Dynamics of Social Cohesion in Post-Conflict Mosul: Group Perspectives, Challenges and Facilitators

1. Introduction

Social cohesion refers to relations between societal units, including individuals, groups, associations, territorial units, and it characterises how members of a community live together (Dragolov et al., 2013). Despite the considerable discussions about its conceptualisation, Schiefer and Van der Noll (2017) demonstrated in a recent literature review that there is substantial agreement on this concept, particularly regarding specific dimensions. These dimensions include social relations, a sense of belonging and identification, as well as an orientation toward the common good.

Social relations involve the quality of individuals’ connections with others, including their families, friends, neighbours and the broader community (Schiefer & Van der Noll, 2017). They can be defined on a spectrum ranging between “bonding” and “bridging” relations. While bonding relations are between people who are alike in certain ways, such as ethnicity, bridging is about relationships with people who differ in some respects or belong to other sub-groups (Geys & Murdoch, 2010). In this seemingly categorical classification, there are numerous overlaps and intersections, depending on the attributes being considered, be it gender, social class, ethnicity or else. A recognized aspect of resilient relations is social trust, defined as the expectation that others’ behaviour is predictable and guided by positive intentions (Morrone et al., 2009). Social trust can be distinguished in “horizontal” trust in other community members and “vertical” trust in authorities and institutions (Fiedler & Rohles, 2021).

Social identification represents a feeling of being attached to and identifying with a specific social or geographical entity, referring more to the community as an abstract entity and having a pronounced link with the historical and cultural continuity (Schiefer & Van der Noll, 2017). It has been considered to be an important determinant of intergroup relations. More specifically, adopting a more inclusive identity that encompasses the members of the outgroup in the larger overall collectivity, is widely recognized to be helpful in bridging communities (Al Ramiah et al., 2011). The concept of identification is closely related to the sense of belonging, defined as the need to be socially embedded and personally involved with the social environment (Fuchs et al., 2021). Level of acceptance and belonging of minorities is a particularly relevant aspect of cohesion in culturally diverse communities.

An additional pertinent indicator of cohesion is the orientation toward a common good that refers to the individuals’ or groups’ willingness to cooperate for purposes that transcend their individual interests but benefits others or the community as a whole (van Oorschot & Komter, 1998). It entails different aspects like solidarity and helpfulness, civic and political participation, collective action, and compliance to the social rules (Dragolov et al., 2013).

1.1. Social Cohesion and Social Relations After Conflict: Theoretical and Empirical Considerations

There are several theoretical perspectives that support the understanding of the effects of conflict-affected crises on social cohesion, as discussed by Fiedler & Rohles (2021). There is the psychological approach of Post-traumatic Withdrawal (Sillo et al., 2023), which has its roots in psychology and describes psychological consequences of traumatic experiences such as emotional numbing or loss of interest in relationships (Figley, 1985). As Hirschberger (2018) pointed out, collective traumatic experiences weaken social ties and disrupt pre-existing sources of support. A comprehensive review of social cohesion in the aftermath of armed conflict confirmed this assumption (Fiedler & Rohles, 2021), particularly pointing to a reduction of trust in social relations (Ingelaere & Verpoorten, 2020; Rohner et al., 2013). In contrast, post-traumatic growth theory (Tedeschi & Calhoun, 2004) suggests that experiencing a crisis can contribute to positive changes such as a greater appreciation of personal relationships and participation. Political participation was associated with a pattern of post-traumatic growth, while some studies indicated that cooperation tends to increase, although the evidence is inconclusive (Fiedler & Rohles, 2021).

Aforementioned approaches assume a general shift in social cohesion due to changes in individual wellbeing or perception, without taking into account various forms of social relations such as bonding and bridging interactions. This distinction is significant when it comes to conflicts that occur between distinct social groups. Here the Social Identity Theory (Tajfel et al., 1979) provides reference points, which assumes that group categorisation naturally leads to distinctions between “us” and “them”. In the context of conflict this translates into a reinforcement of group identification which fosters “bonding” and hinders “bridging” relationships. Although research is scarce, the existing evidence strongly suggests that conflict increases the importance of group identities and inhibits the identification with inclusive groups (Fiedler & Rohles, 2021). While identification with subordinate groups is prone to ingroup bias, salience of inclusive identity is of great significance in reducing it (Al Ramiah et al., 2011).

Integrated threat theory (Stephan et al., 2016) also offers valuable insights for understanding the dynamics of intergroup relations in the aftermath of conflict. This theory emphasises that identification with one’s own group leads to two different forms of threat perception towards the out-group. The symbolic threat is associated with discrepancies in perceived belief systems, values, religion, or ideology, while the perception of a realistic threat is related to physical safety, the distribution of resources and power. Perceived realistic and symbolic threats have different implications for emotions and behaviour (Martínez et al., 2022), thus implying a differential approach to establishing peaceful coexistence (Rupar & Graf, 2018).

While the predominant body of literature focuses on challenges to social relations in post-conflict contexts, the smaller body of literature on its facilitators recognizes several factors that contribute to it. Some studies emphasize the historical and cultural commonalities shared by groups living in the same place—such as values and philosophy of life (Rumahuru & Gaspersz, 2021), religious, traditional and healing practices, as well as leadership structures (UNDP, 2022). Given the prominence of conflict in people’s experiences, transforming conflict experiences into greater mutual respect, truth-telling initiatives, and coming to terms with the conflicted past can play a role in post-conflict social cohesion (Arriaza & Roht-Arriaza, 2008; Rumahuru & Gaspersz, 2021). In addition, the availability of physical and media spaces for intergroup encounters, may be relevant for (re)building trust (Rumahuru & Gaspersz, 2021).

1.2. Social, Cultural and Political Context of Mosul City

The city of Mosul is with its almost 1.8 million inhabitants the second largest city in Iraq and the capital of Nineveh Governorate. It has been known for its diverse and rich cultural, ethnic and religious history (Larkin & Rudolf, 2023a), being home to various ethnic groups (Kurds, Assyrians, Turkmens, Shabak), non-Muslim minorities (Christians, Yazidis) as well as non-Sunni Muslims (Shiites) in addition to its Sunni-Arab majority. The Tigris river separates the west from the east part of the city. In the western side of the city, in addition to the Arabic host community, a number of Turkmen minorities, and in the outskirts the majority of internally displaced persons (IDPs) live. The eastern side of the city is populated by Kurds, Shabakis, and Turkmen, alongside the Arab majority.

Mosul’s history has been marked by conflicts and political shifts that have had lasting impacts on the city. In the last decades, the city saw the US-led invasion, sectarian violence, the emergence of Islamic State (IS) terrorism, the October protests and the pandemic (Meften, 2022). The most shattering experience for the inhabitants of the city was undoubtedly the capture by IS and the establishment of a repressive caliphate causing a humanitarian crisis, forced displacements and tremendous suffering for its citizens that lasted until the so-called liberation in 2017 (Isakhan & Meskell, 2023). The IS rule, but also the combat to free Mosul, substantially damaged citizens’ individual lives, the social ties, economy, institutions and infrastructure (Lafta et al., 2018). The west side was in this process more affected, due to longer times of occupation, heavy airstrikes, street fights and massive damage during the battle for the city (Nummenmaa & Allaw, 2023).

In the time before the IS occupation, the city of Mosul, greatly inhabited by Sunni Arabs, was controlled by Shia-dominated structures (Abdulrazaq & Stansfield, 2016). The desire to be liberated from these structures led a considerable number of Mosul inhabitants to join or support IS, together with the lack of awareness of the impending terror under the IS caliphate, and fear of retaliation which caused massive involuntary submission to the IS (Abdulrazaq & Stansfield, 2016; IOM, 2019). What contributed to the complexity of the social landscape was the influx of the villagers to Mosul during the IS occupation in the region (in further text IDPs). They settled in the poorest parts of the city bordering Mosul, without having adequate access to services (Larkin & Rudolf, 2023a; UN Habitat, 2016). In the context of this marginalisation, some of the newcomers supported IS. This led to a tendency by other inhabitants of the city to associate responsibility for the occupation to this particular group (Larkin & Rudolf, 2023a). Since then, political destabilisation, militia presence, and discriminatory policies contribute to sectarian divides that have been historically challenging in Iraq.

While co-existence of different ethno-religious groups has a long historical presence in the Nineveh region, following the US-led invasion in 2003, Nineveh was subjected to power struggles between the Shia-led central government of Iraq, the Sunni majority population and the Kurdistan Regional Government (Bourhrous & O’Driscoll, 2023) leading to tensions and undermining the status of minorities. The IS occupation further exacerbated the difficult situation, with many members of the minorities being killed, kidnapped, enslaved, sexually abused, forced to convert to Islam or forced to move out of Mosul (Mousa, 2020).

Research evidence of social dynamics, drivers of discord or cohesion in Mosul is, with a few exceptions, limited and mostly focuses on the surrounding regions. The IS occupation reportedly had damaging consequences on relations, and eroded trust in society (Nummenmaa & Allaw, 2023), particularly toward Sunni community members (Bourhrous & O’Driscoll, 2023). It has also been reported that retaliation practices such as revenge killings of people suspected of having collaborated with or joined IS or their relatives, pose a significant risk of triggering new cycles of violence (IOM, 2019). Among ethno-religious minorities, particularly Yazidi and Christians, there is a widespread sense of abandonment, injustice, having been harmed and excluded and the perception that Muslims were complicit in the crimes against members of their in-group (Larkin & Rudolf, 2023a; Mousa, 2020). Additionally, upon the return of those previously displaced, tensions over land ownership, non-governmental aid, livelihood opportunities, and presence of militias related to specific communities, are part of the post-conflict reality in the region (Bourhrous & O’Driscoll, 2023). Some studies have highlighted the everyday activities that facilitate bonding and intergroup relations. These include commercial exchange and agricultural activities, cooking and sharing meals as an expression of solidarity and altruism, traditional and religious practices, and religious reciprocity within festivals (Bourhrous et al., 2022; Bourhrous & O’Driscoll, 2023; Larkin & Rudolf, 2023a). Some studies draw attention to the importance of post-war heritage reconstruction in Mosul’s community for resistance, healing, distancing from the violent past, and sense of belonging (Bourhrous et al., 2022; Larkin & Rudolf, 2023b).

2. Present Study

Due to the complexity of the social landscape in the city of Mosul, there is theoretical and practical relevance in understanding the nature and dynamics of social relations between different population groups. However, there is a scarcity of knowledge, especially post-conflict. Thus, the first aim of this study is to provide a contextual exploration of neighbourhood cohesion, acceptance of other groups, sense of belonging and identification with an inclusive group, trust in other community members and the government, as well as solidarity and civic engagement. While doing this, it is significant to be aware of the tendency of minorities perceiving quality of social relations often less favourably (Dovidio et al., 2012) and that the role of a group in a conflict implies differential needs and perspectives in the post-conflict period (Shnabel et al., 2023). Therefore, it is essential to examine potential discrepancies in perception of social cohesion dimensions, particularly between groups with majority and minority status and differential involvement in the past conflict. In the particular case of Mosul, this includes the host Arab community in west and east Mosul, the IDP community and ethnic minorities.

Complementing this, the aim is to investigate local mechanisms that hinder social cohesion through a contextualisation of the Integrated Threat Theory approach (Stephan et al., 2016). This also includes to clarify the role of conflict in the current state of social cohesion, with reference to the theoretical approaches of Post-Conflict Withdrawal, Post-Conflict Growth and Social Identity (Sillo et al., 2023; Tajfel et al., 1979; Tedeschi & Calhoun, 2004). However, while frequently overlooked, conflict-affected areas also possess vital mechanisms of recovery (Hirschberger, 2018). If this is not taken into account, it can result in a reductionist discourse of post-conflict areas as “fragile” contexts, bearing the risk of contributing to neo-colonial approaches and epistemic violence (Sandhu, 2021). Therefore, the additional objective of this study is to analyse community resources and the localised facilitators of social cohesion.

3. Methodology

3.1. Quantitative Survey

3.1.1. Procedure

Quantitative data was collected through a household survey conducted in Mosul city in June and July 2024. The use of quota sampling ensured the inclusion of a sufficient and equal number of individuals from the four social groups of interest. Separate lists of neighbourhoods for each of the four groups were created by the local expert team and 5 neighbourhoods per group selected from each list. For the areas mainly inhabited by host communities and ethnic minorities, neighbourhoods were selected by randomization. They were divided in economically more advantaged or disadvantaged neighbourhoods, and a randomization process was implemented with the goal of having balanced proportions in each group. For areas inhabited by IDPs, a purposive selection of neighbourhoods was employed as it was unlikely that the required number of assessments per day could be achieved in neighbourhoods with lower proportion of IDPs. In this case, the local expert team identified five neighbourhoods in which acquiring a planned number of assessments was deemed feasible.

Two assistant pairs of mixed gender, who took part in a training and regular supervisions, collected the data. A “buddy system” was implemented as a safety procedure, with female and male assistants being paired, who simultaneously collected data in the same neighbourhood close to each other and were in regular contact. In order to reduce the bias when selecting households, in neighbourhoods with host communities, the teams followed a randomised walking procedure. They began from a central position in the neighbourhood, then turned right on the street, and visited every third household. While starting at initial houses next to each other, they followed the procedure and communicated regularly to prevent the collection of data from the same household. Due to the lower proportion of households with IDPs and ethnic minorities in the selected neighbourhoods, in these cases purposeful and snow-ball sampling was applied. The overall response rate was 86.4%. The questionnaires were translated from English to Arabic by a bilingual speaker and programmed in and administered by the KoboTools software.

3.1.2. Measurements

Neighbourhood cohesion. In order to measure neighbourhood cohesion the Brief Sense of Community Scale (BSCS) (Peterson et al., 2008) was used. This scale assesses different dimensions, including needs fulfilment, group membership, influence and emotional connection. The BSCS consists of eight items rated on a 5-point Likert scale from “Strongly agree” to “Strongly disagree” with total average scores ranging from 1 to 5. An example item is “I feel like a member of this neighbourhood”. Cronbach’s Alpha of the scale is 0.916.

Acceptance, belonging and identification with inclusive group. In order to explore how much participants feel accepted and that they belong, two self-formulated single items were rated on a scale from 1 (“Not at all”) to 5 (“To an extremely large extent”). Social identification with the inclusive group was measured with a single item, asking about the identification with Mosul (being a Mosul citizen). Single item measurement of social identification has proven to be a valid and adequate operationalization, considering the homogeneousness of the concept (Postmes et al., 2013).

Social trust. Different forms of social trust were measured by the single-item question “To what extent do you trust:”, with rating on a Likert scale from 1 (“Not at all”) to 5 (“To an extremely large extent”). They were asked about generalised horizontal trust in Mosul community members as well as generalised vertical trust in representatives of local public authorities and of public institutions like schools, courts etc.

Civic engagement. Three questions from the Integrated Questionnaire for Measuring Social Capital (SC-IQ) were selected (Grootaert et al., 2004). The first question was a yes/no query on respondents’ willingness to support the community, with two sub-questions regarding contributions of time and money: “If a community project/initiative doesn’t directly benefit you, but has benefits for many others in the city, would you contribute time or money to the project?”. Following this, two yes/no questions addressed participation in collective discussions and actions: “In the past 12 months, did you participate in any communal activities where people discuss community issues together/came together to do some work for the benefit of the community?”.

3.2. Qualitative Interviews

While bolstering the validity of the quantitative findings (Erzberger & Prein, 1997), the qualitative component had a distinct focus on identifying localised, idiosyncratic mechanisms of community cohesion and tensions that often remain overlooked in operationalizations with structured questionnaires. Furthermore, sensitive topics related to intergroup relations or experiences with the IS that the survey could cover only to a limited extent were expected to be explored in more depth within the interviews.

In total, 30 semi-structured interviews were conducted (Mage = 39.1, 33.3% female), with 8 participants from the host Arabic community from west and 11 from east Mosul, 5 ethnic minorities (3 Kurdish, 1 Shabaki, and 1 Turkmen) and 6 IDPs. The sample consisted of housekeepers and employees in the private sector, civil organisations and public institutions, and some community influencers. The interviews were implemented by three local research assistants, previously trained in qualitative data collection. The interviews were facilitated in Arabic, in the community centre or in private houses of participants, with assured privacy. The participants were recruited by purposive sampling, relying on connections of the local team, followed by snow-ball sampling with the considerations of inclusion of a diverse range of people.

The agenda consisted of questions about factors and their roles in facilitating social cohesion such as traditional, everyday practices, initiatives, groups, activities, occasions, physical places and community personalities relevant for the community. Subsequently, questions on obstacles or challenges for social relations, including conflict experiences, social practices recognized, and open or unspoken tensions between groups were asked. The questions were firstly formulated openly in order to allow for spontaneous answers, and followed by specific prompts.

The analytic approach employed was thematic analysis, with the objective of identifying common patterns of meanings (Braun & Clarke, 2012). Following an initial familiarisation with the transcripts, coding scheme was developed in the MAXQDA software by the first author. The codes were then assigned to sub-themes and overarching topics, which were subsequently reviewed and adapted in light of new readings and forthcoming data. The interviews were firstly individually analysed, with a transversal analysis implemented afterwards, to find similarities and differences across transcripts.

4. Results

4.1. Quantitative Analysis

Statistical analyses were performed using SPSS Statistics v.26. A total of 317 Mosul community members participated in the survey, of which 135 (42.6%) were female. The mean age of the participants was 40.28 (SD=12.44). While host communities and IDPs were all of Arabic ethnicity, among the 81 (25,6%) participants who belonged to ethnic minorities 15 (18.5%) were Kurdish, 33 (40.7%) Shabak and 33 (40.7%) Turkmen. All participants were Muslims. A detailed description of the sample can be found in Table 1.

One-way ANOVA tests with Tukey’s HSD test as post-hoc analysis were performed to compare the social relations dimensions between the four groups of interest. Descriptive statistics for dependent variables and F-test results can be found in Table 2. The first one-way ANOVA revealed a statistically significant difference in neighbourhood cohesion between communities, with significantly lower score in IDPs than in the other three groups (p=.00; p=.00, p=.00), and lower scores in host-west than in the ethnic minorities and in the host-east community (p=.03, p=.00). Significant differences were found in perceived acceptance by others, which was significantly lower among IDPs than in ethnic minorities, host-west and host-east communities (p=.00, p=.00, p=.00). Similarly, significant differences were found in belonging, with lower scores among IDPs than in host-west, ethnic minorities, and host-east communities (p=.01, p=.04, p=.02). Differences were found in terms of identification with the city, which again was lower in IDPs compared to host-east, host-west and ethnic minorities (p= .00, p= .02, p= .01).

Significant differences were found in horizontal trust. The host community in the east showed the highest trust levels, compared to west, and ethnic minorities, as well as IDPs (p=.01, p=.00, p=.00). Significant differences were found also in trust toward local authorities, again higher among host-east community than in host-west, ethnic minorities and IDPs (p=.00, p=.03, p=.00). Significant differences characterised as well the trust in representatives of local institutions, higher among host-east community, than in host-west, ethnic minorities and IDPs (p=.00, p=.04, p=.00).

In order to examine the participation in collective discussions and actions, and willingness to support the community with time and money, several Chi-square tests of independence were conducted, with frequencies and percentages to be found in Table 3. Significant differences between groups in community discussions were noticed, χ²(3, N = 316) = 11.764, p=.008; Cramér’s V, was 0.193. A Z-test for comparison of column proportions revealed that collective discussions are higher in host-west communities, compared to ethnic minorities and IDPs. The second test showed a significant difference between groups in terms of collective actions χ²(3, N = 317) = 10.178, p =.017; Cramér’s V, was 0.179. The Z test revealed higher collective action in west-host communities compared to IDPs and ethnic minorities. The willingness to support the community with money was comparable in all four groups, χ²(3, N = 253) = 3.10, p=.376, or with time, χ²(3, N = 253) = 3.96, p= .266.

4.2. Qualitative Analysis

Through the thematic analysis of facilitators and barriers of social cohesion, three main topics emerged: “IS occupation”, “Formal institutions and political elites”, and “Cultural heritage and community resources”.

4.2.1. IS Occupation

Experiences with IS

The majority of participants indicated that conflict dynamics remain a persistent challenge to social relations, albeit with a perception of improvement compared to the initial years following the occupation. Previous conflict dynamics are reported to affect relations between neighbours, family members and relatives until today. While some attributed the divides between families and other community members to different attitudes towards IS during the occupation, the majority mentioned that the current tensions are due to people having been accused of being associated with the IS being reported to the authorities by others, sometimes even by family members. Also, the family members—wives and children—of those who joined the IS reported to face exclusion, discrimination, bullying, and isolation, even though some improvements in terms of their acceptance in the community are also mentioned. Links between families of people having been killed due to their beliefs or values, so-called “martyrs” and families of IS associates are described as particularly absent or strained. In addition to the past experiences, economic struggle and lack of governmental support for people who were directly affected by occupation, hinders acceptance of those previously associated with IS.

Some areas, especially poorer ones, still struggle with accepting families who were connected to ISIS, particularly those who were directly affected by the violence. They feel neglected by the state. (P4, host east, 45)

Unfortunately, it hasn’t returned to normal. For example, my husband is a policeman, and his younger brother was involved with ISIS. One of our relatives reported him to the police. This caused a lot of tension in the family, and it worsened relationships, as many people were falsely accused and suffered as a result. (P28, IDP, 40)

Only ISIS — it’s still a source of constant tension. I still get insulted, but I forgive them. What else can I do? (...) For example, my girls were bullied because their father was in ISIS, and during Ramadan, when aid was being distributed to orphans, we were denied aid because my neighbour said my daughters were children of an ISIS member. This is the kind of stigma we’ve faced. (P13, IDP, 53)

Relations between sectarian and ethno-religious communities are also reported to be, as a result of the conflict, more distant and formal, with less intergroup contact. Some interviewees indicated that Sunnis are still stereotyped as IS supporters and that their suffering during the occupation is still neglected. The rejection of displaced persons due to their associations with IS is also mentioned.

Previously, we couldn’t even differentiate between Sunni and Shia. Relationships between Kurds and Arabs were good, with many intermarriages. Before the war, Arabs and Yazidis in Sinjar lived together harmoniously. However, the war has caused separations, and now these communities are living more separated, especially the Yazidis. (P22, host west, 53)

The situation for refugees is quite poor. There is a lot of discrimination, and many in the community fear that refugees might be connected to ISIS. (...) If society could embrace the entire community and recognize that these children and women bear no guilt, things could improve. (P18, IDP, 32)

Displacements

The internal displacements inflicted by the conflict, particularly from the west to the east of the city or outside, affected ties between family members, relatives, and neighbours that were separated, while bringing challenges for their participation in new neighbourhoods. In general, the occupation experiences led to a loss of trust to unfamiliar individuals and reduction of acceptance of newcomers in neighbourhoods. In particular, there are fears that those moving in are connected to IS and there are reports of lack of acceptance of returnees from families of IS supporters. Sometimes the existent cultural differences between settlers and newcomers or returnees bring additional challenges.

There was no such thing in our neighborhood. Before ISIS, if a stranger came to the neighborhood, all the neighbors would ask him why he was there, meaning everyone knew each other and supported one another, especially in times of need. After ISIS, this connection was lost, and not even uncles open their houses for their nephews. (P23, host west, 57)

Solidarity in Hardship and Post-Conflict Unity

The majority of participants described having experienced or witnessed solidarity and mutual support, during or after the conflict, and as a result also reported the strengthening of neighbourly connections. Importantly, this was reported to also happen in diverse neighbourhoods, with solidarity between members of different ethnic-religious groups. For many community members, conflict experiences led to an increased awareness of the importance of unity and mutual acceptance, with observed efforts dedicated to restoring the community such as voluntary groups.

Our neighbors’ house was partially bombed, so they moved their supplies to our house, on the second day it was completely bombed and destroyed so they moved in with us, in addition to them being our neighbors from the seventies, we lived with them for about 30 days under bombardment, so; naturally this strengthened the relationship between us in spite of the difficulties. (P3, host east, 27).

Even after returning, our bonds grew stronger. For example, if we needed water, we would ask our neighbors, and they would do the same for us, despite the diversity in our area.(...) After the conflict, many people have become more focused on healing and unity. There’s a real sense of solidarity and brotherhood now. (p24, Shabak, west, 20)

Yes, solidarity is strong. An example is the old area where the destruction led the residents to come together before even receiving help from the government. Neighbours acted like one family, supporting each other. (P13, IDP, 53)

4.2.2. Formal Institutions and Political Elites

Inequalities, Discrimination and Favouritism

Most interviewees highlighted that unequal distribution of employment opportunities, resources and services and unfair policies, currently bring challenges for intergroup networks. While favouritism and inequalities affect all groups, they particularly strain inter-sectarian relations. For example, some introduced policies, such as officiating Sunni’s Eid of Ghadeer, without officiating Shia’s counterpart celebration led some Sunni’s to feel discriminated against and being concerned that their religious beliefs could be potentially suppressed. Additionally, some host participants from the west and almost all IDP participants described unequal access to services as well as institutional discrimination, e.g., in school enrolment. While discrimination is partly based on class differences, families of IS members face particular bureaucratic obstacles and disadvantages in receiving services.

Yes, everyone has their different sectarian agenda and for example gives the job opportunities to their sect, increasing division. Yes [there are tensions]. Like, why would this person get a job just because they belong to a party or sect? It can make people feel marginalized (P9, Kurd, 34, west)

There’s a bit of sensitivity between the sides. For example, I went to register children in a school on the eastern side, and they initially refused because we were from the west. Eventually, they accepted us after realizing we were related to someone they knew. The eastern side tends to prioritize its own people because they view themselves as more educated. (P10, IDP, 38)

Disparities in jobs, political appointments, and other opportunities create resentment and continue to fuel divisions. (...) While the people of Mosul are generally very respectful and educated, minorities and displaced individuals sometimes feel a sense of injustice. (P14, host east, 38)

Self-Serving Interests of Political Elites

Many participants expressed lack of trust in political figures and the role of some of them in spurring tensions and deepening divides for their own particular interests. As a result, cultural or religious initiatives and actors associated with political elites are often rejected.

When someone participates in these rituals [from another sect], I am more likely to join in if I see them acting on their own, without support from the government. If a Shi’a person is simply expressing their faith and following their traditions spontaneously, I would happily participate. (P19, Turkmen, east, 58)

As ordinary citizens, there’s no problem between them. However, political leaders might have different agendas that create gaps. (P1, host east, 31)

Public Resources for Cohesion

Public institutions and services are perceived to facilitate relations and stability, even though there are differences in perceptions of their usefulness. Regarding the Mukhtars (neighbourhood leaders), some recognized their role in maintaining safety, and occasionally solving disputes, whereas others ascribe to them only an administrative role. Similarly, in terms of the role of community police, some acknowledged their capabilities in solving societal problems, such as family issues or problems regarding electronic blackmail, whereas others reported that they have no visible function. In particular, the families of IS associates are perceived as relying less on institutional resources and more on civil organizations for fear of institutional responses.

For instance, when a new individual arrived in the area, they were required to introduce themselves to the neighborhood leader, which served as an indication of peaceful intent. (...). This demonstrates the important contribution of neighborhood leaders to maintaining safety and security in the community. (P19, Turkmen, east, 58)

Although neighborhood councils are state employees and are supposed to help, I haven’t personally seen any significant impact beyond formalities. (P17, host west, 57)

The social police, established after 2019, plays a clear role in Mosul. I’ve seen them handle cases related to harassment, theft, and family problems. (P15, host east, 26)

Displaced families and those connected to ISIS may hesitate to approach the police or intelligence services due to fear of repercussions. (P7, host east, 48)

4.2.3. Cultural Heritage and Community Resources

Cultural, Religious and Traditional Heritage

Long-lasting, intergenerational ties between some families reportedly serve as a bridge between different groups and have survived during and after the period of conflict. The practice of traditional and religious customs, such as food sharing and joint celebrations, continuously are described to strengthen bonds within, but also between groups. However, it has also been observed by some interviewees that such customs were partly affected by the conflict and that they currently mostly remain in traditional neighbourhoods and the west side of the city.

(...) Even on the familial level, we have friends from the eighties even, Kurds and Yazidis, and you see it in honorable stands amongst us, you see who abandons you and who has your back. My grandfather’s friend, just after my grandfather passed away, after a few days, he came to us less than a week after, and said, “What do you need and how can I help?” The relationship between us extends from the eighties, so the relationships endured. (P3, host east, 27)

Of course. You feel happy when a neighbour or relative brings you food or anything, even if you don’t need it. It definitely strengthens social cohesion and the bond between us. However, we don’t see this much nowadays, especially on the east side. Back on the west side, we had such events and celebrations much more often. (P23, host east, 55)

Differences in cultural and religious background sometimes represent a challenge for mutual acceptance. Cultural differences in social norms and lifestyle are reported to affect the level of the host communities’ acceptance of IDPs. This is reflected in the existing distinction between “Maslawis” and “Arabs”—the former relating to the more educated, higher class and long-term citizens of Mosul, and the latter to the newcomers from rural areas, often less affluent. Another example relates to the differences in sectarian beliefs. Even though participants reported relying on the Islamic principle of “You have your religion and I have mine” that leads them to accept the differences in sectarian or religious beliefs, it was also mentioned that some individuals try to impose beliefs within the intergroup contact, which creates tension.

But there are people who I do not know or recognize their cultural civil backgrounds. (...) For example, they [rural IDPs] even have different traditions. I mean for example in Mosul no one knocks on the door at night, they get embarrassed to do so. They [host community] don’t have the existing social relations that justify visiting people every day at any time for dinner or whatever, they have different customs and traditions from our guests in Mosul. (P3, host east, 27)

Traditional Structures

Religious institutions are recognized to be a space where a general message of unity, peace, and solidarity is conveyed by religious leaders and occasionally as a place for dialogues and for intergroup contact. Some interviewees emphasised on the personal engagement of religious leaders in community issues and disputes. Also, tribal leaders are instrumental to turn to in case of problems and conflicts. Particularly when it happens between members of different tribes, they are in the position to prevent conflict from escalating. However, some interviewees perceived specific influential figures as potentially following agendas that may reinforce divisions.

Religious leaders play the most significant role through their speeches, lectures, and proper guidance. They work to dispel sectarianism and violence, and their influence helps unite different groups. (P30, host west, 74)

Tribal connections are especially strong. If a conflict arises, tribal leaders mediate and resolve the issue before it escalates, showing the strength of their relationships. (P4, host east, 45)

Religious leaders are especially important because many people follow them. Social leaders are also crucial, but we must question whether they genuinely believe in the initiatives they lead. Some social leaders have been co-opted by political parties to push certain agendas. (P21, host west, 21)

Community Hubs

There are various places where people can meet including public halls, mosques, parks and popular cafes. However, IDPs reported gathering in private houses, due to a lack of options available for them. While such places serve predominantly for bonding relations, some of them are mentioned to be used by different groups. Educational and work environments are a platform for intergroup contact, despite the segregation that still exists. Occasional cultural events, such as the Spring festival or Heritage festival, were cited as important for the sense of collective identity and bringing people from different communities together. However, some participants noted that contact between groups within such events is limited and that they do not have a direct impact on the quality of social relations.

Social and cultural centers, mosques, churches, and temples all play crucial roles in strengthening social ties. Schools and universities also provide spaces for youth to interact and build relationships. Public places, even informal ones like squares or parks, are vital for community activities. (P6, host west, 42)

No, we don’t meet in mosques. The mosques and schools are far from us, so we gather in homes. (P26, IDP, 35)

After the liberation and all the challenges we’ve faced, people now feel each other’s pain and no longer differentiate based on sect. For example, at university, when we go on trips, we all contribute food, share the same table, and laugh together. We forget who is Sunni, Shia, Kurdish, or Arab. We simply enjoy each other’s company. (...) There is a slight difference, as Kurds tend to group with Kurds, and Arabs with Arabs. But when we go out together, everyone communicates and gets along. (P24, Shabak, west, 20)

Many young people have misconceptions about others due to a lack of interaction. (...) Each group mostly keeps to itself, but economic relationships sometimes bring them together. (P21, host west, 28)

5. Discussion

The first aim of this study was to explore the level of social cohesion indicators in different communities in post-conflict Mosul. Neighbourhood cohesion, perceived acceptance and belonging to the inclusive group were high. Identification with Mosul was strong, inconsistent with the findings reported for the time during occupation (Barrie, 2021). However, the horizontal and vertical trust were scoring in the moderate range, consistent with evidence on affected trust in Mosul and other contexts (Ingelaere & Verpoorten, 2020; Nummenmaa & Allaw, 2023; Rohner et al., 2013).

Importantly, social cohesion indicators show different pathways in social groups. Firstly, the results reflect the IDPs challenges to integrate in the city (World Bank, 2022) and their socio-economic marginalisation. They are feeling the least neighbourhood cohesion, acceptance, belonging and identification with Mosul. The qualitative part of the study unravelled that reasons for their unacceptance by some host members are possibly due to the fear of their potential association with IS, the perceived pressure of their presence on the city’s resources (Larkin & Rudolf, 2023a; World Bank, 2022), and their different cultural background. The results shed light on the much more tenuous and protracted situation of the west side during the occupation and the subsequent battle for liberation, which has caused a civilian catastrophe (Amnesty International, 2017). In the host community in the west, neighbourhood cohesion was moderate, and the qualitative results stressed the negative effects of conflict and IS affiliations on neighbourly relations. In addition, horizontal and vertical trust have been assessed as higher among the host community in the east compared to the other three communities. This is in line with previously reported perception of unequal resource distribution between city sides, directed more toward the east (IOM, 2019), and it is reflecting the lower involvement of this part of the city in the previous conflict and its currently favourable status within institutions. Ethnic minorities in the east side expressed a sense of acceptance and belonging and neighbourhood cohesion comparable to that of the host community in the east, indicating their long presence in and attachment to the city. However, they also expressed lower general horizontal and vertical trust compared to the host community in the east, possibly due to past treatment of ethnic minorities and current perceived institutional discrimination, an assumption that was bolstered by the qualitative findings. Openness to solidarity and civic engagement was evident, with the majority of people, irrespective of group affiliation, willing to contribute to the community with their time. While this reflects the collectivity and long-lasting traditions ingrained in the Iraqi and west Asian culture in concepts such as Sadaqah (Hassan et al., 2017), the interviewees confirmed that this spirit of being there for one other was fostered by the experiences of the recent conflict. This is in line with evidence on increased readiness for participation after armed conflict (Fiedler & Rohles, 2021), while it is debated if this concerns mainly the cooperation with the in-group or extends also to out-groups (Bauer et al., 2016). The interview testimonies suggest that this was primarily a matter of neighbourly cooperation, while the willingness to participate did not necessarily translate into actual behavior in community discussions or collective actions, especially among IDPs and ethnic minorities. The reasons for this can only be surmised, such as the remote position of the neighbourhoods where these groups reside, the fewer opportunities of engagement or their less pronounced feeling of belonging to the inclusive group.

The second goal of this study was to explore barriers and facilitators of social cohesion. The results predominantly support the thesis of post-conflict withdrawal (Sillo et al., 2023) by demonstrating the significant negative impact of the conflict on cohesion, affected ties and trust on the west, toward residents in general, and particularly toward post-conflict newcomers. Qualitative findings supported this, by the testimonies of affected relations at family, neighbourhood and intergroup level, due to IS support and victimisation in the community. Displacements, as one of the consequences of the conflict, are reported to break ties through physical distance and contribute to resource strain, recognized to lead to tensions (Pabayo, et al. 2020; World Bank, 2022). In addition to the aforementioned factors, this study points to challenges in mutual acceptance due to sometimes existing cultural differences between old residents and newcomers or a general loss of trust and fear that the newcomers are involved with IS. In other words, newcomers were perceived by some members of the host community as both a realistic threat to resources and security and, to some extent, a symbolic threat to the city’s culture. This finding validates the Integrated threat theory in recognizing and distinguishing the types of threat (Stephan et al., 2016), but also shows their simultaneous coexistence.

Connections between different groups were particularly negatively affected by the conflict, as previously reported in other post-conflict contexts (McKeown & Psaltis, 2017), while bonding relations were partially strengthened. This supports the Social identity approach (Tajfel et al., 1979), assuming that conflict would increase the salience of in-group identity and the negative bias toward the outgroup. However, some insights of the study imply that the boundaries between in-group and out-group are blurred and overlap with other affiliations. It clearly emerged that the conflict and associations with the IS had a negative impact on some familial and neighbourhood ties. In particular, the perceived out-group of IS-associates and the need of many to distance themselves from it, seems currently to be more important for people than family, neighborhood, ethnic or religious affiliations. This is consistent with the black sheep effect—particularly harsh judgments of in-group deviations from relevant norms, in order to protect positive self-image or in-group image (Marques & Paez, 1994). Simultaneously, instances of neighbourly help, occurring both within and between different groups in diverse neighbourhoods, suggest particularly salient identification with the neighborhood. In that way both bonding and bridging relationships were reinforced during conflict, provided contact is not with strangers but people with whom there were long-standing connections.

Perceived fairness in resource allocation contributes to cohesiveness, loyalty and better intergroup relations, while unfairness hinders it (Valcke et al., 2020). This study showed that perception of unfair distribution in power, opportunities and resources and the implementation of discriminatory policies is currently also one of the main drivers of discord in Mosul. Political agendas perceived to serve interests of particular groups currently represent prominent sources of resentment and drivers of intergroup tensions, across different groupings, and particularly noticeable for sectarian relations. Tensions are perceived to be less pronounced in everyday interactions and less due to the cultural differences, presumably reflecting the legacy of mutual acceptance and the city’s history of diversity. However, cultural and class differences do represent challenges for relations to some extent. The symbolic threat is particularly pronounced in the host community’s perception of rural IDPs. In terms of sectarian relations, symbolic threat is also noticeable, but perceived as predominantly driven by the political decisions and actions, and less by tensions in everyday interactions and practices in the community. In conclusion, the current intergroup tensions in Mosul are primarily driven by the perceived realistic threat, originating from the institutional and political levels, distributive injustice and occasionally safety concerns, but to some extent also by the symbolic threat due to cultural, sectarian, and religious differences (Stephan et al., 2016).

As previously evidenced (Bourhrous et al., 2022; Bourhrous & O’Driscoll, 2023; UNDP, 2022), religious and traditional practices, intertwined with solidarity and sharing, particularly in more traditional neighbourhoods, continue to facilitate bonding as well as bridging relations. Traditional structures, like religious, tribal and neighbourhood leaders contribute to cohesion by conveying values of peace, offering space for dialogues and intergroup contact, and occasionally engaging directly to support solving disputes. However, perceptions vary, presumably depending on the commitment of the specific leader with whom they have experience with. A legacy of good relations and mutual acceptance between sectarian and ethnic-religious groups serve as a buffer for potentially harmful socio-political events that might spur tensions. A multitude of public spaces, predominantly those traditionally accepted, such as mosques, public halls, parks, bazaars, and cafés, cultural events and festivals, as well as work and education environments, serve as platforms for social bonding and, in some cases, for intergroup contact. This is in line with studies in other contexts (Rumahuru & Gaspersz, 2021). However, some community members recognized that intergroup contact between ethno-religious groups is limited within such spaces, while resources are less accessible to and used by IDPs, resulting in perception of their segregation. Besides traditional resources, new resilience mechanisms generally appear after conflict (Hirschberger, 2018). In Mosul, experiences of solidarity during and after the occupation and newly established structures, such as community police, initiatives of the civil sector and voluntary groups, contributed to cohesive relations. In this regard there is a line with the assumptions of the post-conflict growth theory, in regard to the growth of appreciation of relationships and stability, humanitarianism, and an openness to engagement and cooperation (Hirschberger, 2018; Tedeschi & Calhoun, 2004).

5.1. Limitations

Despite efforts to include all community groups, it was difficult to reach representatives of some minorities. Consequently, Yazidi, Assyrian, Armenian and Kakai perspectives are not represented, while Shia perspectives are underrepresented in the results. The sensitivity of topics related to occupation was bearing some challenges too, as it potentially led to “socially desirable” responding and underreporting of negative experiences and perceptions, particularly regarding sectarian relations or IS experiences. The questionnaires used in the quantitative surveys were not validated in Arabic, which is recommended for future studies to ensure their cultural appropriateness.

5.2. Conclusions

While the results vary across dimensions of social cohesion and between groups, the affected relationships in the conflict-destabilised western part of Mosul city are particularly striking. The findings imply the importance of examining perceptions of majority and minority groups separately when measuring social cohesion, as it shows the disadvantaged position of minority groups (Dovidio et al., 2012), which can be a catalyst for resentment and a significant risk for future tensions (UNDP, 2016). In addition to the past conflict, the unjust political agendas and social inequalities, and to some extent cultural differences are recognized as the main drivers of conflict. While the results support the relevance of classical theories to social relations in (post-)conflict contexts, they also highlight their restricted scope to certain aspects of social dynamics, indicating the need for more comprehensive models. In particular, the fluidity of in-group and out-group identification and the intersection between types of threat should be taken into account. Finally, this paper responds to narratives that portray post-conflict contexts as “fragile” (Sandhu, 2021) by showing a certain degree of resilience and the mechanisms of recovery inherent in the community, as well as examples of positive post-conflict change, despite the predominantly harming effect of the conflict on social ties.

Table 4. List of selected neighbourhoods for data collection.

Community group	West Mosul	East Mosul
Arab Host community	1. Al abar 2. Mosul aljadeda 3. Nabe shet 4. Al ghazlane 5. Al sena’a al shemalya	1. Al sharqiya 2. Nabe younis 3. Al Mothana 4. Al hadba 5. Al qadeseya
Ethnic minorities	1.Al Rashidiya*	2. Hay Al samah 3. Al bakir 4. Qahira 5. Al shalalat
IDPs	1. Al Rashidiya* 2. Al Tanak 3. Rajm Hadid	4. Al entesar 5. Sadam
*All neighbourhoods are visited by only one pair of researchers, with the exception of Al Rashidiya, which appeared on both the IDP and ethnic minority lists and was selected for both.

Table 4. List of selected neighbourhoods for data collection.

Community group	West Mosul	East Mosul
Arab Host community	1. Al abar 2. Mosul aljadeda 3. Nabe shet 4. Al ghazlane 5. Al sena’a al shemalya	1. Al sharqiya 2. Nabe younis 3. Al Mothana 4. Al hadba 5. Al qadeseya
Ethnic minorities	1.Al Rashidiya*	2. Hay Al samah 3. Al bakir 4. Qahira 5. Al shalalat
IDPs	1. Al Rashidiya* 2. Al Tanak 3. Rajm Hadid	4. Al entesar 5. Sadam
*All neighbourhoods are visited by only one pair of researchers, with the exception of Al Rashidiya, which appeared on both the IDP and ethnic minority lists and was selected for both.