Le vote minoritaire dans les démocraties musulmanes : entre engagement, intégration et protestation

par Max-valentin Robert

Projet de thèse en Science politique

Sous la direction de Raul Magni-berton et de Jean Marcou.

Thèses en préparation à Grenoble Alpes , dans le cadre de Sciences de l'homme, du Politique et du Territoire , en partenariat avec Pacte, Laboratoire des sciences sociales (laboratoire) et de Sciences Po Recherche (equipe de recherche) depuis le 01-10-2015 .


  • Résumé

    Contexte : Mon travail de thèse s'inscrit dans un champ d'études explorant les conditions de réussite ou d'échec des processus de démocratisation [Haggard, Kaufman, 2016]. Au-delà des explications traditionnelles relatives à la fréquente prégnance de l'autoritarisme dans ce qu'il est convenu d'appeler le « monde musulman », nous proposons une interprétation de ce phénomène en testant l'hypothèse d'une influence des hétérogénéités/homogénéités linguistique et confessionnelle sur la nature des régimes politiques : en effet, il a été constaté que, si la diversité religieuse d'un pays accroît les probabilités de voir celui-ci être administré par un régime autoritaire, l'hétérogénéité linguistique augmente les chances de voir la société en question bénéficier d'un système démocratique [Gerring, Hoffman, Zarecki, 2018]. Par conséquent, nous émettons l'hypothèse que l'autoritarisme affectant le « monde musulman » serait alimenté par une homogénéité linguistique et une hétérogénéité religieuse particulièrement forte. Cette hypothèse sera testée dans la première partie de notre travail doctoral, à travers des analyses de régression portant sur un échantillon regroupant l'ensemble des pays majoritairement musulmans. Nous proposons également d'explorer cette relation entre composition ethnoculturelle d'une société et nature de son régime politique en partant de l'idée que l'identité minoritaire peut constituer une fonction de « raccourci informationnel » pour la prise de décision sur le plan électoral, contribuant ainsi à une routinisation des préférences partisanes et à une consolidation démocratique plus rapide [Birnir, 2006]. La réussite de la transition démocratique serait donc alimentée par une cristallisation des comportements électoraux de la part des groupes minoritaires. A l'inverse, cette absence de cristallisation et/ou d'accès à la sphère gouvernementale peut alimenter l'instabilité du régime, et même entraîner le recours à la violence extraparlementaire dans une fraction du groupe concerné. La seconde partie de notre thèse est donc consacrée à une étude de cas, portant sur un pays majoritairement musulman ayant eu la particularité de connaître une alternance de périodes démocratiques et de phases d'autoritarisme : la Turquie. Eu égard à notre volonté d'analyser les conséquences différenciées des hétérogénéités linguistique et confessionnelle, le choix de la Turquie s'imposait d'autant plus, car ce pays a la particularité de renfermer une double division relative à sa composition socioculturelle : une division exclusivement religieuse sunnites/alévis [Massicard, 2005] et une division exclusivement linguistique turcophones/kurdophones [Güneş, Zeydanlıoğlu, 2014]. Par conséquent, l'enjeu de notre travail sera de tester également les hypothèses mentionnées ci-dessus, à travers l'utilisation d'outils relevant de l'analyse de régression et de l'approche écologique du vote.

  • Titre traduit

    The minority vote in Muslim democracies : between commitment, protestation and integration


  • Résumé

    Context : My project of thesis is to question the existence of an Alevi electorate in Turkey. Indeed, this heterodox Muslim community is often considered as a strong supporter to left-wing parties (from center-left and Kemalist movements to far-left organizations). This work would be part of a larger trend in political sociology, which is to analyze the possible relationships between religious background and political views. For instance, in the 1940's United States, the political scientists Bernard R. Berelson, Paul F. Lazarsfeld and William N. McPhee (1954) showed that a Protestant background can contribute to make someone an electoral supporter of the Republican Party, whereas a Catholic background can push someone to vote for the Democrat Party. Likewise, in France, Guy Michelat and Michel Simon (1977) demonstrated that, in the1970's France, the atheists were prone to vote for the left-wing parties, while devout Catholics were more in favour of right-wing movements. More recently, Vincent Tiberj and Sylvain Brouard (2012) focused on French electors of North-African, Turkish or Sub-Saharan origins to question the existence of a « Muslim voting » in France. Besides, Gilles Kepel (2014) led interviews with Muslim candidates for the legislative elections of 2012, with intention of understanding the influence of religious values in their political commitment. More generally, many studies showed an influence of religion in many different fields, in the general social behaviour. For example, Robert J. Barro and Rachel McCleary (2003) underlined an existing positive correlation between religious beliefs and economic growth, and an economic negative correlation between economic development and assistance to worship. The sociologist John H. Evans (2014) highlighted that « In every country, every religious group has less faith in science producing meaning than do the non-religious ». The psychologist Gordon W. Allport (1966) also showed that religious people are more enclined to be prejudiced towards the otherness. Furthermore, Samuel A. Stouffer (1955) explained that religious people are more un-tolerant than unreligious persons. According to Philip S. Morgan (1983), believers would be more sociable : « prayerful people do seem more friendly and cooperative. » Likewise, as Thomas Piazza and Charles Y. Glock explained (1979), people who believe in God would have a more altruistic behaviour. Gerhard E. Lenski (1966) also said that religious individuals are more in favour of a humanitarian orientation of scholar integration and external assistance than atheist people, even if they express more liberal ideas about these issues. Moreover, the political scientist Dilek Yankaya (2013) demonstrated that, in Turkey, the « new Islamic bourgeoisie » mobilizes Muslim values for establishing business networks based on a common religious identity. Lastly, Pippa Norris and Ronald Inglehart showed (2004, 2011) that in Muslim countries, public opinions are less in favour of gender equality and sexual liberalism than in Western countries. Therefore, I would like to apply the current methodology of the electoral sociology to the Turkish example, with the intention of analyzing a possible link between Alevism and voting behaviour. Indeed, I think that it is necessary to apply this kind of methodology even for Middle Eastern countries, beyond« Orientalist »prejudices towards them (SAÏD, 1978, 2003). Moreover, I think that focusing on the Alevi community is necessary, because this population is characterized (if we compare them with other Muslim communities) by a peculiar relationship with religion and secularism : as Ruud Koopmans (among other things) underlined (2014), the Alevi immigrants in Europe express more detachment towards their beliefs than Sunni immigrants. Thus, we could verifiy if this phenomenon also is visible in Turkey, and if it implies a stronger support to secularist parties. Nevertheless, this Shia population is under-studied : when Emmanuel Todd and Youssef Courbage (2007) allude to the influence of Shia Islam on inheritance rules, they refer to the local Shia community of Lebanon, but they do not allude to the Alevi population of Turkey. Likewise, Pippa Norris and Ronald Inglehart (in the chapters of their 2004's book in which they deal with Muslim countries) do not evoke this population, whereas it could be interesting to compare the perception of gender equality in Sunni and Alevi Turkish communities. Of course, Elise Massicard (2005) wrote a relevant thesis about the emergence and the politicization of Alevism in Turkey. Nonetheless, I think that it is fundamental to analyze this issue with an electoral focus (as it was previously did about Western States and religious communities), beyond its historical and sociological aspects. Besides, it could be necessary to update the current (scarce) literature about the Alevi population, because Turkey is led (since 2002) by an explicitly Sunni conservative movement : the Justice and Development Party (Adalet ve Kalkınma Partisi, AKP), which is building a new national identity, based on the rehabilitation of the Ottoman Past and Islamic values : the « Muslim Nationalism » (WHITE, 2013). Moreover, the Gezi movement of 2013 showed a strong commitment of the Alevis (thus, 70 % of the protesters who were arrested by the police were from this community). Lastly, the Turkish policy concerning the Syrian civil war could be interpreted as an AKP's contribution to the latent growing clash between Shia and Sunni Muslims in the Middle East, and it could be relevant to study its consequences for the Turkish internal political field (TREMBLAY, 2015). In a nutshell, it could be relevant to verify if the AKP's policies are polarizing Sunni and Alevi electors, and if we are currently seeing a radicalization of the Alevi voting. Research question or aims : Consequently, during the writing of my thesis (which will be co-directed by Jean Marcou), I would like answer to the following problematic : are we seeing an increasing polarization between Sunni and Alevi voters since the takeover of the Justice and Development Party ? These other questions can be considered as linked to our general interrogation : Is there currently a growing radicalization of the Alevi voters ? Are Alevi activists motivating their partisan choices by communitarian and religious factors ? Are we seeing peculiar electoral behaviours in provinces and districts in which members of this heterodox community are numerous ? Indeed, as Robert D. Putnam explained (2004), the religious identity is a social capital which can be exploited in many circumstances, such as economic and political fields. For instance, Dilek Yankaya showed that in Turkey, the new conservative businessmen used their religious capital and their membership to the AKP-linked MÜSIAD (a conservative association of industrialists) as a way to ensure the prosperity of their firms, thanks to preferential markets and mutual aid between devout Sunni bosses. This theoretical framework could be applied to voting and partisan fields : I will try to see if Alevi voters or parties' supporters are using their Alevi identity as a social capital for justifying their political behaviour. This study could help us to see if this specific vote is communitarian or not. Empirical research methods : All along this work, I wish to use quantitative and qualitative methods for building my reasoning. Concerning qualitative methodology, I have the intention to lead semi-directed interviews with members of political parties who are from an Alevi background. Indeed, during my participation to a research seminar at the French Institute of Anatolian Studies (Institut Français d'Etudes Anatoliennes, IFEA), between February and June 2014, I wrote a study about the campaign which was made by the Republican People's Party (Cumhuriyet Halk Partisi, CHP : the main opposition and Kemalist party) in the Istanbul's district of Şişli, for the local elections of March 2014. Now, I realized that many CHP's activists that I met were born in Alevi families. I also have contacts with two other Istanbul's local branches of this party : Beyoğlu and Zeytinburnu. Thus, I will do semi-conducted interviews of Alevi activists who are from CHP's sections in which there are numerous Alevi activists : Şişli and Beyoğlu. Furthermore, I will interview habitants of the leftist neighborhood of Okmeydanı, in which is living a huge Alevi community. I would like to record semi-conducted interviews with Alevi habitants of this neighborhood who are simple electors of left-wing parties (as moderate as radical organizations), people who are members of semi underground movements (that is to say, armed movements such as CEPHE or DHKP-C) and persons who are responsible of the local place of worship : the Okmeydanı Hacı Bektaşi Veli Cem Evi. Beyond the Okmeydanı simple example, I would like to interview Alevis who participated to the Gezi protest movement of 2013. Besides, I wish to talk with Alevi activists of the pro-Kurdish party BDP (Barış ve Demokrasi Partisi, Peace and Democracy Party) of Tunceli, a town which is mainly populated by Alevi Kurdish people : thus, I would like to see if, in the Turkish political sphere, it is the religious identification which dominates the ethnicity or if it is the ethnical self-definition which transcend the religion. Furthermore, I would like to lead interviews with Alevi immigrants of Grenoble and Paris, for verifying if they show different political attitudes to Sunni immigrants. I would also see if a peculiar political consciousness is transmitted from generation to generation, despite the fact of living outside Turkey. However, I think that my project of work would be incomplete if I did not use quantitative data and methodology. Between September 2014 and January 2015, I could test (thanks to the software Stata) the eventual correlation between the secularization of a society and its support to the Welfare State. Thus, I wish to use this software also for seeing if there is a link between the proportion of self-identification to Alevism and the electoral support for left-wing parties. Likewise, I will verify if there is an existing positive correlation between a high support for left-wing parties and a huge Alevi population : I intend to use data which were collected at neighborhood level (for instance, in Okmeydanı and Şişli) and province scales (like in Tunceli, Sivas and Malatya). Reversely, I could also study the possibility of a negative correlation between a big Alevi community and an electoral support to right-wing parties. Referring to the Turkish political history (and to the massacres of Alevis which were committed in Sivas, Çorum and Maraş ), we also could expect that the Alevi voting in favour of the Nationalist Action Party (Milliyetçi Hareket Partisi, MHP) and the Felicity Party (Saadet Partisi, SP) is even lower than the support of the Justice and Development Party. The analyze of a possible correlation between this electoral phenomenon and this sociological factor will be determined thanks to the use of the software Stata. Nonetheless, I will use control variables (such as the level of education, the age and the socio-professional background) with the intention of escaping to a skewed view of this problematic. Moreover, I would like to use another software : R. Indeed, following to the 1971 coup d'Etat, the main radical left party (Türkiye İşçi Partisi, Turkey Workers' Party) was forbidden by the military power. Now, Alevi were deemed to be strong supporters of this Marxist organization. After the come back to democracy, the Republican People's Party seems to have benefitted from the votes of ex-TİP electors : thus, thanks to the software R, I wish to see if there were (in mainly Alevi-populated provinces and districts) transfer of votes from the radical left to the moderate left. Moreover, I would like to verify if (since the rural exodus of the 1960s and 1970s) the left-wing organizations electorally benefitted from the arriving of Alevi electors in the metropolitan cities of Istanbul and Ankara. Lastly, I want to study the evolution of the perception of Alevis in the Sunni community, thanks to opinion polls and surveys (like Eurobarometers or World Values Surveys), and vice versa. Reflections : All along the writing of this thesis, maybe I will have empirical and practical obstacles for performing my research. Indeed, even if I have already contacts inside the CHP's sections of Zeytinburnu, Beyoğlu and Şişli, I do not have contacts now in the HDP's branch of Tunceli. Now, if I cannot obtain interviews with pro-Kudish activists of this city, maybe my analyze will be biaised : it could imply a mono-geographical (Istanbul) and a mono-party (CHP) view about Alevi activism. Likewise, by interviewing some people that I already know, maybe it could be dangerous for my sample representativeness and for (more generally) the deontological framework of my research. Furthermore, as a French student, I could have a distorted perception of the reality : maybe I could project my own prejudices, my own pre-conceived opinions on the Turkish political and social situations. During my work, I will have to beware of analyzing the Alevi voting without pre-established ideas : my research will focus on empirical proofs and scientific data. I will base my reasoning of these elements for building my thesis, and I will be careful of not trying to distort the reality for confirming my prejudices. It will be necessary to not impose my Western perception on my topic, which will be the best way to express a detachment about the possible results of my research. Besides, as French-speaking native, I think that improving my knowledge of the Turkish language will be indispensable before to begin the writing of my thesis. In addition, I fear that my project of thesis has a disproportionate aspect : I think that I suggest too much qualitative methods and not enough qualitative data. Moreover, I have the feeling that my theoretical references are ethno-centered : beyond consulting the existing French-speaking and English-speaking literatures about the Alevi issue, the political sociology and the voting behaviour, I have to know more Turkish-speaking references. Conclusions : In a nutshell, I think that developing a political analyze of the Alevi issue in Turkey is a necessity, because there is currently a lack of electoral studies about Middle Eastern countries : indeed, this part of the world often is decrypted through academics who are specialized in international relations, but not from a point of view which would more focus on the electoral behaviour. Furthermore, I would like to pursue the work which was initiated by Elise Massicard, but from a more focused attention to the voting question and to the partisan activism. More generally, it is necessary to develop an analysis of Turkish Alevism's political aspects, because this issue is progressively coming back in this country (and shows the return of the implicit antagonism between Shia and Sunni Islam in the Middle East).