and last form of action drawing inspiration from the consultative process that we bring here into the discussion of European policy on religious matters - taken in the widest sense - is that of the impact assessment, and more in particular the stakeholder consultation. Regular dialogue (Art. 17 TFEU) and stakeholder consultation (Impact Assessment) are interrelated issues, fitting in well with the tradition of the Commission’s Culture of Consultation and Dialogue (2002).

In 2010 the European Commission developed its Charter Strategy, outlining the way in which the Commission plans to implement the EU Charter of Fundamental Rights. The Operational Guidance on Fundamental Rights (2011), one of the outcomes of the EC Charter Strategy, is a specific instrument for EC staff preparing impact assessments, providing guidance on how EU legislative proposals may have impact on fundamental rights (and, when policy options raise fundamental rights issues, DG Justice staff will be invited to the Impact Assessment Steering Group).

Impact assessments are an important tool for demonstrating that in the development of new EU legislation, full account is taken of the human rights protected by the EU Charter. For this purpose a Fundamental Rights Check-List has been introduced. The EC decided that human rights issues should be addressed in the early-stage preparatory consultations. According to the procedure, stakeholders should give their input during these consultations. As a result, EU policy options that would violate fundamental rights should be discarded. This also applies to the compatibility with respect for the protection of the freedom of religion and belief enshrined in Article 9 ECHR and in Article 10 of the EU Charter. Yet, the concern should also be that where EU institutions are acting within the confines of EU competencies, EU policymakers take account of existing domestic (national) arrangements for relations between the State and religion (ranging from regimes offering full direct State support to religious groups and communities of conviction to the regimes that maintain only indirect or hardly any State support). EU policies (and legislation) cannot ignore the commitments and obligations which Member States historically have taken upon them with regard to the freedom of religion and belief, but also the freedom of association, and when relevant also property rights.

After three years of research on the commitments that come with the obligation both for Members States and the EU institutions to provide appropriate protection to individuals and communities claiming their freedom of religion and belief, the RELIGARE project suggests three Impact Assessment procedures: (1) one with a view to involve stakeholders in EU policy preparation (consultation and advice) on topics that are relevant for religious and non-confessional communities of conviction; (2) a second on the position of religion and belief in its collective dimension; and finally (3) one on EU law and its transposition/implementation by Member States ( EU Guidelines for national governments).

Involving stakeholders in EU policy preparation (consultation - advice) on topics that are relevant to them.

This would require to organise, in an early stage of the EU legislative process, transparent consultations with all relevant stakeholders, including representatives of religious and philosophical organisations, with a view to inform them in timely fashion and assess, with their help, (un)intended consequences of the new legislative measures/instruments that are in the pipeline. Stakeholder consultation should include smaller and/or less well known groups, and not just traditional churches and the humanist movement: it will offer a more complete picture of the possible effects of new EU policy measures on groups of stakeholders - large and small. The right to equal treatment and the principle of non-discrimination (as well as the principle of justice as even-handedness) are at stake here.

Impact assessment for the position of religion and belief in its collective dimension

Impact assessment is an instrument that is not completely new in the context of state legislation on religion. In 2004, the Venice Commission produced a series of concise general guidelines for national authorities engaging in new legislation affecting religion. The Venice Commission is known for being a prominent external advisor to national authorities in matters of minority protection. In the same vein, Integrated Impact Assessments, as a structural component in the EU political process should also address matters related to freedom of religion and belief in its collective dimension (the human rights position of churches and non-confessional communities of conviction). It could help prevent unintended side effects of new EU policies or legislation, such as producing new or reinforcing existing tensions between groups and communities within the national context of some Member States in particular.

Impact assessment on EU law and its implementation/transposition by Member States - EU Guidelines for national governments

Member States are responsible for the implementation/transposition of EU legislative instruments into their internal legal order. To help Members States ensure that the freedom of individuals and communities defined by religion and belief is respected and complies with the European Human Rights requirements, the EU (more specifically, the EU Human Rights Agency) could develop ‘Guidelines’ suggesting the introduction of similar techniques (consultation, impact assessment) to the national authorities as far as the implementation of new EU legislation is concerned.

These guidelines could be: (a) general guidelines, (b) guidelines developed case by case (thus focusing on the subject concerned), or (c) guidelines that take into account the different existing types of constitutional models with regard to the relationships between the State and religions.

3.1.3 The use of expertise

3.1.3.1 Individual experts for examining particular questions

Finally, a third type of action or initiative that would make it possible for the EU institutions to strengthen their policies regarding the various issues identified by the RELIGARE research is the use of expertise, whether in the form of studies commissioned on chosen subjects as was the case with the RELIGARE project, or through setting up expert groups at the service of the EU institutions to help examine particular questions in greater depth and that would in a sense have a consultative competence. The Directorate-General Justice, for instance, invites experts who wish to provide their specific expertise on certain topics to the EU institutions to express their interest. Such a database of independent experts allows DG Justice to make contracts with experts for occasional tasks (drafting of brief reports, assistance with their events, etc.) when needed. With regard to, in particular, issues of religious discrimination throughout the EU the RELIGARE project recommends that the Commission set up a permanent working group of experts: given the diversity of the issues involved and the complexity of the various commitments and obligations of Member States to guarantee that a due balance is struck between, on the one hand, the freedom of religion and belief of individuals and communities and, on the other, the rights of others to be free from discrimination, involvement of several areas of expertise is required. Such permanent body of experts may prove particularly useful, if not indispensable, over the medium term in order to move beyond the inadequacies of ad hoc policy and help the Commission work out adequate legislative instruments to balance between freedom of religion or belief and other discrimination grounds.

3.1.3.2 Expertise that draws from the RELIGARE project. Three research instruments: a legal database, thematic and country templates, sociological country reports

The RELIGARE project has yielded three particular research instruments: (1) a database of relevant case law for the 10 countries involved in the research; (2) a series of thematic templates that summarise relevant legislation, court cases and controversies in the various countries (these reflect, in synthetic fashion, the arguments made both by the legislative branch and by the courts and tribunals, in order to address a number of specific situations); (3) sociological reports drawn up for the six countries where interviews were conducted.

These three instruments, each in its own way and in combination, allow for a systematic investigation of the existing legal frameworks within the different Member States combined with a sociological study. They make possible a critical assessment of national practices and policies, their divergences and convergences – some of the (national) legal frameworks in force are outdated – and a measurement of their impact on the key elements of EU policy for combating religious discrimination. The instruments, in combination, also show that there is no real consensus or clarity on what is or should be the core of religion or belief. A number of paradigms inherited from the past regarding (majority) religion remain unquestioned in the case law; therefore other perceptions of religion and its role in society, as identified in the sociological data and shown in the templates, are side-lined or ignored. While it is generally accepted that religion and belief are fairly well protected throughout the EU, the database as well as the templates show that religious discriminations are still a serious problem. Moreover, the variation among national policies also impacts at EU level: some national policies in the area of religion and belief indeed place hurdles to the free movement of persons and services. It is to be expected that in the years to come not only the ECtHR but also the ECJ will be requested to play a greater role in the area of protection of the freedom of religion and belief by expanding the applicability of EU fundamental rights standards to certain types of action taken by the Member States, with a view to invalidate Member State action where it falls short of the level of protection afforded to fundamental rights under EU law. The sociological interviews for their part show where there is need for clearer and more explicit justification on the part of legislators and why in recent years some actions in particular have given cause for critical comments.

The RELIGARE research assessed four areas (employment, family issues, public space and State support) and nine EU Member States, plus Turkey. The templates, in combination with the database and the sociological reports, offer a threefold research tool that makes it possible to demonstrate for the countries and the topics under study the discrepancies and lacunae in various policies on freedom of religion and its protection. Together, the data thus collected and presented schematically give a relatively complete and up-to-date image of what has already been done both at the EU level and by the Member States to combat religious discrimination and indicate where the priorities should be in the years to come - in which sector of social life in particular - in order to give religion and belief appropriate protection without adding to existing tensions that are clearly evident from the sociological investigations carried out.

The three research instruments created by RELIGARE were conceived with a view to permitting comparison, as systematically as possible, and thus to enabling decision-makers (EU and national level) and those seeking to achieve a balance between protection for religious freedom and other fundamental rights to see how questions that are present in every Member State, albeit to varying degrees, are handled in the respective domestic legal orders. Ideally, a similar approach, with detailed templates, should be followed for the 27 Member States and should cover all the topics affected by religious discrimination.

The templates show that there is no one ‘right way’ to derive policy solutions from the principle of ‘inclusive’ neutrality. The justification of the policy choice made by a government or by a judge need to be explained by reference to context-related arguments (see 17(1)(2) TFEU). The templates, database and sociological reports indicate the conditions under which taking a contextual approach indeed permits one to achieve (inclusive) neutrality; the RELIGARE research shows that what guarantees appropriate protection is not so much justice in the sense of a ‘hands-off’ stance but as seeking ‘even-handedness’ between competing interpretations of the freedom of religion. To have the law on one’s side is not enough, people also seek justice.

3.2 The wider policy implications of the RELIGARE findings

The templates as they were elaborated and filled in by the RELIGARE research teams represent no doubt a major research instrument for assessing the extent of the protection granted to freedom of religion in EU Member States. Their relevance goes beyond the research project, each templates places similar themes in the different national orders and shows how either legislation or case law handles them, both national and international courts. The latter is all the more relevant since all EU Member States are bound by the ECtHR case law, and in future will be increasingly bound by that of the ECJ as well. The case law of the two courts is itself far from consistent. “The case law of the Strasbourg courts that have direct relevance to the current issue of religious pluralism is but fragmentary and casuistic”, the Dutch human rights expert Titia Loenen wrote a few years ago. The way in which the ECtHR has viewed the protection of religious freedom to date is indeed difficult to summarise in a few sentences, precisely because of its fragmentary nature. The templates help indicate how the domestic courts and public authorities in the Member States cope with this fragmentation. One trend that appears from both national and ECtHR case law, is the differentiation between two aspects of religious freedom: the so-called forum internum (internal forum) that refers to adherence to a religious or other belief, and the forum externum (external forum) that refers to the various ways in which one can give expression to that conviction. The difference is important for a good understanding of the scope of the protection that the law has thus far given to religious freedom. The templates illustrate this clearly. Courts in general take a closer look when the forum internum risks coming under threat. Where, by contrast, it is a matter of the manifestations of a religious belief (forum externum), the ECtHR in particular is more inclined to trust the contracting States and to allow them to determine for themselves when and how their policy should impose concrete limitations on freedom of religion and belief, and how lenient they should be towards the protection of certain practices dictated by religious or other philosophical beliefs. In the case law of the ECtHR regarding the wearing of the Islamic headscarf, this tendency is clearly present. The ECtHR states in this regard: “Where questions concerning the relationship between State and religions are at stake, on which opinion in a democratic society may reasonably differ widely, the role of the national decision-making body must be given special importance.”

On the scope of the margin of appreciation that the ECtHR allows to the States, one particular case recently has been the cause of vivid debate among legal scholars: the Lautsi case. On 3 November 2009, a Chamber of the Second Section of the ECtHR ruled that Italy may no longer display crucifixes in public schools. The fallout from this ruling was unprecedented. As a reaction to the judgment, two fronts, so to speak, emerged. On the one hand, the critics: not only in Italy, but in the rest of Europe as well, the decision of the Court was subject to sharp criticism. On the other hand, a number of authors agreed with the judgment and saw it as a step in the right direction. Italy went on to request - and obtained - a referral of the case to the Grand Chamber. In its second judgment, of 18 March 2011, the ECtHR ruled that Italy was justified in making it mandatory to display crucifixes in public school classrooms. This decision, too, elicited mixed reactions. The judgment is based on the principle of the margin of appreciation. It leaves the contracting States extensive freedom to determine how they wish to deal with religious and philosophical pluralism within their internal legal order, given their own specific political, economic, social and cultural situation and in the light of the evaluation of the situations on the ground.”

The Lautsi case and, more recently the Eweida case, are just two examples, but the templates show that in various Member States, similar issues are at stake, where the question is whether all the conditions have been met for an even-handed balancing in the event of a conflict played out between the majority group within society and a minority that opposes mainstream opinion. ‘Local’, context-oriented arguments also play a role in determining how (concretely) the principle of inclusive neutrality may be applied, and the demands of toleration requires that one takes into consideration not only the interests of the majority but also the (justified) demands of all those concerned. If public authorities (including the courts) are reticent in accurately balancing, in an even-handed way, the interests at stake, the weaker parties risk drawing the short end of the stick. The right of minorities - whether religious or other - to religious freedom comes under pressure if the majority in a diverse society puts the imprint of its own conception of the “good life” upon legislation. Freedom of religion and belief is also and should remain, if not primarily, a fundamental right of minorities.

It is the latter concern that brought the RELIGARE project to advance the two yardsticks of ‘inclusive State neutrality’ and ‘justice as even-handedness’ at this (still) early stage of EU policy making in the search for the right balance between the requirements of non-discrimination and equal treatment and the protection of freedom of religion and belief taken in the broad sense. The standards set by ‘inclusive State neutrality’ and ‘justice as even-handedness’, in combination, confirm that the national margin of appreciation is not unlimited - it is mainly about due process and contextualised argumentation. It goes hand in hand with a European oversight both of the laws themselves and of their implementation. That is, when the ECtHR investigates whether a national measure is justified in principle, it also looks at whether the measure is proportionate to the objective sought. In a sense, the two yardsticks of ‘inclusive State neutrality’ and ‘justice as even-handedness’ underscore the fact that national governments (legislators as well as courts) have increased responsibility to ensure that, within the framework of their specific constitutional context, there is a careful and responsible weighing up of whether or not to allow and/or restrict specific forms of manifestation of religious or other beliefs. ‘Inclusive State neutrality’ and ‘justice as even-handedness’ do not stand in the way of the relationship between public authorities and religions and beliefs being seen from the perspective of subsidiarity. Furthermore, it does not necessarily mean that stricter demands as regards compliance with constitutional principles related to the freedom of thought, expression may not be imposed.

Legislators should also systematically, as far as possible, take an approach that opts for inclusive even-handedness. Failing which, any restriction imposed by the law on freedom of religion is likely to be seen by some as discriminatory and to nourish negative feelings either towards public authorities or others groups and communities. What should not happen is that in the EU, a significant number of persons see themselves as increasingly subject to restrictions that they consider unjustified and that, in their view, constitute discrimination against them, which is not a trustworthy blueprint on which to build when rethinking the interface of religion and secularism in the Europe of the future.

List of Websites:

www.religareproject.eu

final1-religare-final-publishable-report-nov-2013-word-version.pdf