Ethnic Boundaries in Turkish Politics. Zeki Sarigil

2013).

      One might expect that similar developments and processes have taken place in the Turkish polity. Indeed, several political formations and movements have adopted a more Islamic discourse and position in the post-1980 period, partly because of the declining influence of ideology politics (i.e., left-right divisions and polarization) in the Turkish polity and the increasingly welcoming attitude of the Turkish state toward Islam and religious orders. Within such a political environment, a shift toward Islam among right-oriented political formations is not that surprising. It is much more striking, however, to see such a transformation taking place within strictly secular, leftist formations, such as the Kurdish ethnonationalist movement. As stated earlier, the Kurdish movement, which is rooted in strong secularism and Marxism, has been flying ethnoreligious colors, especially since the early 2000s. We do not really see such a systematic effort to accommodate Islam and Islamic actors among other leftist circles in the Turkish polity. We do see some initiatives by other leftist political formations or movements to appeal to conservative or pious voters (e.g., efforts to co-opt conservative politicians from the center-right or having female party members wearing headscarves). However, as the empirical chapters of the book illustrate, such cosmetic or superficial and short-term initiatives are not really comparable with much more systematic, comprehensive, and long-term efforts of the secular Kurdish movement to reinterpret Islamic ideas, values, and principles in line with its ethnonationalist interests and objectives and to co-opt Islamic actors. This difference is probably because of the fact that the constituency of the Kurdish ethnonationalist movement has been relatively more religious (for more on this, see chapter 3).

      Despite this striking change, there is a very limited number of works directly dealing with the increasing role of religion in the Kurdish ethnopolitical movement in the Turkish context, for example, works by van Bruinessen (2000a, 2000b), Houston (2001), and Gürbüz (2016). However, in addition to being outdated, the works by van Bruinessen and Houston do not really provide a comprehensive theoretical and explanatory analysis of the secular Kurdish movement’s evolving relations with Islam. Gürbüz’s study indirectly touches on pro-Kurdish groups’ relations with Islam by examining competition or rivalry among social movement organizations in southeastern Turkey (e.g., pro-Islamic groups such as Kurdish Hezbollah and secular pro-Kurdish civil society organizations). In other words, this recent work does not provide a detailed and thorough account of the Kurdish ethnopolitical movement’s Islamic opening in the past decades either. In brief, it is surprising that this intriguing development in secular Kurdish ethnopolitics and its theoretical and practical implications still remain an understudied and undertheorized phenomenon in the fast-growing scholarly literature on the Kurdish issue.

      Thus, it is worth raising the following research questions: Why do we see the aforementioned, unexpected shift? What does it mean? How can we explain this interesting development in secular Kurdish ethnopolitics? For what reasons (ideational and/or material) do ethnopolitical leaders and entrepreneurs develop new attitudes and discourses toward religion and/or ethnicity or nationalism? Why and how do political elites swing between “religionism” and “nationalism”? What are the general causes and mechanisms of such swings? What might the enabling and/or constraining factors be in ethnopolitical elites’ efforts to mold ethnic categories and movements? Further, how are ethnonational boundaries affected by such elite actions and discourse? What roles do elites play in ethnic boundary-making processes? Why, when, and how do they demarcate, maintain, and transform the symbolic and social boundaries of an ethnic category or movement? What might the implications of the case of the Islamic opening of the secular, leftist Kurdish movement be for the broader theoretical debate on ethnicity and nationalism, particularly for ethnic boundary-making processes?

      Methodological Approach

      Focusing on the Islamic opening of the secular, left-oriented Kurdish ethnopolitical movement in Turkey in the past decades, this research project conducts a case study. Defined as “the intensive study of a single case for the purpose of understanding a larger class of similar units” (Gerring 2007, 20) or, similarly, as an “extensive and in-depth analysis of some social and political phenomenon” (Yin 2009, 4), the case study is quite useful for investigating novel phenomena, analyzing complex causal mechanisms and processes, and assisting with hypothesis generation as well as theory development and refinement (see George and Bennett 2005; Gerring 2007; Levy 2008; Brady and Collier 2010).

      With regard to the units of analysis, this study qualifies as a single-case study, and so I focus on within-case variation (i.e., shifts in the Kurdish movement’s attitudes toward Islam and Islamic actors). In addition, I conduct process tracing, which is a particular type of within-case analysis. Associated with small-N, case-oriented research, process tracing is advantageous for uncovering causal mechanisms, processes, and sequences underlying observed phenomena or outcomes (see George and Bennett 2005). Process tracing is also quite helpful for analyzing the sequence of events, decisions, and steps over time that links the initial events and conditions to outcomes (i.e., the Islamic opening of the secular Kurdish ethnonationalist movement in Turkey).

      To obtain more credible, valid, and persuasive results, and thus a more comprehensive and deeper understanding of the issue, I also use triangulation as a research strategy, which simply means “the observation of the research issue from (at least) two different points” (Flick 2004, 178). Triangulation is achieved by using multiple research techniques and a wide range of data sources (Rothbauer 2008; S. Hastings 2010; Stoker 2011). Such a combination (qualitative and quantitative evidence) is expected to increase inferential leverage and confidence in the ensuing results and so generate a more accurate and complete picture of the phenomenon under investigation (Bryman 2004; Cox and Hassard 2010; Tarrow 2010; Stoker 2011). I apply data triangulation, which refers to utilizing “data drawn from different sources and at different times, in different places or from different people” (Flick 2004, 178). Thus, this research project utilizes various approaches to and techniques of data collection and data sources, primarily semistructured, in-depth interviews; ethnographic field research; textual analyses (e.g., discourse analysis of key documents); and electoral data.

      For this study, I conducted 104 interviews with 88 participants in several provinces in Turkey (primarily Ankara, Diyarbakır, Istanbul, and Tunceli) between 2011 and 2015.7 Informants were selected from different political circles (e.g., Islamic and secular), professions (e.g., politicians, local administrators, intellectuals, columnists, activists, civil society representatives), and ethnic origins (e.g., Turks and Kurds).8 I also participated as an observer in the Democratic Islam Congress, organized by the Kurdish movement in Diyarbakır in May 2014, where I had the opportunity to sit in on and make detailed notes of discussions within the Kurdish movement about religious issues. My empirical analyses are enriched with textual analyses that include discourse analyses of the key documents and texts (e.g., publications or political programs of Kurdish organizations) and statements by Kurdish ethnopolitical elites and leading figures of the Kurdistan Workers’ Party (Partiya Karkaren Kurdistan, PKK) and electoral data (i.e., electoral performance of pro-Kurdish political parties at local and general elections).

      Conceptual Issues

      It has been observed that “conceptual wrangles continue to haunt” ethnicity and nationalism studies (Cederman 2002, 422). As such struggles still persist in the literature (see also Hale 2004; Wolff 2006; Chandra 2012b), it is necessary to define the key concepts used throughout this book, that is, ethnicity, ethnonationalism/ethnopolitics, nation, and nationalism. When we look at the existing literature, we see a plethora of definitions and uses of these concepts, some of which are contradictory. Instead of addressing the persisting conceptual challenges or offering the “right” definition, in this volume I simply provide a working definition of these terms. Rather than creating new definitions, I benefited from several definitions of these key terms in the existing literature (e.g., Weber 1968; Schermerhorn 1970; Kasfir 1979; van den Berghe 1981; Gellner 1983; Horowitz 1985; Anderson 1991; Olzak 1992; Cornell 1993; Connor 1994; Hutchinson and Smith 1996; Kellas 1998; Conversi 1999; Cederman 2002; Brubaker, Loveman, and Stamatov 2004; Hale 2004; Chai 2005; Chandra 2006, 2012c; Smith 2006; Cornell and Hartmann 2007; R. Jenkins 2008; Bayar 2009; Brubaker 2009; Wimmer 2013).

      The existing definitions tend to treat ethnicity as a type