Introduction

Life can be truly surprising. As recently as a few years ago, we might have thought that the debates on the future developments concerning freedom of religion or belief in Europe would focus on issues such as the relationship between populism and religion, the ever-growing scope of conscientious objection, or the accommodation of Islam in the European legal space. No one could have foreseen that in 2020, the main challenge to religious freedom would be governmental restrictions to collective worship, and that European churches would only be able to welcome a handful of those who were lucky enough to join the service before the maximum number of participants was reached.

At the onset of the COVID-19 pandemic, the need to curve the spread of the novel coronavirus became everyone’s overriding focus, and the measures which cut people off from their social life did not spare religious freedom. In fact, this freedom became one of their first collateral victim of the pandemic, since churches were commonly identified as hotbeds for infection. The outbreak of the pandemic brought a radical change to the exercise of religious worship and suddenly made a scarce commodity out of what had long been taken for granted.

The chapter addresses restrictions on religious worship. It is rooted in the database collected under the EXCEPTIUS project, but not limited to it, for raw data do not give a full picture of the situation. Freedom of religion or belief is indeed a very special case: any of the imposed restrictions face barriers and challenges of their own kind, and the effectiveness of the law enforcement measures depends on a number of highly specific factors.

In this respect, we believe that the message from the times of the pandemic, although it has received considerable scholarly attention in recent times, has still not been sufficiently understood and processed in academic terms. Many studies on the response to COVID-19 simply reflect the experiences of specific countries. The generalisability of this research is often problematic. Some authors address more general questions and offer considerations as to whether giving priority to public health over freedom of religion or belief is always correct. If detached from the actual numbers, such findings may be difficult to implement. And this is where the EXCEPTIUS dataset can be useful. In this chapter, we aspire to contribute to the extant literature by generalising examples from a number of jurisdictions, thus making it possible to discover patterns in the legal response to COVID-19 within the area under study. This broader picture also allows us to offer policy recommendations based on both pre-existing theoretical considerations and the actual experience of European countries.

The first section of this chapter presents the taxonomy of measures defined within the EXCEPTIUS project, whereas the second section addresses the level of policy responses across Europe, using this taxonomy. This will be followed by a preliminary assessment of the design of restrictions in the light of the requirements established by the European Convention of Human Rights (ECHR). In the next section, the implementation of measures is discussed, with the use of comparative data compiled in the project’s database. The penultimate section will outline the lessons learnt from the response of European countries to the pandemic and explore the potential for comparative research in this field. The final part will briefly conclude the chapter with key policy recommendations.

Taxonomy of Measures

In EXCEPTIUS data, restrictions on religious freedom are classified as a specific restriction of fundamental rights and civil liberties. If the legal act refers to freedom of religion or belief, its content is coded in five categories, which represents the level of burden of restrictions of this freedom. The first category includes situations in which all restrictions are lifted (0). The following categories concern: (1) light restrictions on religious practice, (2) heavy restrictions on religious practice, (3) light restrictions on specific religions, (4) heavy restrictions on specific religions, and (5) light restrictions on religious practice combined with heavy restrictions on specific religions.

The restrictions are identified as heavy if they entirely suspend religious activity or limit the number of people allowed to gather to a maximum of 15. In turn, limitations are considered light if at least 15 persons are allowed to participate in religious activities at the same time, if an obligation to wear masks or keep social distance are imposed, if time or the use of certain objects of worship during religious events are limited or if certain elements of religious ceremonies or worship are affected in certain ways.

Restrictions on religious freedom are identified in 22 out of 25 countries for which the data was collected. Of the European countries under study, Hungary, Ireland, and Luxembourg did not pass legislation that directly restricted freedom of religion. However, this does not mean that such restrictions did not actually affect the believers in these countries. Some constraints imposed on freedom of movement and freedom of assembly, in particular stay-at-home requirements or restrictions on public gatherings (see Chazel’s contribution in this volume), had a direct impact on exercising the collective dimension of religious freedom. Also, face mask requirements, social distancing or similar rules governing social behaviour placed a significant burden on the performance of religious rituals and ceremonies.

With the above in mind, the restrictions on religious freedom discussed in this chapter will encompass both those which specifically target this freedom and those which introduce restrictions on funeral attendance—coded under a specific category in EXCEPTIUS. This approach will make it possible to focus on the characteristics of the restrictions directly related to the exercise of religious worship. Moreover, data collected for the first three waves of the pandemic do not allow the full picture of the shape and effectiveness of the restrictions with respect to freedom of religion or belief to be grasped. This concerns mainly the lack of information regarding sanctions imposed in case of non-compliance with the restrictions and the duration for which the restrictions were established. While it is fair to assume that most penalties took the form of fines (as in the case of the funeral attendance ban, where sufficient data is available) and the potential overstepping of the State’s powers in this area did not affect fundamental rights so much, the period for which the restrictions were adopted should be investigated in greater detail, since the arbitrary extension of this period was one of the major threats to human rights resulting from the legal response to COVID-19.

Restrictions on Religious Freedom in Numbers

Turning now to the conclusions emerging from the data analysed, the most important observation is that light restrictions on religious freedom dominated during the first three waves of the COVID-19 pandemic. As Table 9.1 shows, they were imposed twice as often as heavy restrictions on religious practice. Heavy restrictions were introduced in Belgium, Croatia, Cyprus, Estonia, Finland, France, some German federal states (Länder), Greece, Italy, Poland, Portugal, Romania, Spain, Switzerland, and the United Kingdom. Furthermore, there was no single case of restrictions, light or heavy, aimed at specific religions.Footnote 1 This might give grounds for optimism: it appears that the European governments did not shape these restrictions in a way that discriminated against any particular religion, especially religious minorities.

Table 9.1 An overview of the light and heavy restrictions on religious practice in analysed countries
Full size table

As for country-by-country data, it follows that particularly intense lawmaking in the subject area has been observed in Germany, Spain, Cyprus and, in a lower degree, in Greece, Switzerland, Italy, and Belgium. In the remaining countries, the legislative activity in this area was marginal. The high intensity of legal response in Germany and Spain corresponds with the data concerning the legal form and spatial coverage of restrictions introduced, as indicated in Table 9.2. Regarding Germany, all acts imposing restrictions of religious freedom were implemented at the subnational level by the state (Länder) authorities (see Magni-Berton’s chapter for an analysis of decentralised patterns of pandemic management). The case of Spain is similar, as the restrictions introduced by the State authorities concerned the regional level of autonomous communities (comunidades autónomas). Considering the remaining countries, the limitations were generally enforced by national executive bodies, and they covered uniformly entire jurisdictions.

Table 9.2 An overview of restrictions of religious practice in terms of territorial coverage
Full size table

There is also a difference between unitary and non-unitary countries in terms of the frequency with which restrictions were adopted. In federalised and quasi-federalised countries (with the exceptions of Greece and Cyprus), the number of legal acts adopted was significantly higher. These countries, most likely due to their decentralised organisation, adopted new regulations more frequently, which allowed them to be more flexible and modify the legal response in accordance with the pandemic situation. This adaptative approach contributed to safeguarding freedom of religion or belief, as it allowed the State to react in a proportional manner, especially during the first periods of the pandemic, when little was known about the routes via which the virus was transmitted.

What is also characteristic of the first three waves of the pandemic is the almost complete lack of legal review of the restrictions. The relatively highest intensity in this respect is observed in Spain, where nine acts were subject to judicial review and one act to constitutional review. In Malta, one act was scrutinised by administrative bodies, while in Romania and Italy two acts were subject to parliamentary approval. Given the almost complete absence of ad-hoc legislation it is plausible to assume that this deficit of legal review is best explained by the common social readiness for tolerating even far-reaching restrictions during the first periods of the pandemic.

Restrictions that specifically targeted funeral ceremonies were introduced in 22 out of the 25 countries analysed to the notable exception of Estonia, Norway or Sweden. As illustrated in Table 9.3, restrictions were mostly introduced by national executive authorities. It was only in France and Spain that two national acts were adopted by the legislative authority. Any breach of the ban on funeral attendance would have resulted in fines (Belgium, Denmark, France, Germany, Hungary, Ireland, Netherlands, Switzerland, UK) or, alternatively, a fine and imprisonment (France, Denmark). Considering the scale of the funeral attendance ban, in most cases (353 out of 539 legal acts), the number of people who could attend the funerals was fixed at approximately 10–50. Less burdensome restrictions were introduced in Austria, Belgium, Denmark, Germany, and Spain, where the number of people allowed to participate in funeral ceremonies ranged from 60 to 250. To a small extent, governments chose to impose restrictions that limited the number of participants to a maximum of nine. While very strict restrictions remained in place throughout the three waves of the pandemic, acts introducing medium restrictions (10–50 people) were more numerous during the first and second waves of the pandemic.

Table 9.3 An overview of restrictions on funeral attendance in terms of authority
Full size table

Assessing the Design of Restrictions Through a Human Rights Lens

The greater and more direct the threat, the more likely people appear to be tolerant of restrictions on human rights in general, especially restrictions on religious freedom (Hill, 2021, p. 94). This does not mean, however, that the pandemic offered a carte blanche to policy makers. According to Article 9 para. 2 ECHR, “[f]reedom to manifest one’s religion or beliefs shall be subject only to such limitations as are prescribed by law and are 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.” It follows from this provision that in order to be permissible, restrictions of this freedom must not only be prescribed by law, but they also must pursue a legitimate aim, at least one of those listed above. In the case of the COVID-19 pandemic, the commonly invoked aim was the protection of public health. The restrictions must also be necessary, which presupposes effectiveness. There is no doubt that religious practices, especially rituals performed in the presence of other people, could directly contribute to an increase in coronavirus cases, so technically speaking, the curtailment of religious freedom could work. However, it is highly questionable whether even a serious and imminent health threat justified such broad restrictions through executive acts. And, more importantly, everything depends on the design of a restriction.

Most of the measures which were aimed at preventing the spread of the coronavirus restricted access to religious gatherings by limiting the number of people allowed to gather or by suspending religious events entirely. The most popular tendency was to make the number of people allowed to participate in a religious event dependent on the size of the religious premises. There are several arguments against this kind of measures due to their lack of proportionality. First and foremost, restrictions which focus on the capacity of religious premises place a very high burden on the enjoyment of freedom of religion or belief, while their effectiveness in preventing the spread of coronavirus appears to be low (for a further discussion see Hau, 2022, p. 37). Second, the governments did not provide an explanation as to why the limit was set at the given size. Third, the restrictions based on limiting number of participants were not responsive to the rapidly changing pandemic situations and the growing body of knowledge on how the virus is transmitted. Additional objections may be raised with respect to the ban on attending funerals: since many funeral ceremonies take place outdoors, where the exposure to infection is considered to be much lower, there was little justification for setting the same restrictions as for indoor events.

The collected data also provide examples of other approaches to restrictions. In Bavaria, religious gatherings could not last longer than 60 minutes. This type of measure represents a different approach, but its proportionality is just as questionable as in the case of restrictions dependent on the size of the premises: it places a significant burden on the enjoyment of religious freedom while its effectiveness is doubtful.Footnote 2

In Cyprus, participation in ceremonies, such as weddings, baptisms, and funerals, was conditional on proof of degree of kinship: only first- and second-degree relatives, but no more than ten people, were allowed to participate. Similar restrictions were also introduced in Greece, where funeral attendance was allowed only with the participation of an immediate circle of relatives.

According to other Greek restrictions, participation in religious ceremonies was only allowed for the clergy, without the presence of the faithful and only with a limited number of participants keeping their distance from one another and complying with sanitary measures. In addition, it was mandatory that the entrance and exit doors to the place of worship be locked, and the clergyman in charge of the religious ritual was responsible for taking all reasonable steps to ensure that no person entered the place of worship while the service was being conducted.

On the other hand, in Spain certain restrictions took a more tailored form as they involved the prohibition of choir services, a ban on the use of blessed water, and the performing of ritual ablution outdoors only. Similarly, in Italy, a law was passed which differentiated the types of restrictions depending on the specific traits and needs of a particular religion (Protocolli per le celebrazioni delle confessioni religiose diverse dalla Cattolica, 2020). Such a tailored model of restrictions, which, based on scientific knowledge, focuses on elements of worship activities which pose an increased risk of transmission, deserves approval. This approach allows the regular operation of the religious ceremonies while protecting public health at the same time. The main objection regarding tailored restrictions is that they can interfere with the freedom of religion or belief, especially by arbitrarily prohibiting certain (potentially crucial) elements of worship. Therefore, the implementation of such measures must be subject to prior approval of religious groups and should involve cooperation with religious leaders at the stage of drafting the measures.

But the key issue remains the proportionality of the restrictions. In a recent Strasbourg decision, the European Court of Human Rights (ECtHR) identified the timespan for which restrictions had been established and the size of the target group as essential factors to be considered when assessing the proportionality of the COVID-19 restrictions. The ECtHR initially found that the restrictions imposed by the Swiss government had violated Article 11 ECHR in both its procedural and substantive aspects (ECtHR, 2022).Footnote 3 The procedural violation resulted from the lack of judicial review of the restrictions, whereas the substantive infringement concerned, among other things, the general character of the restrictions and the long duration of the ban (ibid., para. 91; Smet [2022]). The decision concerned freedom of assembly and not freedom of religion or belief, but it perhaps illustrated the emerging attitude of the ECtHR towards COVID-19 restrictions: those established for a shorter period of time and addressed to specific groups rather than to the entire population were more likely to pass the proportionality test.

The Implementation of Measures

The urgency of the epidemiological situation forced the public authorities to immediately impose a wide range of restrictions, from the obligation to wear face masks to stay-at-home orders. The growing fear of the virus and the accompanying social anxiety fostered an expectation that governments would come up with quick solutions. This, however, all too often resulted in ordinary legislation being bypassed with the issuance of executive acts. Soon enough, controversy arose as to the extent to which executive acts can impose restrictions on fundamental rights, including freedom of religion or belief.

As explained in the previous section, the introduction of restrictions on religious freedom through executive acts deserves criticism, especially when such extraordinary measures were imposed in breach of statutory laws. On the other hand, it needs to be admitted that there was a reason for the sudden spate of executive acts: it would be going too far to say that parliaments had no opportunities to convene during the pandemic, but it is true that their functioning was hindered. More importantly, it takes time for a parliament to pass new legislation. Against this background, it is easy to understand why executive acts became a convenient tool for making and enforcing rapid decisions during the pandemic. But there was a price to pay for this initial efficiency: the use of executive acts for the purpose of limiting fundamental rights, including freedom of religion or belief, had an impact on the States’ ability to enforce compliance and uphold administrative penalties before the courts.

Based on EXCEPTIUS data, it can be confirmed that the restrictions on freedom of religion were implemented for the most part through executive acts (a total of 529 out of 548 acts), both at the national (355 acts) and the subnational levels (174 acts). This is inconsistent with the principle that legislative acts be established by the legislature.

Also, it is worth mentioning that in some countries (Malta, the United Kingdom) restrictions on religious freedom were implemented through advisory acts that had no legal force but simply offered guidelines. In the UK, the government issued non-binding guidelines on the safe use of places of worship during the pandemic. The document included general recommendations on the use of measures such as staggering entry, providing multiple entrances and one-way routes to enter and exit the premises, providing hand sanitizers, and so forth. More importantly, the guidelines encouraged religious leaders to consider changes that could be introduced to religious rituals that typically involve close contact between individuals, and gave examples such as preventing worshippers from touching or kissing religious objects that are handled together, removing books and shared resources such as prayer mats, worship sheets or devotional items from use, avoiding chanting and/or playing instruments, ensuring that any initial washing/ablution rituals be conducted outside the place of worship, and closing off common washing areas. The recommendations might not have been binding, but they provided important guidance for religious associations and worshippers. And besides, this way of crafting the government response ensured transparency, which is best achieved when all segments of the population, including minorities, are involved in decision-making (Ó Cathaoir, 2021, pp. 47–48).

Restrictions on religious freedom were mostly introduced for a limited period. Only in 107 cases they were introduced for an unlimited period. In 100 cases, the restrictions were imposed for less than 30 days. Restrictions on funerals were also mainly introduced for a limited period. In 108 cases (out of 539), they were imposed for an unlimited time. In 94 cases, the restrictions were introduced for less than 30 days. In most cases, the duration of the restrictions was intended to be as short as possible, in order not to suppress the freedom of religion or belief, while also sufficient to ensure safety for a period of time. It seems that the introduction of severe restrictions on religious freedom for a period longer than 30 days is likely to have constituted an unjustified interference with the rights of believers, especially in the light of the dynamically changing pandemic situation.

The Reality Behind the Numbers

As illustrated by the collected data, the views of European States on the need to restrict religious worship during the COVID-19 pandemic were remarkably similar. Collective forms of worship were seen as a threat to public health since they require many people to be present at the same time in confined settings, and involve activities that facilitate the spread of the virus, such as loud singing or reciting prayers in unison. The review of the measures taken to slow down the rate of new infections—or to flatten the curve, as the popular saying went—reveals that the public authorities used a legal toolbox similar to that which ultimately proved useful when dealing with freedom of assembly. Consequently, gatherings were banned or suspended, the number of people allowed to attend services was reduced, strict hygiene rules were required, and social distance policies were enacted. Much less of a threat were individual forms of worship in designated public places, such as individual prayer in a temple. These too, however, were restricted as a consequence of reduced access to church premises. By contrast, the restrictions did not extend to individual forms of worship taking place in private places, as these did not pose any threat to public health.

Yet, to draw complete parallels between freedom of worship and freedom of assembly can be deceptive. In the case of freedom of assembly, mass attendance is, or might be, a value in itself. The more the merrier: a large number of attendees sends a message that the cause for which people have chosen to leave their homes and stand together enjoys broad public support. By contrast, with religious worship, the essence lies in a collective spiritual experience whose value does not depend on its social impact. Moreover, the use of legal measures similar to those that might have been appropriate with regard to freedom of assembly did not really have the same effect when restricting religious worship. It should be borne in mind that the first restrictions were introduced in March 2020, at a time when churches belonging to the Christian tradition, still dominant in the European religious landscape, were preparing to celebrate Easter. Restricting access to religious worship during this period had a particularly severe impact on religious life in Europe (see e.g. Papazoglou et al., 2021, pp. 3219–3220).

Another important difference is that in some cases physical presence at a specific time and place can be a condition sine qua non for the validity of a ritual: not just part of a religious protocol, but the essence of the ritual. While the purpose of a political assembly can sometimes be partially achieved online, or with a slight delay, there are certain religious practices which simply cannot be postponed to another day, moved to the digital sphere, or performed over the phone. One obvious example is the Sacrament of Eucharist in the Roman Catholic Church, which is a consequence of the theological doctrine of transubstantiation and the corresponding belief that, at consecration, the substance of the bread and wine is truly being transformed into Jesus Christ’s body and blood (Martínez-Torrón, 2021, p. 57). Another example is the Sacrament of Penance (confession), as it requires face-to-face contact between the penitent and the confessor. Also, in the Jewish tradition, the recitation of Kaddish is expected to involve the physical presence of those performing the ritual. Interestingly, the prolonged duration of the pandemic prompted religious communities to seek ways to adapt religious beliefs to the new situation (see e.g. Macaraan, 2021, p. 531; McElwee, 2020; Cooper, 2021).Footnote 4

And it is precisely cooperation with religious communities, as many believe, that should be the starting point for the design and implementation of restrictions intended to mitigate the spread of the virus.Footnote 5 There is at least one but powerful argument for this: the need to respect the autonomy of religious groups—an essential component of freedom of religion or belief in its collective dimension (see e.g. ECtHR [GC] 2000: para. 91; ECtHR [GC] 2014: para. 127). The religious groups know best which elements of their rituals are easy to replace and which belong to the core of their religious identity. Any top-down interference in this field is always questionable. It is also unnecessary in practice. It is no coincidence that the initiative to introduce certain liturgical changes, such as distributing the Holy Communion in the hand rather than on the tongue in regions where this was not commonly accepted (Budaev, 2021, pp. 2342–2344), came from the churches themselves.

The data collected for the EXCEPTIUS project do not reveal disparities in the treatment of different religions, that is, legislation adopted in response to COVID-19 did not target particular religious denominations. On the face of it, this offers grounds for optimism. The pandemic might easily have become a pretext for adopting openly discriminatory measures against specific religions, especially minority ones. This would have been particularly likely in countries that provide a differentiated legal framework for the functioning of religious groups—this is actually what most European countries do (Uitz, 2007, p. 87)—especially if there are separate acts that define the relationship between the State and the respective church, such as in Italy or Poland. It is well known that such a model significantly hinders the principle of equal rights for religious groups. Had the anti-COVID measures been introduced individually for each denomination at the time, the risk of unequal treatment would have greatly increased, and in the atmosphere of panic about the new threat, such abuses of power were likely to have gone unnoticed. Luckily, the pressure of time prevented the public authorities from yielding to this particular temptation, for the restrictions on religious worship were mostly enacted in general anti-COVID acts rather than through meticulous amendment of the laws on the exercise of religion.

But quantitative data do not tell the whole story. The measures may have been the same, but the effects they triggered varied from one jurisdiction to another. The absence of discriminatory criteria in the body of the regulation does not necessarily mean that various religions were affected to the same extent by the restrictions (Madera, 2021, pp. 3–4). The true picture of the impact of the pandemic on collective worship only becomes apparent when the religious diversity of European countries is considered.

It is no secret that the various religions, even if they respond to the same deepest human needs, differ profoundly in their doctrine, habits, and attitude towards any earthly power. Activities that pose different epidemiological risks are present in varying degrees of intensity: for some communities, singing, eating together, and kissing objects of religious veneration may be central to the ritual, whereas for others such activities may be merely complementary or even unnecessary. Also, there are variations in the degree of juridisation of churches, a factor which apparently had an impact on their resilience in times of pandemic (Valenzi, 2021). Even more importantly, there are religious groups that, for various reasons—whether due to doctrinal motives or because of negative past experiences in a particular country or region, or both—may be distrustful of public institutions,Footnote 6 which obviously makes cooperation in responding to pandemics much more difficult. To make matters even more complicated, a religious group which dominates in one country and enjoys the support of the government may be a minority in another jurisdiction.

For all these reasons, when trying to summarise and compare the impact that the COVID-19 pandemic had on religious worship, it is necessary to consider this less obvious aspect: not just the measures taken but also their varying effects on different religious communities. From this perspective, the experience of a Roman Catholic in Germany and a Roman Catholic in Portugal may have had more in common than the experience of a Roman Catholic and a Jehovah’s Witness living in the same country or even city. This peculiarity should be taken into account in any future comparative research.

Conclusion

The outbreak of the COVID-19 pandemic profoundly affected the conditions in which religious worship was practised. While individual expressions of devotion at home were possible throughout any lockdown, religious life in a communal setting was affected by severe restrictions. But difficult times are there to draw lessons from. If the experience of recent years is not to go to waste, it may be appropriate to draw up three main policy recommendations for the future. These recommendations are based on the EXCEPTIUS dataset, but they also find inspiration in the rapidly growing scholarship on the subject matter and the relevant views of the Strasbourg Court.

First, public health may take priority over spiritual well-being, yet the latter ought not to be completely disregarded. The data collected show that this simple truth was not acknowledged in all jurisdictions. In some cases, religious worship was heavily suppressed, which suggests that not all States showed understanding for the needs of believers.

Second, it does not appear to be necessary to achieve uniformity in public policies on the access to places and practices of worship during pandemics. The existing tools, including the principle of proportionality, provide sufficient consideration for freedom of religion or belief. But proportionality is to be taken seriously: there is no justification for applying different maximum occupancy rates to churches and beauty parlours if the level of risk is similar. Nor is there any justification for extending restrictions beyond the necessary period.

Third, tailored restrictions, that is, those which focus on high-risk elements of religious worship instead of banning worship in general, may or may not be discriminatory and they may or may not interfere with religious autonomy. We believe that if the policy makers wish to avoid discrimination and interference while keeping the restrictions effective, it is crucial for them to act in consultation with religious communities and their leaders. Continuous dialogue and the willingness on the part of the authorities to explain the purposes of the restrictions increases the likelihood of their actual—and not just apparent or superficial—implementation, since they benefit from the authority of a person who enjoys the trust and obedience of the faithful. Such a consensual approach also reduces the level of suspicion of those who might otherwise think that the restrictions imposed are in fact motivated by hostility to religion or a particular church.