Ethnicity and Elections in Eastern Turkey: What do the Kurds Want?

Abstract

Relying on a survey conducted in 2014 in 12 provinces in eastern Turkey, this article analyses the extent to which (a) ethnicity and religiosity shape Kurds’ electoral choices and (b) the Kurds in the East support the demand for education in Kurdish and the demand for self-rule for Kurds. The findings demonstrate that ethnicity is an important factor in shaping Kurds’ electoral choices and that the demands for education in Kurdish language and self-rule are backed not only by the vast majority of those Kurds who support the pro-Kurdish party but also by a significant number of those Kurds who support the AK Party.

Keywords

Elections ethnicity Kurds Kurdish question Turkish politics

Introduction

The literature on the Kurdish question in Turkey has expanded and diversified remarkably over the last two decades. There is now an extensive literature on various aspects of the Kurdish question in Turkey, stretching from the significance of traditional social structures (Bruinessen, 1992; Klein, 2011) to the displacement of the Kurds (Ayata & Yükseker, 2005; Bozkurt, 2014; Çelik, 2005; Jongerden, 2007), and the role played by women in the Kurdish opposition of the last few decades (Çağlayan, 2012). The Turkish state’s policies concerning Kurds and Kurds’ political and military opposition to the Turkish state, for instance, are well documented (Beşikçi, 1978; Beşikçi, 1990; Güneş, 2013; McDowall, 2004).

Similarly, Kurds’ status in the Turkish national community and how the Turkish state and society have perceived the Kurdish question have been well documented (Saraçoğlu, 2010; Yeğen, 1999, 2009; Zeydanlıoğlu, 2008). Likewise, there is now a growing literature on the peace process which collapsed in 2015, the Ak Party’s engagement with the Kurdish question, and the mutual interactions between the Kurdish movements in Turkey and Syria (Ercan, 2019; Jongerden, 2019; Martin, 2019; Özpek, 2019; Rumelili & Çelik, 2017). However, the literature on the Kurds’ electoral choices and political propensities, as well as the ‘sociological determinants’ of these choices and propensities, is still slim. For instance, how many of the Kurds are supporting the pro-Kurdish parties¹ and how many of them are supporting the other parties are uncertain. So is the extent to which the Kurds endorse the cultural and political rights demanded by the pro-Kurdish parties. What is known about the sociological determinants of the Kurds’ party choices and political inclinations is even poorer. The extent to which such central sociological factors as ethnicity, religiosity, gender, education and income, and so on influence Kurds’ party choices and political tendencies have largely remained untouched.²

Relying on field research conducted between August 2014 and October 2014 in 12 provinces in eastern Turkey, this article provides a sociological analysis of Kurds’ party choices and political propensities.³ In particular, we aim to show (a) how many Kurds in Eastern Turkey, home for the majority of Kurds in Turkey, support the pro-Kurdish parties and how many of them support other parties, (b) the extent to which sociological factors such as ethnicity and religiosity shape Kurds’ electoral choices, and (c) how many of the Kurds in the East support the demand for education in Kurdish language and the demand for self-rule for Kurds. The findings demonstrate that majority of Kurds support the pro-Kurdish parties, that ethnicity is an important factor in shaping Kurds’ electoral choices and that the demands for education in Kurdish and self-rule are backed not only by the vast majority of those Kurds who support the pro-Kurdish party but also by a significant number of those Kurds who support the AK Party.

Research

The research was conducted in three stages. First, we analyzed the results of Turkey’s elections since 1950 to see if there is any significant pattern in the electoral choices in eastern Turkey. Second, during Spring 2014, we conducted 137 in-depth interviews with politicians, academics, NGO workers, opinion-builders, and adult citizens to obtain an idea about those sociological factors that shape Kurds’ party choices and political propensities. Lastly, between August 2014 and October 2014, we conducted a survey with 1,918 individuals in 12 provinces in eastern Turkey to ascertain whom the Kurds vote for and what they demand and to understand if sociological factors such as ethnicity, religiosity, gender, age, income, and education shape Kurds’ electoral choices and political propensities.⁴

Although the Kurds are scattered around Turkey, we conducted our survey in eastern Turkey, home for the majority of Kurds in the country. While official figures are outdated, recent research conducted by qualified scholars and institutions have provided reliable data regarding the current size of the Kurdish population and its spatial distribution in Turkey. The 2008 Turkey Demographic and Health Survey, for instance, shows that the Kurdish population in Turkey is around 17.2 percent of the total and that 60 percent of the Kurds live in the east of the country (Eryurt & Koç, 2015, p. 488). Similarly, research conducted by KONDA in 2010 indicates that the Kurds constitute 18.3 percent of the country’s total population and that 66 percent of the Kurds reside in the east of Turkey (KONDA, 2011).⁵

In other words, while one-third of the Kurds are living outside eastern Turkey, we decided to conduct this research only in that region, home for the majority of Kurds in Turkey, to focus on the electoral choices and the political demands of those Kurds who are residents or who have settled in the area controversially known as ‘Kurdistan’.

The term ‘eastern Turkey’ is traditionally used to refer to two geographical regions—Eastern Anatolia and South-Eastern Anatolia—officially containing 23 provinces,⁶ and the survey was conducted in 12 of them. As the main objective was to understand Kurds’ electoral choices and political propensities, we chose the following 12 provinces where Kurds are known to be a majority or of considerable size: Adıyaman, Ağrı, Bingöl, Batman, Diyarbakır, Hakkari, Kars, Mardin, Şanlıurfa, Şırnak, Tunceli, and Van.⁷ Face-to-face interviews were conducted with 1,918 respondents aged 18 and above in these 12 provinces between August and October 2014; of these, 1,410 (73.5%) were Kurdish, 247 (12.9%) were Zaza,⁸ 172 (9%) were Turkish and 88 (4.6%) were Arab. Respondents were selected through a stratified cluster sampling, and age and gender quotas were applied.

Ethnicity and Electoral Choices in Eastern Turkey: Who do the Kurds Vote For?

Having gathered data on both voters’ ethnic identity⁹ and their party choices, the survey we conducted gives an idea as to whether ethnicity plays a role in shaping voters’ party choices in the towns where Kurds are in majority or of considerable size. In the survey conducted, we collected data regarding the respondents’ ethnic identity, their mother tongue, the political party they voted for in the 2007 and 2011 elections, and the political party that they would vote for if an election were to be held in 2014 when the survey was conducted. The survey results show that ethnic identity plays a significant role in party choices in those 12 towns.

Crossing data on respondents’ ethnic identity with their party choices shows that the pro-Kurdish party has been the leading party of the Kurdish electorate(s) since at least 2007 and that there is a high degree of correlation between ethnic identity and electoral choices in these 12 towns. As seen in Table 1, while the AK Party’s share of Kurdish votes has remained less than 40 percent since 2007, more than half of the Kurdish voters seem to have been supporting the pro-Kurdish party during the same time span.¹⁰ Equally important is that the ethnic groups in Kurdish towns have some clear-cut party choices, indicating that ethnic identity is an important factor shaping voters’ party choices in these towns. The survey results show that while the majority of the Kurdish and Zaza voters prefer not to vote for the parties other than the pro-Kurdish party and the AK Party—in other words, they prefer not to vote for ‘the secular nationalist’ Cumhuriyet Halk Partisi (CHP) and ‘the conservative nationalist’ Milliyetçi Hareket Partisi (MHP)—Turkish and Arab voters prefer not to vote for the pro-Kurdish party.

On the other hand, the fact that a considerable portion of Kurds and a majority of Zazaz vote for the AK Party and not for the pro-Kurdish party suggests that ethnic identity is not the sole factor that shapes party choices in Kurdish towns. This brings us to the question of the impact of factors other than ethnicity on electoral choices in Kurdistan.

However, before analyzing the impact of factors other than ethnicity on party choices, let us first note that the apparent impact that ethnic identity has on electoral choices in these 12 towns had already been hinted at in previous elections. The past elections in these Kurdish majority towns indicated that some of the parties in Turkish politics had received much larger electoral support than they had in the rest of Turkey since the 1960s. For instance, support for the Yeni Türkiye Partisi (YTP; New Turkey Party) was around 25–54 percent in these 12 towns in the 1961 elections, whereas its share in the election nationally was 13.7 percent. Chaired by Yusuf Azizoğlu, a former minister and member of an influential Kurdish family in Diyarbakır, the YTP received between 9.1 percent and 55.1 percent of votes in the same 12 towns in the 1965 elections, while its share of the total votes was a paltry 3.7 percent.

A similar picture appeared in the late 1970s. This time the independent candidates Mehdi Zana, Orhan Alpaslan, and Edip Solmaz, all of whom were supported by the Kurdish socialist movements of the time, won the local elections in Diyarbakır in 1978, and Ağrı and Batman in 1979 respectively, while no socialist candidate managed to win any local elections in Turkish cities during this period. Third, established in 1990, the pro-Kurdish party received between 25 percent and 80 percent of the votes in the local and general elections between 1995 and 2014 in these 12 towns, while its share in national votes hovered between 4 percent and 6 percent.¹¹

Table 1.

Ethnic Groups and Electoral Choices (%)

	2007 Elections			2011 Elections			2014
	DTP	AKP	Others	BDP	AKP	Others	HDP	AKP	Others
Kurds	50.6	37.0	12.4	52.1	38.4	9.5	55.8	38	6.2
Zazas	21.7	64.1	14.2	26.7	66.8	6.5	26.3	60.8	12.9
Turks	0.7	67.6	31.7	1.9	66.3	31.2	2.5	61.9	35.6
Arabs	1.3	83.3	15.4	2.4	87.1	10.5	3.5	85.9	10.6

Source: The authors.

Rendering the results of the elections in the Kurdish majority towns considerably different from the rest of the country, these figures may imply that ethnic identity is an important factor in shaping voters’ party choices in these towns because the basic difference between these 12 towns and the rest of Turkey is that Kurds constitute either a majority or a substantial proportion of the former.

Religiosity, Sectarian Identity and Electoral Choices

While the impact of religiosity on the electoral choices of Kurds is as understudied as the impact of ethnic identity on Kurds’ party preferences, there are several works investigating the impact of religiosity on the political identity of Kurds in the broad sense of the term. The role and place of Islam in general, and the Naqshibendi tariqats in particular, in the formation of Kurdish identity and the Kurdish national movement has been acknowledged in some major works on the Kurdish question in Turkey (Bruinessen, 1992, 2000; Olson, 1989; Yüksel, 1993). Both the AK Party and its predecessors received considerable electoral support in the Kurdish majority towns. The AK Party signaled that it considers the Kurdish question in Turkey within the framework of Islamic brotherhood and it motivated many social scientists in Turkey to pay particular attention to the role of religiosity in the formation of the political identity of the Kurds and the close affiliations between the Kurds and the AK Party and its predecessors (Cizre-Sakallıoğlu, 1998; Duran, 1998; Sarıgil, 2018; Sarıgil & Fazlıoğlu, 2013; Yavuz, 1999; Yavuz & Özcan, 2006).

However, the impact of religiosity on the electoral choices of Kurds, in particular, remains somewhat understudied. One exception in this regard is Zeki Sarıgil’s (2010) work on Kurdish ethno-nationalism, which examines the impact of religiosity (and socioeconomic factors) on Kurdish ethno-nationalism, which is measured by the support for HADEP, the pro-Kurdish party of the time, based on the World Values Survey data of 2009–2010. Sarıgil (2010, p. 542) notes that ‘increases in the level of religiosity do not really reduce the likelihood of support for ethno-political formations’ (i.e., the pro-Kurdish party), meaning religiosity does not have a significant impact on the political identity or the electoral choices of Kurds.¹²

Arguing that ‘Sarigil’s research design lacks internal validity because it studies Kurdish nationalism within a sample that predominantly consists of Turks’, Faruk Ekmekçi (2011b, p. 1609) alleges that Sarıgil’s argument is mistaken. ‘By replicating and expanding Sarigil’s models within a sample of Kurdish-speaking people in Turkey, which was created from the data in … the World Values Survey (2009)’, Ekmekçi (2011b, pp. 1614–1615) argues that ‘the religiousness of a Kurdish person does influence his or her inclination towards Kurdish ethno-nationalism, especially through shaping political ideology’. According to Ekmekçi’s (p. 1615) findings, ‘[T]he more religious people among Kurds in Turkey were found to have a lower likelihood of both adopting a leftist ideology and supporting Kurdish ethnonationalism’.

The findings of our research also disprove Sarıgil’s conclusion but indicate that religiosity and sectarian identity are, alongside ethnicity, the two other factors that shape party choices in Kurdish towns. In our survey, we gathered data concerning voters’ religious practices as well as their sectarian identity. To assess one’s religiosity, the respondents were asked questions concerning their own and family members’ religious practices. More particularly, they were asked if they would consider themselves as religious and, if so, in what frequency they performed salaat, and if any adult female member of their families wore a headscarf.

Out of the answers given by the respondents to these questions, four categories of religiosity were formed. Those who perform salaat five times a day regularly and whose adult female household members all wore headscarf were defined as ‘pious’; those that performed salaat frequently and who had adult female household members with and without headscarves were defined as ‘religious’; third, those who performed salaat infrequently and whose adult female household members did not wear the headscarf were designated as ‘symbolically religious’; and lastly, those that considered themselves as non-religious and who did not perform salaat and whose adult female household members did not wear the headscarf were categorized as ‘secular’.

Once delineated in this way, it became clear that half of the voters (49.8%) in these 12 towns were pious, 39.7 percent were religious, and the remaining 10.4 percent were symbolically religious or secular (Table 2). Comparing Kurds, Zazas, Turks, and Arabs in terms of their religiosity, on the other hand, shows that there is no remarkable difference between these three groups in terms of the distribution of the four categories of religiosity. Only among Turks the pious voters are less than 40 percent (25.1%), and those that are ‘symbolically religious’ or ‘secular’ are more than 10 percent (18.7%).

While there is no remarkable difference between Kurds, Zazas, and Arabs in terms of the distribution of four categories of religiosity, there is a significant difference among pious, religious and symbolically religious, and secular voters in terms of their party choices. A macro outlook indicates that the more religious the voters are, the more they tend to vote for the AK Party, while conversely, the less religious voters are, the more they tend to vote for the pro-Kurdish party. The respondents’ answers to the question ‘Which party would you vote for if there were elections this Sunday?’ indicate that while more than half (56.6%) of the pious voters prefer to vote for AK Party, almost half (51.6%) of the religious voters support the pro-Kurdish HDP. However, it is also noticeable that the AK Party has considerable support among religious voters (37.8%), while the HDP has a similar base among pious voters (35.6%). This demonstrates that while the AK Party is the first among the devout voters and the HDP is the first among the religious voters, they both have significant support from the group in which the other party is predominant.

On the other hand, the picture is different in the case of the symbolically religious and the secular voters. While 54.6 percent of this group support the HDP, only 15.8 percent support the AK Party, making this party third in this group. That 17.3 percent of symbolically religious and secular voters support the CHP can be linked to the fact that a considerable portion of this third group are Zaza and Kurdish Alevis of Dersim (Tunceli), where the CHP has had strong support since the 1960s.

Table 2.

Religiosity in Ethnic Groups (%)

	Pious	Religious	Symbolically Religious + Seculars	Total
Kurds	40.9	51.6	7.4	100
Zazas	43.3	35.9	20.8	100
Arabs	43.2	50.0	6.8	100
Turks	25.1	56.1	18.7	100
Total	39.7	49.8	10.4	100

Source: The authors.

While confirming the claim that religiosity is an important factor that shapes party choices in Kurdish towns and that the more religious voters are, the more they prefer to vote for AK Party, these findings undermine the strong belief in the Turkish public that the religious Kurds do not support the pro-Kurdish party. Given that a vast majority of voters in Kurdish towns are religious Kurds (pious and/or religious) and that the pro-Kurdish party is ahead of the AK Party in terms of electoral support in these towns, it is evident that this view is illogical. The fact is that the pro-Kurdish HDP receives the support of half of the religious voters, but it is also supported by 35.6 percent of the pious voters.

Besides religiosity, sectarian identity also appears to be an important factor that shapes electoral choices in Kurdish towns. The Sunni-Alevi division is one of the main sectarian divisions in these 12 Kurdish majority towns. The size of the Alevi respondents in our sampling is remarkably small (91 out of 1,918), making it difficult to draw reliable conclusions about their electoral choices. However, it can still be understood that being an Alevi makes an important difference in one’s party choices in these 12 towns. The findings of our research indicate that only a minimal number of Alevi voters support the AK party and the majority of them opt either for the pro-Kurdish party or the CHP. The answers given to the question ‘Which party would you vote for if an election were held this Sunday?’ show that only 2.2 percent of Alevis support the AK Party. In contrast, 58.9 percent support the pro-Kurdish party, while 28.9 percent say they would vote for the CHP.

Table 3.

Religiosity and Party Choices (%) (If a national election were held this Sunday, which party would you vote for?)

	HDP	AKP	CHP	MHP	Others/Not Voting	Total
Pious	35.6	56.6	0.5	0.4	6.9	100
Religious	51.6	37.8	2.4	1.3	6.9	100
Symbolically religious and seculars	54.6	15.8	17.3	3.1	9.2	100

Source: The authors.

The second sectarian division in these towns, the Shafi-Hanefi division, is an intra-Sunni division and would not be expected to generate any political outcome because the division itself has not been informed by any political cleavage or dispute in the history of Islam and revolves in the main around certain ‘minor’ differences in religious practices. However, our research findings show that there are some significant differences in the electoral choices between Shafi and Hanafi voters. While 58 percent of Shafi voters support the pro-Kurdish party, 35.7 percent of them support the AK Party.

When it comes to Hanafis, the difference is even more striking. While 65.7 percent of Hanafis say they would support the AK Party, the support for the pro-Kurdish party among Hanafi voters is as low as 17.1 percent.¹³ Much of this difference has to do with the fact that while most Kurds and Zazas in these 12 towns are subscribers to the Shafi school, there are practically no Shafi Turks or Arabs in Turkey. In other words, a majority of Shafis vote for the pro-Kurdish party because the majority of Kurds and Zazas are Shafis. However, many Kurdish and Zaza followers of the Hanafi school and their electoral choices indicate that there are important differences between Shafi and Hanafi Kurds and Zazas in terms of their party choices. For instance, while 59 percent of Shafi Kurds and Zazas support the pro-Kurdish party, only 27.0 percent of the Hanafi Kurds and Zazas support the pro-Kurdish party. Conversely, while 63.1 percent of Hanefi Kurds and Zazas support the AK Party, the AK Party’s support among Shafi Kurds and Zazas is considerably less than this, at only 34.7 percent.

As the Shafi-Hanefi distinction is not informed by any political dispute or cleavage within Islam, the difference in the electoral choices and political identities of Shafi and Hanafi Kurds must be related to another factor that supports or overlaps with the Shafi-Hanefi distinction. In fact, a closer look at the geographical distribution of Shafi and Hanafi Kurds and Zazas reveals that while Hanefi Kurds and Zazas are settled in places such as Urfa and Adıyaman, where the Ottoman state managed to impose direct rule, Shafi Kurds and Zazas are mostly settled in those places where the (Hanafi) Ottomans only managed indirect rule. Hence, one may speculate that the Shafi-Hanefi distinction among Kurds and Zazas itself is an outcome of the differences in the history and the scope of integration of the places populated by Kurds and Zazas and that it may go some way to explaining current differences in their respective electoral choices.

What do the Kurds Want?

Our research demonstrates that the demands for education in mother tongue and self-rule are supported by a considerable portion of the Kurdish voters and that ethnic and political identity are two important factors that shape voters’ support in terms of these two demands. Below are the results related to the questions of ‘what the Kurds want?’ and ‘who wants what?’ in Kurdish towns.

Support for Education in Mother Tongue

The respondents were first asked to answer the extent to which they support the demand for having education in mother tongue in public schools by means of choosing one of the following three options: ‘support entirely’ (SE), ‘support partially’ (SP), and ‘do not support’ (NS). The answers indicate that the majority of the voters in the Kurdish towns support the demand for education in mother tongue. While 78.5 percent of voters entirely and 10.1 percent partially supported this demand, only 11.4 percent of all voters did not support the demand or (actively) opposed it (Table 4). However, the extent of support for the demand for education in one’s mother tongue differs considerably between different ethnic groups and among supporters of different political parties. While almost 95 percent of both Kurds and Zazas support or partially support the demand for education in their mother tongue, nearly half of Turks and one-third of Arabs oppose this idea.¹⁴ However, the difference between the HDP and AK Party supporters is not as dramatic as it is among different ethnic groups. Almost all HDP voters and 82.3 percent of AK Party voters entirely or partially support the demand at stake, indicating that majority of the Kurdish and Zaza voters supporting the AK Party also support the demand for education in the (local) mother tongue.

Support for Self-Rule

To understand the extent of support for self-rule demands, respondents were asked the extent to which they supported the maintenance of the current unitary state, strengthened local administrations, and democratic autonomy,¹⁵ federalism and an independent Kurdish state, by means of choosing one out of the following three options: ‘support entirely’, ‘support partially’, and ‘do not support’.

The findings indicate that a ‘strengthened local administration’ is the most supported and that ‘an independent Kurdish state’ is the least supported option. While 79.4 percent of voters entirely or partially support ‘having strengthened local governments’, 49.9 percent of them entirely or partially support the idea of ‘having an independent Kurdish state’. Supported or partially supported by 66.7 percent of voters, the ‘democratic autonomy’ of the HDP is the second-most supported option (Table 5). The second least supported option, on the other hand, is that of federalism; 49.9 percent of voters entirely or partially support the option of federalism. Lastly, 56 percent of voters support or partially support ‘maintaining the current unitary state’.

Table 4.

Support for Education in Mother Tongue

	Education in Mother Tongue in Public Schools
	Do Not Support	Support Partially	Support
Kurds	7.1	8.2	84.7
Zazas	4.1	13.3	82.6
Turks	44.0	16.3	39.8
Arabs	31.0	17.9	51.2
Total	11.4	10.1	78.5

Source: The authors.

These results point to a few important facts. First, the majority of the voters in Kurdish towns support the idea of self-rule, but in a manner that can be described as moderate. To put it another way, the majority of voters endorsed a moderate form of self-rule, as indicated by the fact that the option of a ‘strengthened local administration’ was the most supported option. The second important fact is that the extent of the support for even the most moderate self-rule is less than the extent of the support for education in the mother tongue. This proves that while a vast majority of the voters in the Kurdish towns endorse the demand for education in the mother tongue, they are not so comfortable with the demand for self-rule. The third important fact points to an entirely different picture. While the AK Party and the pro-Kurdish party are supported by almost 90 percent of all voters and while neither of the two supports the options of federalism or an independent Kurdish state, both of these options are endorsed by almost half of all voters, meaning the support for the options of federalism and independent state is substantially more than expected.¹⁶

The difference between the Kurds and the rest (of the population) and between the supporters of the HDP and the AK Party in terms of the extent of their support for these five options also points to some important facts. As expected, the majority of those who support the option of an independent Kurdistan are Kurdish voters: 61.6 percent of Kurdish voters entirely or partially support the option of independent Kurdistan. In contrast, almost all Arab (94.1%) and Turkish (95.6%) and a considerable number of Zaza voters (67.6%) oppose this demand. This proves that the support for independent Kurdistan’s option has a strong ethnic dimension in Kurdish towns. There is a similar picture in the case of support for federalism. Here again, while an almost overwhelming majority of Turkish (84.1%), Arab (88.0%) and Zaza (55.4%) voters oppose the idea of federalism, 62.1 percent of Kurds entirely or partially support it (Table 5).

The picture becomes slightly more complex when it comes to the option of democratic autonomy. As expected, while a considerable majority of Kurdish voters (76%) support or partially support this option, the majority of Turkish voters (81.2%) oppose it: 65.8 percent of Arab voters also oppose the idea of democratic autonomy. The fact that Arab voters’ opposition to this option is not as high as that of Turkish voters might be because the former is somehow more inclined to have more autonomy at the local administration level. On the other hand, the majority of Zaza voters oppose the idea of an independent Kurdistan or federalism and seem to be comfortable with the option of democratic autonomy: 65. 6 percent of Zaza voters support or partially support this option.

The parallels between Kurds and Zazas increases furthermore in the case of the option of maintaining the current unitary state. While there is a stark difference between Kurdish voters and the rest in terms of their support for an independent state and federalism, Kurds and Zazas seem to be equally uncomfortable with maintaining a unitary state. While the majority of Turkish and Arab voters support the continuation of the current unitary state, almost half of all Kurds (47.1%) and 41.7 percent of Zaza voters oppose it.

Table 5.

Support for Self-Rule

	StrengthenedLocal Governments			Democratic Autonomy			Unitary State			Federalism			Independent Kurdistan
	NS	PS	S	NS	PS	S	NS	PS	S	NS	PS	S	NS	PS	S
Kurds	15.3	11.0	73.7	23.9	13.6	62.4	47.1	15.4	37.5	37.9	13.7	48.4	38.4	15.2	46.4
Zazas	9.8	12.3	77.9	34.3	12.6	53.0	41.7	14.7	43.6	55.4	10.7	33.9	67.6	7.6	24.8
Turks	65.6	15.2	19.2	81.2	10.7	8.1	20.3	20.3	59.4	84.1	8.6	7.3	95.6	3.2	1.3
Arabs	38.7	18.7	42.7	65.8	7.9	26.3	19.7	9.2	71.1	88.0	2.7	9.3	94.1	4.7	1.2
Total	20.6	11.9	67.5	33.2	12.9	53.8	44.0	15.2	40.8	47.4	12.3	40.3	50.1	12.7	37.2

Source: The authors.

What approximates voters from different ethnic groups and parties the most is the option of a strengthened local administration. The only ethnic group of whom more than half oppose this idea are Turkish voters, 65.6 percent of which are not comfortable with the notion of granting more authority to local governments. However, 84.7 percent of Kurdish, 90.2 percent of Zaza, and 61.4 percent of Arab voters support or partially support the option of a strengthened local administration.

The picture is less complex when support for these five options is cross-referenced with support for different political parties. Findings point to a notable distance between the supporters of the HDP and the AK Party in terms of the extent to which they support these five options. As the figures in Table 5 show, the majority of those who do not support an independent Kurdistan, federalism or democratic autonomy are AK Party voters, while the majority of those who support these three options are HDP voters. While 78.3 percent of AK Party voters oppose the option of an independent state and 74.5 percent and 57 percent of them oppose the options of federalism and democratic autonomy respectively, 81.2 percent of HDP voters support an independent Kurdistan, 81.7 percent support federalism, and 92.4 percent voice their support for democratic autonomy.

The reverse is the case when it comes to support for the unitary state. While three-quarters of AK Party voters support the preservation of a unitary state, two-thirds of HDP voters oppose it. The voters of two parties seem to agree only in their mutual support for strengthened local governments. While there is still a distance between the voters of two parties, it is important that almost two-thirds of AK Party voters support or partially support the option of strengthened local governments, supported by 94.5 percent of HDP voters.

Conclusion

The findings of our research suggest that ethnic identity is a significant factor in shaping Kurds’ electoral choices and that majority of Kurds in Kurdish towns support the pro-Kurdish parties. The fact that the Kurds’ support for the two nationalist parties, CHP and MHP, remains at a minimum level and that more than half the Kurdish electorate constantly supports the pro-Kurdish parties suggests that ethnicity is an important factor in shaping Kurds’ electoral choices. This is also confirmed by the fact that the pro-Kurdish parties are supported not only by the Kurds with low socioeconomic status but also by the majority of Kurds with high or higher levels of income and education.

Our findings indicate that the pro-Kurdish party surpasses the AK Party in popularity and appeals not only in the case of Kurds with low levels of income and education but also with Kurds with higher levels of income and education. Indeed, the difference between support for the pro-Kurdish party and the AK Party among Kurdish voters with high levels of income and education is even larger than in those with low levels of the same. While 56.3 percent of Kurds with high levels of income and education (Kurds with a university degree and with a monthly income of more than 2,000 Turkish lira or approximately $920 in 2014) support the pro-Kurdish party; only 31.7 percent say they would vote for the AK Party. It can be seen that the AK Party supplants the pro-Kurdish party only in the case of Kurdish electorates with low levels of education but high levels of income. In the case of this particular Kurds group, support for the AK Party and the pro-Kurdish party is 62.9 percent and 35.1 percent, respectively. While these figures testify to a significant correlation between ethnic identity and electoral choices, the fact that a considerable section of the Kurdish electorate continues to support the AK Party indicates that factors other than ethnicity also play some role in shaping the Kurds’ electoral choices.

Our findings also show that the majority of Kurdish electorates support the demand for education in Kurdish and self-rule. The findings attest to the fact that these two demands are backed not only by the vast majority of those Kurds who support the pro-Kurdish party but also by a significant number of those Kurds that support the AK Party. While almost 80 percent of AK Party supporters approve of the demand for education in Kurdish, a little less than 70 percent support the demand for strengthened local governments, that is, a moderate form of self-rule.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors would like to thank TUBİTAK (The Scientific and Technological Research Council of Turkey) for financing the research project out of which this article is produced.

Notes

Appendices

Appendix 2. Most-Spoken Language in Family<break/>Q33. Which is the most spoken language in your family?<break/>(1) Kurdish (2) Zazaki (3) Turkish (4) Arabic (5) Assyrian (6) Armenian (7) Other

Kurdish		Ethnic Origin				Total
Kurdish		Zaza	Turkish	Arab
Most-spoken language in family	Kurdish	75.3	1.9	2.5	1.1	54.2
	Zazaki	1.3	58.4	0.6	0	8.2
	Turkish	22.4	38.3	95.1	50.6	34.3
	Arabic	0	0	0.6	46.0	2.2
	Other	1.0	1.4	1.2	2.3	1.1
Total		100	100	100	100	100

References

Ayata

Yükseker

(2005). A belated awakening: National and international responses to the internal displacement of Kurds in Turkey. New Perspectives on Turkey , 32(1), 5–42.

Beşikçi

. (1978). Kürtlerin Mecburi İskanı [forced settlement of Kurds]. Komal Yayınevi.

Beşikçi

. (1990). Tunceli Kanunu (1935) ve Dersim Jenosidi [Tunceli law (1935) and Dersim genocide]. Belge Yayınları.

Bookchin

(2006). The ecology of freedom . AK Press.

Bozkurt

(2014). The Kurds and settlement policies from the late Ottoman Empire to early republican period: Continuities and discontinuities (1916–1934). Iranian Studies , 47(5), 82–837.

Bruinessen

M. V.

(1992). Agha, Shaikh and the state the social and political structures of Kurdistan . Zed Books.

Bruinessen

M. V.

(2000). Mullahs, sufis and heretics: The role of religion in Kurdish society. Collected Articles . Isis Press.

Çağlayan

(2012). From Kawa the blacksmith to Ishtra the goddess: Gender constructions in ideological-political discourses of the Kurdish movement in post-1980 Turkey. European Journal of Turkish Studies , 14.

Çelik

A. B.

(2005). ‘I miss my village!’: Forced Kurdish migrants in Istanbul and their representation in associations. New Perspectives on Turkey , 32,137–63.

10.

Çiçek

(2017). The Kurds of modern Turkey: National, religious and economic identities . I.B. Tauris.

11.

Cizre-Sakallioğlu. (1998). Kurdish nationalism from an Islamist perspective: The discourse of Turkish Islamist writers. Journal of Muslim Minority Affairs , 18(1), 73–89.

12.

Dündar

(1999). Türkiye Nüfus Sayımlarında Azınlıklar [minorities in population censuses in Turkey]. Doz Yayınları.

13.

Duran

(1998). Approaching the Kurdish question via ‘Adil Düzen’: An Islamist formula of the Welfare Party for Ethnic Coexistence. Journal of Muslim Minority Affairs , 18(1), 111–129.

14.

Ekmekçi

(2011a). Seçimler, Sonuçlar ve Yanıltıcı Yorumlar II [elections, results and misleading comments]. https://fekmekci.files.wordpress.com/2015/09/kc3bcrt-sorunu-ve-baric59f-sc3bcreci-kitap.pdf

15.

Ekmekçi

(2011b). Understanding Kurdish ethno-nationalism in Turkey: Socio-economy, religion, and politics. Ethnic and Racial Studies , 34(9), 1608–1617.

16.

Ercan

(2019). ‘Is hope more precious than victory? The failed peace process and urban warfare in the Kurdish region of Turkey. South Atlantic Quarterly , 118(1), 111–127.

17.

Eryurt

M. A.

Koç

İ.

(2015). Demography of ethnicity in Turkey. In Sáenz

Embrick

D. G.

Rodríguez

N. P.

(Eds.), The international handbook of the demography of race, and ethnicity, international handbooks of population (Vol. 4). Springer Science and Business Media Dordrecht.

18.

Gunes

(2013). The Kurdish national movement in Turkey: From protest to resistance . Routledge.

19.

Haig

Öpengin

(2014). ‘Introduction to special issue - Kurdish: A critical research overview. Kurdish Studies , 2(2), 99–122.

20.

Jongerden

(2007). The settlement issue in Turkey and the Kurds . Brill.

21.

Jongerden

(2019). Governing Kurdistan: Self-administration in the Kurdistan regional government in Iraq and the democratic federation of Northern Syria. Ethnopolitics , 18(1), 61–75.

22.

Klein

(2011). The margins of empire: Kurdish militias in the Ottoman tribal zone . Stanford University Press.

23.

KONDA. (2006). Biz Kimiz. Toplumsal Yapı Araştırması [Who are we? Social structure research]. http://www.konda.com.tr/tr/raporlar/2006_09_KONDA_Toplumsal_Yapi.pdf

24.

KONDA. (2011). Kürt Meselesi’nde Algı ve Beklentiler [Perceptions and expectations in Kurdish question]. İletişim.

25.

Martin

(2019). The A.K. Party and the Kurds since 2014: A discourse of terror. British Journal of Middle Eastern Studies , 45(4), 543–558.

26.

McDowall

(2004). A modern history of the Kurds . I.B. Tauris.

27.

Mutlu

(1996). Ethnic Kurds in Turkey: A demographic study. International Journal of Middle East Studies , 28, 517–541.

28.

Olson

(1989). The emergence of Kurdish nationalism and the Sheikh Said rebellion . University of Texas Press.

29.

Özpek

B. B.

(2019). The state’s changing role regarding the Kurdish question of Turkey: From consistent tutelage to volatile securitization. Alternatives , 44(1), 35–49.

30.

Rumelili

Çelik

A. B.

(2017). Ontological insecurity in asymmetric conflicts: Reflections on agonistic peace in Turkey’s Kurdish issue. Security Dialogue , 48(4), 279–296.

31.

Saraçoğlu

(2010). The changing image of the Kurds in Turkish cities: Middle-class perceptions of Kurdish migrants in İzmir. Patterns of Prejudice , 44(3), 239–260.

32.

Sarigil

(2010). Curbing ethno-nationalism in Turkey: An empirical assessment of pro-Islamic and socio-economic approaches. Ethnic and Racial Studies , 33(3), 533–553.

33.

Sarigil

(2018). Ethnic Boundaries in Turkish Politics: The Secular Kurdish Movement and Islam , New York: NYU Press.

34.

Sarıgil

Fazlıoğlu

(2013). Religion and ethno-nationalism: Turkey’s Kurdish issue. Nations and Nationalism , 19(3), 551–571.

35.

Sarigil

Karakoç

(2016). Who supports secession? The determinants of secessionist attitudes among Turkey’s Kurds. Nations and Nationalism , 22(2), 325–346.

36.

Yavuz

M. H.

(1999). Search for a new social contract in Turkey: Fethullah Gulen, the Virtue Party and the Kurds. SAIS Review , 19(1), 114–143.

37.

Yavuz

M. H.

Ozcan

N. A.

(2006). The Kurdish question and Turkey’s justice and development party. Middle East Policy , 1313(1), 102–119.

38.

Yeğen

(1999). The Kurdish question in Turkish state discourse. Journal of Contemporary History , 34(4, 555–568.

39.

Yeğen

(2009). Prospective-Turks or pseudo-citizens: Kurds in Turkey. The Middle East Journal , 63(4), 597–616.

40.

Yüksel

(1993). Kürdistan’da Değişim Süreci [The process of change in Kurdistan]. Sor Yayıncılık.

41.

Zeydanlioğlu

(2008). The white Turkish man’s burden: Orientalism, Kemalism and the Kurds in Turkey. In Rings

Ife

(Eds.) Neo-colonial mentalities in contemporary Europe? Language and discourse in the construction of identities (pp. 155–174). Cambridge Scholars Publishing.