Governing Religion and Gender in Anti- Discrimination Laws in Norway and Denmark

This article examines the decisions on religious and gender discrimination handed down by two quasi-judicial monitoring bodies in Denmark and Norway, mapping similarities and differences between the two bodies. While the monitoring bodies tend to arrive at similar results, their modes of reasoning and understanding of what constitutes ‘religion’ for legal purposes differ considerably. Looking in particular at the decisions on religious headgear and handshaking, the article suggests that these differences may be due to a range of different factors, from the legal framework on anti-discrimination in the two countries, to the staffing of the monitoring bodies, and the financial support available for their work.


Introduction
Religion has always been an anomaly in discrimination law. From its exclusion from the draft international convention on racial and religious discrimination that was discussed at the United Nations ('UN') in the 1960s and up to the present, claims of religious discrimination have given rise to doubts and quarrels among legislators, rights defenders and adjudicators alike. 1 Despite its status as an outlier in the field of non-discrimination law, religion has been high on the non-discrimination agenda in recent years as the plight of religious minorities has been highlighted across the world, ranging from the persecution of Ahmadis in Pakistan and Baha'is in Egypt, and to growing Islamophobia and anti-Semitism in Europe. This increased interest notwithstanding, the boundaries between religion and neighbouring grounds for discrimination such as race, ethnicity and national origin, and its potential for conflict with issues concerning gender, sexual identity and sexual orientation, have never been resolved.
Three issues in particular complicate the legal relation between religion and gender, and suggest the need to supplement normative legal analysis with empirical studies of a more socio-legal character. First, the lack of consensus on how to deal with religion legally is in no small part due to the widespread perception of religion in legal circles as an individual elective that has grown out of the emphasis on voluntary 'belief ' as the core and centre of religion for legal purposes. 2 This emphasis sits uneasily with the social scientific literature on the complexities of 'religion' , and has also proven to be troublesome in the development of religious minority rights. 3 Second, most religious traditions have historically distinguished between the rights, roles and positions of men and women in ways that conflict with current anti-discrimination law. Recent migration processes 4 have reopened old debates and initiated new ones in the so-called 'women-friendly' 5 Nordic societies. Secular legislation and jurisprudence on asylum, working, and residence permits and family unification may furthermore affect gender equality in ambiguous ways. 6 In contemporary societies, structured ambivalences surround gendered religious beliefs and practices, and formal ideals are often distinct from the actual performance of gender in families, adding further complexity. 7 Third, although the freedom of religion may be in tension with gender equality, religious values and practices enjoy a somewhat privileged position relative to values and practices legitimated by cultural traditions, which do not enjoy the same level of legal protection, even though religion and culture 'always intermesh' . 8 Taken together, these issues provide fruitful grounds for our analysis, which examines the relationship between gender equality and religious discrimination as it plays out in the monitoring of anti-discrimination laws in Denmark and Norway. The monitoring bodies in these two countries have decided numerous cases where questions of gender equality and religious discrimination have intertwined, highlighting the conflicted and fraught relationship between religion and gender. Until recently, the most controversial and closely watched cases on religion and gender have concerned the rights of Muslim women to wear headscarves (hijabs). Over the course of the last years, however, hijab cases have been joined by a range of cases where pious Muslims have refused to shake hands with people of the opposite 2. The main conceptual framework for legal discussions of religion in countries that have ratified the major international human rights treaties is the freedom of religion or belief  sex, so-called 'handshake cases'. The latter, while still far less numerous than hijab cases, represent an interesting case of boundary maintenance, both within religious minority communities, where the practice is only recognised and maintained by a small minority within the minority, and in larger society, where the practice is considered to be detrimental to gender equality. A number of questions thus emerge. How prevalent is the interrelationship of gender and religion in the caseload of the monitoring bodies? Which religious traditions and social arenas dominate? What are the outcomes of the cases, and what is the role of international human rights law? Which understanding of religion dominates at the Ombud/Board and the claimants? And do the answers to these different questions interact with one another? In the following, we seek to answer these questions.
Previous research on equality boards in Denmark, the Netherlands and Sweden has established that discrimination cases concerning gender and ethnicity and/or religious discrimination mainly regard complaints regarding harassment or discrimination filed by women wearing a hijab. 9 A comparative study of the different modes of the legal management of religious diversity in Denmark and Norway points to the political-judicial resolution of 'headscarf ' conflicts in Norway as a 'best practice' . 10 A key difference highlighted by the authors of that study is the different political roles of religion in the two countries: the actively supportive 'religious dialogue' politics of Norwegian authorities versus the generally dismissive policies of Danish authorities. The difference is evident in efforts of Norwegian authorities to actively single out religion as separate from other types of culture and as somehow recognisable, in contrast to how Danish authorities approach religion as indistinguishable from other 'cultural stuff ' . Next to this social science literature engaging with gender, religion and anti-discrimination law, a legal literature on religion, gender and discrimination has developed in both countries. 11 The aim of this article is not to contribute directly to this literature, but to stress the way decisions may be influenced by framings and contexts, and thereby further reflections on the function and workings of law.
Our analysis is not intended to provide a normative, doctrinal analysis of law and jurisprudence on gender and religion in Norway and Denmark. Rather, our aim is to tease out the ways in which these Norwegian and Danish monitoring bodies have navigated between international human rights law, domestic anti-discrimination law and their interpretation of the claims for recognition and protection lodged by claimants in the cases that we have examined. As such, our analytical approach is inspired by socio-legal studies, and in particular an understanding of human rights as a key contemporary, transnational discourse with numerous local iterations that require different modes of analysis rather than normative examinations. As such, we approach the work at the monitoring bodies as examples of 'social and ethical knowledge practices that appear in discrete places at discrete times with enough autonomy that they can be isolated analytically and studied in what is often described as their "local context"' . Our interest is to examine the influence of the differing local contexts offered by the Danish and Norwegian legal systems on the global register of human rights norms on religion and gender. This entails an examination of how global discourses on human rights are reinterpreted and given new meanings in the vernaculars of the local contexts in which they are given concrete, direct expression through legal decision-making. As such, we do not consider jurisprudential issues concerning precedent and consistency with earlier or future legal decision-making within the two countries. Rather, we use the decisions of the monitoring bodies as points of departure to examine the ways in which international human rights norms on religion and gender have been framed and translated into the local contexts offered by Norway and Denmark.

Legal Framework on Discrimination in Norway and Denmark
The analysis below takes its point of departure in decisions from relevant quasi-judicial monitoring bodies in the two countries. These bodies are the Board of Equal Treatment (Ligebehandlingsnaevnet) in Denmark and the Equality and Anti-Discrimination Ombud (Likestillings-og diskrimineringsombud) along with the recently established Equality and Anti-Discrimination Tribunal (Diskrimineringsnemnda) in Norway. Both countries have robust constitutional and statutory legal provisions in place to prevent discrimination, albeit with important differences. The Danish Constitution mentions equal treatment in a central clause: This clause is essentially unchanged from the first Constitution from 1849; however, 'descent' was added in 1953. Protection against general discrimination is not strongly present in the Danish Constitution, which remains unchanged since 1953.
In contrast, a much broader protection was included in § 98 of the Norwegian Constitution in 2014: § 98 All people are equal under the law. No human being must be subject to unfair or disproportionate differential treatment. to the list, 14 yet the law interestingly mentions belief, but not religion. The law also does not consider unintended behaviour. Protection against discrimination on the basis of gender was legally sanctioned from 1978 with the Act of Gender Equality. In the period between 1978 and 2000, violations of this law could be tried before the Gender Equality Council (Ligestillingsrådet), and between 2000 and 2009 before the Gender Equality Board (Ligestillingsnaevnet). A law specifically directed towards differential treatment on the labour market, the Act on the Prohibition of Differences of Treatment in the Labour Market, was introduced in 1996. According to this law, an employer may not treat employees or job applicants differentially whether at the time of employment, dismissal, transfer, promotion or in terms of pay and working conditions. 15 The law covers differential treatment with reference to race, skin colour, religion, political views, sexual orientation, age, disability or national, social or ethnic origin. An amendment introduced in 2004 adds belief 16 after the reference to religion. The amendment was introduced to ensure alignment with EU Directive 2000/43/EC on anti-discrimination. The definition stated in the bill was that 'belief ' is broader than religion and covers something which is clearly defined yet not restricted to the recognised religious communities. 17 Violations of the law are tried in the general civil rights system.
The 2003 Act on Ethnic Equal Treatment concerns all public and private enterprise in the field of social protection, including social security and health care, social benefits, education, and access to and supply of goods and services, including housing available to the public. 18 This law protects against differential treatment due to race and ethnic origin, but not discrimination due to religion or belief. 19 The background to the adoption of this law was concerns about integration and attention to inclusion on the labour market. When the law was discussed in parliament, the spokesperson from the Social Democrats, Lise Haekkerup, directly stated that '[w]e need the employers to take the responsibility to ensure the integration of all minorities on the labour market and the first step is to end discrimination' . 20 The Complaints Committee for Ethnic Equal Treatment (Klagekomitéen for Etnisk Ligebehandling) was established to handle complaints in relation to this law and worked to this effect from 2004-2009. The Board of Equal Treatment was established in 2009 through the merging of two boards, the Gender Equality Board and the Complaints Committee for Ethnic Equal Treatment. Overall, the Board of Equal Treatment hears complaints in accordance with the Act of Gender Equality, Act on Equal Pay to Men and Women, Act on Equal Pay to Men and Women under Occupational Security Schemes, Act on Entitlement to Leave and Benefits in connection with Childbirth and Act on the Prohibition of Differences of Treatment in the Labour Market ( §1.2; §1.3). It is stated that the Board should furthermore consider com-plaints according to the Act on Ethnic Equal Treatment ( §1.4). The Board considers discrimination in general for some types of discrimination, while limiting its field of expertise to the labour market in others: This construction, where the Board adjudicates according to both the Act on the Prohibition of Differences of Treatment in the Labour Market (which includes religion and belief) and the Act on Ethnic Equal Treatment (which does not), complicates the Danish support for non-discrimination on the basis of religion, as the resolution of the issue will depend on which of the relevant legal provisions a case is considered under.

Norway: The Equality and Anti-Discrimination Ombudsman and Tribunal
The Norwegian Gender Equality Ombud, the predecessor to the Equality Ombudsman and Tribunal (nemnd) was created with the adoption of the 1978 Equality Act, which mandated under § 9 that an office to monitor the implementation of the Act should be established. Unlike the Danish system, the mandate of the Norwegian Ombud has never been restricted to employment and education in regard to religion, but has been applicable for all walks of life, as mandated by § 2 of the Act. Figure 2: Equality and Anti-Discrimination Act (Norway, 2018) § 1 Discrimination on the basis of gender, pregnancy, leave in connection with childbirth or adoption, care responsibilities, ethnicity, religion, belief, disability, sexual orientation, gender identity, gender expression, age or combinations of these factors is prohibited. 'Ethnicity' includes national origin, descent, skin colour and language While the original Act only concerned gender equality, parallel legislative changes in the Penal Act and the acts on housing cooperatives and the working environment have increased the remit of protections against discrimination to a wider range of people. 21 This catalogue has gradually been expanded to go well beyond the group of protected persons in human rights treaties like the European Convention on Human Rights (ECHR) and the International Covenant on Civil and Political Rights (ICCPR). 22 21. The 2005 Penal Act prohibits hate speech ( § 185) and the refusal of goods or services ( § 186) based on skin colour or national or ethnic origin, religion or life stance, homosexual orientation, or reduced functional capacity. The 2003 Housing Cooperatives Act § 1-5 prohibits bylaws that discriminate on the basis of ethnicity, national origin, extraction, colour, language, religion or view of life. The 2005 Working Environment Act § 13-1 prohibits discrimination on the basis of gender, pregnancy, leave of absence in connection with childbirth or adoption, care responsibilities, ethnicity, religion, belief, disability, sexual orientation, gender identity or gender expression, referring the matter to the Equality and Discrimination Act. 22. Article 14 ECHR prohibits discrimination on the basis of sex, race, colour, language, religion, political or other opinions, national or social origin, association with a national minority, property, birth or other status. Article 26 ICCPR prohibits discrimination on the basis of race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.
Despite a broad range of protections across the legal framework, Norwegian law lacked a specific act against discrimination until the passing of the Discrimination Act in 2005, which also inaugurated the present office of the Equality and Anti-Discrimination Ombudsman and Tribunal. After a brief period of amendments, including a short-lived tripartition of the Act into three different pieces of legislation from 2013, a new and more comprehensive Anti-Discrimination Law entered into force in 2018, replacing both the Equality Act and the earlier discrimination acts.
From 1978 to 2018, the Ombud could make decisions, both acting on complaints from the public, and acting on its own initiative, with the Tribunal as a more ad hoc institution to manage appeals against the decisions of the Ombud. With the new law in 2018, however, the Ombud was stripped of its decision-making capacity, rendering it an office for information and consultation. The Tribunal, on the other hand, has been expanded and professionalised to become a court-like entity.
The overall change in the two countries seems to suggest a cross-border convergence between the equality bodies as the Danish Board of Equal Treatment is a '[p]redominantly tribunal-type (quasi-judicial) body' , 23 which co-exists with the 1987-founded Danish Institute for Human Rights, which is a '[p]redominantly promotion-type and legal support body' 24 similar to the (post-2018) Norwegian Ombud. This division of labour means that the Board of Equal Treatment is not engaged in advocacy or awareness-raising activities, but only focuses on deciding formally on complaints, similar to the Norwegian Tribunal. In this article, however, we mainly consider cases from before 2018 and decisions from both the Tribunal and the Ombud are therefore covered. While these cases may be of limited jurisprudential significance for a normative legal analysis, we believe they constitute important sources of data on the 'vernacularisation' of international human rights norms in the two different settings.

Discrimination cases at the Board of Equal Treatment (DK) and the Ombud and Tribunal (NO)
Both institutions maintain websites where their decisions are available for full text search, albeit with a fairly rudimentary sorting function. A search in the database of the Danish Board of Equal Treatment (2009-2019) identified 40 cases categorised under the headline 'religion og tro' (religion and belief). Yet, three cases appear to be mal-categorised, as they do not regard religion or belief at all. An open search on 'religious' yields a number of additional cases mentioning religion, of which some should meaningfully be included in the sample because they concern religion even if religion is not singled out as the basis of discrimination. Omitting duplets (20) the total number of cases is 45. As the total number of cases being tried in the Board of Equal Treatment in the period is over 1600, religion is not a very common topic in the Board of Equal Treatment -only about 4 % of the cases deal with religion.
The most common topic for the Board is gender discrimination, which is one of the grounds of discrimination in almost 60% of the cases. Discrimination cases on age constitute about a third of the cases, while cases on disability constitute almost 20%. About 12% of the cases deal with discrimination on the basis of ethnic background. This number shows that cases on religion are not among the most common topics addressed by the Board even if 23. European Network of Equality Bodies (Equinet), 'Danish Institute for Human Rights' in Equineteurope.org <http://equineteurope.org/author/denmark_dihr/> accessed 10 July 2020. 24. Ibid. cases regarding religion are more common than cases relating to sexual orientation, political views and social background.
The number of cases involving religion generally varies between 1 and 7, with 1 in 2010 and 2 in 2018 as the lowest and 7 in 2017 as the highest. The majority of cases (26) concerns discrimination against Muslims, but there are also cases concerning Jews (4), Jehovah's Witnesses (1) and Seventh Day Adventists. The cases raised by a member of Jehovah's Witnesses and a member of Seventh Day Adventists were both assessed in favour of the claimants, who had both been fired for failing to participate in workplace activities (Christmas celebration and an open house on a Saturday, respectively). None of the cases brought forward by Jewish claimants, who had either been fired or been met with hateful attitudes in a shop, were successful because the Board could not interview witnesses that were able to substantiate their claims. Cases where claimants allege that they have been fired or discriminated against during job interviews on the basis of religion have generally not fared well, except for the two cases mentioned above.
The Board has clearly stated that cases can only be raised by the specific person (or someone representing that person) who has experienced discrimination. For instance, the Board rejected a complaint by a woman stating that a supermarket allowed Muslim female employees to wear a veil, but did not allow kippa or turban (J no 7100201-13). A similar complaint from the same year had a clear political agenda, expressing dismay that an imam could become a lay judge, as pastors within the Lutheran Church of Denmark cannot hold this position (J no 2016-6810-62073), implying that this constituted differential treatment. The case was rejected. A case brought forward by an organisation questioning the legality of a Christian organisation to require employees to be Christians (J no 2500080-11) was dismissed as there was no specific person whose rights has been infringed. 25 In Norway, the Ombud processed between 150 and 200 cases each year from 2007 to 2018. The website of the Ombud does not feature a sophisticated search function, only hyperlinks to each case, which is presented with a title of the main subject matter. Based on a manual count, a total of 48 cases concerning religion have been identified. These cases include instances where religion has been one of several grounds, and instances where religious organisations have been accused of discriminatory treatment. The distribution between different religions is quite different from the Danish material, with 17 cases concerning Christianity, 25 cases Islam, one Sikhism, two humanism/atheism, two veganism and one unspecified. Interestingly, the majority of cases featuring Christianity concerns hiring practices in Christian organisations, whereas the majority of labour-related cases concerning Islam have been lodged by Muslims who have claimed restrictions on their religious practice. Of the 29 cases at the Tribunal, 23 are complaints from earlier Ombud decisions.

Anti-discrimination cases with religion and gender claims
In both countries, a number of cases concern religion and gender, although never as a matter of compound or intersecting discrimination, where different discrimination grounds are invoked in tandem, in turn leading to 'intersectional' discrimination where distinctive discrimination grounds reinforce one another. 26 Rather, cases where gender and religion appear simultaneously are framed in oppositional terms, pitching gender equality and religious discrimination as opposed and antagonistic processes. Throughout the case load in 25. The Board argues that the law limits the access to file a complaint to victims of unequal treatments. 26. The term 'intersectionality' is fraught with conceptual and interpretational ambiguities. In this setting, we use it both countries, gender equality tends to be considered an overarching societal value that should be cherished and protected, while allegations of religious discrimination are presented as a topic of 'special interest' that must yield whenever it risks leading to gender discrimination. Despite this overarching similarity, the way that the two equality bodies approach the interconnection between religion and gender differs in several ways.
The key difference is the types of claims presented in the cases: in Norway, the clear majority of cases featuring religion and gender concern the wearing of the hijab, confirming the findings of Jonker and Halrynjo that this topic still dominates anti-discrimination litigation (see above). The Danish material, on the other hand, features only one clear case on the hijab, and is overall more thematically diverse and often legally quite amorphous. Two cases (of which one was categorised as concerning gender and religion, the other not) concerned a job interview (J no 18-1734) and a job advertisement (J no 20146810-06223) in a Christian social work organisation and in the Church of Denmark respectively, were both tried as cases of gender discrimination. A third case (J no 7100188-13), where a student felt discriminated against by a Muslim grader, who for religious reasons would not shake her hand, is discussed further on in this article.
Strikingly, the Danish Board never goes directly into the details of what constitutes 'religion' for legal purposes: in a case concerning a woman being relocated on an airplane, due to a monk not wanting to be seated next to the woman, the Board just remarked that this is 'due to his religion', but provided no elaboration (J no 2016-6811-38330). The Board decided -with a split vote -that the differential treatment of the woman was legal.
The Danish Board also makes very few references to international law in general and none in the religion cases. In contrast, references to international case law are quite common in the Norwegian material. The fact that the Danish Board is reluctant to isolate religion or belief as an independent ground for discrimination, and even more reluctant to refer to national or international case law, sets the comparison of cases from the two countries in an interesting theoretical perspective: will the verdicts from Norway, which isolates religion as a separate ground of discrimination and places discrimination in the context of international case law, offer a better protection of religious freedom, or will the opposite be the case?
A similarity between the two countries is that there are cases concerning the wearing of hijab as well as a reluctance to shake hands with persons of the opposite sex for religious reasons in both countries, though the numbers differ. The major differences between the two countries best appears through a closer assessment of the reasoning and discussions in the different cases. The next two sections will therefore provide a comparison of the hijab cases and the handshake cases in the two countries.

Hijab cases
Discussing the hijab as a 'paradigmatic symbol of intersectionality' , 27 Halrynjo and Jonker argue that the pre-2009 existence of two separate equality boards in Denmark, the Gender Equality Board and the Complaints Committee for Ethnic Equal Treatment, meant that disin the same way as Kimberlé Crenshaw in her analysis of how women of colour experience different and more severe forms of discrimination than members of the majority population because they belong to two subgroups crimination on the basis of wearing the hijab was construed as either an issue of gender or race/religion -but not both. Yet, this did not prevent hijab cases from being 'recognized as discrimination' . 28 However, with the establishment of the Board of Equal Treatment, it was supposed to become easier to address intersectional discrimination. Though often presented as the most prominent example of the intersection between religion and gender, there is only one Danish hijab case and this was only considered as a potential case of religious discrimination, not of gender equality. In this case, the applicant was asked during a job interview whether she would be willing to take off her veil while at work (J no 7100083-12). She refused, and was subsequently not hired. The company was not convicted, even though the Board has consistently sanctioned questions about religion in job interviews in its earlier case law (cf J no 2016-6810-02947; J no 7100072-12). In this case, however, the question from the employer if the applicant would consider removing her veil was considered legitimate, as the issue was stipulated in a pre-existing clothing regulation. In this decision, the Board referred to the so-called 'Føtex decision' , a Danish Supreme Court verdict handed down in 2005. 29 In this famous case, a Muslim woman working in a supermarket (Føtex) was fired when she decided to start wearing a hijab. She had already been working in Føtex for some years. She was offered another position in the company without customer contact, but was fired when she declined. The Court found that the supermarket was entitled to fire her, as (relatively vague) company clothing regulations, based on a company wish that employees appear neutral, justified a ban. The case set a strong precedent for private employers to institute a formal dress code, which the employees are not eligible to be exempted from on account of their religion. There have been no cases regarding public institutions, but a 2008 Amendment of the Administration of Justice Act restricts the right of judges to wear anything that reveals their political or religious affiliations or conviction. 30 Similarly, a Danish police officer is not allowed to wear a hijab or other religious, political or sexual symbols. 31 The Føtex case is not just a landmark case in Denmark, but also seems to have provided strong legal precedent and is interestingly also referred to by one Tribunal case from Norway, where the Føtex decision is cited as legalising the prohibition of the hijab as part of the uniform for employees (case 2/2014, see below).
In the Norwegian material, cases concerning the hijab, which tend to highlight the tensions between gender equality and religious freedom, illustrate the extent to which Norwegian discrimination law is interwoven with international law. In case 07/627, an employee in a mall lost her job because she wore a hijab. According to management, the issue was not one of discrimination, but a general policy on headgear that was the same for everyone, regardless of religion. Unlike the Danish cases cited above, the Ombud found the dismissal to be discriminatory on the basis of religion. Discussing the claim that the headgear ban was 'unrelated to religion' , the Ombud referenced the Leyla Sahin case (elaborated below) to observe that it was for the claimant to determine what motivated her to wear the headgear in question. Additionally, the Ombud observed that it 'evidently' could not discuss the nuances of the religious obligation to wear headgear. The decision was later upheld by the Tribunal. The case of Leyla Sahin v Turkey (Application no 44774/98) is one of the most paradigmatic and often-cited decisions of the European Court of Human Rights. In this case, the Court allowed a prohibition of religious headgear in Turkish universities, citing the strong tradition of secularism and the 'margin of appreciation' of States in contentious issues with no European consensus as its main grounds. While the claimant lost the case both before the regular Chamber and the Grand Chamber, the Court in both instances recognised her autonomy in deciding for herself what constitutes a 'manifestation' of her religious belief -a principle that has been widely cited in later jurisprudence.
The importance of this principle is evident in several subsequent cases before the Norwegian Ombud. In case 08/363, a bus driver who wished to use a regular turban for religious reasons was required by his employer to use a small turban in combination with a hat. Citing its decision in case 07/627, the Ombud declared that the religious motivation of wearing the turban was to be determined by the claimant, and that it had no competence to review his reasons for doing so. Finding no reasonable cause to refuse the wearing of a regular-size turban, the Ombud found that the bus company had violated the law.
In case 08/1528 concerning the use of religious headgear in the Norwegian police force, the Ministry of Justice argued against such use, highlighting the political overtones of using the hijab. Citing Leyla Sahin once more, the Ombud dismissed the argument, finding that it was for the claimant to determine the motivations for her choice to wear the hijab, and faulting the Ministry of Justice for not elaborating the necessity of a ban. In its consideration of the case, the Tribunal, in decision 8/2010, supported the Ombud, but reasoned differently on the precedent set by Leyla Sahin. Citing the importance of the margin of appreciation for the ECtHR's decision, the Tribunal observed that 'no clear resolution of the issue' was available from the international materials. With this starting point, the Tribunal went on to discuss the interwoven identities signified by wearing the hijab, observing that it could have ethnic, cultural and political connotations. On the basis of this reasoning, the Tribunal underlined its greater 'understanding' for potential high risk situations developing on the basis of police officers wearing the hijab. Despite this extended understanding of the potential downsides to allowing the hijab, the Tribunal supported the Ombud's assertion that the decision violated the discrimination law.
In case 26/2009, the Tribunal found that a laboratory worker had been discriminated against when she was asked to quit her position following her decision to start wearing a hijab at work, thereby upholding a similar decision by the Ombud. 32 Citing Leyla Sahin, the Tribunal avoided any further discussion of the religiosity of wearing the hijab.
In case 13/1837, the clothing regulations of the security company G4S were found to be discriminatory because they prohibited the wearing of religious headgear. Unlike its earlier decisions, the Ombud did not reference international precedent in its decision, settling with the travaux préparatoires of the Norwegian Anti-Discrimination Law, which observes that wearing turbans or hijabs constitute manifestations of religion for certain groups in Norway. The company appealed the decision, and in decision 2/2014, the Tribunal overturned the findings of the Ombud. In its decision, the Tribunal went to some length in its consideration of international materials, citing both the EU anti-discrimination directive, which is not binding for Norway, and the obligation for State Parties to the Convention on the Elimination of Discrimination against Women (CEDAW) to take steps to prevent gender discrimination (see Articles 2(f) and 5(a) CEDAW). Citing the possibility to limit religious 32. We have been unable to retrieve the full decision of the Ombud. manifestations in Article 18(3) ICCPR, the Tribunal went on to cite a selection of the hijab decisions of the ECtHR. Notably, the Tribunal highlighted 'the stereotypical perceptions of gender roles that the use of the hijab formally [sic] rests upon' .
In its assessment of the rationale for the hijab ban, the Tribunal stressed the physical interaction between security officers and travellers, and the potential for a gender-segregated security control as a corollary of allowing the hijab, as the 'indication of women's Islamic faith by wearing the hijab may result in greater conflict situations than today' . Continuing its reasoning, the Tribunal observed that allowing the hijab as part of the uniform 'may transmit signals of acceptance of a particular perception of gendered decency in public space'. Crucial to its balancing act on whether the ban was proportionate, the Tribunal returned to its self-constructed argument that the security control would have to be divided on the basis of gender, which would violate the objectives of the Equality Act.
In decision 15/2036, the Ombud found a hairdresser to be in violation of the law for her dismissal of a customer because she wore a hijab. Without further discussion, the Ombud found the rationale for wearing the hijab to be a manifestation of religion. As in earlier cases to come before the Ombud, the 'actual' meaning of the hijab was questioned, as the claimant alleged that she viewed the hijab as a political, rather than purely religious symbol, and that her refusal to allow customers wearing the hijab to enter her salon was protected by her freedom of expression. The Ombud did not address these issues, and found her refusal to be in clear violation of the law. The case was later brought to trial under the penal law, and the hairdresser was fined.
In decision 16/2271, the Ombud found that a private nursing home violated the law when they adjusted their dress regulations to prohibit 'religious or political garments, such as the hijab or Palestinian scarves' . The ban was intended to provide protection to patients suffering from dementia, who could have difficulties with new impressions. Because the prohibition was of a general nature, prohibiting all religious garments, it constituted indirect discrimination, whereby a seemingly neutral rule disfavoured a specific group. Referring to Leyla Sahin, the Ombud observed that while some may choose to wear headgear such as the hijab or the turban for other reasons than religion, most do so for religious purposes. On this basis, moreover, the Ombud pronounced that wearing the hijab was 'not only a way to indicate religious allegiance, but constituted a manifestation of religion in its own right' .
Following the appeal from the nursing home, the Tribunal came to the same conclusion in its decision 2/2017. In the appeals round, the nursing home stressed that 'religion' in the framework of the discrimination law was distinct from 'manifestation of religion' , which could not possibly enjoy the same protections as 'religion' . This claim was unilaterally dismissed by the Tribunal, which observed that religious manifestations were covered under the law. In its decision, on the other hand, the Tribunal was split in a majority and a minority (4-1). The majority found no clear guidance from the case law of the ECtHR or the CJEU, and found no legitimate rationale from the nursing home for the ban. The minority, on the other hand, cited the ECtHR decisions Eweida and Ebrahimian, and earlier decisions of the Tribunal and Ombud that stressed the need for a 'certain corporate image' and neutrality, and found the hijab ban to be well within the confines drawn up in these earlier decisions. 33 33. The cases of Eweida and others v The United Kingdom (Applications nos 48420/10, 59842/10, 51671/10 and 36516/10) and Ebrahimian v France (Application no 64846/11) both concerned the extent to which the right to manifest religion or belief by wearing religious symbols (the cross in Eweida, the hijab in Ebrahimian) could legitimately be limited through dress code requirements in formal work settings. In Eweida, the Court rejected the prohibition on wearing a cross as a British Airways employee, while in Ebrahimian, the Court accepted the prohibition for a schoolteacher to wear the hijab. The crucial point in both cases was the specific setting for the manifestation of religion or belief.
In decision 17/623, the Ombud found that prohibiting religious headgear that covers the ears does not constitute illegal discrimination on the basis of religion, as the requirement was based on legitimate concerns with secure identification of passport holders, and was not disproportionate. Referencing the considerable margin of appreciation that dominates the jurisprudence from the ECtHR, the Ombud observed that the applicable margin for Norway was 'uncertain' in comparison with more clearly secularist countries like France and Turkey, but that it was still likely to be fairly broad. Additionally, the Ombud found that its competence when faced with decisions and regulations from expert organs was limited. Reviewing the decision, the Tribunal came to the same conclusion in its decision 65/2018, without entering any new premises.
In case 30/2018, the Tribunal determined the case without involvement from the Ombud, which had been stripped of its decision-making powers under the new law from 1 January 2018. The case concerned a temporary employee in a company that provides staff lodging in ships, who alleges she was fired after starting to wear a hijab. Observing that it was 'beyond doubt' that wearing the hijab is encompassed by 'religion' under the law, the Tribunal found that the company had discriminated against the employee on the basis of her religion, as no legitimate grounds for the ban had been offered.
Taken together, the Norwegian case law is clear evidence that the hijab issue continues to be a paradigmatic case of the intersectionality between gender and religion. Unlike in the Danish setting, where the Føtex decision appears to have resolved the issue once and for all, the hijab, and the motivations for wearing it, continue to be contested in the case law of the Ombud. Whether or not the cases are indicative of a larger pattern in the governance of religious practice in Norway, or if they are simply the result of lacking decisions from a higher level of the court system, however, is difficult to determine on the basis of the available evidence.

Handshake cases
As in other Northern European countries, 34 the reluctance of some Muslims to shake hands with persons of the opposite sex has been a controversial public issue both in Denmark and Norway. While in no way a widespread practice, some abide by it, and this practice tends to generate quite strong reactions. In the Netherlands, for instance, a public dispute evolved in 2004 after an imam from the small town of Tilburg refused to shake hands with the Minister of Immigration and Integration. This was interpreted as implying that she was not 'equal' to him. 35 A similar interpretation formed the basis of the refusal of the then head of the Senate in Belgium to meet with Iranian parliamentary delegates in 2005. 36 Overall, the refusal of cross-gender handshakes is considered a break with two principles of Western European public spheres: first, a lack of loyalty to national customs, and second, insistence upon gender difference, contradicting the 'universal' and 'abstract' subject fostered by liberal-secular imaginaries. 37  ted that he had 'great difficulty' accepting the ruling. 39 This reflects a general reluctance to accept the rulings from the Dutch Equal Treatment Commission (especially the concept of 'indirect discrimination') and a tension between the Commission's interpretations of legal equality and the popular conception of equality.
In Denmark, the mandatory handshake at the so-called 'Constitutional ceremony' for new citizens (Law no 1735 of 27 December 2018) has generated debate, while in Norway, controversy arose in 2019 when three female representatives of a mosque refused to shake hands with the crown prince, who was visiting a mosque after a shooting attack. 40 There have been five cases dealing with handshakes in Denmark, and three cases in Norway. Two of the Danish cases have been widely discussed in public. The first of these concerns a Muslim taxi driver, who was de-selected from picking up clients from the airport because he did not want to shake hands with his female supervisor. The taxi driver explained his reluctance to be 'just a religious thing' , but the Board treated the violation as potential discrimination on both religious and ethnic grounds. The complainant won the case (J no 2015-6810-19566).
The second case regards a complaint from a student who had attended an exam, where the grader had announced that for religious reasons he would not shake hands with female students. The teacher therefore sent a message to the students stating 'Our grader has informed me that he does not want to shake hands with women. So, it would probably be a good idea not to try to shake his hand. In any case, you are now warned so you will not be confused about it in the exam situation' (J no 10274-14). One student filed a complaint to the Board on the grounds that 'in an exam situation, it is the performance which needs to be judged, and considerations regarding the student's gender, appearance, race or political conviction should not be expressed, as it has nothing to do with academic performance. Therefore, in order to protect the student, a grader must be neutral in the examination situation, regardless of the private attitudes he may have. An examiner should not bring his or her own opinions and views into the examination, but must leave these in private' (J no 10274-14).
The student received a compensation of DKK 2500. The Board treated the case as a case about gender discrimination but did not also consider it as a question of the freedom of religion of the grader.
A third case regards a female childcare student who was dismissed after the initial meeting in her internship because she announced that she did not shake hands with men (J no 2015-6810-01519). The Board could not adjudicate this case due to insufficient information.
The other two cases in the Danish material deal with the handshake in a larger context. One case regards an interpreter who was dismissed because he mixed his translations from Arabic with general counselling regarding handshakes (Muslim men should not shake hands with women) and swimming (Muslim girls cannot show themselves in swimming suits) (J no 2016-6810-36441). A similar case regards a Muslim social worker who was dismissed on the basis of complaints from three of his clients, who found him trying to impose Islam upon them, including the restriction on cross-gender handshakes (J no 18-55315). The Board did not consider whether the firing was legitimate, but assessed that, if legitimate, it was not discriminatory because the firing was due to malpractice. Even if the handshake has come up in several different cases in the Danish material, the Board tends not to consider it from the perspective of freedom of religion, and there is no reference to other cases and none to international jurisprudence. The cases are not seen as connected and there is therefore no 'handshake policy' being developed.
In Norway, on the other hand, the handshake cases have been dealt with over a short period of time, and in a strongly comparable manner. In decisions 108/2018 and 48/2018, the Tribunal decided two cases concerning the refusal of a temporarily employed teacher to shake hands with women for religious reasons: case 108 concerned the decision of the Norwegian welfare authority to reduce the teacher's social support because of his refusal to shake hands with women meant he was unable to accept a position he was offered, while case 48 concerned the decision of a school in Oslo not to prolong his contract because of his refusal to shake hands with women. Both decisions were split (3-2), but case 108 found in favour of the claimant, while case 48 found in favour of the school.
In both cases, the Tribunal cited Eweida to stress that not all outward expressions of religion are protected under the ECHR, and discussed whether the refusal to shake hands with women should be viewed as a 'cultural' or religious precept. Citing a recent decision from the Swedish labour court where a court-appointed expert observed Quranic prescriptions and later interpretations on cross-gender contact, the Tribunal unanimously found the practice to be a manifestation of religion. In case 108, the majority found that the decision to reduce his social support was directly related to his religious manifestation, and as such constituted discrimination. The minority, on the other hand, found that the decision to reduce his social support was not based on his religious manifestation, but rather on his refusal to take a paid job.
In decision 48, on the other hand, the relationship between his religious manifestation and the discontinuation of his contract was not disputed by either the majority or the minority. The majority found that the aim sought by the school -gender equality for all -was legitimate, and that the discontinuation of his contract was not disproportionate to this aim. Central to the reasoning of the majority was the assertion that the limitation 'did not violate the foundations of the religion, but a violation of what the majority perceives to be a limited and not particularly central expression of the claimant's religious opinions'. Moreover, the majority hypothesised that accommodating 'conservative practices within a religion may be unfortunate for the desired development of society' . The minority, on the other hand, found the discontinuation of his contract to be directly discriminatory, and not pursuing a legitimate aim with proportionate measures.
In decision 18/325, the Tribunal found that a job applicant whose job interview was cancelled because he refused to shake hands with women had not suffered religious discrimination. In a split (2-1) decision, the Tribunal cited its earlier assertion that refusal to shake hands with women constituted religion under the law, and that it was this factor that led to the termination of his interview. The majority observed that the decision to terminate the interview was based on internal ethical rules in the company, and sought gender equality as a general aim. The majority distinguished between 'limiting his religion as such, and limiting an outer expression of religion visualised through the handshake, which can be said to constitute only one aspect of his religious expression' , and other discrimination grounds like gender, which 'are physically impossible to "lay down" during working hours' . The majority also cited the concerns for a 'desired development of society' from decision 48/2018 (see above). The minority, on the other hand, echoed the minority opinion from decision 48/2018, observing that the termination of the interview was directly discriminatory, and lacked a legitimate aim, in particular because the company had not explored how the claimants' religious manifestation could have been accommodated. The Norwegian cases are narrowly decided (3-2, 2-3, 2-1); the centrality of the practice of not shaking hands with persons of the opposite gender is accepted, but also called into question. Elusive notions of gender equality are promoted as legitimate grounds for limitation. The cases cite 'societal development' as a legitimate ground for limitation.

Discussion
The comparison between the Norwegian and Danish cases indicates several differences, ranging from the competence of the monitoring bodies, their institutional resources, and the distribution of cases between the world religions, to the themes dealt with in their case law and the outcomes of the cases. These differences, however, are fairly minor compared to the overarching difference in approach of the two equality bodies: unlike the Norwegian Ombud and Tribunal, which appear to consider 'religion' in discrimination law to be synonymous with the freedom of religion or belief, the Danish Board does not seem to employ a clear concept of religion, or even to find the delineation of the concept necessary to decide the cases that come before it. 41 The Board appears to be conflating religion and ethnicity, as in sentences like 'the defendant deliberately deselected applicants of different ethnic origin and women with scarves' (J no 2017-6810-59283). Also, unlike the Norwegian material, there is no reference to international law and very little cross-referencing. Overall, except for references to the Føtex case, there is no religion law in the Danish material. 42 The question is how this can be explained. There are several possible answers to this question. The differences could be due to the ways cases of discrimination are handled in Norway and Denmark. Where the Norwegian system has worked towards simplification, the Danish system of handling cases of discrimination has developed towards an even more complex structure, managing parallel and partly overlapping provisions on discrimination. As mentioned above, the Board of Equal Treatment was established in 2009, but the civil court case system has continued to be relevant to cases of discrimination, as have the trade unions, which have historically been central for resolving issues on the labour market through the so-called Danish model. 43 The use of juridical and quasi-juridical bodies to resolve conflicts on the labour market is therefore a rupture with traditional ways of conflict-solving in Denmark, where conflict mainly has been resolved through non-legal means and conflict resolution. This approach has been described as less principled and quite pragmatic, 44 and it fits with the general Danish model of conflict resolution outside the court system as well as well with the decentralised structure of the Danish welfare state. 45 A report from Equinet Europe states that one of the barriers for bringing charges in front of the Board of Equal Treatment in Denmark is a lack of knowledge about rights and failing confidence in authorities. Yet, compared to Equality Boards in other European countries, the Danish population is quite well acquainted with the Board (known to 65 % of respondents), being second only to Ireland (67 %), 46 so perhaps the critical issue is the legitimacy of conceiving religion and gender in terms of rights.
The political establishment in Denmark has not been entirely happy with the decisions made by the Board. There were discussions in 2015 to change the Board of Equal Treatment due to political discontent with it, which, according to the politicians, took up 'silly cases', but the threatened change did not go ahead. 47 Because of the general opinion that religions are not equal in Denmark, as reflected by § 4 of the Danish Constitution, and widely stated by Danish politicians as 'Freedom of religion does not equal equality of religion' , it may not be possible to discuss differential treatment between religions as instances of discrimination.
Additionally, the work of the Danish Board of Equal Treatment is not financially prioritised. The Board consists of twelve members, of which the chairperson and the two deputy chairpersons are judges, while the nine additional members are lawyers/professors with expert knowledge on labour market regulations, discrimination and EU law. None of these, however, are full-time and the only permanent staff at the Board is the secretaries.
The weak institutionalization of the Board of Equal Treatment arguably reinforces people's reluctance to bring cases before the Board, as it exacerbates the reasons that people often raise for not bringing cases, which include the constraints on the production of evidence and a general reluctance in Denmark towards enforcing anti-discrimination law in practice. 48 The Norwegian anti-discrimination system, on the other hand, is far more potent than the Danish system, and enjoys cross-political support for its work. While the decision to remove the case-deciding abilities of the Ombud has been criticised, 49 the system remains considerably more developed than its Danish counterpart, both in terms of available resources and staffing, and in terms of the quality of its decisions. This difference is particularly apparent in the tendency of the Norwegian system to act as a 'court-like' entity, developing procedures, principles of interpretation, and standards of proof that closely resemble that of the formal court system. This development may contribute to the relatively unclarified precedent set by the Board and Tribunal: because of the wider discussions of legal provisions and both domestic and international precedent offered by the Norwegian decisions, claimants may be more reluctant to bring their case before the formal legal system. This may in turn prevent the kinds of final clarification of troublesome interpretational issues offered by the Føtex decision in Denmark.

Conclusion
This analysis has revealed some fundamental differences between the ways in which religion and gender are handled legally in the two countries. Interestingly, despite the many differences identified between the Danish and the Norwegian system above, the outcomes of the cases from the two systems are not very dissimilar. Both countries tend to stress the importance of gender equality over concerns related to religion and the freedom of religion. The main difference lies in the way the monitoring bodies arrive at their conclusions, which displays a considerable difference between the two countries. While this difference may have been inconsequential, it is likely to produce different results in the near future: until the Norwegian system receives a decision from the High Court or Supreme Court, the reasoning and outcome of cases that come before the Tribunal will be unpredictable, as is evident from the split decisions it has arrived at in several cases. Similarly, as long as the Danish system continues to operate with a singular body overseeing overlapping legal provisions, the boundaries between ethnicity and religion will continue to be drawn haphazardly. Both of these situations hinder a robust and efficient legal protection for discrimination that occurs at the intersection between religion and gender.