Civil Religion or Nationalism? The National Day Celebrations in Norway

The Norwegian National Day (17 May, also referred to as Constitution Day) stands out as one of the most popular National Day celebrations in Europe. According to surveys, around seven out of every 10 Norwegians take part in a public celebration during this day. This means that the National Day potentially has an impact on the way people reflect upon national identity and its relationship to the Lutheran heritage. In this paper, I will focus on the role religion plays in the Norwegian National Day rituals. Researchers have described these rituals as both containing a significant religious element and being rather secularized. In this article, I discuss the extent to which the theoretical concepts civil religion and religious nationalism can help us understand the role of religion, or the absence of religion, in these rituals. Based on surveys of the general population, I analyze both indicators of civil religion and religious nationalism. The two phenomena are compared by looking at their relation to such items as patriotism, chauvinism, and xenophobia. The results show that civil religion explains participation in the National Day rituals better than religious nationalism.


Background of the Study
In this article, I will discuss the role of religion in the National Day celebrations. Two theoretical perspectives on public forms of religion will be analyzed: civil religion and religious nationalism.
The historical foundation for the Norwegian National Day celebrations, the Constitution of 1814, has both including and excluding features. It was inspired by and meant to characterize liberal group principles, but at the same time, it has traits from a homogenous culture with excluding paragraphs in relation to ethnic and religious minorities. Examples of this are the paragraphs excluding Jews and Jesuits, which were first removed in 1852 and 1956, respectively (Hoffmann 2020). This is the legal foundation of the National Day. But what about the celebrations themselves? What type of profile do they have in relation to the question of including religious symbols and institutions in them?
Because the National Day in Norway is extraordinarily popular, it has attracted the interest of researchers. Three earlier research projects focusing on the Norwegian National Day celebrations will be presented before I examine data from a research project that I have been involved in.
For many years, secularization was the hallmark of many European countries. The development of multicultural societies in recent decades has, however, given religion a more prominent place in the public sphere, as multiculturalism has made religion both more visible and more contested.
On the one hand, religious people react negatively to what they see as the secularization of society, while on the other hand, organized secularists protest the visibility of religion in the public sphere. Here, I will focus on how representatives of the local communities and the general population deal with topics relating to religion in the local celebrations.
I will take a closer look at the relationship between religion and national identity on the local level in Norway. The data material shows us how the local authorities interact with civil society actors in preparing for the National Day celebrations. Using recent survey data, I will examine how people in Norway look upon the role of church and religion in the National Day celebrations and discuss how these findings relate to the theoretical perspectives presented above. Are the religious aspects of the National Day celebrations best seen as related to an inclusive civil religion phenomenon or as a part of a narrow ethnic-cultural form of nationalism? To discuss the role religion plays in the National Day celebrations, I will introduce two theoretical perspectives: civil religion and religious nationalism.
My hypotheses are: (a) Over time religion has been downplayed in the National Day celebrations and has become a more implicit part of the celebrations in most local communities; (b) It is still relevant to use the concept civil religion to describe these elements; (c) Religious nationalism is in general not suited to describe the religious elements of the National Day rituals, its power potential has been undermined by the ongoing secularization process.

Previous Research
Early in the 1990s, the Swedish ethnologist Barbro Blehr conducted a research project on "the 17th of May and Norwegian nationalism" (Blehr 1999a(Blehr , 1999b(Blehr , 2000. Based on fieldwork in three Norwegian towns over a three-year period, she "investigated the transitions of ideas into rituals that take place in Constitution Day celebrations" (Blehr 1999a, p. 28). Her focus was on particular actions that people were interested in doing during the celebrations. In her view, the active bodily involvement of participants is crucial for the outcome of and the possible effects the rituals have on attitudes towards nationhood.
A ritual is a sequence of activities involving gestures, words, actions, or objects, performed in a sequestered place and according to a set sequence. Rituals may be prescribed by the traditions of a community, including religious groups. Even if the term is associated with religion, "ritual" can also be used about a set of actions outside the religious realm (Bell 1997).
Blehr describes the National Day celebrations as a three-part sequence of events. The early morning events consist of commemorative rites, held at memorials or graves of national and local luminaries. Blehr explicitly mentions that this part of the celebrations is marked by morning services in the Lutheran state churches (Blehr 1999b, p. 178). 1 Later in the morning, the children's parade takes place. This is the main event in the celebrations, and parades are held in every local community across the country.
The children march as representatives of their school and their collective identity is presented by a banner bearing the name of the school. The children at the front carry large Norwegian flags. Around noon, right after the children's parade, a public gathering is held which includes the main speech of the day, songs, and other performances. In addition to the children's parade, there are other parades like the "citizens parade" (borgertog in Norwegian) and the "graduating student's parade" (russetog).
The third phase of the day is more informal. The afternoon festivities consist of local neighborhood gatherings held outdoors, preferably in a schoolyard. This is the end of the public part of the celebrations. Later in the evening, some people arrange private parties in their homes or come together at semi-public gatherings in clubs and associations.
In her study, Blehr focuses on the parades taking place during the National Day, especially the citizens parade: "In these parades, Christian groups are highly visible," she writes (Blehr 1999b, p. 180). These groups tend to represent modern, charismatic branches of Christianity, rather than the traditional National Lutheran Church. While Christian groups are very visible in the parades, this is not the case for homosexuals, atheists, and agnostics. According to Blehr such groups lack representation in the parades. When looking at the voluntary associations represented in the parades, she also concludes that ethnic diversity is downplayed, observing that ritual events like the parades produce messages that collide with the values underlined in official programs and speeches. The parades present foreign ethnicity as a status hard to reconcile with belonging to the Norwegian nation. The "undesired message" is that Norwegian nationalism as an ideology is presented as more white-ethnic than what the official rhetoric expresses.
Since Barbro Blehr conducted her study in the early 1990s, Norwegian society has changed in many ways. There has, for example, been an ongoing public debate on the potential for integration of ethnic minorities in the National Day celebrations. This question is related to the role of religion. The more multicultural Norway has become, the more challenging is the presence of majority church-related religion in public events. The ongoing secularization of the Norwegian society may have an impact on the way the National Day is celebrated, and the meaning people attach to the rituals being performed. Both trends may have led to religion being downplayed in the May 17 rituals.
Some 20 years after Blehr conducted her study, another Swedish researcher, the sociologist Elgenius (2011), compared the Norwegian National Day celebrations to similar celebrations in other European countries. Like others, Elgenius describes the Norwegian National Day as a special case because it has enjoyed such enormous popularity over a long period of time. She suggests that the Norwegian case is unique in combining historically significant factors with popular ideological framing. In its formative years, Constitution Day became a symbol of independence before statehood was achieved. Later, the National Day was linked to popular democracy and egalitarianism movements. Elgenius argues for the uniqueness of the "national day design", with the parades and performances of schoolchildren securing the participation of parents, grandparents, and other family members. Unlike Blehr before her, Elgenius points to the overall non-religious character of the central celebrations. In her analysis, the fact that non-Christians are not excluded but take part to the same extent as Christians is crucial for the preservation of the level of participation (Elgenius 2011, p. 178). The description of the Norwegian National Day celebrations as inclusive and underlining national identity in all parts of the population contrasts the description given by Blehr just 15 years earlier. Most strikingly, this concerns the role attributed to church and religion.
The third research project was conducted by the Danish historian Damsgaard (2013). In 2011, she undertook a comparative study of the National Day celebrations in Denmark, Norway, and Sweden. Unlike Blehr and Elgenius, Damsgaard describes the Norwegian National Day celebrations as an example of civil religion. She argues that the chosen hymns and songs have a clear civil-religious message. This includes the Norwegian national anthem, which has an explicit civil-religious character, as it invokes God as a protector and savior of Norway in stanza 7 (which is always sung).
The three above-mentioned studies have rather different takes on the role religion plays in the celebrations. This is in part due to the use of different theoretical perspectives, but it is nevertheless striking how differently the role of religion is described. Even if it is difficult to use qualitative studies as evidence of changing trends, the studies leave the impression that religion has been downplayed over time. In this study, I will use representative survey data to ascertain if the role of religion in the rituals has changed over the last 20-year period on the national and local levels. Then I will discuss if the role of religion in the celebrations is best described as civil religion or religious nationalism.

Civil Religion
The French philosopher Jean-Jacques Rousseau was the first to refer to the concept of 'civil religion', but it was the American sociologist Robert Bellah who applied the concept to the contemporary situation in his article "Civil Religion in America", from 1967. Here, Bellah (1967) describes civil religion not as a state religion but rather as expressions that sacralize national values and turn them into a kind of religion.
There are two aspects of Bellah's original reference to civil religion. The first of the defining elements for civil religion is that it is tolerant and inclusive. According to Bellah, one of the main functions of civil religion is to connect people across confessions and denominations. Then there is the critical-prophetic element, which is activated when people do not live up to the standard expected of a chosen people (Repstad 2009, p. 200). The critical aspect has been connected to the international role of the Nordic countries as small, peace-loving democratic countries with a strong sense of human rights. From this perspective, the Scandinavian countries are seen as realizing a special mission in international society as agents of human rights, peace, and negotiations between other countries (Gustafsson 2000, p. 184). Several Scandinavian sociologists of religion have described the folk church culture as the Scandinavian counterpart to American civil religion (Sundback 2000;Gustafsson 2000;Riis 1989). Their argument is that the folk churches fulfil the same functions as civil religion in the United States.
The Danish sociologist of religion Warburg (2005aWarburg ( , 2005bWarburg ( , 2008Warburg ( , 2009 argues that a transcendent element is needed if we are to use the concept of civil religion. When exploring this theme, the call for God is a key focus area. In her studies of civil religion in Denmark, Warburg looks for references to God in speeches and songs used at national and local celebrations. Her way of addressing civil religion corresponds with Bellah's studies of presidential speeches in the 1960s and 70s.
Even if the formal ties between the national Lutheran churches in Scandinavia and the state have become looser in recent years, it is still fair to say that these churches play a special role in national rituals. This is the case both in events of joy, like the National Day, and when the country is stricken by accidents and disasters. I will focus especially on the role the national church in Norway performs during the May 17 celebrations to see if these rituals qualify as examples of civil religion. Neither Blehr nor Elgenius uses this concept in their studies of the Norwegian National Day. The study conducted by Damsgaard, on the other hand, uses civil religion as a lens through which to study national day rituals.

Religion and Nationalism
Historically and institutionally, it is difficult to make a clear distinction between the development of the nation and the development of the dominant religious tradition. During the twentieth century, some of the Lutheran churches in Scandinavia gradually gained more freedom from the state. Nonetheless, in Norway, the state still has some formal power over the church, for instance over the church budget. In cases of national anniversaries and celebrations, representatives from the church appear with the royal family and together symbolize the unity of the nation. Since the nineteenth century, the national churches in Scandinavia have served as the main religious symbols of the nations and as the caretakers of national ritual.
I will not delve deeply into the historical background of Norwegian national identity. My focus is on how national identity is viewed by individuals today. But before I present my empirical data, I will examine how theorists look at the relationship between religion and nationalism. The historian Smith (1991Smith ( , 2001 finds that the rise of nationalism is closely related to secularization and the development of science. When religion loses ground, new forms of affiliation are actualized. Smith regards nationalism as a substitute for religion and as a potential common connection point for the population. He emphasizes the emotional and cultural aspects of national identity and distinguishes between an ethnic-cultural form of national identity and a political-ideological form. Smith believes that the ethnic aspect is just as important as the political aspect, arguing that many of today's nations are based on ethnic elements from pre-modern times. When an objective ethnic base is lacking, attempts will often be made to construct one. Without an ethnic-cultural foundation, the nation may fall apart. While admitting that economic changes may have an impact, Smith still believes that modern society's most important contribution to the development of nationalism comes from science and secularization. Building on Anthony D. Smith, the Norwegian social scientist Østerud (1991a, 1991b, 1994) discusses the ethnic-cultural and civic-political approaches to national identity. He claims that France and Germany are prototypes of these two forms. The German type of national identity is based on the notion that the nation is a community based on shared ethnicity, language, and culture. The French tradition sees the nation primarily as a politically constructed community built on the idea of rights and responsibilities within the community (Østerud 1991b, p. 200). According to Østerud, among the Scandinavian countries, Norway is the most prominent example of the ethnic-cultural form of nationalism, while Sweden is oriented towards civic-political nationalism, with Denmark in a middle position.
The sociologist Brubaker (2011) operates with four approaches to religion and nationalism. First, religion can be seen as a phenomenon analogous to nationalism. Second, religion can explain nationalism. Third, religion can be treated as part of nationalism, and fourth, is to study a distinctively religious form of nationalism. Religious nationalism points in the direction of seeing religion as an integrated part of nationalism.

Data and Methods
Over the last 20 years, several surveys have been conducted to explore the organizational aspect of the National Day, participation in the different rituals, and attitudes towards the various elements in the celebrations. In 2012, a research team I took part in conducted a survey among the directors of the cultural departments in Norwegian municipalities (kultursjefer). Each local government appoints a committee that has the responsibility for the local National Day celebrations. These committees have members from civil society organizations (sport clubs, school boards, and so on) and political party representatives. The cultural departments perform the secretarial functions in the committees. Since these committees are organized on a year-to-year basis, and are decommissioned after the celebrations, the directors for cultural affairs are assumed to be the officials who have the most information about changes that might have occurred over time. An electronic questionnaire was sent to the cultural directors in all Norwegian municipalities (N = 429). Two hundred and five directors responded, giving a response rate of 47 percent. The sample was checked against data on all the municipalities to see if it corresponded with the population. Table 1 shows that small municipalities are somewhat underrepresented in the sample, while large municipalities with 10,000 or more inhabitants are slightly overrepresented. Nevertheless, medium-sized municipalities dominate the sample. The overrepresentation of large municipalities may be explained by the fact that they are the ones with the best infrastructure and access to human resources. Several of the questions in the survey were formulated as follows: "Based on your experiences of the local celebrations, would you say that there has been any change in the way religious groups or church representatives take part in the rituals at the local level, or has the situation remained the same for the last five years?" By asking retrospective questions, we can trace changes even when dealing with cross-sectional data and not panels. Most questions had fixed categories for answers. There were also, however, open spaces in the questionnaire for additional comments. In addition to the director study, I have access to several representative population surveys about National Day celebrations conducted at different points in time. Two of the surveys are from the 1990s; the ISSP survey of national identity from 1995 and a KIFO survey from 1998 (Botvar et al. 2013). 2 A KIFO survey from 2013 is also used (Botvar et al. 2013). The 2013 survey was carried out as an online survey during the period 22-31 January 2013 by the data collection agency Norstat on behalf of KIFO. Norstat has a panel of approximately 83,500 persons who can receive questionnaires via e-mail. With this point of departure, a nationally representative sample has been achieved. It has become increasingly common to use web panel surveys to categorize population samples, as it has become more difficult to use telephone and postal surveys. In Norway, 98 per cent of the adult population between 16 and 75 years of age use the Internet (figures from Statistics Norway). When using Internet interviews, the term percentage response is not used in the same way as in the traditional postal surveys. Approximately 25 per cent of those who received e-mail invitations to participate in the survey responded. This is a low number, but 95 per cent of those who had "opened" the invitation mail responded. When comparing the non-weighted distributions in the survey with the actual distribution related to gender, age, and place of residence, we find that there is a large degree of concurrence. According to Norstat, these surveys have low weights, and corrections have only been made for relatively small imbalances (Botvar et al. 2013). In this article, I will focus on the most recent survey from 2013 with only some minor comparisons with those from the 1990s. The empirical analyses are dominated by descriptive statistics and the use of cross-reference tables. I will also use ordinary linear regression analyses to ascertain which factors contribute to explaining participation in the National Day rituals.

A Local Example
National celebrations also tend to be local celebrations. On the one hand, the Norwegian National Day is celebrated in connection with national institutions situated in the capital of Oslo. The King and the President of the Parliament are key figures in these events. On the other hand, all municipalities and local communities have their own celebrations. My focus is on ways religion is actualized in the National Day celebrations at the local level.
Religious services held on National Day can be controversial. In recent years, there have been several conflicts related to religious services on May 17, and occasionally local incidents reach the national media. One example is from the village of Kråkstad, south of Oslo. Here, the children's parade used to go from the local school to the church and back again during National Day. A church service was held in the middle of the event. The church did not have room for everyone to take part, and not everyone wanted to attend the service. Thus, the parade split in two, some people went inside to worship, while others waited outside the church for the parade to continue. Many locals found this problematic (Botvar 2017), and this tension escalated when some young activists from the Humanist Association appeared at the celebrations and served buns and juice to those who decided not to enter the church. The young activists, who had come from Oslo, were seen by many as intruders into the local celebrations. Later, the case was discussed in the municipal council and a proposal was tabled by the Socialist Left Party (SV) and the Liberal Party (V) stating that, "religious and philosophical events should not be part of the children's parade". The motion was voted down by the majority (Aagedal 2014, p. 318). Locally, a solution was found when those who did not want to attend the religious service were offered an alternative event in the form of a history lecture in the stable next to the church (Aagedal 2014, p. 319). Kråkstad, thus, continues to have a worship service as part of the local program.

The Cultural Directors Study
The role played by the Church of Norway at the local level was a topic in the survey given to the local cultural directors. We asked them about the ways in which the Church of Norway was integrated in the celebrations at the local level. Table 2 shows that seven out of 10 directors state that a worship service is part of the official program in their municipality, a high figure. Only seven percent state that the local church does not take part in the celebrations at all. It is difficult to state, however, exactly how common a worship service is as part of the National Day celebrations on the local level. On average, each municipality has three to four church buildings related to the Church of Norway, meaning that if there is a service in at least one of the four churches, the director will answer in the affirmative. Furthermore, while these worship services may be an integrated part of the main celebrations, as in the Kråkstad case, they are more often a separate part of the celebrations. A small number of the municipalities, 14 percent, report that there is an event in the churchyard right after the parade. This is not a worship service, but an arrangement where some national hymns are sung, and the local vicar greets the parade by saying a few words (for example about the country being protected by God). A check of public announcements in newspapers in the Oslo area the same year showed that one in three churches had a service on May 17 (Botvar and Holberg 2012). The number of local celebrations and parades is also difficult to calculate. Altogether, there are 2800 primary schools in Norway, but many of them celebrate together with neighboring schools. In the big cities, all schools tend to meet in the city center to celebrate together. This means that the most reliable data about local celebrations that include a sermon come from counting the number of church announcements.
Even though as many as one in three local churches has a worship ceremony on May 17, this does not mean that people who seldom go to church show up on this special day. On the contrary, a population survey conducted at the same time as the municipal survey shows that the number of churchgoers on National Day resembles the statistics showing regular churchgoers, indicating that it is probably the regular attenders who also show up on this particular day. According to population data from 2012, about seven percent attended a service on May 17. This is about the same number as those who visit a church at least once a month. The number of churchgoers on May 17 is much lower than on other holidays, such as Christmas when 23 percent of Norwegians report visiting a church (Botvar et al. 2013).
As shown in Table 2, in almost half of the municipalities (45%) there are commemoration services in the cemetery outside one of the churches in the morning. The most important part of this ritual is the placing of a wreath on the grave commemorating people who lost their lives during the Second World War. In some cases, the vicar in the local church will hold a memorial service as part of the ritual. Again, it is difficult to say how common this ritual is because the focus is on the municipality as a whole and it is not taken into consideration that it will most often have three to four parishes, each with their own church building.
It is quite common for the children's parade to start or stop at a church building. This does not mean, however, that the service is an integral part of the event. Usually, services are held before or after the event and the service is only rarely included in the parade's route, as it is in Kråkstad. In the data material, there is no statistical correlation between the two categories "having a service" and "the parade starts/ends at a church building". This indicates that the church is mostly used as a landmark for practical reasons, especially in small municipalities with only a few public buildings to choose from.
Our survey does not directly tell us how often the May 17 parade enters a church building. But the low number of municipalities that offer an event in the churchyard after the parade (14%) indicates that this happens rarely. When a worship service is perceived as problematic and an intrusive element in the celebration, the event might end up in the national media, as happened with the Kråkstad case. Every year, a few local cases make their way into the national media.
To ascertain if religion is an integral part of the National Day rituals, we also asked about the role the local vicar plays in the celebrations, where it was found that the vicar has a formal role in the public celebrations in some municipalities. Vicars can, as we have already seen, be active in various ways-in connection with the laying of wreaths, holding worship services, or making other types of speeches and appeals. On a direct question, 23 percent of the municipal directors state that the vicar(s) in their municipality had a special task on May 17. This is a much lower number than those reporting that a worship service is part of the celebration.
Asking about the activity of the vicars is probably a better indicator than asking about the church. In many cases there is a worship service in at least one of the churches in the municipality. But these worship services are often not an integrated part of the celebration. This is the case when the worship service is formally part of the program but taking place at the same time, 11 a.m. for example, as other parts of the program, such as the children's parade. This is a signal that the worship service is an event intended for the part of the population that does not have children in school.
The role of the vicar is a more concrete question. If the vicar has a prominent role in the program, this is probably something the directors will know about. In Table 3, the number of "don't knows" is quite high. This indicates that in some cases the program in the churches is something that takes place off to the side of the main event. The finding that about one fourth of the local vicars are actively involved in the celebration is supported by the data from the 2012 population survey. In this survey, 24 percent of the informants also state that the vicar in their local community performs special tasks on May 17.
When including the number of inhabitants in the analysis, we find that it is more common to report that vicars have special tasks on May 17 in large rather than in small municipalities. In large municipalities with several vicars, chances are high that at least one of them is involved in the celebrations program. This supports our argument that most vicars are not actively involved in the celebrations.
A follow-up question about what tasks the vicar performs on May 17 shows that this concerns holding services, giving greetings at institutions (senior citizens homes), holding an appeal/speech, or being a member of the May 17 committee. Only two percent of the municipalities mention that the vicar gives the main speech of the day. This happens very rarely and would require a special reason, such as a historical anniversary relating to the church building. Thirteen percent report that the vicar performs "other tasks". This is probably related to the ritual at the graveyard or the event in the church after the parade. But again, the answers differ significantly when asking about the role of the church and the role of the vicars. When asking about the church, the informants tend to think about the municipality as one unit, but there is one vicar in each parish and several parishes in each municipality.
In the municipal survey, we also asked how religious groups in the municipality relate to the celebration. Fourteen percent of the directors state that some religious groups have their own celebrations outside of the official program. However, only seven percent state that religious groups have their own celebrations in parallel with the official celebrations. This indicates that religious groups tend to join the ordinary events, but that some also have their own events in the evening (for example a social event at their house of prayer). This happens more often in the southern and western parts of the country than elsewhere, in the so-called bible-belt.
The data material gives us reason to believe that over time, it has become less common to include the majority church in the May 17 events. When asked, seven percent of the cultural directors answered that the role of religion in the celebrations has changed over the last five-year period. Answers given to open questions about what kind of changes are seen indicate that what is often mentioned is the service being moved away from the parade and that participation in services is no longer mandatory for children. The inclusion of worship services does not appear to be a controversial topic in most municipalities. Only in a small number of municipalities does the main event of the day, the children's parade, include a sermon in a church. In such cases, no one has raised this as a problem, or for geographical reasons there is no alternative route.
Our study also indicates that local actors try to minimize conflicts related to religion on the National Day. The informants were asked: Have there been conflicts related to the visibility of religious groups in the May 17 celebration? The question can be interpreted as dealing both with conflict over the role of the majority church and the role of religious minorities. In the survey, only three percent of the municipal directors answered that there had been a conflict relating to religious groups in the celebrations during the last five-year period. Even though six percent answered "Don't know", this indicates that religion very rarely leads to conflicts. In the few cases where conflicts have occurred, they relate to the church being too active and visible in the celebrations.
I have now reported on findings from the survey conducted among the municipal directors. The data is somewhat difficult to interpret. Some findings indicate that the church plays an active role in the National Day rituals and celebrations. Other data point in the direction of a minor role. This is especially the case when the role of the vicars is in focus. There is little evidence pointing in the direction of religious groups actively trying to dominate the public events. This would probably have been the situation if religious nationalism had been a driving force behind the celebrations.

General Population Surveys
What about data from the general population? In what way and to what extent is individual participation in public rituals on National Day linked to civil religion or religious nationalism? Which of the two concepts could best explain public support for the National Day rituals?
Several surveys have touched on this theme, but only a few have used the exact same questions. In this article I will focus on the KIFO survey conducted in 2013 (Botvar et al. 2013). This survey includes many of the same items that have been asked in earlier surveys, such as the ISSP survey from 1995. 3 The surveys referred to are all representative population surveys conducted with assistance from professional data collection agencies. My analysis is threefold; first, I will look at items showing participation in the National Day rituals to see if there has been a change in the public support over time. Second, I will look at items that can be used as indicators of participation in the public rituals, as well as indicators of patriotism, chauvinism, xenophobia, church involvement in the celebration, and religious nationalism. In the third step, I will conduct regression analyses to see which factors contribute to explaining public support for the National Day rituals. This will indicate what kind of nationalism is dominating the National Day celebrations and in what way the rituals are related to religion.
First, I want to focus on the position the National Day has in the population and see if support for it is declining. This will tell us about the potential for shaping people's views about the nation. Also, civil religion requires that the rituals are supported by a substantial and varied part of the population.
The two population surveys from 1994 and 2013 asked the exact same question about celebrating the National Day. Even if this represents a time span of almost 20 years, there are only small changes in the way people celebrate National Day. The number saying that they do not celebrate the day has gone up from around five percent to around 10 percent. But this means that about 90 percent of the population celebrates the day. It has become somewhat more common to only celebrate the day in private settings, and the number celebrating publicly has gone down from around 70 to around 60 percent of the population.
Another question that has been asked in several surveys relates to the national anthem. According to the survey from 2013, a majority of 53 percent takes part in the singing of the national anthem on National Day. This means that only a small proportion of the population participates in the public celebrations without singing the national anthem. When comparing the data from 2013 with data from 1998, we find that the number of people taking part in the singing has remained stable over time. According to a KIFO survey from 1998, 52 percent of the total population sang the National anthem that particular year. This is almost exactly the same figure we found in 2013, indicating stable support for the National Day rituals even if there has been a small increase in the number of people that only celebrate the day privately.
The 2013 survey also had questions asking the respondents what view they have of the National Day celebration. As many as 88 percent of the population state that they think it is very important or fairly important to celebrate the day. The figure is higher than the percentage taking part in the public celebration, indicating that some want to take part but are impeded for practical reasons. To obtain a more nuanced picture of the attitudes we also included items formulated in a negative way. This showed that between five and 10 percent of the population agreed with statements like "it is irrelevant whether one celebrates the day or not", "in Norway celebrating the national day is exaggerated", and "it is inappropriate in these international times". A reliability analysis shows that the four statements can be combined in an index measuring support for the celebration (Cronbach's alpha > 0.80).
The three variables described above (participation in public rituals, singing of National anthem, and thinking the celebrations are important) will be used as dependent variables in regression analyses of factors explaining participation in National Day rituals. The aim is to find out if civil religion or religious nationalism are relevant perspectives to use. Only one of the dependent variables is an example of civil religion, namely singing the National anthem. The other two variables measure participating in and support for the public celebrations. As part of the analyses, I will examine whether the dependent variables are related to such explanatory factors as ethno-cultural and political-democratic national identity, church involvement on National Day, patriotism, chauvinism, and xenophobia. In addition, ordinary sociological variables such as gender, age, and education are included.
As outlined in the theoretical section, there are different types of nationalism and national identity. Some of these are open and inclusive and others are more exclusive by focusing on ethnic and cultural elements. The respondents were asked what characteristics they think are important for being truly Norwegian. I have selected two items that represent different types of national identity. The political-democratic item is represented by the statement "to be truly Norwegian it is important that one participates in political elections". About 70 percent agrees fully or partly with this statement. The other statement is, "it is important to be a Christian to be truly Norwegian". This question represents an ethnocultural national identity by making a particular religious identity a prerequisite for being fully Norwegian. Only 13 percent of the population supported this statement. This focus on the religious aspect of national identity has declined since a similar survey was conducted by ISSP in 1994. Then around 20 percent of the population thought that being a Christian was important for being a true Norwegian. This indicator points more in the direction of religious nationalism since civil religion embraces people with different religious affiliations.
Church-involvement on National Day is measured by a variable describing the role of the local vicar in the celebrations. As mentioned above, 23 percent of the respondents state that the vicar in their local community actively takes part in the rituals. This variable is descriptive and does not ask if people think this is a good or a bad thing. However, a positive relationship between this variable and participation strengthens the argument for the civil religion perspective being relevant.
The 2013 survey deals with questions relating to national symbols, national rituals, and religion. Patriotism refers to a positive identification with the nation that accentuates the feeling of belongingness to the country one lives in. The concept of chauvinism is associated with the attitude that a state or its citizens is or are superior and better than others (Kosterman and Feshbach 1989). While patriotism is something positive, chauvinism is referred to as something negative. Xenophobia is another concept that is often mentioned in connection with national identity. The concept refers to anxiety and skepticism towards people from other countries than one's own (Knudsen 1997).
All these concepts are addressed in the survey. The survey questions I use to measure patriotism, chauvinism, and xenophobia have been used in other international studies, such as the ISSP studies from 1994 and 2003. According to factor analyses, three pairs of questions can be used to constitute distinct dimensions in the material. Reliability analyses produce Cronbach alphas in the range of 0.63 to 0.72. The only scale that has weak alpha scores is chauvinism (0.40). The following sets of questions or statements are used to operationalize the theoretical concepts of patriotism, chauvinism, and xenophobia. The respondents were asked to indicate the degree of agreement-disagreement with the statements.

Patriotism
I am proud to be Norwegian.

Chauvinism
The world would be a better place if people in other countries were more like us.
One should support one's country, even if it behaves badly.

Xenophobia
Immigrants are associated with crime.
Immigrants take jobs from people who were born in Norway.
By making indexes based on these pairs of questions we obtain variables with values from 1 to 9. The three indexes are positively correlated, with the strongest correlation between chauvinism and xenophobia (Pearson r 0.44) and the weakest between patriotism and xenophobia (r 0.25). By conducting regression analyses I will ascertain which factor contributes to explaining participation in and support of National Day rituals. Table 4 explains seven percent of the variance in the dependent variable. This reflects that only four out of nine variables have a significant effect. Churchinvolvement and patriotism have a positive effect on participation in public rituals. Xenophobia has a significant negative effect while none of the other variables related to nationalism have an effect. Of the sociological variables, only age has a significantly negative effect, meaning that the younger cohorts participate more than the older ones. This probably reflects the focus on children in the National Day rituals. Turning to the singing of the National anthem (Table 5), the model explains a little bit more of the variance. Six out of nine variables have a significant effect. Again, churchinvolvement and patriotism are the strongest predictors. This time, however, the variable indicating a political-democratic national identity also has a significantly positive effect. Young people more often than older people take part in the singing of the National anthem, and women are somewhat more active than men. The last regression table (Table 6) uses support for the National Day celebration as the dependent variable. This time the analysis explains as much as 25 percent of the variance in the dependent variable. However, only five out of nine variables have a significant effect. Patriotism is the one variable with the strongest explanatory power. Church involvement has a significant effect, as does political-democratic national identity. Again, the nationalism indicators have no effect. As in Table 5, women are a little more supportive than men when it comes to National Day rituals. This time, age has a positive effect, meaning that the older cohorts are more positive than the younger ones. This contradicts the tendency seen in Tables 4 and 5. While young people take a more active part in the celebrations, older people are more inclined to support the idea of publicly celebrating the National Day.

Summary and Discussion
In this article, I have looked at the role of religion in the National Day celebrations in Norway. These celebrations consist of a set of rituals being performed every year on the 17th of May in all Norwegian municipalities. The structure of the rituals remains the same in all local communities throughout the country. Only in a minority of the local celebrations does the church or the vicar in the local Lutheran church have a prominent role in the main rituals of the day. This does not mean that religion is absent from the celebrations, it is only more subtle and difficult to detect.
Some would say that the celebration of National Day, no matter how it is organized, has an element of civil religion in it. A strict way of defining civil religion requires that one looks for religious elements in individual events, such as speeches, songs, and so on. Following such an approach, considerable variation will be found between municipalities. In some cases, neither church nor vicars are visible in the official program. In others, worship services are closely connected to the parades. The phenomenon of civil religion occurs, for example, when the mayor of the municipality introduces the parish vicar as today's keynote speaker and the vicar talks about how God has preserved the nation throughout history. This type of direct link between religion and nation takes place in a few local communities without causing any form of public debate. In these local communities, the National Day celebrations constitute a field where religion is seen to be legitimate and acceptable. In other cases, religion is downplayed in the celebrations, and sermons in the local churches are not explicitly connected to the official program of the day. When religion in the National Day celebrations has been brought into the public eye, this is most often actualized by groups affiliated with the Humanist Association, as in the case at Kråkstad. It is difficult to see that religious actors have much to gain from making the National Day a matter of conflict. The empirical analysis suggests that religion in the National Day celebrations is a field that is not very controversial at the local level. Religion belongs primarily in the civil sphere and to civil-society actors.
Even if the role of the church over time has been downplayed, there are still religious elements in the rituals. This is, for example, the case with the collective singing of the National anthem in which God is mentioned. This makes it relevant to talk of civil religion in a strict sense. Other researchers use civil religion in a broader sense that incorporates the use of national symbols in rituals shared by a majority. The empirical analyses show that religion and nation are linked in different and often subtle ways, even in a secularized context such as the Norwegian one.
The data material I base this article on makes it possible to look at changes over time. Data on participation in the National Day celebrations show that only small changes have occurred between the 1990s and 2010s. Norwegians in general tend to participate in the public part of the celebration. This means that when religion is manifested in the celebrations it will cover most of the population. I have discussed two theoretical perspectives, the civil religion and the religious nationalism perspectives. According to the first, national symbols and rituals bind people together across faiths, while according to the latter, the religious are more inclined to nationalism than non-believers. Data from representative population surveys make it possible to see the extent to which participation in and support for National Day celebrations are related to various kinds of national identity and nationalism. It is possible to look for indicators of both in the data material. To help me interpret what characterizes participation in the public rituals on National Day I included a set of relevant variables in the regression analyses, such as patriotism, chauvinism, and xenophobia. In order to use the concept civil religion about the public rituals, chauvinism and xenophobia need to be toned down. An open and inclusive form of patriotism, however, can be seen as a necessary if not adequate condition for civil religion to be present. Religious nationalism is part of an ethno-cultural approach to the nation. Contrary to civil religion it is exclusive and focuses on Christianity as an aspect of being Norwegian.
There is more empirical support for the civil religion perspective than the religious nationalism perspective. Both the fact that a majority of the population takes part in public rituals containing religious elements, and the fact that most Norwegians think the celebrations are an important national event point in this direction.
The regression analyses also can be seen as support for the civil religion perspective. Variables that had a positive relationship with participation were church involvement, patriotism, and sometimes an inclusive form of national identity. The ethno-cultural approach to national celebrations did not receive much support. If this perspective were important, chauvinism, xenophobia, and religious nationalism would have had explanatory power. On the contrary, what we find is that these variables do not have significant impact on participation. In some cases, these variables were even negatively correlated with participation.
According to researchers, Norway is characterized by an ethno-cultural form of nationalism and xenophobic attitudes to a greater extent than, for example, Sweden (Østerud 1991b;Botvar 2009Botvar , 2016. While this may be correct, these tendencies are not dominating the National Day celebrations. On the contrary, the way National Day is celebrated may be seen as a vehicle for integration of minority groups in society. If one looks at the 200-year history of the National Day celebrations, from the beginning until today, new groups have gradually been included in them. In the beginning, young men from the middle class dominated before boys, then girls, were included. Later, the working class found their place in the celebrations before immigrants began to take part from the 1970s onwards. The struggle for inclusion of new groups in the celebrations contributes to the fact that many still find it meaningful to take part in them more than 100 years after Norway gained its independence. Funding: This research received no external funding. Data Availability Statement: Data used in this study is available from Norwegian Centre for Research Data (NSD) or from KIFO, Institute for Church, Religion, and Worldview Research.

Conflicts of Interest:
The author declares no conflict of interest.