‘My father is Muslim and my mother is Christian: What about me?’ Religious identity and agency within mixed families in Morocco

Introduction

The intensity of contemporary migratory flows and the process of globalisation have led to a multiplication of mixed couples around the world. As a result, we are seeing more and more mixed families settling in Southern and Muslim majority countries, hence the value of looking at their specific experience. Given that the studies which explored mixed Muslim-non-Muslim families focused on the dimension of the couple (Al-Yousuf, 2006; Bangstad, 2004; Igundunasse and Fatunji, 2015; Jawad and Elmali-Karakaya, 2020; Roer-Strier and Ben Ezra, 2006), we have a little knowledge of the day-to-day experience of mixed families and individuals living in these contexts and even less on their religious experience. How the parents cope with the challenges of transmitting (or not) and practicing (or not) their religion, and moreover, how their offspring build their own religious identity in a Muslim country are relevant topics that have been largely overlooked.

Previous ethnographic research on mixed couples revealed diverse religious transmission patterns within mixed families in Morocco (Therrien and Le Gall, 2017). The interviews conducted with the parents allowed me to describe these different patterns and to argue that not transmitting Islam at all, transmitting Islam as a ‘choice’, transmitting two religions or transmitting a religion different than Islam requires mixed families to develop flexibility and adaptability to socially navigate the majority society. In other words, I showed how parents in mixed couples cope with the legal, religious, and social obstacles they face in the Moroccan context (Therrien, 2020a).

This current research on mixed individuals gives insights that enrich the previous data by exploring religious transmission and identity formation from a perspective that is still understudied, that of mixed individuals raised in Muslim–Christian practicing families living in Morocco (such cases are quite rare). Although racial classification of mixed individuals in Morocco has recently been explored (Gilliéron, 2020; Therrien, 2020b), we do not know how religious identity is experienced in their daily life. How does religious transmission work in mixed families where two parents practice a different religion? What is the impact of the Muslim context on their religious identity formation? How do they adapt to family expectations and to the constraints of the religious setting in which they are living? What makes their experience different than that of mixed individuals raised in a non-practicing mixed family living in Morocco or in secular countries?

In addition to bringing new knowledge on the daily life of mixed families living in a majority Muslim country, this article contributes to the growing literature on mixed individuals and to the debate on religious transmission, identity formation and agency within mixed interreligious families. The focus on mixed individuals who have been raised in Morocco by two practicing parents, one Muslim and one Christian, will allow me to argue that despite the constraints of the religious context and their family religious background, mixed individuals who have grown up in a religious country are active agents in their religious identity formation. Studying mixed individuals raised in a non-secular country provides new empirical data to explore the concept of agency and so contribute to the current literature.

In the following sections, I first review the literature on religious identity of mixed individuals and expose a theoretical basis that will be useful for the analysis of the data. The contextual section highlights the social constraints of the Moroccan context to better situate the results of the study. I then explain the methodology I adopted to collect the data. The four cases presented in the results section describe the religious patterns of the families and the adaptive practices they develop to assume their religious choices while negotiating with family and social expectations. The discussion section connects these ethnographic data with the broader scientific literature to enrich the discussion on agency and its interplay with other factors like age, context, family and social expectations, self and social identity, and freedom of choice.

Negotiating religious identity between agency and social constraints

From an anthropological perspective, mixedness refers to exogamy, a transgression of endogamous matrimonial norms. A couple is defined as ‘mixed’ when their union transgresses a collective norm, a symbolic boundary (Streiff-Fenart, 2000) whether in terms of nationality, ethnicity, race, religion, language, and/or social class. As stated by Byron and Waldis (2006), mixedness ‘refer[s] to an unacceptable transgression of what a group defines or considers as its boundaries’ (p. 2). In a recent paper, I argued that the most significant symbolic boundary for white Westerners in Morocco is religion (Therrien, 2020a). Mixed unions are perceived as a challenge to Moroccan social cohesion because they force the endogamous, Muslim, and patriarchal Moroccan society to incorporate elements that do not conform to its traditional model (Addidou, 1990). Mixed unions also shake up Moroccan society because they bring transmission issues into the equation, issues that involve a crucial question: which religious group will the child belong to? This point has been highlighted in different studies on mixed Muslim-non-Muslim families (Al-Yousuf, 2006; Jawad and Elmali-Karakaya, 2020; Therrien, 2014a).

The mixed individuals’ national, ethnic, and racial identifications (King-O’Riain et al., 2014; Paragg, 2015; Song and O’Neill Gutierrez, 2015) as well as their agency (Aspinall and Song, 2013; Rodríguez-García and Reche, 2022; Unterreiner, 2015) are topics that have been addressed by scholars working in European and North American contexts. Studies that addressed religious transmission within mixed families are also grounded in secular contexts (Allouche and Benayoun, 2008; Arweck and Nesbitt, 2010a; Meintel and Le Gall, 2009). However, what mixed individuals think about what have been transmitted to them in terms of religious beliefs, practices, and affiliations and the choices they made themselves in adult life regarding their religious identity remains relatively unexplored (Froese, 2008) and is all but absent in regard to the Muslim world. As mentioned by Cerchiaro (2020), ‘their “voice” is often drowned out, giving centrality only to the negotiations of the couple itself’ (p. 504).

In this article, religious transmission is defined as ‘the way in which parents pass on their religious attitudes and behaviours to their children’ (Arweck and Nesbitt, 2010b: 68), and religious identification as the self-perception of a collective affiliation (Peek, 2005). Religious identity, however, corresponds to ‘a process in which individuals explore and commit to a set of religious beliefs and/or practices’ (Balkin et al., 2009: 420). ¹ Peek (2005) distinguishes three stages of religious identity formation that are relevant for our analysis: religion as an ascribed, a chosen, and a declared identity.

Arweck and Nesbitt (2010b), who have worked on religious identity within interfaith families, recognise two different types of transmission: a ‘passive’ one (informal nurture), where the children internalise parental behaviour and attitudes and reproduce them (or not), and the ‘pro-active’ aspect (formal nurture) which corresponds ‘to the deliberate and often planned activity of parents in perpetuating certain aspects of their faith’ (p. 70). This distinction will be helpful in analysing the different religious transmission patterns found in the field. Researchers have also observed different parental attitudes towards religious transmission: some families decide to stay neutral and not transmit any religion to their children to keep the door open (Arweck and Nesbitt, 2010b), while others decide to offer their children a maximum of symbolic resources in order for them to make a choice later (Meintel and Le Gall, 2009). Regardless of the strategy chosen by the parents, freedom of choice seems to be a common point raised in the literature on mixed families.

The few studies that explore the religious affiliation of mixed individuals question the impact mixedness has on religious transmission. For Arweck and Nesbitt (2010b), who conceptualised religion as a ‘chain of memory’, ² mixed families are of particular relevance for examining the (dis)continuity of religious transmission. They concluded that both fracture and continuity exist within the process of religious transmission. Cerchiaro’s (2020) description of different identification processes has led him to talk about a reshaping rather than a dilution of religious identity. His results contradict existing literature that claims that interreligious marriages weaken religious transmission (Voas, 2003). Allouche-Benayoun (2008) also addressed the question of religious attenuation among mixed Jews-non-Jewish individuals and related this process to what she calls a detachment of the parents from their community of origin. ³

The focus on the Muslim–Christian practicing families living in Morocco was chosen because their case clearly reveals the difference between the process of religious transmission and that of religious formation, thus highlighting the mixed children’s agency, which is until now, an under-researched area. The concept of agency is understood, in this article, as the ability to define one’s goals and to make free choices (Lenoël, 2017). Researchers agree that the exposure of mixed individuals to several languages, religions, or traditions, and their frequent transnational trajectory, leads them to make choices between the cultural references that have been transmitted to them (Therrien and Le Gall, 2012). Consisting of both parental transmission and personal choice, their identity is not a mere reproduction of the parental project (Allouche-Benayoun, 2008; Barbara, 1993; Varro, 1995). They should therefore be considered as full social actors in the identity construction process (Unterreiner, 2017). The distinction brought by Campbell (2009: 409) between agency (‘the capacity for willed voluntary action’) and agentic power (‘the capacity for individuals to act independently of structural constraints’) sheds interesting light on this study. The first type of agency refers to ‘the ability of individuals to implement their will’ whereas the second stresses ‘the individual’s ability to do so against resistance’ or, as already summarised by Lukes in 1974 ‘the power to’ as opposed to ‘the power over’ (in Campbell, 2009: 409). If the second type of agency is necessarily social and relational, the first one ‘is intrinsically private and intra-subjective in nature’ (Campbell, 2009: 417) and has nothing to do with evading the social constraints. Agency and agentic power, both refer to the fact that individuals act freely, and that they make their own choices. However, if agency implies some individual qualities such as autonomy, voluntarism, and creativity, performing voluntary actions does not necessary mean that individuals use their agentic power (Campbell, 2009). Whether their decisions appear to be in conformity with societal expectations or not is irrelevant. The individuals’ agency should not be read and observed in the content of their action (to resist or not; to challenge or not; to modify or not), but in the manner through which their actions are engaged, that is with a willingness to be aligned with oneself. The feminist anthropologist and scholar of Islam Saba Mahmood, who worked on women’s piety movements in Egypt, takes the reflection one step further by inviting scholars to uncouple the notion of agency from that of resistance. For her, ‘the capacity for agency is entailed not only in acts that resist norms but also in the multiple ways one inhabits norms’ (Mahmood, 2006: 12). According to her, we should ‘conceptualize agency not simply as a synonym of resistance to relations of domination, but as a capacity for action’ (Mahmood, 2006: 15). She invites us to move away from this dualistic framework and to think about the multiple ways in which norms ‘are lived and inhabited, aspired to, reached for, and consummated’ (Mahmood, 2006: 18). This article responds to this invitation with a thick ethnographic description of how mixed individuals build their own religious identity in a different but comparable religious context. In this article, the concept of agency and that of adaptive practices, defined by Vigh (2009) as the movement deployed by social actors to constantly adapt to the environment they are living in, complement one another. Indeed, this article highlights the capacity of the mixed individuals we interviewed to make their own choices in terms of religious identity while adjusting (and not resisting) to the social constraints of the majority Muslim context.

The following section, which specifically describes these social constraints, will allow a better understanding of the reasons why mixed individuals have to develop some adaptive practices to socially navigate this particular setting.

The Moroccan religious context

In Morocco, Islam is the official religion of the State under the Constitution and the majority population is Muslim. In total, 99% of Moroccans declare themselves to be Sunni Muslim (Elmoudni, 2016). The Christian population is estimated at 40,000 (with a strong majority of Catholics), mainly made up of foreigners with the majority coming from Sub-Saharan Africa. The number of Moroccans who have converted to Christianity (mainly to Protestantism) is approximately 8000 according to the Observatoire des libertés religieuses (Idrissi, 2019), while the estimated number of Moroccan Jews still living in Morocco is about 2000 ⁴ (MJT, 2020).

Although freedom of religion is legally enshrined in the Constitution, socially this freedom is not granted to Moroccan Muslims (Jelmad, 2014). For example, a Moroccan can convert to another religion without being punished by the law, but it is socially complicated to affirm and live openly with this faith. Morocco’s penal code does not explicitly prohibit apostasy ⁵ but proselytising is illegal (Arbaoui, 2015). This distinction is unclear for many Moroccans, which is what makes the fear of apostasy prevalent in the country. Religious freedom is legally and socially granted to Moroccan Jews and to Christians living in Morocco. First, personal status is related to the religion of each individual. For example, Muslim family law is regulated by the Mudawana (the Moroccan family Code). There are specific jurisdictions for Moroccan Jews and Christians settled in Morocco continue to be subject, for marriage, divorce, and inheritance, to the laws of their country of origin, except when these laws conflict with Moroccan public order (e.g. homosexual marriage; Zeghbib and Therrien, 2016). Second, Moroccan Jews and Christian foreigners can freely practice their religion and frequent the official religious places available in the major cities. However, Moroccans who have converted to Christianity mainly practice their faith in underground churches (Kadiri, 2019). In contrast to the research conducted in secular countries that emphasises the increasing stigmatisation faced by mixed Muslim-non-Muslim families (Cerchiaro, 2022; Giovanna Casa, 2020; Rodríguez-García et al., 2019), in Morocco, these families are not stigmatised (no anti-Muslim sentiment) but they are nevertheless perceived as a challenge for Moroccan social cohesion.

Morocco is not a secular country. The legal challenges faced by mixed families with regard to the Mudawana are a relevant example of the imbrication of State and religion. Being Muslim, or being legally recognised as Muslim – whether practicing or not – is the condition not only to be considered as an insider, but also to benefit from the law (Therrien, 2009). According to the Mudawana, a Muslim man can marry a non-Muslim woman from any of the three monotheist religions. However, a Muslim woman cannot marry a non-Muslim man (Article 39). In practical terms, this means that to marry a Muslim woman, a non-Muslim man has to convert to Islam. Islam is transmitted by paternal filiation. In the eyes of the law, the children of a Muslim father – converted or by birth – are considered Muslim. Transmitting Islam is part of the legal obligation of Muslim families living in Morocco (Article 54). Any mixed marriage that includes one Muslim partner is considered a Muslim family in Morocco. As a consequence of this, Islam is the only religion that mixed Muslim-non-Muslim couples can legally transmit to their children.

In the public sphere, religion is relatively present in Morocco. Beyond the calls for prayer that resonate in every neighbourhood five times a day, religious Islamic education is a subject on the curriculum of all Moroccan schools. There is social pressure concerning religious practice and identification, not only on Moroccans in general, but also on foreigners in mixed couples and on their children. For instance, eating during the daytime in Ramadan (the Muslim holy month) is legally prohibited in Morocco. ⁶ The majority society has internalised this prohibition and applied it to all Moroccans. In this context, it is worth considering how Islam is practised and transmitted within mixed families and how mixed children socially navigate this dominant religious context.

Methodology

This article is based on ongoing research around the identity of mixed children in Morocco. ⁷ In total, 51 in-depth, semi-structured interviews with mixed individuals were conducted involving 42 different households, as some of our interviewees were siblings. Observation also played an important role in this research. To be recruited, the participants had to have been socialised mainly in Morocco (even if they were living abroad at the time of the interview), to have one Moroccan parent and one immigrant parent, and to be above 18 years old. We encountered individuals whose experiences differed in terms of several criteria: nationality (the non-Moroccan parents of our participants came from 30 different countries), gender (22 men and 29 women), ⁸ age (from 18 to 60 years old), socialisation environment (they were mainly raised in big cities in Morocco), place of residence (32 individuals were living in Morocco at the time of the interview and 19 were living abroad), and marital status. In terms of social class, our sample comes mainly from middle- to upper-class families.

In terms of religious family background, the participants of our research can be classified in three different categories that correspond to seven sub-categories of religious patterns: ⁹ (1) households where there was no religious transmission (10/42). This includes non-practicing families (7/42) and families where only one parent is practicing but did not engage in religious transmission (4/42); (2) households where one religion (Islam) has been transmitted (24/42). This includes families where both parents are practicing Muslims (15/42), families where both parents are Muslim but where only one parent is practicing (1/42), and families where the parents are from different religious backgrounds (Muslim and Christian) but only one parent is practicing (8/42); and (3) households where the children grew up with parents from two different religious backgrounds, both practicing, which meant that the children were exposed to two religions (Islam and Christianity) (7/42). In this latter category, we need to distinguish between families where two religions are practised but where Islam is first and foremost actively transmitted (3/42), families where two religions are practised and actively transmitted (3/42), and families where two religions are practised but where there is no active religious transmission (1/42).

For the purpose of this article, I decided to focus on the narratives of mixed individuals who have been raised in Muslim–Christian practicing households (the last category, which corresponds to 7 households and 10 individuals) precisely because they clearly illustrate that being raised in a religious context by two religious parents does not impede the capacity of mixed individuals to make their own choices in terms of religious identity. If I stress the fact that the participants presented in this article had two practicing parents, it is because they do not correspond to the majority of our sample. ¹⁰ As explained by Bangstad (2004), ‘atypical’ narratives may serve to illustrate how normality is constructed in a specific Muslim community, and what the limits of tolerance for ‘atypical’ case are.

Building one’s religious identity while negotiating the family and social environment

The four cases described in the results section are first presented separately to give a clear idea about the specific religious pattern of their family and their religious practice and affiliation. They are structured around four different adaptive practices that show how they adapt to their parents’ and extended families’ expectations, and more broadly to the religious and social norms of Moroccan society.

The four participants selected grew up with a Muslim father and a Christian mother, both practicing. However, the two first cases (Malika and Amir) correspond to families where Islam was the main religion formally transmitted. Christianity was transmitted but in a more informal way. The two other narratives (Abdallah and Lahcen) correspond to families where both religions were proactively transmitted. In spite of these differences in terms of religious transmission, they all share something in common; they developed a capacity to socially navigate their family and social environment between agency and social constraints.

What these cases tell about agency

The first thing that these narratives highlight is the difference between the parents’ identity transmission projects (Meintel, 2002) and the choices made by their offspring, thus highlighting the mixed individuals’ agency. They show that mixed individuals raised in Morocco, as well as those who have grown up in secular contexts, are agents and not just passive receivers of their parents’ legacy (Unterreiner, 2014). The fact that parents’ identity transmission projects only work if the subject assumes this identity by making it their own (Unterreiner, 2014) was also confirmed by this study. As argued by Unterreiner (2014), ‘the child in a mixed couple is an actor in the transmission process and not a sponge whose identity is a strict reproduction of the parental wish’ (p. 98). What has been highlighted in the literature on mixedness concerning agency and ethnic or racial identity has also proved to be true for religious identity.

In light of Mahmood’s (2006) reflection, Malika’s experience shows that agency does not necessarily equate a resistance or a religious disaffiliation. She never asked herself if she was Muslim or not, but she did ask herself questions about what has been transmitted (or not) to her and her own beliefs: were they hers or were they what people wanted her to believe? Malika has no conviction that she knows the ‘truth’. She is totally open to other religions and is critical towards some aspects of Islam – like the religious condemnation of homosexuality – but, like half of our total sample, she never envisioned not being Muslim or endorsing another religion, as was the case for Amir, Abdallah, and Lahcen. From a young age (teenagers), they asked themselves many questions about religion that led them to distance themselves from what had been passed to them in terms of religious background and to claim themselves as Atheists. However, this discontinuity with their parents’ religious identity did not prevent them from developing adaptive practices to adjust to the different family and social contexts. Indeed, the four cases revealed this discrete but powerful ‘social pressure’ that certainly impacts their choices, but mainly encourages them to adapt their practices or, in Mahmood’s terms, to inhabit the norms, one of which is to not give away information. Abdallah was more than discrete regarding his religious affiliation to the point of not even pronouncing the word Atheist. He and his family kept his Christian baptism secret. Fasting when in presence of his family, like Lahcen, respecting his part of the deal and fasting when he was still living at his parents’ home even if it meant nothing to him, like Amir, or pretending to pray by locking herself in her room and doing nothing like Malika, are other adaptive practices that have been described in the previous section.

These cases show another important finding: the influence age and context have on agency. For instance, Malika’s young age and the fact that she was not yet living independently are factors that should be considered when analysing her choices in terms of religious practices. Interestingly, when I talked to Malika a year and a half later, she told me that she had stopped pretending she prayed, that her father knew it and that it was a ‘silent decision’ (she never discussed it with her father, he just stopped asking her to pray). Her narrative uncovers the influence of time on religious identity. As stressed by Arweck and Nesbitt (2010b), ‘individuals’ positions and views change over time’ (p. 171). Indeed, religious identity is dynamic and nonlinear (Griffith and Griggs, 2001). Amir and Lahcen’s narratives also stress the impact age and context have on agency. In Amir’s case, the ‘deal’ (of fasting) lasted until he reached adult age. As soon as he left Morocco to study, he stopped fasting and never fasted anymore even in front of his parents. Both Amir and his parents respected their part of the ‘deal’: he fasted until the age of 18 and they respected his decision of not fasting when he reached adulthood. Lahcen also stopped fasting when he went to study abroad (at the age of 18), but contrary to Amir, he decided to keep it discrete in front of his parents (like his consumption of pork and alcohol). Abdallah, for his part, does not pray or fast (even in presence of his extended family). Nor does he drink alcohol or eat pork, but he said that if they were more openly available, he might do so. All these examples show the impact that age, time but also context have on mixed individuals’ choices, factors that have already be highlighted as significant in other research (Arweck and Nesbitt, 2010b; Peek, 2005).

The data also confirm what we know about the Moroccan context, that the majority society assumes that mixed individuals are Muslim from a social and legal point of view, and that their parents are expected to transmit Islam (Therrien, 2020a). However, this does not prevent mixed individuals from making choices that do not conform to social expectations. Even though they were all considered Muslim from a religious and legal point of view, only 5 out of the 10 mixed individuals raised in Christian–Muslim families we interviewed declared themselves to be practicing Muslims. Four defined themselves as Atheist or agnostic and one declared himself as deist. None of them declare themselves to be culturally Muslim (Jawad and Elmali-Karakaya, 2020) or identified as Christian or as Muslim–Christian. Indeed, the difference between the social (ascribed identity) and the personal dimension of one’s identity (chosen identity; Peek, 2005) is a significant result. Despite the fact that Abdallah is socially, religiously, and legally considered Muslim, he does not affiliate himself with this religion. What is more complicated for Abdallah is what Peek (2005) described as the third stage of religious identity: the declared identity. Being an Atheist and declaring yourself as an Atheist are two different things in a dominant religious context. For example, Abdallah never pronounced the word during the interview, but Amir openly declared himself as an Atheist from a young age. As for Lahcen, he chose the people and situations where he feels at ease to declare his atheism. Previous research on parents’ identity transmission projects showed that a context strongly regulated by a dominant religion produces parental projects that serve as a social display and places mixed identity projects in the private sphere (Therrien and Le Gall, 2017). This new research goes in the same direction by showing that context also influences the ways in which the religious choices of mixed individuals are displayed. They learnt to remain discrete when their religious identity is not consistent with that of the majority group. Nevertheless, they made individual choices regarding their religious beliefs, affiliation, and practices, but also the adaptive practices they decided to develop to adapt to the different settings. To come back to the distinction put forward by Campbell (2009), the four cases presented in this article were not resisting or challenging the social constraints (agentic power). They demonstrated willingness and an autonomy to make their own religious choice, while adapting to the different norms and expectations.

Freedom of choice – another common element of these narratives – was coloured not only by the structural context (legal obligation to be Muslim if the father is Muslim), but also by the specific religious pattern of these families. The possibility of ‘choosing’ which religion they affiliate themselves to was offered to Malika and Amir. However, their narratives clearly show that a choice in terms of identity markers and religious practices has been made during their childhood, that of proactively transmitting (Arweck and Nesbitt, 2010b) those concerning principally Islam. The idea of a choice given to the children in terms of religious identity was also present in Abdallah and Lahcen’s narratives, but with a significant difference. In the two first cases, the children were exposed to both religions, but the religious practices that were proactively transmitted to them were Muslim. In the two last cases, their parents transmitted Christianity and Islam without any distinctions. Not only both parents feel free to practice their religion, but they also feel free to actively transmit it to their children. The latter were also free to practice both religions as they wished. In these households, there seems to be no implicit or explicit agreement that Islam will be the main religion transmitted to the children; we could say that the freedom of choice was ‘less oriented’. Freedom of choice is an element that is often highlighted in the literature on mixed families. Like many other parents in mixed couples, the parents of our interviewees were open to the fact that their children might make a different choice at an adult age. However, contrary to the families studied by Arweck and Nesbitt (2010b) and by Meintel and Le Gall (2009), the parents referred to in this article did not want their children to be without any religious transmission. Whether or not they transmitted the religious model of the majority society or exposed their children to both religions indistinctively, what really matters to the parents is that their children receive a religious education. So, although freedom of choice is something that my interviewees have in common with mixed individuals living in secular countries, my findings also qualify these results by connecting them to Arweck and Nesbitt’s (2010b) conclusion: ‘parents with a strongly religious stance are likely to want to pass this on to their children’ (p. 84).

Conclusion

The focus on Muslim–Christian practicing families living in Morocco not only contributes to the literature on mixed families living in Muslim countries, but also revealed some important findings regarding mixed individuals’ religious identity and agency. One such finding is the significant influence that context plays on religious transmission and identity. The religious constraints of the Moroccan Muslim context have been clearly presented in this article: imposed religious identity, legal obligation to fast, imbrication of the family code and religious law, religious education at school, social pressure, social control, fear of apostasy, and so on. We saw that this context influences mixed individuals’ daily life and required them to find adaptive practices to adjust to the different family and social environments. But this did not necessarily impact their personal convictions, whether similar to or very different from the path being taken by their parents. Although they develop strategies to cope with the different social and family expectations, living in a Muslim country does not restrict their capacity to make their own choices in terms of religious practices and affiliation. Indeed, this study allows me to highlight the agency that mixed individuals living in Morocco share with those living in secular contexts (Arweck and Nesbitt, 2010b), while showing the specificity of their experience of growing up in a Muslim country and more specifically in a mixed Muslim–Christian practicing family. This agency is further highlighted when studying mixed individuals living in a religious context and within an interreligious family. They developed their religious identity between agency and social constraints, a dialectic that was equally central in other studies on Muslim–Christian couples living in Muslim regions, like in Bangstad’s (2004) analysis. Some factors impacting religious identity and agency, like the influence of the contexts (Therrien and Le Gall, 2017), the impact of age and time (Griffith and Griggs, 2001), and the freedom of choice offered within these mixed families (Arweck and Nesbitt, 2010a), have been also emphasised in this article, as well as the difference between the ascribed, the chosen, and the declared identity (Peek, 2005).

The comparison between the two different types of religious transmission highlights another important finding: the freedom to choose, although different depending on the type of religious transmission (formal and informal), did not lead any of them to affirm a multiple religious affiliation. The main observation shows a mix of continued affiliation with the religion of the father (Islam) and a discontinuation of any form of religious identification. In line with Cerchiaro (2020), Kurttekin (2020), Edwards et al. (2010), and Froese (2008) who worked on interreligious families, none of the 51 mixed individuals we met identify themselves as being mixed religiously, although they almost all identify as being nationally, culturally, and linguistically mixed, and none of them identify as being Christian, two important findings that need to be further explored. Half of our participants also define themselves as non-Muslim. This significant result can certainly be connected with what Allouche-Benayoun (2008) calls the choice of a ‘third space’ path. ¹³ The other half declared themselves to be Muslim (most of them practicing although some of them specified they were culturally Muslim). So this current research goes in Cerchiaro’s (2020) direction by contributing to a new interpretation of the existing theories of religious dilution within mixed families (Voas, 2003) and by analysing mixed individuals’ religious affiliations as being more of a reshape of their mixed legacy rather than a loss from it.

The data collected also showed that gender dynamics have to be taken into consideration when analysing the religious choices of mixed individuals. In regard to gender and religious affiliation, this article contradicts some previous research. Arweck and Nesbitt (2010b) and Al-Yousuf (2006) highlighted the influential role of mothers in religious transmission. According to these studies, more children of interfaith marriages identify themselves as Muslims when their mother was the Muslim partner, which is the opposite of what I found in Morocco. Indeed, the empirical data collected through this research show that when the Muslim father practises his religion, the children are more likely to identify with his religion and have more chance of practicing it themselves. Our data confirm what other research highlighted: that in terms of religious transmission, the patrilineal lineages often prevail (Allouche and Benayoun, 2008). As explained earlier, Islam is the only religion that it is legally possible and obligatory to transmit in Morocco when one parent is Muslim. If, as some studies have already shown, the Muslim mother has a significant role in transmitting religion in everyday life (Mossière, 2022), in mixed Christian–Muslim couples when the mother is Christian, the responsibility of the Muslim father is significant. Not only is he the only one who can religiously and legally transmit a religious identity to the children (which is passed on by paternal filiation), but he also has the duty of transmitting Islamic practices. Gender and the structural condition of religious transmission are significant elements of this study that require an in-depth analysis.

To sum-up, the analysis of these atypical narratives allowed me to illustrate how normality is constructed in a Muslim country (by highlighting the symbolic boundaries and the constraints of the context) and what the limits of tolerance for ‘atypical’ case are (by describing some adaptive practices they developed to negotiate their religious identity within their environment). As highlighted, the adaptative practices described in this study were actions ‘independent of the constraining power of social structure’ (Campbell, 2009: 416). In other words, agency, in this article, does not refer to an action that goes against resistance, it is simply about ‘implementing will’. If we conceive agency outside of ‘the binary logic of the doing and undoing of norm’ (Mahmood, 2006: 20) and more as how norm is ‘lived and inhabited’ (p. 19) by the social actors, we can then really be open to observe, describe, and comprehend the social reality of our research participants and all its diversity and richness without being tied to any political engagement that could restrain our responsibility of analytical complexity.