Enemies in Iraq, Human Beings in Norway: ‘Multilocal’ Boundaries between Radicalised Sunni and Shi‘a Muslims

This article employs symbolic boundary theory to investigate how a sample of radicalised Sunni and Shi‘a Muslims in Norway make meaning of their views, behaviour and interactions with denominational others. It draws on 23 interviews with individuals who legitimise violence or are willing to use violence themselves to achieve political change. Research on boundary construction shows that the relative importance of boundaries varies across geographical contexts for different individuals. This article introduces the concept of ‘multilocal’ boundaries, which vary across contexts for the same individuals. The findings suggest that despite the boundaries being firmly agreed upon, their impact is limited by participants’ avoidance of religion and politics, their need to stand together as a minority and the enforced laws in Norway. At the same time, these boundaries can remain largely undiluted in another context. The war in Syria and Iraq seems to strengthen the influence of boundaries.


Introduction
A growing body of sociological literature is concerned with how context influences boundary construction (Alba, 2005;Lamont, 1994Lamont, , 2000;;Lamont et al., 2016;Wimmer, 2013;Zolberg and Woon, 1999).These studies have contributed terms and developed theories to study variations and changes in ethnic, racial and religious boundaries (Alba, 2005;Lamont et al., 2016;Wimmer, 2013;Zolberg and Woon, 1999).While the existing literature shows that the relative importance of boundaries varies across geographical contexts for different individuals and groups (Alba, 2005;Lamont, 1994Lamont, , 2000)), this article contributes by introducing the concept of 'multilocal' boundaries, which implies that boundaries vary across contexts for the same individuals.To illustrate, I investigate how a sample of radicalised Sunni and Shi'a Muslims in Norway make meaning of their views, behaviour and interactions with denominational others.In doing so, I depart from the existing literature on boundary work, which mainly studies the boundaries between majority and minority groups, and add to the growing literature on minority-minority relations.In particular, I contribute to the increasing literature on Muslim-Muslim relations that addresses distinctions relating to extremism and the Sunni-Shi'a divide.
Atrocities and conflicts between self-defined Sunni and Shi'a groups in Muslimmajority countries have influenced some European Muslims.Sunni-Shi'a narratives are part of the recruitment strategies of wars (Isakhan, 2020), and approximately 6000 Western Europeans joined the wars in Iraq and Syria (Cook and Vale, 2018).Cases of bigotry and violence have also occurred in Europe.In 2012, a Shi'a imam died in Belgium in an arson attack on a mosque conducted by a Salafist (BBC, 2012), and Britain witnessed violence between Sunni and Shi'a Muslims in 2013 (Crossley, 2014).Violence towards other religions and schools of thought are essential aspects of interpretations of Islam related to the radicalisation of Muslims (Precht, 2007;Silber and Bhatt, 2007).Still, we know little about how radicalised Muslims who support or use violence towards denominational others create meaning of their views, behaviour and interactions.I address this issue by critically examining 23 in-depth interviews with self-defined Sunni and Shi'a Muslim men in Norway in 2015, 2018 and 2019.In this study, the term 'radicalised' refers to individuals who support and legitimise violence by non-state actors or are willing to use violence themselves to achieve political change in society (Borum, 2011).
By drawing on symbolic boundary theory (Lamont and Molnár, 2002), I demonstrate how Muslim men in this study made meaning by actively drawing boundaries between themselves and denominational others.In doing so, the participants referred to two strongly intertwined cultural resources: religious sources (Islam, Quran, traditions of Prophet Mohammed and holy shrines) and their knowledge of political decisions, events and conflicts.The argument here is that the boundaries the Muslim men in this study draw may be 'multilocal'.How these boundaries constrained participants' behaviour and patterns of social interaction in Norway seemed to be limited by avoidance of religion and politics with denominational others, their need to stand together as a minority and a functioning legal system in which breach of the law could result in a prison sentence.At the same time, this did not dilute the influence of these boundaries in the context of Muslim-majority countries.On the contrary, the context of the war in Syria and Iraq seemed to strengthen the influence of boundaries.Although the participants may have firmly agreed on these boundaries, context proved to be vital for how they constrained identity and patterns of social interaction.This finding implies that context plays a crucial role in the strengthening, weakening, transporting and crossing of boundaries.
The aim of this article is threefold.First, it offers insights into how the intertwinement of religion and politics influences radicalised Muslims.More specifically, it illustrates the heterogeneous processes of boundary making among radicalised Muslims.Second, introducing the concept of multilocal boundaries adds to the literature on symbolic boundaries by demonstrating the importance of context for understanding how minorities manoeuvre behaviour and interactions.Third, this article makes a significant contribution to terrorism studies and its subfield of radicalisation studies by providing primary qualitative empirical data, which is scarce (Neumann and Kleinmann, 2013;Schuurman, 2020).
In the following, I describe the Sunni-Shi'a divide before situating my research in the literature on boundaries.After the description of the data and methods, the analysis of the religious and political boundaries and how these boundaries vary and are challenged by context follows.Finally, this article concludes with a discussion of the findings and suggestions for future research.

Sunni-Shi'a Divide
Debates over succession, a Muslim leader's role and the religious grounds for leadership in Islam divided Muslims after the death of Prophet Mohammed in 633 AD (Hashemi and Postel, 2017).Some of Mohammed's followers, later known as Sunni, claimed that a leader should be chosen from the Prophet's companions.Others, later known as Shi'a, argued that Mohammed's successor must be from his close family.This debate triggered the creation of two 'overlapping yet distinct theologies and contending interpretations of history' (Hashemi and Postel, 2017: 1).Some of the modern theological divisions include beliefs about the essence of God, free will, predetermination and fallibility of the Prophet.A critical difference today is related to the authority of scholars and legal theory.While the Sunni school of thought closed 'the gate of Ijtihād', jurists' independent interpretation of the sacred text to deduct a legal ruling, in the 10th century, Shi'a maintain that 'the gate of Ijtihād' is open and still used by contemporary Shi'a jurists.Many Shi'a Muslims worldwide follow the legal rulings of Shi'a grand jurists in Iraq and Iran (Ali-Karamali and Dunne, 1994;Vogt, 2007).According to the Pew Research Centre (2011) estimates, the majority Sunni constitute approximately 87-90% of Muslims worldwide, and the minority Shi'a approximately 10-13%.
The Sunni-Shi'a divide is often observed through the actions of political actors in Muslim-majority countries.Pakistan's political leaders have mobilised along the Sunni-Shi'a divide to gain or maintain political power, including permitting activities by extreme groups engaging in violence and killings.The Saudi monarchy feared the Iranian revolution's toppling of the regime in 1979 and branded the revolution a product of dishonest Shi'a faith (Hashemi and Postel, 2017;Nasr, 2000).Similarly, Iran and Saudi Arabia use this divide as a tool in their rivalry for leadership of the 'Muslim world'.This rivalry has often played out in other Muslim-majority countries, where Iran and Saudi Arabia support different non-state political actors.Struggle for political power and manipulation of the boundaries between the two main branches also play a role in contemporary conflicts, such as the wars between self-defined Sunni and Shi'a groups in Iraq, Syria and Yemen.The political actors involved in the wars blame extreme factions of the other group for atrocities, target Muslims and holy sites of the opposite group and mobilise through religious frames (Hashemi and Postel, 2017;Isakhan, 2020;Nasr, 2000).
This divide also influences the Muslim diaspora in the West.It often plays a role in framing Muslim identity and in interactions among Muslims (Esposti, 2019).Polemics, bigotry and violence between Sunni and Shi'a Muslims have all occurred in Europe (BBC, 2012;Crossley, 2014;Linge, 2016).The aim here is to explore the Sunni and Shi'a divide by analysing how the Muslims studied draw distinctions between themselves and denominational others.

Symbolic Boundaries, Boundary Variations and Muslims
The process of categorisation and the drawing of symbolic boundaries are vital parts of social life.Lamont and Molnár (2002: 168) defined symbolic boundaries as 'conceptual distinctions made by social actors to categorise objects, people, practices and even time and space.They are the tools by which individuals and groups struggle over and come to agree upon definitions of reality.' Broadly agreed-upon symbolic boundaries may influence and constrain identity formation and patterns of social interaction.Individuals and groups use conceptual distinctions, such as cultural attitudes and practices, to draw similarities with those included and differences with those who should be excluded.Conceptual distinctions are significant in constructing, maintaining and deconstructing groups (Lamont and Molnár, 2002;Lamont et al., 2015;Pachucki et al., 2007).The concept of symbolic boundaries may be useful for studying the Sunni-Shi'a divide, and examining symbolic boundaries between radicalised Sunni and Shi'a Muslims may provide insights into their identity and meaning-making processes.
Contemporary social scientists utilise symbolic boundaries to study various themes.Barth (1998Barth ( [1969]]) viewed ethnic identities as boundaries and argued that ethnic groups were not predefined collectivities, but constructs made through everyday interaction in the interplay between self-ascription and ascription by others.Further developed by Jenkins (1997), Barth's ideas highlight the changing and varying nature of boundaries.Building on Bourdieu, Lamont (1994, 2000) compared French and American upper-middle class and working class and demonstrated how they constructed identity and competence by entangling moral, socio-economic, cultural, class and racial boundaries.One major finding of Lamont's work is that the relative significance of boundaries varies across contexts for different individuals.For example, members of the American uppermiddle class gave more importance to socio-economic and moral boundaries than the French upper-middle class, who emphasised cultural boundaries.According to Lamont (1994), differences in cultural resources explain this variation.Lamont and colleagues also illustrated variations in other forms of boundary processes between different individuals and groups (Lamont et al., 2016; see also Alba, 2005;Wimmer, 2013).However, the question of how boundaries influence differs according to context for the same individuals has not yet received scholarly attention.
Existing research has primarily dealt with minority-majority distinctions and boundary making among the majority (Alba, 2005;Bourdieu, 2010).However, research on boundary making among minorities in the West is growing, as some scholars have begun to focus on conceptual distinctions and categorisations between Muslims, and between Muslims and wider society (Abbas, 2019;Esposti, 2019;Furseth, 2011;Sandberg and Colvin, 2020).For example, Furseth (2011) studied the boundary work of hijab-wearing and non-hijab-wearing women in the Los Angeles area and explored how religion framed their identities.Abbas (2019: 261) analysed how divisions between Muslims in Leeds and Bradford were influenced by state interventions to counter extremism and identified the dual distinctions of 'the suspected extremists for Muslims to look out for and suspected informer who might report fellow Muslims'.Similarly, Sandberg and Colvin (2020) examined how a sample of Norwegian Muslims drew multiple distinctions by rejecting the mainstream projection of Islam as an aggressive religion and, at the same time, supporting the mainstream in their exclusion of jihadists.
Scholars have also examined distinctions between Sunni and Shi'a Muslims.Isakhan's (2020) analysis of Shi'a mobilisation against the Islamic State (IS) shows that non-state Shi'a actors from Syria, Iraq and Lebanon drew religious distinctions to mobilise, which involved 'the shrine protection narrative'.Shi'a Muslims regard the tombs and shrines of Prophet Mohammed and his family with reverential respect and honour them with a ritual act of devotion.Esposti (2019) studied how affiliation and belonging to the Iraqi Shi'a political and religious scene influenced the boundaries between Shi'a and Sunni Muslims in London, stressing the importance of global and local contexts in boundary processes.
Some scholars have further developed terms and theories to explain variations in boundaries.Zolberg and Woon (1999) distinguished between the crossing, blurring and shifting of boundaries.Boundary crossing occurs when an individual transcends the distinctions between insiders and outsiders without changing the distinctions or structure of society.Boundary blurring occurs when the characteristics of the structure of society change.Boundary shifting involves reconstructing a group's identity, resulting in a relocation of differences between members and non-members towards more inclusion or exclusion.Alba (2005) introduced the concept of bright and blurred boundaries: the former refers to clear distinctions between the majority and the minority, while the latter refers to unclear distinctions between the majority and the minority.These concepts help understand various forms of boundaries, which 'cannot be conceptualised in terms of a single set of processes' (Alba, 2005: 21).
I complement the existing work on boundary variation by introducing the concept of multilocal boundaries.The extant literature does not explicitly conceptualise the significance of contexts for how boundaries constrain behaviour and interactions.Existing concepts for understanding boundary processes do not adequately explain the boundary variations observed in this study, indicating that there is a need for the concept of multilocal boundaries.Concepts that fit the empirical findings 'elucidate them as simply and adequately as possible, and allow us to explore their implications' (Barth, 1998(Barth, [1969]]: 10).'Multilocal' implies that despite a firm agreement about boundaries, the same individuals can act and interact differently according to the multiple local contexts with which they identify.A shift in individuals' local contexts can change how boundaries constrain their behaviour and patterns of social interaction.Religious and political boundaries that the men in this study draw are multilocal in the sense that although they firmly agree that these boundaries exist, the ways in which these boundaries constrain their behaviour and social interactions towards denominational others vary in different contexts, that is, in Norway, Syria and Iraq.Lamont and Molnár (2002: 187) called for more research 'in terms of exploring the conditions under which boundaries generate differentiation and dissolve to produce hybridity or new forms of categorisation'.This article contributes to theoretical insights into similarities and differences in cultures and contexts in which boundary processes occur and are strengthened or weakened.Notably, it provides knowledge about transportation and crossing boundaries across cultural and geographical contexts.

Methods
I draw on 23 semi-structured interviews with 18 radicalised, self-defined Sunni and Shi'a Muslim men in Norway in 2015 and 2018-2019.Five participants were interviewed twice, due to the opportunity and unique access to these guarded individuals and their roles as leaders, motivators and ideologues.The semi-structured interviews do not necessarily provide information about actual events and experiences, but interpretations of experiences, ideas and practices.This information may offer insights into social processes and individual motives (Dalgaard-Nielsen, 2010;Sageman, 2014).Participants' views, interactions and actions towards denominational others reflect knowledge about categories and perceptions in their cultural, social and religious contexts that were a part of their values, identities and communities (Furseth, 2011;Sandberg, 2010).
The interviews were conducted during 22 months of fieldwork in and around the Norwegian capital, Oslo, due to the relatively high density of Muslims.In total, I conducted semi-structured interviews with 84 Muslim men, including radicalised Muslims, Muslim inmates and self-defined practising Muslims who were neither radicalised nor incarcerated.The data also included field conversations and participatory observations.Here, I focus on the interviews with radicalised Muslims because including the other data did not add to the analytical and conceptual points.
I began the fieldwork by spending time in mosques, study circles and other meeting places to establish contact with potential participants.The sample was expanded through purposive and snowball sampling to gain further access to relevant participants.The sample criteria were second-generation immigrant Muslim men between 18 and 36, as this group constitutes the individuals most vulnerable to experiencing a radical change in identity (Precht, 2007;Silber and Bhatt, 2007).However, due to access to individuals in essential roles as ideologues and motivators and with knowledge and life experiences relevant to the study's theme, I included participants who did not meet the initial criteria.To explore the lives of Muslims with different experiences, I conducted fieldwork in and outside four different prisons in 2018-2019.Norwegian Correctional Services approved this project. 1Prison priests and correctional officers facilitated contact by introducing me to Muslim inmates.The average interview time was two hours and 48 minutes.One participant did not permit me to audiotape the interview, so I took handwritten notes with his consent.The rest of the interviews were recorded and transcribed in full.Only the quotes used to illustrate the analytical and conceptual points were translated into English from Norwegian.
The participants varied in age from 18 to 50, with the majority between 18 and 36.Four were single, 13 were married, 11 were married with children and one was a single parent.Six had finished high school, five upper-secondary, five had bachelor's degrees and one had completed his master's education.The educational background of one participant was unknown.The majority were unemployed or in and out of workingclass jobs.Four held other occupations.Eight belonged to first-generation immigrant Muslims and nine to second, while one did not have an immigrant background but was a convert.
Seven had South Asian backgrounds, three North African, six Middle Eastern, one Eastern European and one Norwegian.The majority of the participants (13) were Sunni Muslims, of whom the majority adhered to Salafi-jihadi understanding (an interpretation that entails an intolerant and violent methodology), and the rest (five) were Twelver Shi'a (the most prominent group within Shi'a).All the participants supported and legitimised non-state Sunni and Shi'a actors' actions in the wars in Iraq and Syria.Some had participated or attempted to participate in battles, and the sample included men who had attempted to conduct political violence in Norway.At the time of the interviews, three participants were incarcerated.Most participants belonged to two small loose-knit groups that support self-defined Sunni and Shi'a political actors in Muslim-majority countries.The Salafi-jihadi group overtly endorsed the IS and motivated others to participate in the wars in Syria and Iraq.
All participants received an information sheet explaining the aims of the research project, the handling of data and the participants' rights.Consent was requested and obtained verbally, since saving signatures could potentially compromise anonymity.I promised confidentiality, with the clause that I was obliged to notify authorities about acts that could compromise security and life.In line with this, I reminded the participants that their participation was voluntary.However, due to imprisonment, surveillance and other security measures, future attempts to participate in war or political violence were considered unlikely, and if any such actions were planned, it was implausible that I would be informed.Information about participation or attempted participation in wars was provided by men who attempted these acts when it was not unlawful before July 2016 (Criminal Law, 2016), or by men serving or who had already served a prison sentence for these actions.I used services for sensitive data at University of Oslo, which offers security measures, such as two-factor login, that fulfil state regulations for handling and saving sensitive data.I deleted recordings, anonymised data and used pseudonyms after transcriptions.The safety of the participants was also preserved by focusing on patterns, categories and themes to understand meaning making, rather than individual life stories that could expose them.Information considered harmful to anonymity, such as the names of prisons, was omitted.Norway's Social Science Data Services, which handles ethical approvals for University of Oslo, approved this project. 2 I presented myself to the participants as a sociologist interested in understanding the lives of Muslim men in Norway, with a particular interest in Sunni-Shi'a identities, interactions and relations.I informed the participants that this interest included questions about why and how some adopt and maintain radicalised views and actions while others reject them.In the interviews, I asked them open-ended questions about their identities and views on denominational others: how do you view Muslims as belonging to the other main branch of Islam?How have your experiences been with denominational others?How would you describe the nature of Sunni-Shi'a interactions in Norway?Influenced by the existing literature on radicalised Muslims, I asked concrete questions about their religious and political stances towards others (Juergensmeyer, 2017;Roy, 2017;Sageman, 2017).These questions were not necessary in all cases, since the participants readily addressed these issues in the interviews.I used NVivo 12 software to code participants' answers about denominational others before I categorised the codes into themes, followed by a revision in line with the study's aim and symbolic boundary theory.
The participants may have responded to me as a second-generation immigrant Muslim.My affiliation with Islam seemingly functioned as a door-opener and led to trusting relationships during interactions.I was addressed as 'brother' and treated as a part of a collective: 'we Muslims' and 'our community'.Still, my religious and political affiliations and views were under scrutiny, and I was never completely viewed as one of them, but as a researcher with a similar background.This perception seemingly resulted in the belief that I might understand them and their answers that were not socially accepted.However, participants may still have presented a particular self to be liked or to gain attention.

Findings
Next, I discuss the identified patterns in which boundaries are constructed, how these boundaries constrain behaviour and how patterns of social interaction are challenged by context.

The Boundary of Religion
By referring to important historical figures, religious symbols and their understanding of Islam, the men in this study provided religious justifications for the boundaries they drew.Murthada believed that his soul had accepted the divine leadership of Ali ibn Abu Talib before he was born.Ali ibn Abu Talib was the first imam of the Shi'a and the fourth caliph in Islamic history, whom the Shi'a believe was chosen by God as the rightful heir of Prophet Mohammed.In line with the 'protect the shrine' narrative (Isakhan, 2020), Murthada cited religious motivation for his wish and failed attempts to engage in the civil war in Syria: I want to go there to protect my muqaddasat, my holy places.Tomorrow, I can come with a white face [equivalent to proud or honourable] in front of Rasulullah [referring to meeting Prophet Mohammed in the hereafter].I have protected your daughter [shrine of Syeda Zainab bint Ali, the granddaughter of Prophet Mohammed, in Damascus], your daughter's grave, which I believe is her house.That's my belief.
Murthada's account draws multiple distinctions.He categorised himself with specific traits, namely, accepter of divine leadership (i.e. the truth) and among the honourable in the hereafter through being the protector of the Prophet's family.This identification implicitly means that denominational others were those who denied the truth and were attacking the house of the granddaughter of the Prophet, and would therefore be dishonoured in the hereafter.
Religion seems to justify distinct identities and categorisations of oneself and denominational others.This was the case for most of the participants.For example, Ali hated Sunni Islam and thought about travelling to Syria to engage in the fight against the IS.Like Murthada, he conformed to the 'protect the shrine' narrative.Ali emphasised the importance of shrines for his wish to travel and fight in the Syrian civil war: 'As you know, the most important thing for us [Shi'a] is our shrines.And when I see that one of our shrines is in danger, I am not able [to be silent].' Ali's and Murthada's explanations were in line with the symbolic boundaries drawn by many of the participants, including the Salafi-jihadis.Omar viewed Shi'a as 'kuffar' [unbelievers] and actively engaged in motivating Norwegian Muslims to participate in the Syrian civil war.He legitimised his exclusion of the Shi'a from Islam by categorising himself as a follower of the Quran and traditions of the Prophet (i.e. the truth): 'We say that we cannot do our actions [we cannot act] if there is no proof [justification for those actions] from the Quran and sunnah [traditions of the Prophet].'Sufiyan engaged in the same activities as Omar, excluding Shi'a Muslims from his monotheistic religion of Islam by defining their beliefs as polytheism: 'Shi'a has shirk [polytheism] in their aqeedah [creed].' Zubayr engaged in Sunni-Shi'a debates on social media platforms and motivated attendees at the local mosque to participate in the wars in Syria and Iraq.He distinguished the believers, Sunni Muslims, who upheld the rules of Islam, and the unbelievers, Shi'a Muslims, who broke them: 'Islam also has rules [religious doctrines].And the Shi'a break these rules.' Participants' religious interpretations defined them and drew lines between them and others.By referring to Islam, the Quran, sunnah or shrines, the participants presented Islam as a source that guided their views and behaviour and explained their experiences.An illustration is Saif, who hated Sunni Muslims.His experiences with Sunni Muslims involved harassment, death threats and being thrown out of a Sunni mosque.Saif's understanding of why these events took place was religious in content: They [fellow students] did not have a clue about my beliefs.They thought that we worship humans and do not have the same Quran.A lot of them stopped spending time with me.They looked at me as an infidel because they thought that we curse the companions of the Prophet and his wives.This affected them because they had seen YouTube clips about Shi'a and heard [about us] from their scholars.
Saif did not refer to explanations offered by his personal religious beliefs about his experiences with Sunni Muslims.Instead, he provided insights into his assumptions about why he thought some Sunni Muslims treated him badly.These assumptions included a belief that the Sunni Muslims with whom he had had negative interactions thought that Shi'a Muslims should be excluded, harassed and threatened.The reference is to the Quran, beliefs and historical personalities -wives and companions of the Prophet revered by Sunni Muslims who did not hold the same position in Islam for many Shi'a Muslims (Pinault, 2001;Spellberg, 1994).He further described his experiences of being thrown out of a Sunni mosque as an interaction in which the individuals involved applied religious distinctions: We prayed, and they said, 'they are kuffar [unbelievers]; do not let them pray here'.When we finished the prayer, he [one of the Sunni Muslims at the mosque] began badmouthing, and we also said some stuff to them.They said things like, 'You worship stones, and you curse Aisha [wife of the Prophet] and the sahabah [companions of the prophet], you are kuffar, [and] you are not welcome here.'So we got aggressive and said: 'You are wahabi.'And we cursed Muawiyah.
Muawiyah Ibn Abi Sufiyan became the caliph after Ali ibn Abu Talib, the fifth caliph of Islam (Watt, 1974: 12).Many Sunni Muslims revere him, but Shi'a views Muawiyah as a usurper of the rights of the family of Prophet Mohammed.Even after Ali's death, there were minor uprisings against Muawiyah (Donner, 1999: 16).Wahabi refers to the conservative Islamic creed followers established by Mohammad ibn Abdul Wahab, adopted by the Saudi family in 1744.This religious understanding within Sunni Islam opposes Shi'a Islam (DeLong-Bas, 2007: 83-90).All the religious tropes -accepter/follower of the truth, protector of the Prophet's family, honourable in the hereafter, believers, dishonoured in the afterlife, kuffar/unbelievers, infidels, polytheists, breakers of religious rules and wahabi -draw a boundary between the participants and who they see as denominational others.Participants defined themselves through this boundary construction, including those who shared similar beliefs and excluding denominational others.

The Political Boundary
The participants also drew on their knowledge of political decisions, events and conflicts when drawing boundaries between themselves and denominational others.Addressing the conflicts in Syria and Iraq, Zubayr explained that the Shi'a stood at war with Sunni Islam: 'Shi'a, they view Sunni as [someone] who should be exterminated; they do not say it, but we know it.'He viewed denominational others as those who must be attacked and defended against due to their actions against his group in Muslim-majority countries.The men in this study commonly provided explanations of this type.This can be seen as a powerful trope for boundary drawing.The others were excluded as a threat who wished to 'exterminate' 'us' and must be fought.Each participant identified their own group as victims under attack whose defence was legitimate.
Hasan argued that the conflict in Iraq and Syria was about the Shi'a 'survival' in the Middle East.He supported the Shi'a militia by openly saying that he was Shi'a, by speaking up against injustices and through symbolic gestures, such as wearing parts of the war gear of the Shi'a militia operating in Iraq and Syria.Hasan defined the conflicts in the Middle East as significant for the survival of his group, which was attacked unjustly.The denominational others were described by Hasan as unjustly fighting to remove Shi'a Muslims and Shi'a Islam from Syria and Iraq.Therefore, he viewed the fight against denominational others as a legitimate fight for their survival.Participants' views that political conflicts would determine the survival or 'extermination' of their religious group suggest a strong link between participants' religious and political understanding.Some also linked religious and political sources in their boundary work more explicitly.Omar emphasised that Shi'a states were fighting against Sunni Muslims: We already know that the Iranian and Lebanese regimes, [Hezbollah] are fighting against Muslims in Syria.We also know that they have taken over Yemen.And they say that they are going to conquer Mecca.But I believe that some things will happen in the coming years because everything started in Sham, Syria.
In his interpretation of political conflicts, Omar referenced important religious sites -Mecca and Sham [present-day Syria, Lebanon, Israel, Jordan, Palestine and parts of Turkey] -in Islam, which also hold significance in the end of time prophecies (Khalaji, 2008).Omar's statement implies a view of the others as non-Muslim and strong political actors -states -fighting Muslims and wishing to conquer Muslim land, including important religious sites.These religious interpretations of political conflicts in the 'Muslim world' created boundaries between the participants and denominational others.
The participants also drew political boundaries by allying enemies, the West, and denominational others.Zubayr was convinced that the Norwegian state sided with the Shi'a: 'Of course, we think it's silly, at least they [the Norwegian state] could have been neutral.It can lead to conflicts in Norway because we have a huge percentage of Sunnis and Shi'a in Norway.' Contrary to this, Murthada argued that the Norwegian state sided with Sunni Muslims: How can one come with the Islamic State flag in the middle of Oslo without being afraid?Can a Shi'a take the Iranian flag and go on the street?He will get beaten on the spot.But the others have done it because they know the truth that the entire West is with them.
Murthada referred to an incident in which individuals drove through Oslo with an IS flag draped from their car (for more on this, see Lohne et al., 2014).His statement was political, and he was not merely allying the West with IS, but also viewing Iran as a symbol of Shi'a Islam.By this logic, his group was also a political actor.Implicitly, these statements identify the others as strong political actors who conspire with the enemy and are supported by them.These descriptions also legitimatise the exclusion of denominational others because they have abandoned the Muslims.A political boundary is drawn between those who stand with the enemies -the West -and those who stand with the Muslims.Some were more extreme in their statements when they spoke about the alliance between the others and the West: 'I treat a Shi'a the way I would have treated an unbeliever' (Khalid), or, 'It's the same rules for [use of violence against] all the enemies of Islam' (Razzaq).When allying the enemies, Khalid and Razzaq implicitly identified themselves as believers and representatives of Islam and categorised the others as unbelievers and enemies who deserved the same treatment.
Again, a boundary is constructed by intertwining religion and politics and providing religious framing for political decisions, events, conflicts and alliances.By drawing these boundaries, the participants constructed binary identities for themselves and the others, in which the others were excluded as an enemy, a threat and unjust, powerful conspirators and unbelievers.Their own groups were constructed as true and pure believers who represented Islam and fought for a just cause.

Multilocal Boundaries: 'Here in Norway, We Are Human Beings'
When the radicalised Muslims interviewed spoke about their views, behaviour and experiences with denominational others, the boundaries they drew were often contextualised.Few viewed the impact of the boundaries as absolute.Talha offered an illuminating illustration: he supported the Salafi-jihadi groups and hated anyone whose actions opposed his beliefs.He had no relations with Shi'a, and he said he would love to kill them given the opportunity.Although Talha believed that it was an obligatory religious act to kill Shi'a, he stressed the importance of the Norwegian context for not doing so: I did not know that he was a Shi'a, but he started to curse when he began to talk.We were in Norway, so I could not cut [off his head].So, I distanced myself from him, viewed him as a non-Muslim and stopped saying salam [Peace, a term used for greeting] to him.
Although he viewed the Shi'a he interacted with as non-Muslim, Talha did not use violence against him because, as he explained, the interaction took place in Norway.He further emphasised that a functioning legal system, which can punish Norwegian citizens with a prison sentence, made him constrain his behaviour because he did not want to spend his life without family and friends and without the opportunity to seek knowledge about Islam -an opportunity he assumed did not exist in the penitentiary.This implied that violence could have been an option if the context had provided the opportunity to get away with it without being imprisoned.
Khalid held negative views towards Shi'a, and made several attempts to get involved in the war in Syria to fight self-defined Shi'a groups.Still, he believed that positive interactions with Shi'a were possible in Norway.Khalid argued more overtly than Talha that different contexts made different behaviours and interactions possible: We, Ahl-e-Sunnah-Wal-Jamah [literally, the people of the (prophetic) tradition and the majority], let the place [context] tell us how to be.Here in Norway, we are human beings.So, we can let the politics be.But if I were in Iraq or another place, I would have treated them the way the situation required.They are enemies; that is it.They are enemies of ahl-e-Sunnah [abbreviation of Ahl-e-Sunnah-Wal-Jamah]; they want to get us.We will take them [down] before they take us.
Khalid emphasised context as crucial for behaviour and interactions.While the situation in Iraq required that denominational others be treated as enemies, the Norwegian context made it possible to view one's enemies as human beings, since it was possible to avoid politics: It is very complicated.What is the situation we are talking about?Is it war?Is it peace?Here in Norway, Shi'a should be treated as human beings.If they do good, they get good.If they show respect, they will get respect.Just as everyone else.We socialise, we have our transactions and everything should happen in a peaceful manner, as long as we adhere to human affairs, do not go into politics and do not go into religion.We are a people who can say ok; you are a Shi'a here in Norway, we are peaceful, we talk, we can have a cup of tea, we can discuss transactions; however, as soon as we set our foot in Iraq, I am after you, do you understand (laughs).You go from one condition to another.
Parts of Khalid's account signify the relational aspect of boundary construction (Barth, 1998(Barth, [1969]]): 'If they do good, they get good.If they show respect, they will get respect.'However, it also emphasises the importance of context and the conditions that make different behaviours and interactions possible.While positive interactions with denominational others are likely in Norway, a secular state where one can adhere to human affairs without going into politics and religion, in the context of Iraq, a war-torn country in which non-state self-defined Sunni and Shi'a actors have been active in fighting each other, Shi'a are his enemies, and he would act towards them accordingly.Another contextual explanation for this variation may also be that due to conflicts, these boundaries may be widely agreed upon in Iraq and Syria and thus strongly influence participants' actions in comparison with Norway.
The majority of the interviewees applied similar rationales to explain the variation in their practices towards other Muslims in Norway compared with Muslim-majority countries.Anas stressed the importance of context for variations in views and behaviour: 'Differences [in views and behaviour] are dependent on where one comes from, and what kind of preconditions one has for having relations with those people.'Likewise, Omar focused on context when interpreting religious texts and statements addressing Shi'a: 'This comment [one statement about killing Shi'a by an ideologue] does not mean that you are suddenly supposed to go and kill Shi'a.' The image of denominational others as enemies was part of the boundaries the participants drew and could be strengthened or weakened depending on the context.Emphasis on context was found among those who attempted to participate in wars, supported selfdefined Sunni and Shi'a groups fighting in Syria and Iraq and returned as foreign fighters.The latter had positive interactions with denominational others during their time in prison.The inmates emphasised the need to stand together as a minority and not show the majority that Muslims were against each other: In prison, I do not want to give the others [the majority] the image that we are against each other.(Abdul) You feel that we [Sunni-Shi'a] must be united in the West because other powers are after us.(Yaqoob) These examples illustrate that some participants did not want Muslims to be perceived as separated by the majority population in Norway.This discussion implies that despite the participants' agreement about the boundaries, they constrained behaviour and social interactions differently due to varying contextual conditions.

Concluding Discussion
As noted, social scientists stress that boundaries change over time and vary across contexts (Alba, 2005;Barth, 1998Barth, [1969]]; Lamont, 1994Lamont, , 2000;;Midtbøen, 2018;Wimmer, 2013).Some studies have demonstrated variations in the relative importance of boundaries across contexts for different individuals and groups (Alba, 2005;Lamont, 1994Lamont, , 2000)).Others have shown how multiple contexts, ancestral and local, influence boundary processes among the diaspora in specific contexts (Esposti, 2019;Furseth, 2011).By drawing on symbolic boundary theory (Lamont and Molnár, 2002) when analysing interviews with 18 radicalised Sunni and Shi'a Muslim men in Norway, I found that boundaries could vary for the same individuals across cultural and geographical contexts.This signifies the importance of contexts for strengthening, weakening, transporting and crossing cultural and communal boundaries.
The participants in this study constructed their identities and defined denominational others by drawing on two intertwined cultural resources: religious sources and their knowledge of political decisions, events and conflicts.Although strong agreement existed among the participants about the boundaries that were central for how they made meaning about themselves, their reality and denominational others, the geographical and cultural context in which the participants were functioning and their interpretation of it were vital for how strongly the boundaries constrained behaviour and social interactions.In other words, the boundaries the participants drew were multilocal in the sense that they varied from context to context.The boundaries appeared seemingly weaker in Norway, while being more critical in Iraq and Syria.This is not to say that bigotry and harassment do not occur among Muslims in Norway (Acharki and Mordt, 2013;Alsabeehg, 2014), but the context of Syria and Iraq increased the possibility of extreme behaviour and negative social interactions for the same participants.
These findings indicate that boundaries are not merely changing through everyday interactions (Barth, 1998(Barth, [1969]]); they also exist in multiple ways simultaneously.Individuals who look upon the opposing group as the enemy and might be willing to use violence against them in one context may reject this sort of extreme behaviour in another.This provides a multidimensional, multilocal, varying and flexible understanding of boundaries.Future research should illuminate our understanding of boundaries by exploring themes, research questions, individuals and groups related to multiple contexts.Such a focus should also seek to develop frameworks and concepts to conceptualise various forms of boundaries.