This is an open access article distributed under the terms of the Creative Commons Attribution License (In this study, we use the term 'Turkish–Kurdish conflict,' common in social science and psychology literature, while acknowledging its oversimplification. With this term, we refer to a complex web of social and political practices shaped by colonial state oppression against Kurds and Kurdish resistance, structural inequalities, exclusionary daily practices, and mutual resentment between the parties involved. Kurds in Turkey have been subjected to various studies in anthropology, sociology, political science, and other disciplines in the social sciences, yet they have not been very popular for research into the social psychology of intergroup relations (see Göregenli, 2010; Şen et al., 2023; Uluğ & Çoymak, 2017; Uysal et al., 2024). Nevertheless, research on the role of intergroup contact in the context of the Turkish-Kurdish ‘conflict’ has been on the rise in the last decade (Bagci et al., 2018, 2020; Tropp et al., 2021; Uluğ & Cohrs, 2017a). Although social and political psychology research on intergroup contact can provide well-founded knowledge for understanding and resolving conflicts, adapting conceptual frameworks generated in WEIRD contexts (Western, Educated, Industrialized, Rich, Democratic: Henrich et al., 2010) to non-WEIRD conflict settings without a critical examination of the conditions created by their historical and geopolitical context limits the contributions that social psychological research can make. Therefore, a full grasp of the role of intergroup contact in intergroup conflict would also require a re-examination of the available literature in particular contexts, considering the potential socio-political biases and norms that affect researchers’ position and understanding of the topic.
First, studies conducted in specific contexts may not be directly transferable to other intergroup conflict situations. One significant factor contributing to this is methodological nationalism, which occurs when the nation-state is viewed as the natural and necessary representation of modern society (Chernilo, 2011). Methodological nationalism manifests when national discourses, agendas, loyalties, and dominant historical narratives are unquestioningly accepted without critical examination or analysis (Wimmer & Schiller, 2002). The influence of national identity, norms, and dominant conflict narratives on the execution, framing, and interpretation of psychological research varies across contexts due to unique state-building processes and conflict narratives. Consequently, researchers in different countries may exhibit different biases. Therefore, a re-evaluation of the existing literature within the Turkish-Kurdish conflict context is essential to comprehend the specific dynamics, limitations, and embedded biases inherent to this historical and geopolitical setting. By scrutinizing studies within this context, researchers can adjust and enhance existing theories and methodologies to better suit the complexities of this conflict. This may facilitate the development of critical, context-specific theories and approaches that offer practical solutions for the conflict, transcending dominant state narratives.
Moreover, current political and scientific perspectives on discrimination, prejudice, societal conflict and contact have predominantly been explored within the American or Western European contexts, where efforts to grapple with a colonialist and racist history are ongoing. These societies possess established narratives regarding prejudice and discrimination against minority groups—a narrative that, although problematic in many aspects, generally acknowledges the historical marginalization of minorities and aims to improve intergroup relations in the present and future. In contrast, Turkey presents a unique context characterized by an ultimate rejection of historical and contemporary colonial practices in Kurdistan, intermittent ongoing armed conflict, and a prevalent dominant "terrorism" narrative regarding conflict among both institutions and the population (see Uluğ, 2023). Attempting to replicate American- or Eurocentric conflict studies within this framework fails to capture the nuanced dynamics of the context and does not yield systematic knowledge conducive to practical reconciliation and emancipation efforts. Therefore, it is imperative to build a foundation for understanding prejudice, discrimination, racism, and intergroup contact within this context by examining existing literature and delineating its limitations.
Hence, this paper, as the first meta-analytical review on the role of intergroup contact in the Turkish-Kurdish conflict, aims to do four things: i) determine the most common social psychological processes and variables that were examined as an outcome of intergroup contact in the social psychology research on the Turkish-Kurdish conflict, ii) conduct meta-analyses to examine the correlations between intergroup contact and these common variables and determine the potential moderators that impact the magnitude or valence of these relationships, iii) provide an overall examination of how social psychological research approaches intergroup contact and conflict and the limitations of these approaches, and finally, iv) discuss how dominant biases in the literature such as methodological nationalism and epistemic violence influence research practices and knowledge-production systems in conflict and contact studies conducted under state repression. To do so, in this paper, we focused on four research themes on the role of intergroup contact in this context and conducted four meta-analyses: 1) intergroup contact and outgroup attitudes, 2) intergroup contact and support for minority rights/multiculturalism, 3) intergroup contact and perception of discrimination against the minority group (i.e., Kurds), and 4) intergroup contact and ethnic identity.
Overview of the Historical and Political Context
Understanding the study of Kurdishness in Turkey necessitates familiarity with its historical, political, and geographical contexts. Kurds are an ethnonational group with populations extending across the borders of four nation-states: Turkey, Syria, Iran, and Iraq. Despite the lack of formal international recognition, Kurds collectively identify with the ethno-national territory of Kurdistan, which stretches from southeastern Turkey to northwestern Iran. In Turkey, Kurds represent the largest ethnic minority, comprising an estimated 15-20% of the population, although there are no official statistics available (Koc et al., 2008). Since the establishment of the Turkish Republic in 1923, Turkish identity has been enforced as the dominant ethnic and national category, marginalizing other ethnic identities, notably the Kurds (Güneş, 2019). The assimilation of Kurdish identity under Turkishness has been central to political narratives and discourses surrounding national identity (Ünlü, 2016, 2018). Turkish national identity was constructed as part of a modernist, Western-inclined state ideology, often depicting Kurdishness as a threat to Turkish national identity and symbols of backwardness that hinder the country's development and modernization (Cagaptay, 2004). The Kurdish ethnic and cultural identity was systematically denied by the state until the mid-1990s, with the denial often accompanied by explicit state violence (Yavuz, 2001).
Authoritarian, centrist state policies and violence elicited various forms of resistance and uprisings from Kurds. The state responded by deporting Kurds to different regions of the country, both to quell rebellions and prevent future ones and to initiate assimilation efforts. Forced resettlement expanded through policies such as changing and Turkifying the names of Kurdish villages and cities (Jongerden, 2001). With the 1980 military coup, state violence towards all minorities and stigmatized groups in Turkey cultivated and reinforced assimilationist ideologies, deepening divisions and grievances within Kurdish communities, including torture and enforced disappearances of Kurdish prisoners and activists, and the ban on the Kurdish language in public spaces (Yavuz, 2001). The contemporary armed conflict between the Turkish military and the Kurdish movement peaked with the armed insurgency of the Kurdistan Workers' Party (PKK) in 1984, leading to an asymmetrical conflict characterized by violence, political tensions, securitization policies and polarization. The peace process initiated in the 2010s, partly driven by Turkey’s EU candidacy, brought hope for reconciliation and opened space for psychology to finally research this issue. However, the process ultimately faltered in 2015, leading to renewed tensions and violence. This brief window in psychology research, however, did not hinder the impact of epistemic violence and methodological nationalism on the production of knowledge regarding Kurdishness within the literature of "Turkish" psychology (Bayad & Şen, 2023). Therefore, we aim to critically review studies on intergroup contact in this context to identify methodological and theoretical limitations, which may not be unrelated to political biases and methodological nationalism.
The Role of Contact in Intergroup Conflicts
In 1954, Allport suggested that intergroup contact would reduce prejudice and discrimination under particular circumstances. Since then, the social psychology literature has presented considerable evidence for the role of intergroup contact in prejudice and discrimination reduction through cross-sectional (for reviews, see, Pettigrew, 2016; Pettigrew & Tropp, 2008), longitudinal (Christ et al., 2010; Dhont et al., 2014; Swart et al., 2011), and experimental studies (Brannon & Walton, 2013; Ioannou et al., 2018; Vezzali et al., 2012) in various contexts. Research over the years has shown that intergroup contact reduces prejudice towards members of the outgroup and negative intergroup behaviour even if the particular circumstances such as common goals or institutional support outlined by Allport (1954), are not met (see Pettigrew & Tropp, 2006).
A growing body of literature has examined the consequences of intergroup contact in predicting intergroup attitudes and behaviours, both in low- and high-conflict contexts. For instance, intergroup contact reduces negative outgroup attitudes through reduced intergroup anxiety among British children and adults (Turner et al., 2013, 2014) and Italians (Voci & Hewstone, 2003), as well as reduced intergroup threat among Europeans (Schmid et al., 2014) against immigrants. Intergroup contact not only reduces negative outgroup attitudes but also promotes positive outgroup attitudes, such as outgroup trust among British high school students (Turner et al., 2013) and Italian children (Vezzali et al., 2012), cooperation among German university students (Kuchenbrandt et al., 2013) and Italian adults (Pagotto et al., 2013), intergroup forgiveness among Bosnian adults in the postwar context (Voci et al., 2017) and Catholics and Protestants from Northern Ireland in the sectarian conflict context (Voci et al., 2015), humanization among Italian psychology students (Capozza et al., 2017), empathy among children in Italy (Vezzali et al., 2017) and South Africa (Swart et al., 2011).
The positive impact of intergroup contact on attitudes towards outgroups is not limited to low-conflict societies. Instead, the vast majority of studies have demonstrated the positive role of intergroup contact in shaping attitudes towards outgroups in conflict and post-conflict societies. For example, positive contact between Protestants and Catholics in Northern Ireland and between Greek Cypriots and Turkish Cypriots in Cyprus emerges as an important predictor for stronger future contact intentions and improved attitudes towards the conflicted outgroup (McKeown & Psaltis, 2017). High-quality contact between Protestants and Catholics in Northern Ireland, and between Whites and Blacks in South Africa predicts more positive intergroup attitudes, increased trust, improved perceptions of outgroup intentions towards peacebuilding, and greater involvement in reconciliation efforts (Tropp et al., 2017). In light of these findings, our meta-analytical review aims to explore the relationship between intergroup contact among Turks and Kurds and attitudes towards outgroups in Turkey.
Intergroup contact research has consistently demonstrated the positive effects on intergroup relations for both minority and majority groups. However, the original theory highlights the importance of group status in determining the strength of these effects (Allport, 1954, for a review see also Al Ramiah & Hewstone, 2013). It is well-established that members of minority groups generally derive less benefit from intergroup contact and cross-group friendship compared to their counterparts in the majority group (see Pettigrew & Tropp, 2006; Wright et al., 2017). Given these insights, we aim to conduct a moderation analysis to explore whether the relationship between intergroup contact and outgroup attitudes varies between Turks and Kurds.
Furthermore, evidence has shown that intergroup contact reduces perceived discrimination among minority group members (Tropp et al., 2012). This effect of intergroup contact on minority groups is often criticized for producing a “sedative effect” (Çakal et al., 2011; Reicher, 2007) and “irony of harmony” (Saguy et al., 2009; for meta-analytical review, see Reimer & Sengupta, 2023). Dixon et al. (2007) suggested that intergroup contact is associated with decreased support for minority rights among minority groups. Wright and Lubensky (2009) also highlighted that intergroup contact might reduce ingroup identification and salience of intergroup inequality; in turn, it may lead to lower intention to participate in collective action for social change. Subsequent longitudinal research by Tropp and colleagues (2012) has shown that greater contact with Whites leads to lower intention to engage in activism for minority rights through decreased perceived discrimination among racial minority group members in the US. Therefore, our study investigates the impact of intergroup contact between Turks and Kurds on these two key factors as well: perceived discrimination against the disadvantaged group (i.e., Kurds) and support for minority rights in Turkey. Once again, we intend to explore the role of group status as a moderator in shaping these relationships.
In addition to attitudes toward outgroups, perceived discrimination against minorities, and support for minority rights, our study also examines the relationship between intergroup contact and ethnic identity in the context of the Turkish-Kurdish conflict. Ethnic identity plays a pivotal role in intergroup dynamics, particularly in conflict settings, where tensions often stem from differences in cultural and ethnic backgrounds. Ethnic identity has been identified as a significant predictor of various factors such as perceived threat (Verkuyten, 2009), negative attitudes towards minority groups (Verkuyten, 2011), decreased support of majority group members for minority rights (Uluğ & Uysal, 2023, see also Uysal et al., 2024). Therefore, we aim to investigate the relationship between intergroup contact and ethnic identity among Turks and Kurds.
Method
Inclusion Criteria and Literature Search
To be included in this review, studies had to meet the following initial inclusion criteria: indexed in Web of Science, Scopus, or PsycINFO; written in English or Turkish; conducted in Turkey with Kurdish and/or Turkish samples; testing the role of intergroup contact in this context. Despite the absence of a Kurdish search option in these databases, we conducted a thorough search in the Nûbihar Akademî journal, which, to our knowledge, is the only peer-reviewed and indexed publication featuring original research in Kurdish on Kurds in Turkey within the realm of social sciences. However, none of the Kurdish articles met the inclusion criteria or aligned with the scope of our review. We also excluded articles focused on the Kurds in the diaspora (e.g., Germany, the UK, or Holland) or other countries (e.g., Iraq and Iran) where Kurds populated vastly. After the full-text examination and thematic analysis, we decided to include articles for four themes and conducted separate meta-analyses for each theme with relevant articles.
The present study followed the PRISMA guidelines (Liberati et al., 2009). We performed a comprehensive search in the databases in April 2020 (see Figure 1). In this search, we used the related search terms on the Kurds and the conflict in Turkey, and their variations in the keywords of articles: (Conflict OR Ethnic Minority OR Kurdish Issue OR Kurdish Question OR Reconciliation OR Contact OR Identity OR Identification) AND (Turk OR Kurd OR Turkey OR Turkish OR Kurdish OR Turkish-Kurdish OR Kurdish-Turkish). The initial database search yielded 373 articles. We refined the results by limiting the categories. Because a huge range of categories is presented by the database from engineering to medicine, we made sure that only the directly relevant categories were included in our search, such as social psychology, in this phase. First, we removed 16 duplicated articles. We also called for unpublished data from various channels (e.g., EASP and ISPP websites, email groups, and social media accounts) in English and Turkish.1 The authors then independently screened articles by title and abstract to identify articles that were relevant to the scope of our research, and 296 articles were removed based on inclusion criteria and the focus of the review. Where the authors did not initially agree on the articles that should be discarded, conflicts were resolved via consensus.
Figure 1
PRISMA Flow Diagram for Identification of Included Studies
After a full-text examination of the remaining 70 (61 published and nine unpublished articles), we decided to include 54 studies in the review for common themes and variables in the social psychology research on the Turkish-Kurdish conflict. These 54 studies were analysed thematically to determine themes and articles related to intergroup contact (Attride-Stirling, 2001). All studies were read repeatedly by all authors, and codes regarding the focus and variables of the studies were noted. The codes were discussed and grouped into clusters. We used an inductive approach to themes. We discussed the findings amongst ourselves and then formed themes accordingly. Finally, we created four themes which reflect the most common variables that were examined with intergroup contact: a) outgroup attitudes, b) perceived discrimination, c) support for minority rights or multiculturalism, and d) ethnic identity. Accordingly, 26 studies that were not assigned any of these final themes and focused on relevant but different topics in the context of the Turkish-Kurdish conflict such as reconciliation (e.g., Baysu & Coskan, 2018; Baysu et al., 2018), conflict narratives (Adelman et al., 2016; Uluğ & Cohrs, 2017b, 2017c), collective action (Tropp et al., 2021), values (e.g., Bayad & Cesur, 2018), and trust (Çelebi et al., 2014) removed from the final list. Hence, we identified 28 studies and conducted four meta-analyses using this pool. Some of the articles are used for multiple meta-analyses since they reported more than one relationship among those we focused on (see Table 1). Except one (Bagci, Piyale et al., 2021), all the papers were written in English. All the unpublished papers were either published during the writing process of this paper or discarded after thematic analysis. Therefore, we did not conduct a publication bias analysis since there was no unpublished study in the final data. Details including article lists and raw data are accessible in Uysal et al., 2024S.
Table 1
Descriptive Information of the Samples Included in Meta-Analyses
| Study | Sample Size | Ethnicity of the Sample | Measured Variables | Theme(s) | Correlation coefficient(s) | Sample Characteristics |
|---|---|---|---|---|---|---|
| 1. Bagci & Çelebi, 2017 (Study 1) | 153 | Kurdish | Perceived conflict, cross-group friendship, outgroup attitudes, outgroup responsibility, multiculturalism | 1, 3 | r1 = .18 r3 = .12 | Student |
| 2. Bagci & Çelebi, 2017 (Study 2) | 320 | Turkish | Perceived conflict, cross-group friendship, outgroup attitudes, outgroup responsibility, multiculturalism | 1, 3 | r1 = .21 r3 = .18 | Student |
| 3. Bagci & Çelebi, 2018 (Study 1) | 329 | Turkish | The quality of cross-ethnic friendship, r-ethnic identity, national identity, cross-ethnic friend’s ethnic identity, cross-ethnic friend’s national identity, outgroup attitudes, multiculturalism | 1, 3, 4 | r1 = .15 r3 = .13 r4 = -.07 | Student |
| 4. Bagci & Çelebi, 2018 (Study 2) | 178 | Kurdish | The quality of cross-ethnic friendship, r-ethnic identity, national identity, cross-ethnic friend’s ethnic identity, cross-ethnic friend’s national identity, outgroup attitudes, multiculturalism | 1, 3, 4 | r1 = .07 r3 = .10 r4 = -.15 | Student |
| 5. Bagci & Turnuklu, 2019 | 376 | Turkish | Positive and negative contact, ingroup identification, relative deprivation, perceived discrimination, outgroup attitudes, collective action tendencies, psychological well-being | 1, 4 | r1 = .45 r4 = -.16 | Student |
| 6. Bagci & Turnuklu, 2019 | 151 | Kurdish | Positive and negative contact, ingroup identification, relative deprivation, perceived discrimination, outgroup attitudes, collective action tendencies, psychological well-being | 1, 2, 4 | r1 = .39 r2 = -.21 r4 = -.18 | Student |
| 7. Bagci et al., 2017 (Study 1) | 356 | Turkish | Contact, perceived minority statues, perceived discrimination, perceived threat, support for multiculturalism, outgroup attitudes | 1, 2, 3 | r1 = .26 r2 = -.20 r3 = .26 | Student |
| 8. Bagci et al., 2017 (Study 2) | 82 | Turkish | Contact quality, support for multiculturalism, affective outgroup attitudes, attitudes towards culture maintenance, attitudes towards contact | 1, 3 | r1 = .74 r3 = .49 | Student |
| 9. Bagci, Stathi, et al., 2021 (Study 1) | 384 | Turkish | Direct contact, extended contact, ingroup attitudes, outgroup attitudes | 1 | r1 = .31 | Student |
| 10. Bagci et al., 2018 (Study 1) | 47 | Turkish | Contact, ethnic identification, outgroup attitudes, intergroup anxiety, perceived group threat, support for multiculturalism | 3, 4 | r3 = .66 r4 = -.06 | Student |
| 11. Bagci et al., 2018 (Study 2) | 107 | Turkish | Contact, ethnic identification, outgroup attitudes, perceived group threat, intergroup anxiety, support for multiculturalism | 1, 3, 4 | r1 = .49 r3 = .42 r4 = -.20 | Student |
| 12. Bagci et al., 2018 (Study 3) | 55 | Kurdish | Contact, ethnic identification, perceived discrimination, outgroup attitudes, intergroup anxiety, perceived attitudes of the majority group | 1, 2, 4 | r1 = .53 r2 = -.18 r4 = -.03 | Community |
| 13. Bagci, Piyale, et al., 2021 | 142 | Turkish | Intergroup contact, perspective taking, ingroup identification, competitive victimhood | 4 | r4 = -.15 | Student |
| 14. Bagci, Piyale, et al., 2021 | 92 | Kurdish | Intergroup contact, perspective taking, ingroup identification, competitive victimhood | 4 | r4 = -.18 | Student |
| 15. Bagci et al., 2019a | 243 | Kurdish | Cross-group friendship, outgroup attitudes, ethnocentrism, social dominance orientation, cognitive-flexibility, perspective-taking and emphatic concern | 1 | r1 = .39 | Community |
| 16. Bagci et al., 2019a | 240 | Turkish | Cross-group friendship, outgroup attitudes, ethnocentrism, social dominance orientation, cognitive-flexibility, perspective-taking and emphatic concern | 1 | r1 = .59 | Community |
| 17. Bagci et al., 2019b (Study 1) | 80 | Kurdish | Imagined contact, perceived discrimination, ethnic identification, relative deprivation, outgroup attitudes, collective action tendencies | 1, 2, 4 | r1 = .14 r2 = .33 r4 = .03 | Community |
| 18. Bagci et al., 2019b (Study 2) | 127 | Turkish | Imagined contact, perceived discrimination, ethnic identification, relative deprivation, outgroup attitudes, collective action tendencies | 1, 4 | r1 = .39 r4 = -.08 | Student |
| 19. Bagci, Piyale, Sen, & Yildirim, 2019 (Study 2) | 66 | Kurdish | Acculturation preferences, perceived discrimination, belongingness, social acceptance, ethnic identification, contact | 2, 4 | r2 = .11 r4 = .48 | Community |
| 20. Bagci et al., 2020 (Study 1) | 175 | Turkish | Intergroup contact, prejudice, avoidance behavioral tendencies | 1 | r1 = .39 | Community |
| 21. Bagci et al., 2020 (Study 2) | 194 | Turkish | Positive intergroup contact, negative intergroup contact, prejudice, social distance towards the outgroup | 1 | r1 = .30 | Student |
| 22. Bikmen & Sunar, 2013 | 163 | Turkish | Contact with Kurds, contact with Armenians, ethno-national identification, religious identification, social dominance orientation, belief in cultural diversity, perceptions of discrimination, commonality vs. power talk with Kurds, commonality vs. power talk with Armenians | 2, 3, 4 | r2 = -.04 r3 = .00 r4 = -.15 | Student |
| 23. Bilali et al., 2018 | 1361 | Turkish | National identification, religious identification, religious practice, intergroup contact, perceived threat, empathy, social distance | 1, 4 | r1 = .51 r4 = -.13 | Community |
| 24. Çakal et al., 2016 (Study 1) | 289 | Turkish | Ingroup identification, perceived threat, intergroup contact, collective action tendencies, outgroup evaluations | 1, 4 | r1 = .57 r4 = -.09 | Community |
| 25. Çakal et al., 2016 (Study 2) | 209 | Kurdish | Ingroup identification, perceived threat, intergroup contact, collective action tendencies, outgroup evaluations | 1, 4 | r1 = .34 r4 = -.12 | Community |
| 26. Çelebi et al., 2016 | 376 | Turkish | Ethnic identification, national identification, cross-group friendship, discrimination of Kurds, outgroup national unity, Kurdish language rights | 2, 3, 4 | r2 = .09 r3 = .26 r4 = -.19 | Student |
| 27. Çelebi et al., 2016 | 320 | Kurdish | Ethnic identification, national identification, cross-group friendship, discrimination of Kurds, outgroup national unity, Kurdish language rights | 2, 3, 4 | r2 = -.27 r3 = -.23 r4 = -.30 | Student |
| 28. Firat & Ataca, 2020 | 168 | Turkish | Ethnic identification, national identification, prejudice, contact frequency | 1, 4 | r1 = .39 r4 = -.21 | Student |
Note. Theme 1: intergroup contact – outgroup attitudes, Theme 2: intergroup contact – perceived discrimination, Theme 3: intergroup contact – support for minority rights, Theme 4: intergroup contact – ethnic identification.
Analytical Strategy
We conducted analyses with the MAJOR package (version 1.2.0) of jamovi software (Hamilton, 2020). To consider the sample sizes, we transformed correlation coefficients into Fisher z correlation coefficients for analysis and back-transformed them for results. Since we consider that effect sizes vary according to different samples in studies and there are contradictory findings in intergroup contact studies on the Turkish-Kurdish conflict, we used the random-effects model, which assumes that effect sizes vary across studies.
Although there were many potential moderators that may impact the relationship between our core variables such as valence of contact or conceptualization and measurement of outgroup attitudes, we decided to focus on the moderator variables that were categorically well distributed among studies. For instance, as only a few studies measured negative contact and all of the others focused on positive contact, contact valence was not a suitable moderator variable for reliable subgroup analysis. However, the categories of group status, type of contact measure, and sample composition are very well distributed across studies. Therefore, we tested moderator roles of 1) group status (Turks vs. Kurds) 2) type of contact measure (single item quantity measure, single item quality measure, and multi-item scales), and 3) sample composition (student vs. community sample) for applicable themes.
Results
Theme 1: Intergroup Contact and Outgroup Attitudes
Across the 22 studies (N = 5,624), seven of the collected data were from Kurdish samples, and fifteen were from Turkish samples (see Figure 2). The average correlation between intergroup contact and outgroup attitudes was rz = .42 (95% CI [.327, .503], z = 9.28, p < .001). This result indicates that there is a moderate to strong (Cohen, 1992) positive relationship between intergroup contact and positive outgroup attitudes in social psychological studies of the Turkish-Kurdish context. That is, more frequent contact between Turks and Kurds is associated with more positive attitudes against the outgroup.
Figure 2
Forest and Funnel Plots for Theme 1 (Contact – Outgroup Attitudes)
Moderator analyses showed that group status or ethnicity is not a significant moderator in the relationship between intergroup contact and outgroup attitudes (r = .16, 95% CI [-.021, .343], p = .084). In other words, the relationship between intergroup contact and outgroup attitudes did not significantly differ among Turks and Kurds. Moreover, we tested the contact measures in the studies as a moderator. Accordingly, we categorised the contact measures into three groups: contact quantity or frequency with a single item (e.g., How many Kurdish friends do you have?), contact quality with a single item (e.g., To what extent is your contact with Turks positive?), and multi-item measures that consisted of different aspects or contexts of intergroup contact (e.g., How often do you talk to your Kurdish friends?; How often do you spend time with them socially?; How often do you visit them at their home?). The type of contact measure did not significantly moderate the relationship between intergroup contact and outgroup attitudes (r = .05, 95% CI [-.042, .148], p = .274). Finally, we tested whether studies recruited from student or community samples differed in terms of the relationship between intergroup contact and outgroup attitudes. However, sample composition did not moderate the relationship as other moderators (r = .10, 95% CI [-.075, .282], p = .255).
Theme 2: Intergroup Contact and Perception of Discrimination Against Kurds
We included eight studies (N = 1,567) for the meta-analyses on the relationship between intergroup contact and the perception of discrimination against Kurds (see Figure 3). We did not include studies that measured the perceived discrimination of the majority group member. Only the studies that measure perceived personal discrimination of minority members (Kurds) and majority group members’ (Turks) perception of discrimination against minority group members were included. Our decision not to focus on the perceived discrimination experiences of Turks was based on the fact that examining the perceptions of victimisation among the dominant group requires additional framing (such as competitive victimhood, see Noor et al., 2017), given that it is a qualitatively different phenomenon (i.e., competition-driven, instead of based on prejudice), which goes beyond our scope. We also believe that such a meta-analysis might not be feasible or reliable, given that only a few studies report on discrimination perceived by the majority group for comparative analysis.
Figure 3
Forest and Funnel Plots for Theme 2 (Contact – Perceived Discrimination)
Accordingly, five of eight studies collected data from a Kurdish sample and three from a Turkish sample. The average correlation between intergroup contact and perception of discrimination against Kurds was found as rz = -.06 (95% CI [-.199, .088], z = -0.76, p = .450) which implies that the relationship between contact and perceived discrimination was not significant, corresponding with the conflicting findings. Moreover, moderator analyses showed that group status or ethnicity did not moderate the relationship between intergroup contact and perception of discrimination against the minority group (r = .01, 95% CI [-.313, .323], p = .974). In other words, the relationship between intergroup contact and perception of discrimination against Kurds did not significantly differ among Turks and Kurds. Furthermore, sample composition (student vs. community sample) did not moderate the relationship between intergroup contact and perception of discrimination against the minority group (r = .14, 95% CI [-.048, .510], p = .105). Since seven of the eight studies used single-item contact quantity measures, we did not conduct a moderator analysis with contact measures.
Theme 3: Intergroup Contact and Support for Multiculturalism or Minority Rights
Across the 11 studies (N = 2,431), three were conducted with Kurdish samples and eight with Turkish samples (see Figure 4). The average correlation between intergroup contact and support for multiculturalism and minority rights was rz = .22 (95% CI [.069, .373], z = 2.82, p = .004). This result indicated that there is a small to moderate positive relationship between intergroup contact and support for multiculturalism or minority rights. That is, more frequent contact between Turks and Kurds is associated with greater support for multiculturalism or Kurdish minority rights. Furthermore, moderator analysis showed that the group status (i.e., ethnicity of participants) moderates the relationship between intergroup contact and supporting minority rights or multiculturalism (r = .31, 95% CI [.025, .602], p = .033), indicating that the relationship is significant only for majority group members (i.e., Turks; rz = .35, 95% CI [.194, .495]) while the relationship was nonsignificant among minority group members (i.e., Kurds; rz = -.01, 95% CI [-.241, .220]). That is, while Turks’ contact experience with Kurds was associated with their support for Kurdish minority rights, there was no significant relationship between Kurds’ contact experience with Turks and their support for Kurdish rights. The type of contact measure, on the other hand, did not moderate the relationship between intergroup contact and supporting minority rights (r = .08, 95% CI [-.122, .280], p = .439). Since all the studies collected data from student samples in this theme, we did not conduct moderation analysis for a sample composition.
Figure 4
Forest and Funnel Plots for Theme 3 (Contact – Support for Minority Rights)
Theme 4: Intergroup Contact and Ethnic Identity
Across the 19 studies (N = 4,636), eight were conducted with Kurdish samples and eleven with Turkish samples (see Figure 5). The average correlation between ethnic identity and intergroup contact was rz = -.13 (95% CI [.019, -.007], z = -4.23, p < .001). This result indicates that there is a small negative relationship between ethnic identification and intergroup contact. That is, higher identification with the ethnic group is associated with less frequent (or worse quality) contact between Turks and Kurds.
Figure 5
Forest and Funnel Plots for Theme 4 (Contact – Identity)
Moderator analyses showed that group status or ethnicity did not moderate the relationship between intergroup contact and ethnic identity (r = -.03, 95% CI [-.166, .097], p = .606). In other words, the relationship between intergroup contact and ethnic identity did not significantly differ among Turks and Kurds. While the type of contact measure did not significantly moderate the relationship between intergroup contact and ethnic identity (r = -.03, 95% CI [-.094, .043], p = .466), sample composition (student vs. community sample) moderated this relationship (r = .16, 95% CI [.039, .270], p = .009), indicating that the relationship between ethnic identity and intergroup contact was significant only among studies that collected data from student sample (rz = -.18, 95% CI [-.228, -.135]), not significant for studies collected data from community sample (rz = .00, 95% CI [-.177, .184]).
Discussion
This meta-analytical scoping review examined social psychological studies on intergroup contact in the Turkish-Kurdish conflict. We tested the relationships between a) intergroup contact – outgroup attitudes, b) intergroup contact – perception of discrimination against Kurds, c) intergroup contact – support for minority rights, and d) intergroup contact – ethnic identity. We conducted four meta-analyses on these four themes from 28 quantitative studies conducted with Turkish and/or Kurdish samples in Turkey (see summary in Table 2). Our findings showed a moderate to strong positive relationship between intergroup contact and positive outgroup attitudes. This was the strongest relationship that emerged among the four themes we examined. It is fair to say that social psychological studies conducted in Turkey and focused on the link between intergroup contact and outgroup attitudes in the context of the Turkish-Kurdish conflict provided strong, concrete findings regarding this relationship in general.
Table 2
Summary of Meta-Analyses
| Theme | Number of Included Studies | N | rz | Moderators | |
|---|---|---|---|---|---|
| 1. | Intergroup Contact – Outgroup Attitudes | 22 (7 from Kurds; 15 from Turks) | 5,624 | .42 | – |
| 2. | Intergroup Contact – Perceived Discrimination | 8 (5 from Kurds; 3 from Turks) | 1,567 | -.06ns | – |
| 3. | Intergroup Contact – Multiculturalism / Minority Rights | 11 (4 from Kurds; 7 from Turks) | 2,431 | .22 | Group status: the relationship is significant only for majority group members (e.g., Turks; rz = .35, 95% CI [.194, .495]) |
| 4. | Intergroup Contact – Ethnic Identification | 19 (8 from Kurds; 11 from Turks) | 4,636 | -.13 | Sample composition (student vs. community): the relationship is significant only among studies who collected data from student sample (rz = -.18, 95% CI [-.228, -.135]) |
Note. N = Total sample size; rz = Average correlation coefficient; ns = non-significant.
On the other hand, studies on intergroup contact and perception of discrimination against Kurds suggested inconsistent and null results. We also found a small to moderate positive relationship between intergroup contact and support for multiculturalism or minority rights. However, moderation analysis showed that this relationship was only significant among Turks, while it was non-significant among Kurds. This finding suggests that while intergroup contact is useful for Turkish majorities’ support for minority rights, it is an ineffective, even sometimes adverse, predictor for Kurdish minorities, as argued in the “irony of harmony” literature that focuses on the negative impact of intergroup contact on minorities’ social change efforts (e.g., Dixon et al., 2007; Saguy et al., 2009; Wright & Lubensky, 2009).
Another inconsistent set of results was provided by research on the relationship between intergroup contact and ethnic identity. Although we found a small negative relationship between identity and intergroup contact, the sample composition of studies (i.e., a student vs. community sample) moderated this relationship, indicating that the relationship between identity and intergroup contact was significant only among studies that collected data from university students. This finding reveals the problem of generalizability of the studies on Kurdish identity and conflict in Turkey (for a broader discussion, see Uysal et al., 2024). Considering the influence of liberal education on the dynamics related to prejudice (Duckitt, 1992) and the problems with the student samples in terms of generalizability and external validity in research, together with the political vulnerability of the public discourse of the Kurdish issue, the use of student samples should be addressed carefully. It also reveals that the contact experience and its consequences may differ across disadvantaged subgroups. Hence, further studies and methods that consider the heterogeneity of disadvantaged groups, such as latent class and profile analyses, are needed.
On the other hand, most studies with or without student samples were conducted in the western parts of Turkey (i.e., Turkish-majority parts of the country) with the measurement devices in Turkish. None of the studies in this meta-analysis included items in the Kurdish language, with very few conducting research in Kurdish-majority regions (but see Bagci & Çelebi, 2017; Bagci et al., 2019b; Bilali et al., 2018 for some exceptions). There may be several reasons that the researchers have chosen to conduct research in Western Turkey. One possible reason may be due to their own connections and networks – without colleagues working in those regions, they may have difficulty collecting data. Alternatively, they may also find that they can collect data in those regions if they can afford to hire someone to do so. This could also be related to their own biases in going to those regions - the perception of difficulty, danger, and risk may prevent some researchers from making the trip. Finally, the political risks of conducting research in Kurdistan, especially after the end of the peace process in 2015 and the reinstatement of securitization discourse and militarization policies by the government, might be a factor that causes researchers to refrain from research in Kurdish-majority areas.
Another potential issue lies with Allport’s (1954) conditions for intergroup contact. Among other conditions, he suggests that equal status is necessary between those in contact. Establishing equality in any situation can be difficult, but in a context where there is little to no institutional support for the equal recognition of Kurdish identity, it becomes even more difficult. In the studies included in the meta-analysis, even student populations come from universities in Turkish-majority cities or regions. Therefore, Kurdish students in these contexts are still beholden to “Turkifying” themselves in order to fit in. Among the narratives of the conflict, the question of a “common identity” is still unresolved (Somer, 2022), where the commonality is dominantly defined via the majority identity. Participants in Kurdish-majority universities or cities could have provided more information as to whether the condition of equality could be established in these studies.
Another critical point in the studies we reviewed was that in most of the quantitative studies on dynamics such as discrimination, conflict, contact and identity, which have a strong historical and political content, information on when the data were collected was not given. However, in studies conducted in a country where the political agenda is highly intense and variable, the knowledge of the period in which the data were obtained is essential to evaluate the intergroup dynamics studied within the historical and political context.
Our results showed there are robust findings only for the relationship between intergroup contact and outgroup attitudes, while contact’s impact on the broader construal of intergroup relations or efforts for peace is either non-significant or small. This corresponds with Burrows et al. (2022), which demonstrated that intergroup contact has a relatively stronger effect on intergroup attitudes toward ethnic outgroups, while its effect on the broader construal of relations between groups in society remains limited in the contexts of inter-ethnic conflicts like Bosnia. We argue that intergroup contact studies that do not engage in structural analysis of historical narratives and political projects that shape social and cultural reality will be far from a comprehensive understanding of triggers for social change and large-scale peacebuilding. For instance, the dominant "terrorism" narrative in Turkey, the lack of (political, discursive, and scientific) acknowledgement of colonial practices in Kurdistan, and the ongoing armed conflicts appear to be an important obstacle to extending the temporal and individual positive effects of contact to the group and societal level. The Kurds, who are involved in the Kurdish struggle in some way, and even politicised in areas such as women, LGBTQA+, climate, or workers' struggle outside the Kurdish movement, encounter the narrative of terrorism, hence, they are stigmatized (European Parliament, 2023; Uluğ & Uysal, 2023). Whether the contact experience with Kurds in these studies can be extended to the experiences with the politicised, stigmatised, and criminalized Kurds remains an important question.
In none of the studies in our review, we could not find any statement on reflexivity and positionality. The researchers’ identity and their insider, outsider, or in-between positions influence the research question we raised, the epistemological and methodological preferences we adopted, and the conclusions that we derived (Türkmen, 2023). In addition, the reader should be informed about the relationship between the participants and the researcher, how the participants felt during this relationship, which participants preferred to participate in the research and which ones chose not to participate.
During all the socio-political developments we summarised in the historical overview, contemporary psychology in Turkey has shown limited attention to societal issues, political struggles, and ethnic minorities like Kurds (Bayad et al., 2022). With the founding Kemalist ideology of the state, Western-inclined modernization around homogenous Turkish national identity has always been one of the greatest aims, and psychology has an important role in supporting these state aims (Bayad & Şen, 2023). It’s possible given that state ideology is embedded in scientific discourse that psychology has deliberately overlooked these methodological issues in its approach to Kurdishness. The peace process, initiated and supported by the Turkish government in the 2010s and failed in 2015, created space for psychology to explore Kurdishness and conflict in an environment that was more supportive of this work. However, the production of knowledge surrounding Kurdishness has been influenced by epistemic violence, with research often constrained by methodological nationalism, limiting perspectives to fit within nation-state narratives (Wimmer & Schiller, 2002). Methodological issues and political biases often constrain research, hindering a nuanced understanding of the Kurds on these issues which may result in studies of intergroup contact between Turks and Kurds predominantly reflecting majority (Turkish) perspectives of Kurds, for instance. It can also lead to the conceptualization and measurement of national and ethnic identities in a way that fails to fully consider the perspectives and dynamics of the minority (Kurdish) party in the conflict, due to the embedded state discourse in the scientific literacy that limits the perspective and identity expressions of minorities (Uysal et al., 2024).
Although the history of the conflict between Turks and Kurds living in Turkey goes back to the 1920s when the republic was founded (and it is possible to trace the roots of the conflict in the Ottoman Empire to earlier times), we can come across the first empirical social psychological articles centred on this conflict in the 2000s. This strongly overlaps with Said’s (1978) analysis that indigenous peoples are not accepted artistically, academically, politically, historically and legally, which he sees as one of the basic building blocks of colonial societies. Ünlü (2016, 2018) states that the Kemalist founding ideology in Turkey prevents producing information about the place and role of Kurds and other minority groups in social life, with the implied contract he calls the "Turkishness Contract". According to Ünlü, the lack of information on Kurds and other groups outside the tacit contract is not just an academic problem but basically an ideological preference. Therefore, it is important to examine this symbiotic relationship between academic interest and ideological approach in its reflections on social psychology studies on the Kurdish issue through decolonial critical approaches (Coşkan & Şen, 2023, 2024; Uysal et al., 2024).
Limitations and Future Research
Due to the quantitative nature of meta-analysis, we only focused on quantitative studies. Therefore, this study perhaps does not provide a full view of the type of research on Kurdishness that has been conducted in Turkey. Future research should also evaluate qualitative research into Kurdishness. It may be that qualitative research has produced more nuanced analyses of Kurdish identity than the evaluated quantitative research has (e.g., Coşkan & Şen, 2023, 2024; Kışlıoğlu & Cohrs, 2018; Uluğ & Cohrs, 2017b). We believe some guidance for studies with Kurdish participants should be laid out for future studies. While some previous work has discussed this (see Uluğ et al., 2021), concrete steps for working with marginalized populations should be considered for Kurdish participants in Turkey as well.
A significant limitation of this study is its lack of papers written in Kurdish. Despite efforts to search for Kurdish-written articles, none were included in the final list of articles. We believe that this limitation goes beyond a mere language barrier and warrants discussion within the context of power dynamics and colonial mindset within the academic sphere in Turkey, as well as the distinction between insider and outsider researchers in conflict contexts (Uluğ et al., 2021). One clear outcome of methodological nationalism in this case is the lack of prestige for Kurdish-language publication, and the general lack of support for publishing in Kurdish in Turkey. Moving forward, it is imperative for future studies to not only incorporate Kurdish-language papers but also engage insider researchers who may offer unique perspectives and positionalities for Kurdish-language knowledge production. The concept of an "insider researcher" highlights the significance of shared characteristics or perceived closeness to the studied community (Kirpitchenko & Voloder, 2014). This distinction becomes pivotal in understanding research dynamics in conflict zones where researchers' identities influence outcomes and participants' perceptions. Additionally, reflexivity in research is essential, prompting researchers to consider contextual factors such as the timeframe and historical events shaping participants' experiences and responses (Türkmen, 2023). This becomes particularly crucial as most studies in our review did not report the date of data collection, despite the ever-changing political landscape and intermittent armed conflicts, necessitating an understanding of the study period for interpreting findings and conclusions.
Our study focused solely on research within the Turkish context of the Kurdish conflict, excluding studies from the Kurdish diaspora in Europe or Kurds in other Southwest Asia countries. While this allowed for a more targeted analysis of dynamics within Turkey, it may limit the generalizability of our findings. Additionally, variations in political power dynamics across geopolitical regions in Southwest Asia, such as those where the Kurdish freedom movement has established self-governed areas like Rojava, highlight the complexity of the conflict and the importance of considering diverse contexts in analysing intergroup dynamics and conflict resolution strategies. Also, many Kurds and Turks live in diasporas, such as Germany, Belgium, the Netherlands, and the UK. Examining the intergroup relations between Turks and Kurds in contexts where their distinct ethnic identity combined with similar immigrant identity may provide interesting findings on the intersectional nature of this conflict (e.g., Baysu et al., 2018; Ufkes et al., 2015; Ünal et al., 2022).
Due to Turkey’s political context, it is risky to study the Kurdish issue. Uluğ et al. (2021) showed that political context strongly influences how participants decide to participate in research about this conflict context. The researchers' identity is another important reason to decide whether to participate in research. Overall, since the conflict is still ongoing and has consequential effects on racism, prejudice, and negative intergroup attitudes, 1) collecting data is more difficult, 2) researchers are reluctant to travel to Kurdish regions, and 3) participants are hesitant to take part. It could be one reason why there are so few studies in general related to this issue and in social psychology in particular. Considering the tumultuous nature of the political context in the region, and as our meta-analysis is sensitive to this context, we also believe it is important for this meta-analysis and review of the literature to be conducted again in the future.
Our meta-analysis did not incorporate studies from grey literature, such as unpublished works and theses. One potential reason for this limitation on grey literature could be the prevailing political oppression in the Turkish academic sphere, which hinders the easy examination of the Turkish-Kurdish conflict, especially for postgraduate students who are already in a precarious position in the oppressive Turkish academy. However, there is a clear need for further review studies to delve into unpublished papers and theses, as they could offer more comprehensive insights into scientific knowledge regarding the Turkish-Kurdish conflict.
Lastly, there is a need to conduct interviews or surveys with social and political psychologists in Turkey and ask them about their experiences and goals around researching the conflict and Kurdishness: What challenges did they face? What sorts of things prevented them from wanting to do research in this context? There also needs to be a discussion about how different dynamics are measured. Indeed, this is a broader epistemic discussion, but in Turkey, it may become contextual. Since researchers don't just be afraid to conduct research in this context, they are also afraid to measure Kurdish identity in particular ways, and we need to dive into researcher experiences in performing research and conceptualization.