The AKP’s clientelist–machine politics and the role of Kurdish brokers: the case of Bağcılar

Abstract

Although there are numerous studies showing the significance of clientelism in the electoral mobilization of the poor in Turkey, scant attention has been paid on the specific clientelistic strategies employed by the Justice and Development Party (Adalet ve Kalkınma Partisi; AKP) among the Kurdish population in İstanbul. Addressing this gap, this study focuses on the AKP’s clientelist politics among Kurdish voters in Bağcılar, a lower-income district on the European side of İstanbul. Drawing on fieldwork, this study argues that the AKP successfully mobilizes Kurdish voters through machine politics and relational clientelism, which are facilitated by co-ethnic brokerage and intermediary linkages. By analyzing different modalities of Kurdish brokers – Kurdish (male) entrepreneurs, Kurdish housewives and Kurdish associations – the study discloses the intricate dynamics of the AKP’s clientelist–machine politics at the local level. It reveals that the AKP’s co-ethnic brokers not only distribute material benefits but also cultivate enduring personal relationships with Kurdish voters, providing problem-solving networks for the Kurdish poor in everyday life. The AKP’s co-ethnic brokerage relations foster trust, care, solidarity, and affectivity among Kurds. The study concludes that the AKP employs both material and symbolic dimensions of clientelism to mobilize the Kurdish electorate in İstanbul.

Introduction

In the last two decades, there has been a competition between the Justice and Development Party (Adalet ve Kalkınma Partisi; AKP) and pro-Kurdish parties (Barış ve Demokrasi Partisi [BDP] until the mid-2010s, and then Halkların Demokratik Partisi [HDP]) for the Kurdish vote (Grigoriadis and Dilek 2018). While the HDP achieved remarkable electoral success in both Kurdish-majority provinces and western metropolises and became the third largest party in the parliament in the June 2015 general elections, the AKP maintained its electoral success through repression and securitization since then (Jongerden 2019; Şahin 2021). ¹ Although many studies analyzed the reasons behind Kurdish voting behavior at the country level (e.g. Sarigil and Fazlioglu 2013; Bilecen 2015; Çiçek 2016), there are few studies addressing how the AKP garners support from Kurdish voters in İstanbul (Yörük 2014; Arıkan Akdağ 2015, 2016; Günay and Yörük 2019). In addition, there is insufficient attention paid to the AKP’s mobilization of Kurdish voters in the peripheral districts. To fill these gaps in the literature, this study focuses on the AKP’s clientelistic strategies for mobilizing the Kurdish electorate in Bağcılar, a low-income peripheral district of the city. Based on fieldwork in Bağcılar between 2014 and 2018, I argue that the AKP became successful in appealing to Kurdish voters through its clientelist–machine politics. In addition, I show that the AKP’s clientelist mobilization of the Kurdish electorate depends on the performance of co-ethnic brokers and intermediaries in the district. This study draws attention to three different types of brokers that carry out the AKP’s clientelist–machine politics among Kurdish voters: Kurdish (male) entrepreneurs; Kurdish housewives; and Kurdish associations.

This study suggests integrating the perspectives of “relational clientelism” (Nichter 2018) and “machine politics” (Scott 1969; Levitsky 2003) to understand the AKP’s clientelist mobilization among Kurdish voters. In accordance with recent literature, I prefer to define clientelism as brokerage relationships between political patrons, brokers, and clients that lie beyond electoral times (Hagene 2015; Holland and Palmer-Rubin 2015). Stokes et al. (2013, 100) note that “only locally embedded agents of the machine command the knowledge of voter preferences and partisan inclinations” on the ground. Likewise, the AKP manages its clientelist–machine politics by means of locally embedded Kurdish brokers and intermediaries in Bağcılar.

This study shows that material distribution is a required, but not a sufficient, condition of clientelism. My ethnographic findings highlight the significance of the symbolic aspect of clientelism. Non-material relations (e.g. community-making practices) are as influential as material relations (e.g. social assistance) in the AKP’s mobilization of the Kurdish electorate in Bağcılar. The AKP’s Kurdish brokers and intermediaries, who are well-positioned and well-connected within the local communal networks, provide practical problem-solving for and create dyadic and durable personal relationships with the Kurdish poor in everyday life. These relationships involve care, trust, reciprocity, and affectivity. The party’s ethnic brokers and intermediaries aim to incorporate the Kurdish poor, who feel that they were discriminated, excluded, and marginalized in the past, with communitarian discourses revolving around the Islamic solidarity principle. These brokers and intermediaries strive to produce “the morality of communitarian life-world” for the Kurdish poor in everyday life by trespassing the rational exchanges of the electoral market (Hagene 2015, 148). However, it is also necessary to note that the symbolic dimension of clientelism is, to some extent, shaped by the brokers’ neoliberal ethos and, to a large extent, constrained by structural economic factors.

In the following sections, after discussing the literature, methodology, and the field, I will outline the AKP’s clientelist–machine politics in Bağcılar and specifically examine the clientelist relations and practices carried out by the AKP’s Kurdish businessmen (male brokers), Kurdish housewives (female brokers), and Kurdish hometown (“hemşehri”) associations.

Literature review

The scarce empirical literature on the subject indicates that religiosity is the most salient cleavage of Kurdish voting behavior (Sarigil 2010; Sarigil and Fazlioglu 2013; Bilecen 2015). These studies argue that while religious–conservative Kurdish voters support Islamist parties, (Welfare Party [Refah Partisi; RP] and the AKP), the secular Kurdish electorate supports pro-Kurdish parties. In a recent study, Günay and Yörük (2019) show that the AKP’s Islamic articulations and the ummah (i.e. Islamic universalism) ideology are appealing to Kurdish voters in the poor neighborhoods of İstanbul. While acknowledging the significance of religiosity, this study argues that focusing exclusively on this factor prevents us from understanding the AKP’s actual mobilization strategies in everyday life. Therefore, I suggest shifting attention towards the AKP’s clientelist linkages that play upon the interactions between Kurdish ethnicity, poverty, and religiosity.

The AKP’s clientelism has dimensions resembling both “machine politics” (Scott 1969; Levitsky 2003; Stokes et al. 2013) and “relational clientelism” (Auyero 2012; Hagene 2015; Nichter 2018). In basic terms, machine politics refers to a political system in which a political party (i.e. patron) manages patronage resources and exchanges over a specific territory in a highly organized fashion, whereas relational clientelism refers to political relations in which brokers and intermediaries provide material and non-material benefits to individuals or groups in exchange for their loyalty, support, or votes. Unlike traditional clientelism, which focuses on short-term material interests, relational clientelism focuses on long-term and personal ties between patrons, brokers, and voters.

Machine politics have become meaningful considering the rapid urbanization vis-à-vis the massive migration experience of the Kurdish population in the last three decades in Turkey. Forced internal displacement and the collapse of traditional subsistence agriculture displaced millions of Kurdish citizens in the eastern and southeastern provinces in the 1990s. With the subsequent Kurdish migration waves, İstanbul has become the “largest Kurdish city” in the world (Keyder 2004; Kalaycı 2023). Yükseker (2009) points out the overlapping processes of the neoliberal restructuring of the Turkish economy and the forced displacement of Kurds. While a small fraction of the Kurdish migrant population has been able to climb the social ladder by exploiting the emerging opportunities of the informal economy and rentier capitalism in urban areas (Yılmaz 2014; Küçük 2019), most Kurdish migrants have turned into a source of cheap labor force and formed an informal proletariat in the western metropolises of Turkey (Yörük 2014).

Scott (1969) argues that immigration and urbanization processes constituted the basis of machine politics by transforming the loyalty patterns of urban migrants from traditional patronage to particularistic inducement in the United States. Likewise, starting from the 1950s, political parties provided patronage to win the votes of both peasantry and new urban migrants in Turkey (Sunar 1990; Güneş-Ayata 1994; Sayarı 2011). In particular, conservative right-wing parties, such as the Democratic Party (Demokrat Parti; DP) in the 1950s, the Justice Party (Adalet Partisi; AP) in the 1970s, the Motherland Party (Anavatan Partisi; ANAP) in the 1980s and the True Path Party (Doğru Yol Partisi; DYP) in the 1990s, as well as leftist parties, such as the Social Democratic People’s Party (Sosyaldemokrat Halkçı Parti; SHP) in the 1990s, resorted to patronage politics (through urban machineries and pork-barrel services). In the 1990s and the 2000s, the political machines of the Islamist RP and AKP provided particularistic benefits to poor migrants in urban peripheries (Ayata 1996, 54; Arslantaş and Arslantaş 2022). While establishing patron–client relations, the Islamists have been able to interact with the poor. As a matter of fact, the AKP’s brokerage networks succeeded the RP’s “vernacular politics” that denotes face-to-face interactions and door-to-door activism performed by the RP brokers in poor neighborhoods during the 1990s (White 2002). In this process, Kurdish migrants became the main target of political machines of the RP and the AKP, respectively, in urban peripheries because of their poverty and material insecurity. Positioning themselves against the secular Turkish elites, Islamists have easily interacted with the traditional–conservative habitus and local idioms of poor Kurdish migrants and achieved notable success in Kurdish-populated neighborhoods in İstanbul (Kuran 2020, 266–271). ²

Turkey and Latin American countries, such as Argentina, exhibit similarities in their clientelistic politics (Sayarı 2014, 662). Both the AKP in Turkey and the Peronist Party (PJ) in Argentina share historical and organizational characteristics. Firstly, both countries underwent neoliberal economic transformations in the 1980s and 1990s. Secondly, the role of brokers was crucial in mobilizing urban poor through clientelism. Levitsky (2003) notes the rise of machine politics in Latin America during the neoliberal era, coinciding with increasing poverty and the expansion of the informal economy in urban areas in the 1990s. During this period, the PJ shifted towards clientelist mobilization of the poor instead of traditional working-class politics. Similarly, the RP and the AKP established clientelist relationships with urban migrants during the neoliberal transition (Doğan 2017). The Peronist machine provides material benefits (such as employment, pensions, food, medicines, and social assistance), as well as effective problem-solving networks and long-term reciprocal relationships with the urban poor (Auyero 2001; Nichter 2018). Likewise, the AKP’s clientelistic machine politics involve “practical problem-solving” and “long-term personal reciprocity” for the poor (Ark-Yıldırım 2017, 482).

In analyzing Peronist clientelism in Argentina, Zarazaga (2014) emphasizes the pivotal role of multitasking brokers in impoverished Buenos Aires neighborhoods. These respected figures possess deep local knowledge, addressing the poor’s issues while offering tangible benefits like food, medicine, and infrastructure improvements. Simultaneously, they cultivate enduring personal connections with voters, engaging in grassroots clientelism (Zarazaga 2014, 37–39). Other ethnographic research in Argentina (Auyero 2001, 2012, 2014) aligns with Zarazaga’s (2014) findings, highlighting clientelism’s dual nature, encompassing both material and non-material aspects. Material distributions serve utilitarian purposes, while non-material elements, such as reciprocity, recognition, and community building, are essential for sustaining long-term clientelist politics. Offering material benefits alone falls short in perpetuating clientelism. Political parties must integrate both dimensions to maintain clientelist dynamics in everyday life. Likewise, Hagene (2015) emphasizes how party brokers establish enduring clientelist relationships with poor voters in poor Mexican neighborhoods. These brokers forge personal ties and local solidarities, fostering affective bonds among the poor. Hagene (2015) distinguishes between vote-buying and political clientelism, with the latter grounded in reciprocal obligations and deemed legitimate by the poor. The poor perceive clientelism as a form of “fictive kinship,” ³ viewing it as “help,” “social support,” or a means of “reaching the state” (Hagene 2015, 152–156).

Despite the significance of the non-material/symbolic dimension of clientelism, scholars have mostly focused on vote-buying in analyzing the AKP’s clientelism in Turkey (Sayarı 2014; Çarkoğlu and Aytaç 2015; Cinar 2016). Some highlight the effectiveness of the distribution of cash and benefits in-kind (e.g. food and coal) (Yıldırım 2017) and public employment (Kemahlıoğlu 2012), whereas others emphasize the instrumentalization of social policies and assistance (Aytaç 2014). In general, mainstream clientelism literature in Turkey overlooks the AKP’s actual brokerage strategies and networks at the grassroots. Nevertheless, in recent years, there has been another line of ethnographic scholarship that scrutinizes AKP clientelist mobilization in specific urban localities (Ark-Yıldırım 2017; Yıldırım 2020; Arslantaş and Arslantaş 2022). These latter studies show that AKP brokers have succeeded in establishing personal relationships with voters at the neighborhood level. The AKP’s electoral success is attributed to its dense organization and strong grassroots presence, as well as its problem-solving networks and long-term relations of reciprocation provided by the party brokers (Ark-Yıldırım 2017).

By following the footsteps of Latin American scholarship and recent ethnographic studies in Turkey, I argue that the AKP has been able to mobilize the Kurdish electorate through the performance of locally embedded co-ethnic brokers and intermediaries in Bağcılar. In addition to party brokers, Kurdish hometown associations that rely on co-locality and kinship play a critical role in the clientelist mobilization of Kurdish voters. Therefore, this study not only tackles party brokers but also addresses “organizational brokers” in the form of local associations (Holland and Palmer-Rubin 2015). Beyond material exchanges, the AKP’s co-ethnic brokers establish personal relationships with Kurdish voters and provide practical problem-solving for the Kurdish poor in everyday life. These co-ethnic brokers and intermediaries construct the “symbolic and affective identification” with Kurdish voters in everyday life (Küçük and Türkmen 2020, 251).

A brief note on methodology

This study uses a political ethnographic method. Political ethnography corresponds to “a continuum of procedures for collection of evidence, from intrusive to inobtrusive [such as] the in-depth interviews, conversation, participant observation, passive observation of interaction, covert observation of interaction, and inobtrusive observation concerning residues and consequences of interaction” (Tilly 2006, 410). In a recent study on poor neighborhoods in Recife, Brazil, Koster and Eiró (2022, 223) offer that ethnographic fieldwork is the most appropriate method in exploring local clientelistic relations because of public “sensitivity” and “tacit understandings” on this issue. Likewise, Hagene (2015) argues that the ethnographic method is effective in uncovering the moral values and rationality behind clientelistic practices. She underlines that ethnography “provides insights into the mysteries of clientelism difficult to obtain by other methods, while it also presents us with the messiness of everyday life” (Hagene 2015, 157).

Utilizing this method, I conducted fieldwork in Bağcılar from 2014 to 2018. The extended duration was due to Turkey’s political turmoil in the mid-2010s, with ethnic and religious conflicts resurfacing amid terror attacks in major Turkish cities after 2015. During this period, there were bombings targeting civilians by the Islamic State of Iraq and Syria (ISIS) and the Kurdistan Workers’ Party (PKK), as well as armed attacks and violence against Kurdish communities in İstanbul. Political repression of dissident groups, the collapse of the “Kurdish resolution process,” escalating violence between the military and the PKK, the 2016 coup attempt by the Gülenists (a global Islamic community led by Fethullah Gülen) and the subsequent declaration of a state of emergency (OHAL) hindered interviews and observations. Security issues posed gatekeeper and trust problems, leading to repeated efforts to reestablish contacts.

Nevertheless, I conducted twenty-eight in-depth personal interviews, two focus-group sessions at AKP and HDP offices and engaged in participant observation during election campaigns in Bağcılar. The personal interviewees comprised six AKP brokers (three Kurdish housewives, three Kurdish businessmen), an AKP social affairs officer (male), a Bağcılar Municipality officer (male), three pro-AKP Kurdish textile workers (one female, two males), two pro-AKP Kurdish Islamist shopkeepers (males), a pro-AKP Kurdish Islamist university student (female), a pro-AKP Islamist academic (male), and a pro-AKP Islamist publishing house worker (male). For the HDP, interviewees included co-chairs (one male, one female), four activists (engineer/female, sociologist/female, anthropologist/male, realtor/male), a Kurdish Islamic scholar from the HDP’s Religious Scholars Assistance and Solidarity Association (male), and three pro-HDP textile workers (males), along with Socialist Solidarity Platform (SODAP) and Socialist Party of the Oppressed (ESP) representatives (both males). In finding interviewees, I initially utilized my previous contacts in leftist and Islamist circles, followed by snowball sampling. The study predominantly relies on the narratives of the AKP’s Kurdish brokers. Throughout the article, the term “broker” is used for both municipal and party brokers due to their overlapping roles. However, I try to define the affiliations of brokers when quoting. Profiles of the brokers cited in this study can be found in the Interviews section following the References. Last, but not least, it is necessary to emphasize that all names used in this study are pseudonyms.

Introduction to the field – Bağcılar

There were several reasons for me to choose Bağcılar. Firstly, when I started my fieldwork in 2014, Bağcılar was the most populous district in İstanbul with a population of 754,623. Secondly, both Islamist and Kurdish movements have been well organized in the district since the 1990s. Third, both the AKP and HDP have received higher percentages of votes than their provincial averages for İstanbul in this district in past elections (see results of the elections in Figure 1).

Figure 1. Electoral results in Bağcılar and İstanbul.

Source: T.C. Yüksek Seçim Kurulu (2019).

Bağcılar, located on the European side of İstanbul (see Figure 2), was an underdeveloped district that lacked infrastructure until the late 1980s (Temurçin 2012). Many interviewees remarked that it became a “modern city” with the efforts of Islamist RP local government in the 1990s. Nevertheless, the narrow surface area (22 square kilometers) and the high population density (33,582 people per square kilometer in 2010) have resulted in unplanned urbanization with informally built attached apartment buildings (which are called apartkondus by the locals) along narrow streets in most neighborhoods of the district. The district’s population increased by more than threefold between the mid-1980s and mid-2000s and has stabilized around 700,000 since then (see Figure 3).

Figure 2. Map showing the location of Bağcılar.

Source: İstanbul Metropolitan Municipality (2023).

Figure 3. Population of Bağcılar 1985–2022.

Source: Merkezi Dağitim Sistemi (2023).

The analyses of the election results (Figure 1) and birthplace proxy (see Figure 4) ⁴ indicate that Bağcılar has one of the largest Kurdish populations of any district in İstanbul. Kurdish communities migrated to the district mostly in the 1990s. As of 2015, the largest group in Bağcılar was the Bitlis community (mostly Kurdish), followed by the Sivas community (mostly Turkish) – respectively 43,283 and 39,605 people. Nonetheless, the number of migrants from non-Kurdish provinces is still more than migrants from Kurdish provinces in Bağcılar.

Figure 4. Birthplaces of Bağcılar residents (2015).

Source: Merkezi Dağitim Sistemi (2023).

The electoral results and participant observation reveal that the Kurdish population is especially concentrated on the northeastern axis of Bağcılar, particularly in Demirkapı, Fatih, and Kemalpaşa neighborhoods, which are the most populous neighborhoods (TURKSTAT 2022). Although there are no clear-cut spatial boundaries between different ethnic groups, some neighborhoods are stigmatized because of their strong Kurdish presence; for example, Demirkapı is stigmatized as “little Kurdistan” by the Turkish populace.

In Bağcılar, the HDP increased its votes considerably in the last decade and reached its peak in the June 2015 general elections. Although the AKP’s votes decreased between 2011 and 2015, the party recovered its votes by taking advantage of the securitization of the November 2015 general elections in the district (Şahin 2021).

The AKP’s clientelist–machine politics on the ground

Although it seems that the AKP provides some material benefits to the poor in exchange for votes and support at first glance, there are indeed intricate clientelistic relationships between the AKP, the party brokers, and poor voters (including Kurds) in Bağcılar. In interviews, the AKP officials stated that both Turkish and Kurdish poor asking for various favors frequently come to the party offices in the district. Nevertheless, party officials also claimed that they are against any form of favoritism and emphasized their “professionalism” in serving citizens. The following quotation from a local AKP official exemplifies this claim.

I started my political life at AK Party’s youth branches. Before that, I had never thought about the possibility of demanding some favor from a political party. However, people who visit our party demand various favors. I think this is quite inconvenient. People demand all sorts of things – money, jobs, apartments, husbands, wives, etc. They tell their problems. I am listening, making suggestions, and trying to find solutions. But they do not listen. Some of them said, “I am a member of this party since its foundation. You must do me a favor.” I told them, “Whether you are a member or not, your demand is inappropriate. You’re an ordinary citizen, and I’m a political organization.” We are against individual favors. I tell our visitors that they cannot get direct help from the party. The district governor shall provide assistance on behalf of the state and the municipality shall provide assistance on behalf of the local government. But our party is the wrong address. I always tell them where and how to apply for help. The current condition is not that bad, and we have achieved some success in professionalization. (Gökhan 2015, interview)

Despite the discourse on “professionalism,” AKP officials continue to provide particularistic benefits to citizens in the district. If the supply does not meet the demand in clientelism, clients are expected to stop asking for favors. Instead, there is mutual dependence between the patron, brokers, and clients at the local level. Indeed, the AKP’s Bağcılar Social Affairs Officer stated that they try to provide monetary aid to the poor coming to the party offices in “emergency cases.” This coincides with the findings of Tafolar’s (2015) study that investigates the nature of social programs in Bağcılar. Accordingly, non-clientelistic/formal and clientelistic/informal assistance are implemented together.

One of the most critical aspects of the AKP’s clientelist politics is the provision of social aid. Although AKP officials stated that they channel any request for social aid to the municipality and the district governorship (i.e. the Social Solidarity Foundation), some officials did not hesitate to talk about their continuing contact with the relevant authorities especially to provide food and cash to the poor. The officials make phone calls to the municipality or the governorship and ask for quick help in urgent cases. As an AKP official puts it, the relationships between the party, the municipality, and the governorship are like “professional inter-company relations.” There is continuous coordination between the government offices, municipal authorities, and party brokers in managing social aid.

Bağcılar Municipality, which is governed by a mayor from the AKP and whose city council has an AKP majority, provides considerable monetary and in-kind aid to the poor. For instance, the municipality provides “soldier aid” to those families whose children are doing their military service and financial support to those families affected by natural disasters (e.g. fires or floods). It distributes food packages, shopping coupons, and the Ramadan meal (iftar) tickets. It is difficult to assess whether such assistance is discretionary or not. However, in the interviews, both municipal authorities and party brokers stressed that the municipality does not discriminate in the distribution of aid and suggested that the Kurdish households comprise the largest group of recipients of municipal assistance in the district. AKP officials emphasized that Kurds from Bitlis are the primary recipient group, which is not surprising because they are the largest community in the district.

It should be noted that private-sector donations are often distributed directly by AKP brokers in the district. Although the municipality controls and keeps records of private donations distributed by the party brokers, it is almost impossible for any researcher to access the volume and composition of donations or the list of donors. My informants said that private donors delivered considerable resources to the party to alleviate poverty in the district. The businesspeople who provide these donations are presumably supporters of the AKP or those who have business interests with the party. Refika, who is a Kurdish housewife and one of the AKP’s neighborhood officers, claimed that they were impartial in distributing private donations.

Outsiders believe that we are only helping party supporters. This is false. We treat all groups equally. We also help Syrian migrants. If someone needs help, such as those living in the basement or housewives whose husbands cannot work because of illness, we provide assistance to them for sure. [Are these municipal aids?] No, these are aids from the private sector. State authorities perform their duties properly. However, we also distribute donations from the private sector under the control of municipal authorities and the party’s district executives. We keep records of the households we deliver food and money to. Volunteers donated a significant amount. Businessmen, especially large ones in the food industry, trust us and bring their donations. In Ramadans, with God’s blessing, businessmen make huge donations. (Refika 2018, interview)

The distribution of material benefits allows the party brokers to come into direct contact with the poor in the district. While distributing assistance, the party brokers specify and address all sorts of problems of the poor people. In the interviews, the party brokers highlighted that they are in the field every day, all year round, and personally visit voters at regular intervals. These visits and encounters facilitate the intimate relationship between the party (patron), brokers, and voters (clients). It could be said that the party brokers strive to gain the trust of the poor through long-term clientelistic relations. A Kurdish shopkeeper, who manages the AKP’s brokerage relations at the neighborhood level, described the intimate encounters between the party brokers and poor voters as follows:

We do not distribute food packages only during election campaigns or Ramadan; we always support the poor and needy people throughout the year. Our party has an organization in every neighborhood. These organizations try to keep up with the urgent needs of people. We do not consider religious, racial, ethnic, or linguistic differences in providing assistance to the poor. This assistance could be both material and moral. If a voter has a particular problem, we try to solve the problem. Thus, the success of the AK Party lies behind this fact. … We interpret the act of caring for the poor as our holy mission. We always care about deprived, injured, or oppressed people. Let us assume that citizens experience some sort of injustice in Bağcılar. It is just enough for citizens to reach our neighborhood representatives. If someone is seeking a solution from the municipality or the state, the rest is our duty. We solve problems. We convey the problems we specified in the neighborhoods to our superiors – to the deputy mayors. We do not visit people from one election to another like other parties. On a regular basis, we meet with the employers and headmen of the neighborhoods. We meet with the school principals in every two to three months. We try to reach orphans and poor children. (Kemal 2018, interview)

In addition to material distribution, the party’s brokers provide problem-solving networks for the poor in everyday life. Thanks to the continuous performance of party brokers, the AKP has achieved a strong grassroots presence at the local level. Benefiting from long-term intermediation, the AKP is constantly catching up with the various problems, concerns, and expectations of the poor.

The AKP’s problem-solving networks were formalized to some extent when the Bağcılar Municipality initiated the Gönül Bağı (“Love Bond”) project in 2017. Gönül Bağı is a “citizen-centered urban movement for solidarity and cooperation,” which helps the poor and needy in the district (Gönül Bağı 2023). The project was initially carried out by the party brokers, then became quite popular and turned into the largest publicly sponsored civic initiative in İstanbul with more than 10,000 volunteers. Gönül Bağı activists regularly visit households in all streets and corners of Bağcılar to ask about their needs. These activists had visited 160,000 homes as of November 2019. They distribute food, clothing, and basic supplies, which are provided by the municipality and private donors, and try to provide moral support to excluded, marginalized, and neglected people. In addition to Gönül Bağı, the municipality has various social projects, in which food, basic supplies, and healthcare are provided to the poor and elderly. For instance, through the Life Support Project and Charity Bazaar, the municipality meets the basic needs of thousands of low-income households (Yeni Şafak 2017). Under the Life Support Project, the municipality regularly cleans homes, washes laundry and dishes, and provides special care services (e.g. haircuts) for the elderly.

The AKP’s clientelist–machine politics extends beyond elections. The party’s female brokers conduct home visits to patients, the elderly, pregnant women, and mothers with newborn children. The party’s male brokers organize gatherings with the pious poor and meet regularly with village associations. The brokers especially use communitarian moral discourses to appeal to the poor pious voters in Bağcılar. These opportunities create a symbolic universe and affective community grounded in shared religious values. The symbolic aspect of the AKP’s clientelism resonates with the pious Kurdish voters as well because they also felt discriminated. In interviews, many Kurdish informants expressed their discontent with the secular Turkish elite. Against this background, the AKP’s relational clientelism provides a sense of integration and belonging to poor religious Kurds. Güneş-Ayata (1994) underlines that clientelist relationships produce a sense of belonging to an imagined community for migrants. Considering the “uprootedness, anxiety and insecurity” and “the feeling of helplessness” of migrants in urban areas, the recognition dimension of clientelism becomes crucial in their lives (Güneş-Ayata 1994, 21). The symbolic aspect of clientelism is vital for Kurdish migrant groups who experienced poverty, precarity, and vulnerability in the past. Kurdish migrants have felt discriminated against, stigmatized, and excluded for a long time, and this “sense of distance,” some argue, alienated the Kurdish poor from secular political parties (Pérouse 2016). One may therefore surmise that the relationship that AKP brokers have established with the Kurdish poor, based on the principle of Islamic solidarity, has been at least as effective as ethno-national ties.

However, it should be noted that the clientelistic relations between the AKP and the Kurdish poor are not always tangible but depend upon the favorable political–economic conjuncture. An HDP activist told the story of her Kurdish friend who lost his job in the municipality after the AKP’s defeat in the July 2015 general elections as follows:

In the last election, I was the supervisor of the HDP in Bağcılar. There was also this AKP supervisor at my school. I knew him very well because he was our neighbor, a Kurdish fellow from Adıyaman. We met each other several times in the school corridors and talked a little bit. … He has been working in the municipality. I heard that they fired him immediately after the elections. Later, I saw him on the street. He said that they were packing to move back to Adıyaman. I could not help but ask him: “My brother, did you really approach AK Party to find a job?” He said, “We are living in İstanbul, life is not easy here. We did not get as educated as you did, so we had no better options. I had to take shelter somewhere and had no choice. I gave AK Party my labor and efforts, but they fired me without mercy. I deeply regret my decision now.” … Later, I heard that he could not find a decent job in İstanbul and migrated back to his hometown. (Dilan 2016, interview)

Until the mid-2010s, Turkey’s economic growth allowed the Kurdish poor to benefit from the AKP’s patronage resources. However, with the economic slowdown in the late 2010s, the “AKP’s class-based inclusion project significantly weakened among the Kurdish poor. With the decline in economic growth, the government’s capacity of welfare provision and resource distribution also declined” (Günay and Yörük 2019, 39).

The AKP’s male brokers: Kurdish entrepreneurs

Deep-rooted patronage connections exist between the Bağcılar Municipality under the AKP’s administration and Kurdish businessmen. A small but effective group of Kurdish businessmen has emerged since the early 2000s in the district. The high volume of trade in the textile sector and the boom in the construction sector helped these businessmen to make considerable profits. This group is mostly composed of private contractors undertaking urban transformation projects or subcontracted municipal services. In addition, hundreds of Kurdish small business owners (mostly shopkeepers and tradesmen) support the AKP in Bağcılar. I call these groups “Kurdish entrepreneurs.” Their significance lies not only in their economic roles but also in their pivotal contributions to the AKP’s local organization. They function as brokers in the AKP’s mobilization of Kurdish voters at the local level. Kurdish entrepreneurs have articulated their support of the AKP government in terms of political stability and economic growth, at least until the late 2010s. A case in point illustrating this perspective is that of a Kurdish contractor who leased a tea garden from the municipality for a duration of twenty years:

We know very well that hundreds of businessmen in Bağcılar support the AK Party today. My Kurdish brothers, who have companies in the district, support our party. … Alhamdulillah, today we have economic opportunities owing to the government. As a Kurdish employer, I appreciate the government. I have thirteen to fourteen employees here in the tea garden; most are Kurdish. I also employ almost eight to ten workers in the construction business. I create employment for twenty-five people and contribute to Turkey’s economy. Let me tell you, my brother, I have never discriminated against any of my employees because of their ethnicity. This is not because of my Kurdish identity; I believe in justice and equality. Alhamdulillah, we have achieved great business success thanks to economic stability. Let’s assume that if all those favorable conditions become reversed in the future, God forbid that, then we will collapse. We know that very well. There are hundreds of Kurdish employers in our district who think in the same way. From time to time, we meet our colleagues. There is the Bitlis Platform, in which Kurdish businessmen come together every fifteen to twenty days, discuss economic issues and evaluate our contributions to the national economy. (Sabir 2018, interview)

Kurdish entrepreneurs assert that their profits have increased within the conducive political–economic climate of recent decades. Some Kurdish entrepreneurs remark that the municipality has played a supportive role in enhancing their businesses. Their narratives illuminate the municipality’s efforts to mitigate potential ethnic tensions and foster a favorable business environment at the local level. This is achieved through the convening of meetings that bring together networks of hometown associations representing various regions.

The municipality brought Turkish and Kurdish businessmen from the Black Sea and Southeastern regions together and declared their associations as sister institutions in Bağcılar. It was an excellent task to integrate them. Approximately a year ago, the municipality brought the Bitlis Platform and the Zonguldak Platform together in a massive meeting. Almost 5,000 people were gathered in a large hall. The MPs and mayors of the region arrived at this meeting. Ethnic prejudices have disappeared. Twenty years ago, there may have been some discrimination against Kurds. I mean, the businessmen from the Black Sea region approached Kurdish people with caution. Today, we are in the process of integration, thanks to the initiatives of the municipality. Ethnicity no longer matters. Kurdish businessmen in Bağcılar appreciate the current conditions very well. (Sabir 2018, interview)

Most of the Kurdish entrepreneurs I interviewed were well-recognized brokers. With extensive experience in local politics, they have actively participated in various stages of the AKP’s organizational structure. Some were active participants in the Islamist National Outlook Movement (Milli Görüş) and the RP’s organizations during the 1990s. A significant number of these Kurdish entrepreneurs have resided in their neighborhoods for over two decades, making them repositories of local knowledge. Owing to actual or fictive kinship ties, Kurdish voters frequently visit their workplaces seeking advice or assistance. In the evenings, Kurdish workers often drop by these workplaces for a chat or to request aid while sharing a glass of tea. The circulation of people, especially after working hours, is substantial. Kurdish entrepreneur–brokers genuinely engage in problem-solving for the disadvantaged, guiding them to relevant authorities or resources. Kurdish poor often seek assistance in finding employment, housing, resolving workplace or neighborhood issues, among other matters. However, Kurdish entrepreneurs faced challenges in securing public sector employment for their fellows due to limited opportunities, primarily reserved for educated individuals within the AKP’s inner circle. These brokers frequently assist their fellows in finding jobs within the private sector through their personal connections. Furthermore, it is necessary to remark that the workplaces of Kurdish entrepreneurs serve as informal party hubs where neighborhood representatives carry out grassroots operations for the AKP. Especially before and during election periods, these spaces are bustling with individuals engaged in various campaign activities, including recruiting new members from the neighborhoods, conducting phone checks of existing members and voters, and putting up campaign posters and banners.

Amid the prevailing neoliberal ethos, ethnic brokers’ “entrepreneurship stories” evoke respect among the Kurdish poor. Birelma (2019, 53) highlights that transitioning from “wage work” to self-employment is seen as the best path for social mobility among the working-class. During several visits to Sabir’ office, a pro-AKP Kurdish private contractor from Elazığ, I observed that as he shared his life story – starting from humble beginnings, doing manual labor, saving money, establishing a small construction company, and expanding his business – poor Kurdish workers listened attentively, eager to glean career insights. It is possible to say that the entrepreneurial narratives of Kurdish brokers are appealing to the Kurdish poor who aspire to acquire wealth quickly. Many Kurdish workers dream of becoming their own bosses while admiring hardworking relatives who have established successful workshops and looking down upon poorer relatives as “lazy” and “idle.” This neoliberal ethos propagated by the Kurdish entrepreneur–brokers seems effective among the Kurdish workers.

However, Kurdish entrepreneur–brokers are careful not to appear self-centered or egoistic to the poor. They prioritize Islamic solidarity, viewing assistance to the needy as a fundamental duty of every Muslim. They actively engage with hometown associations, mosque communities, Islamic foundations, and Quran schools. They contribute to these organizations by making donations, offering scholarships to students, and providing financial aid to the Kurdish poor, using their own resources or donated funds. This phenomenon is indeed in line with the Islamists’ “vernacular politics,” which can be defined as a “value-centered political process rooted in local culture, interpersonal relations, and community networks, yet connected through civic organizations to national party politics” (White 2002, 27).

The AKP’s female brokers: Kurdish housewives

The majority of the AKP’s Kurdish female brokers are housewives. During the interviews, Kurdish female brokers underlined they were much more influential than their male counterparts at the neighborhood level. While male brokers often expressed their aspirations for higher positions within the party’s administrative hierarchy, none of the female brokers articulated such ambitions in the interviews. In contrast to the male brokers, who frequently engage in formal political activities such as attending large congresses and meetings, female party brokers remain constantly active in the field (“sahada olmak”) and maintain direct interactions with voters. This phenomenon aligns with earlier research findings that associate the successful mobilization of Islamist political machines in poor urban peripheries with the door-to-door activism of female brokers (White 2002; Eligür 2010). Emel, a Kurdish housewife and the AKP’s neighborhood officer, depicted their grassroots activism as follows:

We are always in the field. We do not visit our electorate from election to election like other parties. We are always among voters. We hold weekly neighborhood meetings. Each neighborhood has a party admin. There is a group of executives in each neighborhood. In my neighborhood, I am in public relations; there is a friend of mine who conducts social affairs; and another friend is responsible for the media. We are doing such a division of labor. In the weekly meetings, we decide what we will do in the following week, who will be responsible for the operations, how we manage the activities, and so on. Our officers seek needy, disabled, and elderly people in every neighborhood. … Unfortunately, I moved to this neighborhood a year ago, so I am quite new. I did not have any prior knowledge of the neighborhood. After moving, the very first thing I did was that I went to the variety store (tuhafiyeci), introduced myself to the saleswoman, and tried to get preliminary information about the neighborhood. The saleswoman was so helpful and told me the names and addresses of the needy people in our neighborhood. I took notes regarding those people, and transferred my notes to the administrator, who then informed the municipality. In such cases, after we specify the needy people, the municipality comes in and provides food, coal, supplies, etc. [So, you become some sort of intermediary?] Of course, we send information about the neighborhood to our administrators. We have strict coordination. The admins report this information to the AKP Bağcılar branch. (Emel 2018, interview)

A Kurdish female broker of the AKP, Behice, said that she had been very active at every stage of the AKP’s local organizations for the last five years – from the recruitment of new members by visiting every apartment to attending “the democracy watches” at the Atatürk Airport after the failed coup attempt in 2016, from the distribution of the party’s gifts by knocking on every door in the neighborhood to watching the party’s election campaign stand in Bağcılar’s central square. Behice commented that:

Women in this party seem more faithful and sincere to me, perhaps because I see the enthusiasm of these women every day. Compared to men, women have limited opportunities but look at how we manage these limited resources and become successful. We spend a long time collecting the necessary knowledge and mobilizing the electorate in the neighborhoods. (Behice 2018, interview)

Kurdish female brokers pay special attention to the mobilization of poor pious Kurdish housewives. Ark-Yıldırım (2017) notes that the symbolic and affective identification established between the AKP’s female brokers and pious housewives over the Sunni-Islamic norms could not have been achieved by other political parties. In this respect, there is a similarity between the Turkish and Kurdish cases. As in the Turkish case, the AKP’s Kurdish female brokers, who act in accordance with the hegemonic religious norms, easily adapt to the traditional–conservative habitus of Kurdish housewives. The AKP’s Kurdish women brokers visit Kurdish housewives when their husbands are not at home and invite them to “only ladies” seminars, talks (“sohbetler”), and meetings. Similar to their Turkish counterparts, the AKP’s Kurdish female brokers deploy Islamic references in their home visits, prefer religious greetings (“selamünaleyküm”), celebrate Islamic rituals, holy nights (kandil), and Ramadan meals (iftar), and dress modestly (Doğan 2017). These traditional–religious references are effective on pious Kurdish housewives. However, beyond the religious identification, there is an important difference between the Turkish and Kurdish cases: the common language (Kurmanji) spoken between the Kurdish female brokers and housewives is very critical in the formation of the symbolic identification. In the interviews, the Kurdish female brokers frequently mentioned the advantage of common language, which is an integral part of a common morality and lifeworld.

These female brokers’ home visits do not necessarily have direct political purposes, as they often meet with Kurdish housewives for home gatherings (“gün yapmak”) and recite the Quran and mawlid in Kurmanji. In the interviews, the female brokers said that they try to regularly visit all apartments in their neighborhoods during Ramadan and organize communal gatherings for housewives. These visits and gatherings serve the purpose of Islamic solidarity. Most female brokers emphasize their Islamic motivations; some even claimed that their main mission is showing “the true path” and conveying “the holy cause.” A female broker of the AKP narrates her Islamic mission as follows:

For the time being, I work in the women’s branch of the AK Party, and I am also in the district administration. Many people are curious about whether we are paid or not. People often ask how much we get paid. In the name of God, we never receive any money. We are volunteers. It is a kind of love affair in our hearts. I believe that this is a spiritual thing. For example, in the month of Ramadan, I hosted thirty guests in one day and made meals for all of them. The same day, I went to my fieldwork. So, we wandered the neighborhood again and again, checked the records of our members in each street, and tried to recruit new members. I thought that I could not have done all those things within one day if it had not been Ramadan. I was hungry and thirsty. Even on a normal day, when a guest is coming, I might feel frustrated because of the preparations. However, on that day I did not become tired, Alhamdulillah. We did not pursue personal interests. It is about our motherland, nation, and flag. God gives us special power, Alhamdulillah! This is a sacred duty. (Behice 2018, interview)

In articulating Islamic morality, Kurdish female brokers often refer to God’s sake (Allah rızası), God’s blessings (Allah’ın lütfu), and God’s will (Allah’ın takdiri). Some emphasized that they do not expect the rewards of their efforts in this world but expect to get the fruits of their deeds in the afterworld. They said that they pay special attention to the Islamic principle of preaching (tebliğ and irşad).

When we go to homes, some of them slam the doors in our faces. They scream at us: “We do not want you here, get out.” I often understand from their face that they support the opposition parties. However, we still try to present small gifts, such as coffee and flowers. However, we encounter serious difficulties. … I am not saying this because I am an AK Party member, but please believe me, we try to communicate with everyone, we try to embrace each and every individual for God’s sake (Allah rızası). (Behice 2018, interview)

Considering the challenges Kurdish female brokers faced during the political campaigns and the rewards they claimed to expect in the afterlife, it is not possible to explain the AKP’s brokerage relations through self-interest and cost–benefit calculations in the short term. In the case of the AKP’s Kurdish female brokers, clientelistic exchange becomes an opportunity for constructing a moral–symbolic universe.

The AKP’s organizational brokers: Kurdish hometown associations

Kurdish hometown associations play a crucial role as intermediaries connecting the AKP with Kurdish voters in Bağcılar. According to official estimates, there are 490 hometown associations in the district (Ministry of Interior 2022). Notably, the Adıyaman Foundation (Adıyamanlılar Vakfı) and the Bitlis Platform (Bitlisliler Platformu) stand out as the two largest associations, both heavily influenced by Kurdish entrepreneurs supportive of the AKP.

These associations actively contribute to their Kurdish communities by offering financial aid to the poor, needy individuals, and students. They prioritize principles of Islamic solidarity and engage in community-building initiatives. Various activities, including iftar programs during Ramadan, historical peninsula excursions, and seminars on current political–economic affairs, are organized by these associations. The AKP officials underline that, despite limited financial resources, smaller village associations exhibit more dynamism compared to larger hometown associations. While some village associations transformed into coffee houses, most remain effective in community-building, orchestrating events such as wedding ceremonies and condolences, as well as mawlid programs.

The municipality plays a supportive role by providing essential goods and financial backing to village associations. There is a coordinated effort between municipal authorities, party officials, and hometown association executives. The municipality actively engages in bringing together Kurdish and Turkish associations from different regions through regular consultation (“istişare”) meetings held every five to six months, involving executives of hometown associations. A Kurdish businessman describes the clientelistic ties between the municipality and hometown associations as follows:

The municipality wants to know what we can do together and how we can serve our people in the best possible way. For example, let’s assume that I am from a village in Diyarbakır and the village has an association here. At istişare meetings, I have the opportunity to report which public goods and services are lacking in my village. I state my request and get them from the municipality in a short period. The municipality not only helps my villagers living in Bağcılar but also sends goods and services to my villagers back in Diyarbakır. Upon my association’s request, the municipality can build a children’s park in the village, send the necessary medical equipment to the village clinic, and renovate the village school. The municipality helps all hemşehri associations without discrimination. Nowadays, we are in the month of Ramadan. The Municipality gathered all associations and said, “Dear Friends, please make iftar organizations to your hemşehris, and let me help you with the meal, tables, and chairs. You just have to gather your people together and we’ll provide all necessities.” I mean our only mission is to communicate with our people (“hasbihal”). Yesterday, I attended a grand iftar organized by eight different associations. Most of the participants were Kurdish. (Sabir 2018, interview)

Facilitated by the municipal support, Kurdish hometown associations actively engage in reproducing clientelist–machine politics at the grassroots level, effectively operating as organizational brokers. While ongoing coordination is maintained, the frequency of meetings between Kurdish hometown associations and party executives intensifies during election periods. Serving as intermediaries, these associations play a pivotal role in directing Kurdish votes towards the AKP, employing a combination of material benefits and symbolic relationships to consolidate their influence.

Conclusion

It is necessary to move beyond conventional theories of clientelism, which predominantly focus on vote-buying – a straightforward exchange of votes for material gain – to understand the AKP’s clientelist–machine politics on the ground. This study calls for a more nuanced analysis, uncovering that the AKP’s mobilization of Kurdish voters in İstanbul’s peripheries relies on both machine politics and relational clientelism. Specifically, it delves into the pivotal roles played by three categories of AKP brokers: Kurdish (male) entrepreneurs; Kurdish housewives; and Kurdish associations. These co-ethnic brokers cultivate enduring and intimate relationships with Kurdish voters, offer problem-solving networks for the Kurdish poor in everyday life, and promote communitarian discourses and practices rooted in Islamic principles to integrate the Kurdish population. Furthermore, this study underscores that the “success stories” articulated by Kurdish (male) entrepreneur–brokers resonate with impoverished Kurdish workers, the traditional conservative values instilled by the AKP’s Kurdish housewife–brokers influence other Kurdish women, and Kurdish associations serve as organizational brokers for the AKP at the local level. These findings show that the AKP’s clientelist–machine politics encompasses multifaceted strategies that extend beyond mere vote-buying. This multifaceted approach not only addresses immediate material needs but also resonates with the moral–symbolic values embedded in pious Kurdish groups.

The AKP’s clientelist–machine politics is particularly effective among pious Kurdish migrants who have previously experienced discrimination, exclusion, and marginalization in western metropolises. The co-ethnic brokers strategically endeavor to establish a shared moral universe for these Kurdish groups by interacting with their traditional–conservative habitus and idioms and creating reciprocal obligations especially among the Kurdish poor in peripheral neighborhoods of İstanbul. By invoking Islamic solidarity, these actors construct an imagined community for the Kurdish poor, fostering trust and a sense of indebtedness. Therefore, the AKP secures Kurdish votes by employing both material and moral–symbolic dimensions of clientelism.

Nevertheless, it is crucial to acknowledge recent developments that have significantly constrained the AKP’s clientelist capacity. Firstly, there has been a notable shift in the AKP’s discourse from Islamic communitarianism to Turkish nationalism. This shift has to some extent excluded Kurdish voters from the previously established clientelist networks. Secondly, the economic crisis in the late 2010s, coupled with the AKP’s loss of the Metropolitan Municipality of İstanbul in the 2019 local elections, has led to a substantial reduction in the party’s financial resources. In this context, Kurdish businessmen who were once significant financiers of the AKP might not transfer sufficient resources to the party. These structural constraints have diminished the AKP’s mobilization capacity among the Kurdish poor. Recalling that the symbolic dimension of clientelism depends on the availability of material resources, it can be argued that the effectiveness of the AKP’s mobilization strategies among Kurds might decrease in the future.