Skip to navigation – Site map

HomeThematic issues34Politics of Notables versus Natio...

Politics of Notables versus National Machine: Social, Political and State Transformations, Party Organizations and Clientelism during AKP Governments

Toygar Sinan Baykan and Murat Somer


This article reviews the existing body of scholarship and draws on original qualitative research that compares the party organizations of four major Turkish parties. By doing so, it highlights the causal relationships -mediated by political agency– between socioeconomic structures and party politics. Clientelism remained part and parcel of Turkish party politics and state-society relations since the transition to multi-party system in mid-20th century through AKP governments. However, important changes occurred in terms of who effectively supplied it, how, and to whom on the basis of broader socioeconomic transformations of Turkey. Political parties did not equally or homologously adjust and respond to social change. The AKP’s successful organizational responses to socioeconomic transformations had important implications for Turkey’s political regime. In pre-AKP clientelism, the state and powerful local notables within parties were key to the supply and distribution of clientelistic benefits. We argue that this contributed to the emergence of electoral democracy under tutelage of military-bureaucratic state actors who faced weak resistance from a fragmented civilian political class. In turn, during AKP governments, the locus of patronage shifted from state to party and from local notables to the AKP party machine: national machine politics. This helped the party to pacify military-bureaucratic tutelage and achieve dominance in party politics, but, the lack of a categorically pro-democratic party ideology and the paradoxes of national machine politics together led the party to transform Turkey to an electoral autocracy instead of developing democracy. Further, the AKP reshaped the Turkish state, which on the surface began to look like a party-state. By revising and providing nuance to this observation, we discuss the emergence of a popular broker-state mediating between national and local state institutions and local communities through its control over resources. We maintain that all these changes facilitated also the party’s internal decay. This generates important opportunities for opposition parties, who however continue to rely heavily on pre-AKP forms of clientelistic politics. Without refashioning their party organizations and party-voter linkages, they may not be able to defeat the AKP electorally, and even if they are, they may not be able to secure successful democratization. They also face difficulties of altering how they form clientelistic linkages with society without controlling state resources at national level.

Top of page

Full text

We would like to thank Élise Massicard, Ceren Ark-Yıldırım, Sinem Adar, Serdar Tekin, and Pınar Bedirhanoğlu for their comments and suggestions in the workshop organized for this special issue, and Aytuğ Şaşmaz for additional comments. We also would like to thank two anonymous reviewers for their criticisms and recommendations. A substantial part of this paper is based on the data and insights acquired in a TÜBİTAK Research Project (Turkish Scientific and Technical Research Council, project no: 118K279). We are indebted to the research assistants of this project, Seval Gülen and İlker Kocael, for their contributions.


1How have the agents who distribute clientelistic benefits and those who receive them evolved in Turkish politics since the 1950s? How have political parties responded to socio-economic and spatial transformations and how have these adaptations influenced party organizations, problems of democratization, and Turkey’s political regime? How has the AKP (Adalet ve Kalkınma Partisi – The Justice and Development Party) been different from other parties in terms of its management of party-voters, and later, party-voters-state linkages, and how has this shaped Turkey’s political regime and the internal functioning of the AKP itself?

  • 1 We acknowledge that this research was funded by TÜBİTAK (Turkish Scientific and Technical Research (...)

2To address these questions, we focus on the intersection of party politics, socio-economic changes, and the historical evolution of clientelistic politics from the 1950s through the AKP era, by drawing on our qualitative field research on comparative party organizations in Turkey.1 Our resulting arguments acknowledge the important roles that agency plays in shaping political transformations. However, our main theses come together as a three-fold, multi-dimensional, and historically-structured explanation. We highlight the interactions among a) the progressive decline of state elites and assertive local notables in sustaining clientelistic state-party-society linkages as a result of socio-economic changes (most notably, increasing urbanization and the rise of neoliberalism), b) the rise of the AKP as a “national machine” filling the vacuum left by this transformation and c) the fall of military-bureaucratic elites as veto players as a consequence of these transformations. We argue that these factors prepared the ground for autocratization under civilian rule and should be part of an overarching explanatory framework that accounts for the current moves back and forth between autocracy and democracy in Turkish politics.

  • 2 We explain, and to a certain extent reinterpret, this term proposed by Ellen Lust (2009) in the rel (...)
  • 3 In this paper, we use the term “notable” in a loose fashion for referring to powerful political fig (...)

3We derive theoretical and empirical insights regarding the links between socio-economic transformations, party organizations, especially concerning how party-voter linkages are built and managed, how local and central party elites relate to each other, and broader political-institutional changes at national and party system levels. Based on a large body of research, we observe that clientelism has been a major, constitutive component of how almost all Turkish political parties (except profoundly ideological minor parties) have been building and operating party-voter linkages. “Competitive clientelism”2, thus, has been a decisive mechanism of party competition and power dynamics in Turkey. We also observe that major parties other than the AKP (partially, and for different reasons, also the HDP-Halkların Demokratik Partisi - The Peoples’ Democratic Party) have heavily depended on a “politics of notables,” 3i. e. a long-standing political class of local power holders to develop and maintain their ties with citizens, especially in the province.

4As a result, we note that, under the appearance of all-powerful leaders who dominate party organizations, these parties have had partially decentralized organizational structures and have depended on the personalities and social and economic resources of local notables in order to be able to function at local levels. While major structural changes have been perceptually undermining the power of local notables at least since the 1980s, we argue that the parties’ dependence on local personage (and, relatedly, mayors) has remained intact for the central administrations of parties like the CHP (Cumhuriyet Halk Partisi - The Republican People’s Party). In addition to these parties’ own cultural and organizational legacies, the fact that they remained in the opposition throughout the 2000s has also contributed to this dependence. Party centers, thus, lacked national government resources to distribute to local organizations and this enhanced their reliance on local elites.

  • 4 By this term we refer to the most dedicated and active elite core of party political activity inclu (...)
  • 5 Here, we would like to draw attention to a classic study by James C. Scott in which he describes a (...)

5By comparison, the AKP partially drew its cadres from a more recently developed “political class”4 with less local recognition and resources. Hence, the party was able to manage clientelism based on a highly centralized and bureaucratic “national machine politics”5 where party officials think and work as party functionaries rather than local power holders. This organizational model helped the AKP gain the upper hand in the politics of competitive clientelism and pacify military-bureaucratic tutelage. It also contributed to the party gaining predominant party status in Turkish politics overall, which the party then used to create an uneven playing field for its rivals (for opposition parties as well as powerful non-party elites embedded in military, state bureaucracy and media) and to generate unfair advantages for itself and for its supporters (most notably for its voters and party members but also for a large cluster of supportive business people).

6In the background of these developments lie two-way and non-deterministic causal connections between broader socio-economic transformations, party organizations, and politics. The rapid industrialization, urbanization and rural-urban migration that Turkey has been experiencing since the 1960s have restricted the opportunities and effectiveness of the politics of notables. At the same time, it has generated opportunities for new types of clientelistic politics since the 1970s, through the rise of “well-educated brokers” in urban and newly urbanizing settings (Ayata 2010; Sayarı 2014; Unbehaun 2005). Yet, we would like to highlight that the reflections of these changes on political parties have not occurred automatically. Presumably, national machine politics was not the only possible product of these transformations. We first see the mobilization of political Islamist mass parties in the 1990s, and, then, the emergence of a splinter party at least partially as a by-product of authoritarian interventions by pro-secular veto players in elected party politics (Somer 2014). This new party, the AKP, appears to have tapped into long-occurring socio-economic transformations by developing the “national machine politics,” which in turn helped the party build a more capable organization in mobilizing its supporters and distributing clientelistic benefits effectively. Thus, the AKP strategically outcompeted the opposition parties who could not regenerate and compete with it.

  • 6 Having said that and to contextualize this point, we also note that “dissent” has a distinct intern (...)

7What have been the consequences of this highly centralized and bureaucratized organization on internal party democracy and Turkey’s political regime and state-society relations? In tandem with the party’s increasing domination by Erdoğan’s persona, national machine politics stifled internal dissent6 within the AKP and produced a potential for growing authoritarianism. While the AKP turned increasingly authoritarian and eroded Turkish democracy by engaging first in stealth and nuanced authoritarian policies, then explicitly democracy-eroding ones; it also transformed itself into a major interface and intermediary between the state and citizens, who became grateful and captive audiences for the party. As the cumulative result of democratic backsliding that had started much earlier, by 2018, major political regime indices in the world no longer classified Turkey as an electoral democracy but categorized it as an “open autocracy” or “competitive authoritarian” regime (Esen, Gümüşçü 2016; Lührmann, Lindberg 2019; Somer 2016; Sözen 2019a; 2020).

8Hence, the AKP’s dominance of state institutions such as the judiciary branch and other parts of bureaucracy increasingly resembled a “party-state.” However, in general, we argue and discuss that the AKP developed a more nuanced role vis-à-vis state-citizen relations: what we call a “popular broker-state.” In turn, we observe that the CHP and the İYİ Parti (The Good Party) mainly preserved the politics of notables. At the same time, the İYİ complemented this with new recruits from urban professional classes due to reasons explained below. Finally, the HDP stands out as a distinctly ideological mass party, which, however, is by no means immune from clientelism.

9We also note that in recent years the AKP organization has been losing its competitive edge and spirit. The highly personalistic fashion in which the AKP built its national machine politics by concentrating powers in the hands of Erdoğan in the absence of meaningful mechanisms of accountability within the party might have gradually resulted in over-reliance on particularistic incentives for mobilization while collective incentives largely diminished. Such transformations have gone together with high-profile exits (e.g. the splits of Babacan and Davutoğlu factions) from the party. These factions recently turned into new parties and threatened the AKP by becoming alternative platforms for office seeking conservative elites. Yet another major development weakening the AKP organization has been the creation of an authoritarian “hyper-presidential system” in 2017 (Özsoy-Boyunsuz 2016; Sözen 2019a). This shifted power from the party to Erdoğan’s presidential administration. Meanwhile, opposition parties such as the CHP and İYİ Parti have been trying to renew themselves through alliance-building, discursively, and, partially, organizationally, thereby posing greater electoral challenges to the AKP. However, we do not find that these parties have yet shed their reliance on the politics of notables.

10In the following second section, we explore the historical contours of party-voter linkages in Turkey before the AKP. Following this, we analyze the AKP era. We conclude with a summary of our findings, arguments, and their theoretical implications to the topic at hand.

Party-voter linkages in Turkey before the AKP

11Since the transition to the multi-party system, material exchanges among parties, their members, and supporters have been “usual business” underlying inter and intra-party competition. Beginning with the pioneering works of Sayarı (1975; 1976), Sunar (1985) and Ayata (2010) based on field research conducted in the 1970s, studies on Turkish parties have regularly referred to patronage politics as an integral part of party competition. More recently, quantitative research by a younger generation of scholars has demonstrated that clientelistic party-voter linkages continue to prevail in contemporary politics (Çarkoğlu, Aytaç 2015; Çeviker-Gürakar 2016; Çınar 2016; Kemahlıoğlu 2012; Marschall et al. 2016).

The politics of Notables

12The predominant state-society dichotomy in the analysis of Turkish politics usually led researchers of Turkish party politics to focus on top leadership groups of political parties, national discourses, and ideologies. To a great extent, this tendency has inhibited a thorough understanding of the local sources of national party politics and its ills. These conventional analyses shaped by the center-periphery approach (Mardin 1973) or “state-society dichotomy” (Turam 2012) have, until recently, defined the ways in which we interpret the early Republican period and the following multi-party era. The idea of an almighty center and state in contradistinction with the segmented, heterogeneous, passive periphery and society has long dominated the works on Turkish politics until the rise of various critical theoretical approaches in the 2000s (Açıkel 2006; Navaro-Yashin 2002; Turam 2012; Massicard, Watts 2013; Somer 2014; Aymes et al. 2015).

13In the 1970s, 1980s, and 1990s, a specific cluster of scholarship on Turkish local politics had already presented a picture of party politics in the periphery that was remarkably different from the “strong center-weak periphery” frame (Akarlı, Ben-Dor 1975; Çitçi 1989; Schüler 1999; Varol 1989). More contemporary works have complemented these studies by demonstrating the widespread clientelistic and other practices in Turkish local politics that escape the control of central rational-legal bureaucratic authorities (Ark-Yıldırım 2017; Arslantaş, Arslantaş 2020; Açıkel, Balcı 2009; Arıkan-Akdağ 2014; Doğan 2016; Joppien 2018; Kılıçdaroğlu 2019; Kurtoğlu 2004; Sayarı 2014)

  • 7 For the power of local elites in the late Ottoman Empire see Yaycıoğlu (2016).
  • 8 One should also note, however, that the Republic with its greater coercive capacity diminished the (...)

14An important account in this regard is Michael Meeker’s seminal work (2001), A Nation of Empire, which is an excellent historical demonstration of local power structures in an Eastern Black Sea town.7 Contrary to narratives exaggerating the power of the Early Republican regime, Meeker’s account demonstrates how local social oligarchies have kept dominating the town of Of (Trabzon) even after the founding of the Republic.8 While at times these local power-holders fought against the pressures exerted upon them, they usually sought compromises and agreements with the center and its forces. In return, local power holders have been transformed into agents of the center while retaining a considerable degree of local autonomy that they defended against the center whenever they deemed necessary. In the multi-party era, these local power holders found an even larger room to manoeuvre since there were now more parties with whom they could negotiate. In another excellent historical and ethnographic study, Unbehaun (2005) reveals the considerable degree of local autonomy that Tuhfezâdeler enjoyed in Datça from the late Ottoman period until at least the 1970s.

15Even in the 1970s, well after Turkey started its rapid industrialization and urbanization in the 1960s, works by Sayarı (1975; 1976), Kıray (2006: 272-302) and Ayata (2002; 2010), reveal the importance to party politics of powerful local figures across Anatolia. Sayarı (1976), for example, demonstrates how it was futile to search for party organizations in distant towns of eastern Turkey. Instead, specific personalities with traditional symbolic and economic resources dominated the parties and local party politics. As a result, parties in these settings were mostly extensions of local power struggles. Ayata (2010), in her comparison of an industrializing town and a more rural local setting towards the end of 1970s observed similar patterns.

The rise of well-educated brokers in the 1970s

16Ayata’s (2010) seminal work on the CHP, however, reveals an important transformation of party-voter linkages in the 1970s. The socio-economic changes started to undermine the power of traditional notables, even in small towns, and paved the way for the rise of a new type of local elite that depended more on parties for their power, particularly in industrializing localities. This took place in the context of an import substitution economy, a highly fragmented party system, and shifting coalition governments, which together provided parties with more national resources to distribute to their clientele via loyal local party organizations and brokers. Compared to the dominant local elites who had access to considerable economic and symbolic resources of their own, these new distributers relied on their expertise and brokerage skills for political prominence. They had mainly become ideologically loyal representatives of the CHP, which had embarked on a more programmatic path at that time (Ayata 2010).

17But socioeconomic and sociocultural transformations of Turkey were far from complete compared to advanced capitalist societies. Turkey was still remarkably rural in the 1970s. In 1975, only %41,8 of the population lived in provincial and sub-provincial urban settings (Tekeli, Gülöksüz 1983: 1232). As Mübeccel Kıray’s classical study demonstrates (2006: 272-302), while rural Turkey, especially suburban towns, was increasingly being incorporated into industrial urban centers and traditional local notables (such as aghas) were losing ground, a new type of still autonomous local notables kept dominating the politics of rural Turkey. Urbanization was still a new process, and rural and local power structures were still able to reproduce themselves in urban politics and in urban peripheries in such metropolitan areas as İstanbul, Ankara, and İzmir.

18From a broader point of view of regime dynamics, one important consequence of the “politics of notables” in Turkey had been powerful veto players embedded in the state apparatuses such as the army and the judiciary, who found a large room for political influence and interference in governance. “Politics of notables” and the rifts and shifting loyalties these created within and between the parties sustained a highly fragmented political class and party system that were vulnerable against the interventions of (ideologically and organizationally) relatively more unified establishment elites (military, judiciary, academia and high-bureaucracy). This continued until the consolidation of the AKP’s domination of the political system in the 2010s. Until this moment, Turkey’s regime was an “electoral democracy” (Schedler 2002; Lührmann, Lindberg 2019) with free and fair elections, but also contained powerful and politicized bureaucratic veto players whose role kept Turkey from consolidating liberal, full democracy.

The social changes since the 1980s and their consequences for party organizations

  • 9 , accessed: 10.12.2020. Accordin (...)

19In the 1980s, urbanization accelerated, and in the mid-1990s, the rural-urban balance in Turkey changed in favour of urban areas. In 1990, 56% of the entire population lived in urban areas (Işık 1996: 785). According to the World Bank development indicators, the percentage of the urban population in Turkey increased from 65% in 2001 to 75% in 2019.9 It is unsurprising that this social transformation had profound implications for party-voter linkages, party organizations, and the regime type. The politics of traditional local notables became increasingly anachronistic. But reliance on the new politics of powerful local brokerage highlighted by Ayata was not necessarily conducive to building party-voter linkages in such a way that, for example, strong programmatic parties could establish full democracy free from military-bureaucratic tutelage either. Nor did such politics enable political parties to respond to the changing societal needs successfully. In the 1980s and ‘90s, significant portions of an urbanized society with improving indicators in areas such as education levels faced insecurity, inequality, and impoverishment due to neoliberal economic policies.

20Local patrons struggled handling the poverty generated by neoliberal, consumerist cities. Urban poor’s expectations exceeded far beyond a squatter house and informal employment without being subject to harassment (“başını sokacak bir ev ve ekmek”). They now included expectations such as better health services, education, infrastructure, and decent accommodation. Further, local patrons and educated brokers – lacking the backing of a strong government and lavish public spending – progressively lost ground themselves. They increasingly lost touch with local constituencies in an economy based on service-cum-construction sector and the alienating/anonymizing impacts of life in neoliberal cities (increasing individualization of even the urban poor, deterioration of the traditional neighbourhood community in which face-to-face relations supported the survival of urban poor). Even in a city such as Diyarbakır where local, individual and non-partisan resources of candidates are seen as vital, the trend in the mid-2000s was towards the weakening of the impact of local and personal resources of Diyarbakır political elite and an increasing role for “partisan state machinery” (Dorronsoro 2005: §56).

  • 10 We would like to thank Elise Massicard for drawing our attention to this important factor.

21While these transformations paved the way for a new party such as the AKP with a new clientelistic strategy, this certainly was not the only possible outcome. The other parties – including new parties that had been founded – do not appear to have adapted to the changing environmental conditions the way that the AKP did. One factor that led these parties to continue their reliance on old methods of organization and increasingly weakened notables even in the 1990s might have been the party system’s high degree of fragmentation and volatility. These were often reinforced by personal rivalries among major political actors such as that between the center-right Doğru Yol Partisi (DYP-The True Path Party) of Tansu Çiller and Anavatan Partisi (ANAP-The Motherland Party) of Mesut Yılmaz as well as the rivalry between the center-left CHP and the Demokratik Sol Parti (The Democratic Left Party) of Ecevit.10 This suggests that many causal mechanisms and intervening variables such as political agency, ideology, organizational incentives, and party system formation lie between socioeconomic changes and their consequences on party organizations.

22National machine politics was only one of many possible strategies for constructing party-voter linkages that a new political party like the AKP could pursue through adaptation or innovation in this new era. The possible responses to socio-economic dislocations and transformations could have been more, less, or differently clientelistic. Among potential responses were a reinvented Fazilet Partisi (The Virtue Party), the rise of a more programmatic center-right party than the clientelistic AKP or, in a markedly different scenario, the rise of a social democratic programmatic party. Addressing the counterfactual questions of why these collective actors did not or could not pursue these alternatives and how and why the founding elite of the AKP developed “national machine politics” is certainly pertinent. Yet, such an endeavour reaches beyond the pointed scope of this paper.

The rise of a political Islamist mass party and regime troubles in the 1990s

23As discussed above, the politics of clientelism in the 1960s and 1970s had also empowered bureaucratic veto players in the military and judiciary who were deprived of local resources but endowed with expertise and bureaucratic power. Insofar as they could maintain some level of ideological unity and coordinate their actions, they remained strong vis-à-vis existing political parties and elected officials. Adding insult to injury, the organizational characteristics of the parties of 1960s and ‘70s did not allow their elites to mobilize the masses against bureaucratic elites when bureaucratic veto players trespassed and interfered with the realm of democratic politics. They had not built direct linkages with their constituencies based on the party centers’ ideology, identity, and program (–i.e., autonomous from local elites and brokers).

  • 11 Züğürt Ağa, a brilliant Turkish comedy movie by director Nesli Çölgeçen, first screened in 1985, na (...)

24In the 1990s, two deviations from these organizational structures emerged with the introduction of new and professional political marketing techniques and the rise of the Islamist Refah Partisi (RP - The Welfare Party), a dynamic mass party-like entity (Özbudun 2001). At this point, immigration already became a decade-old phenomenon and had already made traditional rural patronage structures hard to reproduce.11 Still, however, new parties –most of which were actually refurbished versions of parties in the 1970s-- were far from grasping these transformations in organizational and strategic terms and kept mobilizing their supporters through powerful local intermediaries in urban and rural contexts. Only the RP, the predecessor of the AKP, was busy constructing a tightly controlled massive grassroots membership organization consisting of ideologically devoted members (Özbudun 2001; White 2002; Eligür 2010; Delibaş 2015). In the middle of the 1990s, the ideological “mass party” model (Özbudun 2001) that the RP embraced which bypassed the powerful local intermediaries whenever possible, proved to be a winning strategy.

25Hence, the RP threatened the electoral democracy under military-bureaucratic tutelage not only because of its ideology, i.e., passive-revolutionary political Islamism oriented toward state conquest (Tuğal 2009; Somer 2016). As a mass party able to form different and tighter links with its base than its center-right and center-left counterparts, the RP was also challenging the power of the military-bureaucratic veto players with its unprecedented mobilization capacity. Yet, the latter considered themselves as defenders of the Republic and saw the veto powers they enjoyed against elected politicians as sine qua non for maintaining the unitary and secular characteristics of the state.

The rise of “national machine politics” and the AKP as “the popular broker-state”: the bases of the new authoritarianism

26The AKP, to a great extent, embraced the fundamentals of the RP’s organizational strategies. But, the AKP also renovated these strategies in congruence with an increasingly mediated political climate. Public opinion researchers and professional advertisers (Baykan 2018) as well as recruits from center-right parties at both local and national levels (Ocaklı 2015) increasingly became part and parcel of the AKP politics. In addition, links with, and later grip on, media and business, and the increasing concentration of power in top leadership, most notably Erdoğan, complemented the conventional methods inherited from the RP. This model helped the AKP keep its core, devout and largely low-income constituency, together with relatively wealthier median voters. While a massive grassroots organization mainly targeted the core, median voters were targeted via more professional techniques including a partisan media. Combining direct contact with the electorate through devoted and loyal party members and new methods of political marketing (Baykan 2018), this model was a successful electoral adaptation to Turkey’s social and economic transformations – especially when the other parties failed to adapt to their new environment. Nevertheless, it became, in a decade, the primary vehicle of an increasingly authoritarian regime. This organizational evolution was also a drift away from the old politics of notables towards a new strategy of clientelistic politics for the AKP: the national machine politics.

  • 12 See interview 4C7.

27In our fieldwork on party organizations in Turkey, we met various AKP activists and members in different contexts. Their backgrounds reflected a mixture of national machine politics and vestiges of the politics of notables, but the former clearly represented the dominant, if not constitutive, element within the party. In addition, we also observed that figures within the AKP with a proclivity to autonomous behaviour and a type of old-fashioned personal patronage had to conform with the general scheme of organizational behaviour in the AKP. As a youth branch chair of the party in a metropolitan sub-province (ilçe) of İstanbul asserts “no one can become a party executive in the AKP by donating big sums to the sub-provincial organization”.12 It should also be noted that widespread acceptance of interventions by the AKP elite in the AKP activist base has always existed. This acceptance has resulted from a combination of selective (access to state resources) and collective incentives (trust in the leadership and devotion to the populist-conservative cause) (Ayan-Musil 2010) as well as specific organizational arrangements (Baykan 2018).

28The following two interviewees are similar to many others representing remarkably different characters and positions within the party. The first example, Sebahattin Bey (a pseudonym), is a well-educated elite coming from a wealthy, provincial notable family, – one of the founders of a local AKP organization. He represents older notable politics which quickly failed within the ranks of the AKP. In turn, the second, also a male interlocutor, is a neighbourhood representative coming from a financially underprivileged family without a graduate degree who worked on the administrative board of an AKP local organization and then as a neighbourhood representative in the city center. He represents the logic of “national machine politics” embraced by the AKP as the “popular broker-state”. The rather more predominant profile in the AKP is the latter since the party tasked talented and submissive brokers in local contexts with mediating the distribution of municipal and/or national government resources to supporters instead of local elites who deploy their own resources. But as the following Profile 1 reveals, this does not mean that the AKP excludes local elites with personal resources as long as they remain submissive to the role assigned by the party.

Profile 1: A failed notable/patron in the context of AKP’s national machine politics: Sebahattin Bey (1C5).

Sebahattin Bey, a wealthy man well over his middle age, received us in his private office, which initially gives the impression as if were a public office belonging to a high up politician or bureaucrat. He hails from a provincial notable family involved in politics for generations in a small provincial town in the Aegean region, which turned into a popular tourist destination in the 1970s (pseudonym “Seyidiye”). Sebahattin Bey entered politics in the 1970s through the youth branch of the Adalet Partisi (The Justice Party). Afterward, he was close to center-right parties in power in the 1980s and 1990s and became a candidate for the parliament from the lists of the center-right DYP. He was also a founder of the AKP branch in a sub-province of İstanbul. Although Sebahattin Bey was from Seyidiye, after graduating from a prestigious university in İstanbul he started living in İstanbul.

Sebahattin Bey compared the AKP era with the previous decades and argued that local politicians and candidates had more say over party decisions then. But in the AKP era, “everything is in between the leader’s lips”. There is no doubt that Sebahattin Bey was a figure created by the “politics of notables” who prevailed in the decades before the AKP. When comparing these different periods, it was evident from his mood, wording, and narrative that he had some nostalgia for this previous way of “doing politics” and was a bit reproachful toward the new ways.

Later, it became clear that Sebahattin Bey invested a considerable amount of his wealth in his hometown. He told us that they, as a family, had investments worth 25-30 million Turkish Liras in a halal, non-alcoholic touristic facility in Seyidiye, and he helped many people in his hometown. Thus, although he migrated to İstanbul, Sebahattin Bey was still employing and engaging through other patronage practices that helped his stature in his hometown. But such contributions did not enable Sebahattin Bey to rise into prominence within the ranks of the AKP. Overall, the profile of Sebahattin Bey manifests an ambitious local patron who was side-lined within the AKP machine politics.

  • 13 In our field research, we encountered many similar figures among the AKP local elites who came from (...)
  • 14 Two other interviewees -from the party’s executive board in two remarkably different provincial cit (...)
  • 15 Certainly, this may also have something to do with “in-group trust” in Islamic organizations. See A (...)
  • 16 In Turkish, using the second person plural (siz) is the proper and polite way to formally address s (...)

Profile 2: A loyal functionary of “the popular broker-state”:13 a neighbourhood representative in an Eastern Black Sea city (7C5).

This interview demonstrated to us the multi-dimensional relationships between state institutions, local party organizations, voters, and the AKP HQ. Muhammet Bey (a pseudonym) was born in 1975 in this Eastern Black Sea city as the child of a low-income, remarkably devout family with many siblings. His father provided the livelihood of this large family as a lorry/bus driver. Under economic difficulties, Muhammet Bey started to work and got involved in petty commerce already in his childhood. As he could not make it into the university, his commercial activities intensified. He also gained experience in various paid positions as a street vendor, pharmacist assistant, and jewellery and leatherwork salesperson in a touristic city on the Mediterranean coast. But “the lifestyle and culture that was far from spirituality” (“maneviyattan uzak yaşam biçimi ve kültür”) in that city led him back to his hometown after he fulfilled his mandatory military service.

In 2013, Muhammet Bey met people from the sub-provincial (ilçe) administration board of the AKP. As their relationship developed, he got closer to the party. In the first sub-provincial convention, after he established personal connections with the local party elites, he was invited to take part in the party’s executive committee in the central sub-province of his hometown. Nevertheless, Muhammet Bey noted, some contentious relations within this particular party administration that led the AKP HQ to demand the withdrawal of the administration board. Muhammet Bey told us that most of the board members, including himself, found it legitimate and followed this demand by the people who previously selected them as members of the board in the first place: “What I saw in politics is this: it is very hierarchical. This is, of course, not blind obedience, but it is futile to try to hold on to some positions.”

When he was in the sub-provincial executive board Muhammet Bey was the sub-provincial vice-chair responsible for “social affairs” (sosyal işler başkanlığı), which is one of the key positions within the party’s local organization for connecting the party to the poorer segments of its constituency: “for one and a half, two years, by relying on the state’s power, I helped many people and received many prayers in return. We were content when we were poor but I know very well the difficulties of being in need” (“Biz mutlu yoksullardık ancak muhtaçlık durumunun zorluklarını da çok iyi bilirim”). According to Muhammet Bey the mediation of the party is necessary, in specific cases, for “solving the problems of citizens”.

Nevertheless, Muhammet Bey added that people could abuse the party’s “brokerage/mediation” function. In our interview, he gave an example of such a situation, which perfectly demonstrated the two-dimensional function of the AKP’s local organizations between local society and local institutions of the central state. Muhammet Bey related the story of a citizen in detail whose disabled son’s social aid from the state was cancelled. Muhammet Bey took this “uncle” (amca) to the local directorate of social services: “When the director saw the uncle, he smiled. I did not have power, of course. I was there with the power of the party, I had the backing of the party’s power. Therefore, since I went there as a party representative, the director had to be interested in the problem. Many types of income in [city removed] are not officially registered. The director of social policy examined the uncle’s case and found that he registered all of his tea fields on to his daughter (“bütün çaylıklarını kızının üstüne yapmış”). So, I told the uncle that ‘uncle, if you want this money take it but this is another person’s right/benefit (“bu başkasının hakkıdır”); but if you want it, if you digest this (“içine sinerse”) I kindly ask (“rica edeyim”) the mister director and you take this money’”. According to Muhammet Bey’s narrative this “uncle” got embarrassed after this encounter and withdrew his request.

One of the points that should be emphasized regarding this incident is that Muhammet Bey was well aware of the fact that the power that he deployed in this particular encounter was by no means stemming from his own resources but entirely the outcome of his position within the party. He also had a sharp awareness of the leverage of the party vis-à-vis local state institutions. Muhammet Bey was almost sure that his “kind request” in this particular encounter as one of the AKP sub-provincial executives would be taken seriously, almost as if it was an “order”. But, with a remarkably bureaucratic attitude, Muhammet Bey refrained from “kindly asking” this favour and respected the director’s expertise and bureaucratic records.

Thus, Muhammet Bey also felt the need to address potential abuses of this power of the party, particularly in the lower echelons of local organizations: “The neighbourhood administrations should be service-oriented too. This job does not go well with personal expectations. For example, the guy calls the chief physician in the public hospital and tries to get his business done by telling them that ‘I am a member of the neighbourhood administration’. Such incidents of course reach to the HQ, provincial and sub-provincial administrations and these higher authorities intervene in such personal abuses. Party executives are frequently required to make such interventions.” 14

Muhammet Bey’s following comments are very important for a better understanding of the AKP’s collective broker role in citizen-state relations and the function of this role in AKP politics: “Most of the time, applying to political parties for their problems is the last resort for people. And these problems are usually tough to solve. They are, in fact, unsolvable. And most of the requests are unfair. But ‘patience’ is key. The citizen (“vatandaş”) would be sorry, would yell at you; you have to tolerate”.

In the following period, Muhammet Bey did not take part in the sub-provincial executive board of the party, but he was appointed as a neighbourhood representative to one of the important central neighbourhoods of this provincial city. It is crucial to note that the accepting/obedient attitude of Muhammet Bey regarding his change of position within the party, which appears as a demotion at first glance, is widespread among the AKP executives/activists.15

At the end of our interview, Muhammet Bey compared the political elites from ANAP and DYP in the province with the AKP executives: “In the 1990s, executives of the parties like ANAP and DYP were much stronger. The guy was like a governor. But now people talk to the provincial chair very informally, by calling him ‘sen’. (“Ama şimdi adam il başkanı ile senli benli konuşur.”)16 The reason behind such attitudes is this: the present executives of the party are those people who lived through these problems in that period, who were looked down upon [by the elites of that era]. And they usually learned positive lessons from this experience of maltreatment. … In addition, the control within the party is so tight. The HQ knows what is going on here and this brings success.”

  • 17 For a few such takes on the issue, see the following commentaries arguing that we now have a kind o (...)

29As our interviewee, Muhammet Bey, and other research participants represented, the AKP fulfilled a two-way role between citizens and the state. On the one hand, the party exerts tight, comprehensive control over the local state bureaucracy and brings problems of social sectors –no doubt those close to the party- to bureaucracy. Yet, on the other hand, the party organization, which recruits its “loyal personnel” mostly from popular sectors, supports and legitimizes bureaucracy by informing local society -mostly poor constituencies- about bureaucratic restrictions and functions and convincing people to respect them. In other words, party officials fulfilled a double representational function by representing citizens in their encounters with state institutions and vice versa. Therefore, describing the Turkish state under the AKP as a “party-state” is too simplistic, if not simply wrong.17 Instead, the AKP should be seen as a “popular broker-state” – largely separate from the state bureaucracy – which locates itself in a privileged position between state institutions and societal sectors in localities.

  • 18 Ceren Ark-Yıldırım drew our attention to this point. Even people with the capacity to act as local (...)

30While Sebahattin and Muhammet Beys exemplify two distinct typologies – local notables type and loyal and submissive party brokers type –the two types of party elite can in practice sometimes overlap when some notables accept to function as loyal brokers.18 The dominance of the second type of elites, namely the loyal broker, within the AKP seems necessary to sustain the party’s country-wide, centralized and pervasive clientelistic operations, i.e. “national machine politics”. Hence, the party’s power and hegemonic impact might rather stem from a liminal position resulting from its double representation function based on loyal functionaries. A de facto or de jure merger of the party with the state institutions would undermine its credibility and hegemonic position. By contrast, the capacity to simultaneously act as if the AKP were an oppositional force by representing unorganized social sectors in local and national level state institutions and generating consent to the decisions of state institutions among the citizenry, while exerting tight control over the functioning of these very institutions helps the party to maintain credibility and good relations with both state institutions and society.

National Machine Politics, Tutelary Elites, and Political Regime

31The AKP also struggled with powerful elite groups occupying entrenched positions in state apparatuses immediately after it came to power. After defeating these elite contenders based on electoral victories and democracy-eroding methods such as state-capturing, executive aggrandizement, creeping authoritarianism, and political lawsuits (in alliance with Gülen Community), the party started to more skilfully exploit and control public resources in local and national contexts. Thus, we observe an even more effective employment of national machine politics and the beginning of the internal transformations of this political structure itself.

  • 19 See interview 1C2. This particular interviewee gave us details on how the AKP manages clientelistic (...)
  • 20 These people could be “warden administration board members” (“nöbetçi yönetim kurulu üyeleri”) or i (...)
  • 21 Here we should also note that our observations in the field in eight different local contexts acros (...)
  • 22 Because the AKP distributes market cards to its clientele that could be used only in certain big re (...)
  • 23 As Arslantaş, Arslantaş (2020) relates, construction businesses close to the party recruit workers (...)
  • 24 See Lord (2018) for the clientelistic linkages between the AKP, pro-AKP religious civil society org (...)
  • 25 Again we would like to thank Ceren Ark-Yıldırım for drawing our attention to this important point.

32The party not only distributes in kind (such as food packages and coal) or cash benefits to its clientele, but also establishes these clientelistic linkages in localities in a remarkably impersonal way by distributing “market cards” for example. These “impersonal” instruments of clientelistic exchange such as “market vouchers/cards” carrying party or other insignia sustain the “national machine politics” and inhibit the association of benefits with specific figures in local branches. The party collects CVs in its local branches and records them to a city-wide “pool” in İstanbul, for example.19 The party also provides its clientele with other benefits such as mediating their relationship with state institutions including social service directorates and hospitals. Wherever “people in need” (“ihtiyaç sahipleri”) visit an AKP branch, there are always people to provide them with information and guidance.20 Thus, in the eyes of the party’s clientele, it is the party that provides benefits. In other words, in the AKP’s machine politics, the sole patron is “the party” (“parti”) and Erdoğan.21 It should be also highlighted that such “impersonal mediation” between party branches, local constituencies, and local state institutions creates “multiple dependencies” among various social groups, including economic elites in retail22 and construction sectors,23 and some pro-AKP civil society organizations24 as well. As a result, the entire clientelistic arrangement becomes highly durable.25

  • 26 It is difficult to define the sequence between the rise of the popular broker-state and the fall of (...)

33When the party consolidated its power at the beginning of 2010s (Sözen 2019a), this party of “loyal servants” (who were recruited from popular sectors and mostly with modest roots) became what we call the “popular broker-state”. It tightly controls state apparatuses and resources, locates itself in between citizens and bureaucracy, and distributes state offices and resources around a centralistic clientelistic logic in favour of the party (not necessarily in favour of local party actors and functionaries).26 The “popular broker-state” could be defined as a type of populist and clientelistic party that recruits its “broker elite” primarily from among the popular sectors and occupies a privileged position between state institutions and society by undertaking a centrally tightly controlled clientelistic function. We, however, also argue that national machine politics became a double-edged sword for democracy. With the rise of the AKP, the power asymmetry between a fragmented political party system based on internally contentious parties including many powerful figures and ideologically and organizationally more coherent veto players gradually dissolved. When this happened, this created greater opportunities for the AKP’s ongoing tendency toward authoritarianism.

34Once again, this was not a moot outcome. The changing power relationship between bureaucratic and incumbent party elites also generated an alternative path Turkey could have followed, had the AKP leaders chosen to utilize a potential for democratization that was unleashed through the empowerment of elites subject to electoral accountability vis-à-vis technocratic elites. On paper, this political power could be employed to build a more democratic social and political “center,” had the AKP had the political will and ideational repertoire to do so (Somer 2014). However, it is now clear that a cluster of predominant AKP leaders lacked a categorically pro-democratic and pluralist ideology that informed and shaped their policies. Moreover, factors such as confrontations with social and political opposition such as the Gezi protests, the fall out with the Gülenists followed by the coup attempt in 2016, and the geopolitical turmoil that resulted from the Arab Spring might have only deepened the AKP elite’s antidemocratic proclivities. Arguably, a government with fully democratic intentions might have responded to all of these domestic and international challenges without becoming authoritarian, in fact, by perhaps strengthening the democracy. Thus, already existing powerful populist, conservative, and nationalist authoritarian ideological trends (Sözen 2019b) fully unfolded throughout this process. Hence, the AKP embarked Turkey on a very different path. This also enabled the AKP to erode electoral democracy by the erasure of institutional checks and balances and to develop open autocracy through the creation of a popular broker-state. The popular broker-state has a double function: at times, it instrumentalizes state bureaucracy on behalf of its clientele, and at other times, it coerces its clientele to obey the undesirable decisions of the bureaucracy (as demonstrated in Profile 2). But ultimately, as the popular broker-state, the party does not consolidate a relatively autonomous state bureaucracy, a sine qua non for liberal democracy, but a highly dependent one submissive to the party.

The Roles of Other Parties

  • 27 We return to this theme of “competitive clientelism” in the section below.

35Which roles did the other parties play – if largely unintendedly – in consolidating the AKP-led autocracy? Despite their good intentions and, from time to time, tactical victories, these parties’ organizational and strategic culture were not appropriate for effectively challenging the AKP’s predominance in “competitive clientelism” (Lust 2009).27 As we have outlined previously, while the Islamist RP was investing in building a massive membership organization in the 1990s, other parties in the Turkish party system were blending old habits of Turkish party politics with political marketing techniques. Nevertheless, in organizational terms, these “old habits” inherited from the period preceding the 1980s were decisive in shaping organizations of center-right parties as well as the CHP.


  • 28 For an important analysis of factionalism in a provincial CHP organization, see Massicard (2013).

36Internal conflicts and publicly visible factionalism have always characterized the CHP organization.28 Powerful local and national figures in the CHP frequently challenged the party leadership. News regarding these internal rifts have been frequently publicized in national and local media. Indeed, we observed during our fieldwork this “contentious” intraparty relationship within the CHP. In two sub-provinces of İstanbul, for example, we witnessed verbal quarrels of local party elites in front of our eyes, and, many times, we heard critical stories from our interviewees regarding other party members in the same branch.

  • 29 In contrast, see Kemahlıoğlu, Özdemir (2018) for how the AKP center exerts a tight grip over AKP mu (...)

37In an extreme example in a small Thracian town, we witnessed how a powerful local boss divided the party’s local branch organization in the central sub-province and was elected as an independent mayor (who returned to the party later on). In another example from a central, wealthy district in İstanbul, we heard the story of a previous CHP mayor who dominated this central sub-province for four consecutive terms (almost twenty years). The crucial point here is that he dominated the party branch in this district and effectively challenged the party HQ by relying on municipal resources.29 At some point, he constructed his loyal grassroots organization in this district alongside – and occasionally in competition with – CHP neighbourhood organizations and created an independent power base of his own depending on municipal resources. Hence, this mayor, who quickly turned into a local boss within the overall structure of the CHP organization, created internal rifts within the local party branch. While, at the national level, these internal rifts have usually been presented as ideological factionalism (for example factionalism between ulusalcılar and social democrats), in fact, particularly at the local level, these rifts are usually more about personal conflicts regarding status and power than profound ideological disagreements.

  • 30 A similar observation regarding the divisive role of competitive formation of local ruling bodies i (...)

38The contentious nature of the intraparty organizational life in the CHP may have some ideological and historical-institutional roots related to the CHP’s being the country’s oldest surviving party. In fact, the party is the founding party of the secular Turkish republic (Sözen 2010; Esen 2014). Further, among the reasons might be the internal rules regulating the formation of local administrative bodies of the party, most notably “the system of neighbourhood delegates” (“mahalle delegeliği sistemi”), which fuels internal conflicts in the formation of sub-provincial administrative bodies of the party across Turkey.30

  • 31 Ayan-Musil (2010) underlines the lack of resources – that may be extracted from national office – i (...)
  • 32 For a detailed qualitative and quantitative historical analysis of the CHP combined with a local et (...)

39But an arguably important reason behind this contentious intraparty politics, which usually inhibits the party from effectively mobilizing its supporter base in elections and present the party as a united front to local and national constituencies, might be the scarcity of resources that could be extracted from the office due to the lack of control over national government and restrictive impact of competitive authoritarianism.31 In the absence of vast national-level resources that could back a more potent party center, keeping contentious politics at the local level under check becomes remarkably harder for the CHP HQ. The main consequence of these circumstances is the culmination of deeply contentious intraparty relations in general as well as the predominance of powerful local bosses within the party, most notably the mayors (Açıkel, Balcı 2009), in the command of only available external resources to the party, in particular. These factors together diminish the party’s capacity to run effective campaigns and challenge the AKP hegemony.32 Finally, as a positive indication, one should also note the widespread presence in the CHP of people from professional classes such as medical doctors, lawyers and academics, and recent attempts by figures such as chair of CHP Istanbul provincial organization, Canan Kaftancıoğlu, to strengthen the party’s grassroots basis and ties with the poor and the young.

The minor contenders: İYİ Parti and the HDP

  • 33 Most of the party chairs we interviewed from the HDP had experiences of custody and imprisonment an (...)
  • 34 This was evident in our field work in a South Eastern province where the İYİ Parti branch fell unde (...)

40Apart from the CHP, there are two effective minor contenders in the hegemonic party system created by the AKP: the right-wing İYİ Parti (The Good Party) and the left-wing, pro-Kurdish HDP (Halkların Demokratik Partisi – The Peoples’ Democratic Party). While the former has a considerable degree of pragmatic, centrist inclination, the HDP, as we observed in the fieldwork, is an ideologically oriented party mainly relying on a restricted but a highly motivated and devoted body of membership and activists.33 While this helps the HDP survive under heavy political pressure, it diminishes the party’s appeal beyond its core constituency. Here İYİ Parti is the more remarkable case in the domain of minor contenders. As we observed in our fieldwork, on one hand, organizationally, the party tends to reproduce the old habits of Turkish party politics by incorporating some ambitious, influential local figures.34 But, on the other hand, alongside these influential local figures with their own personal resources, the authoritarian and clientelistic nature of the AKP predominance in Turkey led the İYİ Parti to recruit its elite from among the private sector executives and well-educated professionals instead of local business class that need municipal and central public bids for their trade, particularly in large and medium-sized cities. This, in turn, provided the İYİ Parti with managerial effectiveness, which we observed in our interviews with İYİ Parti chairs across Turkey in various provinces and sub-provinces. But, again, lack of any extensive local or national resources in the hands of these parties under the conditions of “competitive clientelism” and uneven playing field for opposition parties diminishes their appeal to a broad audience.

“Competitive clientelism”35 and a fragile hegemonic party autocracy

  • 35 This paper uses the term “competitive clientelism” in a markedly broader way than Lust (2009). She (...)
  • 36 Our older interviewees with knowledge of party politics in the 1970s, 1980s and 1990s usually conte (...)

41The rising centrality of clientelistic practices deployed by the AKP led other parties, to a certain extent including the HDP, to further strengthen the clientelistic links with their supporters. In our interviews with chairs and members of opposition parties, we learned how it was frequent for these local activists to have demands from local constituencies regarding jobs and aids in kind and cash. In various visits to these parties for conducting interviews we saw piles of CVs on tables in party offices. And almost every effective opposition party, including the HDP, has tried to accommodate these requests. As our interviewees argue, while the AKP is engaging extensively in such practices, it has become difficult for other parties to avoid clientelism because “the citizens got used to this kind of relationship with parties during the AKP era” (4A1). Hence, the national machine politics embraced by the AKP has caused a kind of “clientelistic contagion” that deepened and widened the existing clientelistic proclivities of the Turkish party system.36

  • 37 While there is a consensus on the authoritarian aspects of the AKP-led system in Turkey, at least s (...)

42But one problem in a system of “competitive clientelism”, which, at the same time, operates in an electoral-authoritarian political system is that clientelistic practices drawing on unequal resources and different appropriation routes and strategies create remarkably different outcomes for these parties. The competition in this “clientelistic authoritarian system” is like a sword fight in which one party fights with multiple real swords it owns and the others with daggers they borrowed from somebody else. In other words, while the hegemonic center in this “competitive clientelism” enjoys extensive central and local resources it skilfully distributes as a “clientelistic national machine” or as the “popular broker-state”, contenders either required to rely on resources controlled by powerful mayors (like the CHP), or extremely restricted resources stemming from party members’ and party elites’ limited private wealth (the İYİ Parti and the HDP), grassroots volunteerism, and official aid. These conditions foster a fragile hegemonic party autocracy (Magaloni 2006; Greene 2007). Clients become more dependent on one party rather than comparing and choosing between different party elites who compete for the clients’ loyalty.37

43Finally, the evidence amounts that systematic and centralized modes of clientelism combined with polarizing politics, authoritarian party hegemony and the hyper-presidential system have been changing the membership profile and weakening the AKP from within, as recent revelations of high-level corruption, complaints by senior party founders, and splits from the party testify (Somer 2019). In our interviews, however, we encountered silence and only a few tangential references to such complaints and problems by AKP members. This could be contrasted with rather open criticisms of CHP members of their own party, and suggest that a kind of party discipline built into the national machine politics was still intact within the AKP.

Concluding discussion: the relationship between party-voter linkages, organizations, and regime types

44As the extensive literature on this dimension of Turkish politics attests to, patronage and clientelism has been part and parcel of Turkish party politics since the transition to multi-party politics. Powerful local patronage networks and national elites that these networks generated dominated most parties from the 1950s to the1990s. This “politics of notables” in a largely rural developing country created a deficient electoral democracy with powerful veto players. But as Turkey got more urbanized under the impact of neoliberalism during the 1980s and 1990s, local patronage networks gradually but persistently deteriorated. This created a party system crisis in the 1990s and prepared convenient conditions for the rise of, what we call in this paper, “national machine politics”, an extremely centralized way of clientelism that depends on a loyal, bureaucratic, and highly routinized organization. The AKP, which successfully built its party organization based on this model, dominated Turkey’s party politics and political regime in the 2000s and 2010s.

  • 38 See for example Baykan (2018), Gürsoy (2021), Yabanci (2016), Kaya (2019) and Yılmaz (2015).
  • 39 Here it is important to note that “bureaucratic modes of authoritarianism” in the 20th century, as (...)

45It would be misleading to describe this organization, which began to functions as a “popular broke-state” after the AKP consolidated its power, as a party-state. The party organization plays a major role in representing poor local constituencies in encounters with local bureaucracy and persuading the local society to the unpopular measures of local state institutions. In these encounters, the AKP enjoys a privileged liminal position between state and society. Despite evident authoritarian inclinations, the party carefully avoids a de facto or de jure merger with state institutions. Doing so fits with its populist worldview and praxis,38 which promises responsiveness and immediacy to unprivileged supporters. Acting otherwise would conflict with the image it carefully crafted over years in the eyes of its supporters as the antithesis of an authoritarian-bureaucratic arrangement distant to citizens.39

46Combined with what proved to be AKP governments’ dubious commitment to pluralistic democracy and rule of law, the replacement of contentious “politics of notables” with the “popular broker-state” facilitated the transformation of Turkey’s electoral democracy to a competitive or “new authoritarian” regime that drew on a new model of state-society relationship (Somer 2016). The politics of notables and contentious politics within Turkey’s ruling right-wing and left-wing parties before the AKP era were opening room for the interventions of powerful and ideologically coherent elite veto players and kept the system in check against the predominance of a political party. By comparison, centralized clientelism run by a coherent bureaucratic party structure progressively diminished the role of these veto players. In their place, the AKP implanted “the popular broker-state” recruiting members from Turkey’s popular sectors and uplifted these members with limited resources as local and national elites within a tightly controlled bureaucratic organization. They began to tightly control the public bureaucracy and locate themselves in between public services/resources and citizens. The ultimate consequence was a fragile hegemonic party autocracy that mainly relied on public resources, which contrasted with models of bureaucratic authoritarianism that rely on state elite or of electoral democracy that relies on dispersed local patronage networks and is vulnerable to the coherence of state elites.

47One of the implications of our research is that focusing on national-level power structures and struggles is inadequate for understanding the current authoritarian turn in Turkish politics. The changing clientelistic structures and local politics, and the asymmetrical response of the AKP and other parties to these developments during the last several decades comprise a fundamental factor that facilitated the current authoritarian predominance of the AKP. The broader theoretical implications of this analysis are as follows: there are strong causal links between social change, party-voter linkages, party organizational change, and regime change. Party organizations can be seen as a function of the party-voter linkages (Kılıçdaroğlu 2019). From the viewpoint of organization theory, party organizations should be seen as “open” systems sensitive to outside resources (Scott 2003) with internal conflicts (Pfeffer, Salancik 2003; Massicard, Watts 2013).

48Party-voter linkages are always sensitive to social change, but, as we show in this article, parties are not able to equally or homologously respond and adjust to change. The relative internal coherence or incoherence of parties in Turkey partly reflects how they arrange clientelistic relations in localities. Under the diminishing impact of local bosses/patrons in a rapidly urbanizing society, the AKP elite had the best organizational/strategic proclivity to respond to such an environment. This competitive advantage, nevertheless, gradually turned into authoritarian predominance, where contending parties remained under the burden of their historical legacy and old habits. Hence, there is a strong bond between how clientelistic practices are arranged and the way democratic backsliding emerged, as recent accounts highlighted (Çınar 2019, White 2017). With such party structures and strategic predispositions, it might be difficult for the CHP (or any other opposition coalition) to manage clientelistic aspects of electoral competition in Turkey, even if the CHP (or a coalition led by CHP) were to win national elections in future electoral contests.

49Despite their disadvantages in competitive clientelism in a context of autocratization under a fragile hegemonic party rule, Turkish opposition parties may still be able to defeat the AKP electorally through alliances struck by party leaders and new discourses, as the 2019 local election victories demonstrate. Our analysis here suggests that it may take more than electoral wins to lead sustainable democratization afterwards. Current opposition parties need to reduce their dependence on clientelism and the politics of notables by developing new forms of programmatic politics, recruiting new cadres at national and local levels, and cultivating different organizational dynamics. If non-AKP parties were to win the presidency, the executive resources at their disposal can be expected to strengthen their party centers, but this would not suffice to secure successful rebuilding of democracy.

Top of page



Interview 1C1 (2019). Personal Interview with an AKP sub-provincial chair in İstanbul – Jan 4.

Interview 1C2 (2019). Personal interview with an AKP sub-provincial vice-chair in İstanbul – Jan 9.

Interview 1C5 (2019). Personal interview with an AKP sub-provincial branch founding member in İstanbul – Feb 11.

Interview 2C2 (2019). Personal interview with an AKP sub-provincial administration board member in İstanbul – May 21.

Interview 2C4 (2019). Personal interview with an AKP sub-provincial women’s branch chair and a provincial administration board member in İstanbul – May 21.

Interview 2C5 (2019). Personal interview with an AKP sub-provincial youth branch vice-chair in İstanbul – May 22.

Interview 2C7 (2019). Personal interview with an AKP neighbourhood representative in İstanbul – May 22.

Interview 2C8 (2019). Personal interview with an AKP municipal assembly member in a peripheral sub-province of İstanbul – July 4.

Interview 3A3 (2019). Personal interview with a İYİ Party sub-provincial executive secretary in İstanbul – Aug 29.

Interview 3C1 (2019). Personal interview with AKP sub-provincial vice-chairs in İstanbul – Sep 5.

Interview 3C2 (2019). Personal interview with an AKP sub-provincial chair in İstanbul – Sep 6.

Interview 3C3 (2019). Personal interview with an AKP sub-provincial youth branch chair and a municipal assembly member – Sep 11.

Interview 3C5 (2019). Personal interview with an AKP sub-provincial youth branches administration board member – Sep 18.

Interview 4A1 (2020). Group interview with sub-provincial İYİ Party chair and the executive secretary in İstanbul – Jan 17.

Interview 4C6 (2020). Group interview with the AKP sub-provincial women’s branch chair and an executive board member – Feb 18.

Interview 4C7 (2020). Personal interview with the AKP sub-provincial youth branch chair in İstanbul – Feb 19.

Interview 6C3 (2020). Personal interview with an AKP provincial vice-chair in provincial Turkey– Jul 9.

Interview 6C5 (2020). Personal interview with the AKP provincial women’s branch chair and vice-chair in provincial Turkey – Jul 10.

Interview 7C1 (2020). Personal interview with an AKP provincial vice-chair in an Eastern Black Sea province – Aug 25.

Interview 7C2 (2020). Personal interview with an AKP youth branch chair in an Eastern Black Sea province – Aug 25.

Interview 7C3 (2020). Personal interview with an AKP women’s branch chair in an Eastern Black Sea province – Aug 28.

Interview 7C5 (2020). Personal interview with an AKP neighbourhood representative in an Eastern Black Sea province – Aug 31.

Interview 8C4 (2020). Personal interview with an AKP sub-provincial women’s branch chair and youth branch chair – Oct 7.

Interview 9A1 (2021). Personal interview with the İYİ Party provincial chair in Southeastern Turkey – June 22.

Interview 9C4 (2021). Personal interview with the AKP provincial women’s branch chair in Southeastern Turkey – June 28.

Interview F7 (2020). Online personal interview with a senior political scientist focusing on Turkish party organizations – June 12.

Published References

Açıkel, Fethi (2006). ‘Entegratif Toplum ve Muarızları: “Merkez-çevre” Paradigması Üzerine Eleştirel Notlar,’ Toplum ve Bilim, 105, pp. 30-69.

Açıkel, Fethi; Balcı, Sarp (2009). ‘Mayors as New Political Oligarchs/Magnets in Turkey: Patterns of Transition from Local Politics to National Politics and Vice Versa,’ in Reynaert, Herwig et al. (eds.), Local Political Leadership in Europe, Brugge, Vanden Broele, pp. 91-114.

Akarlı, Engin; Ben-Dor, Gabriel (eds.). (1975) Political Participation in Turkey, İstanbul, Boğaziçi University Publications.

Ark-Yıldırım, Ceren (2017). ‘Political Parties and Grassroots Clientelist Strategies in Urban Turkey: One Neighbourhood at a Time,’ South European Society and Politics 22 (4), pp. 473-490.

Ark-Yıldırım, Ceren (2018). ‘Cash Assistance by Smart Card: From Multiple Functions to Multiple Legitimacies of a Municipal Social Policy Instrument,’ Turkish Studies 19 (4), pp. 593-611.

Arıkan-Akdağ, Gül (2014). Ethnicity and Elections in Turkey – Party Politics and the Mobilization of Swing Voters, London, Routledge.

Arslantaş, Düzgün; Arslantaş, Şenol (2020). ‘How Does Clientelism Foster Electoral Dominance? Evidence from Turkey,’ Asian Journal of Comparative Politics,

Ayan-Musil, Pelin (2010). ‘Authoritarian Party Structures in Turkey: A Comparison of the Republican People’s Party and the Justice and Development Party,’ Turkish Studies 11 (2), pp. 197-215.

Ayata, Ayşe Güneş (2002). ‘The Republican People’s Party,’ in Rubin, Barry; Heper, M. (eds.), Political Parties in Turkey, London, Frank Cass, pp. 102-121.

Ayata, Ayşe Güneş (2010). CHP: Örgüt ve İdeoloji (trans. B. Tarhan and N. Tarhan), İstanbul, Gündoğan Yayınları.

Aymes, Marc; Gourisse, Benjamin; Massicard, Elise (eds.) (2015). Order and Compromise, Leiden, Brill.

Baykan, Toygar Sinan (2018). The Justice and Development Party in Turkey: Populism, Personalism, Organization, Cambridge, Cambridge University Press.

Çarkoğlu, Ali ; Aytaç, Selim Erdem (2015). ‘Who Gets Targeted for Vote-Buying? Evidence from an Augmented List Experiment in Turkey,’ European Political Science Review 7 (4), pp. 547-566.

Çeviker-Gürakar, Esra (2016). Politics of Favoritism in Public Procurement in Turkey, New York, Palgrave Macmillan.

Çınar, Kerem (2016). ‘A Comparative Analysis of Clientelism in Greece, Spain, and Turkey: The Rural-Urban Divide,’ Contemporary Politics 22 (1), pp. 77-94.

Çınar, Kerem (2019). The Decline of Democracy in Turkey – A Comparative Study of Hegemonic Party Rule, London, Routledge.

Çitçi, Oya (1989). Yerel Yönetimlerde Temsil- Belediye Örneği, Ankara, TODAİE.

Delibaş, Kayhan (2015). The Rise of Political Islam in Turkey: Urban Poverty, Grassroots Activism and Islamic Fundamentalism, London, I. B. Tauris.

Doğan, Sevinç (2016). Mahalledeki AKP – Parti İşleyişi, Taban Mobilizasyonu ve Siyasal Yabancılaşma, İstanbul, İletişim.

Dorronsoro, Gilles (2005). ‘The Autonomy of the Political Field. The Resources of the Deputies of Diyarbakır (Turkey): 1920-2002,’ European Journal of Turkish Studies 3, Being a MP in contemporary Turkey,, accessed: 18.9.2021.

Eligür, Banu (2010). The Mobilization of Political Islam in Turkey, Cambridge, Cambridge University Press.

Esen, Berk (2014). ‘Nation-Building, Party-Strength, and Regime Consolidation: Kemalism in Comparative Perspective,’ Turkish Studies 15 (4), pp. 600-620.

Esen, Berk; Gümüşçü, Şebnem (2016). ‘Rising Competitive Authoritarianism in Turkey,’ Third World Quarterly 37 (9), pp. 1581-1606.

Greene, Kenneth F. (2007). Why Dominant Parties Lose: Mexico’s Democratization in Comparative Perspective, Cambridge, Cambridge University Press.

Gürsoy, Yaprak (2021). ‘Moving beyond European and Latin American Typologies: The Peculiarities of AKP’s Populism in Turkey,’ Journal of Contemporary Asia 51 (1), pp. 157-178.

Hopkin, Jonathan (2004). ‘The Problem with Party Finance: Theoretical Perspectives on the Funding of Party Politics,’ Party Politics 10 (6), pp. 627-651.

Işık, Oğuz (1996). ‘1980 Sonrası Türkiye’de Kent ve Kentleşme,’ in Danzikyan Y. (ed.), Yüzyıl Biterken Cumhuriyet Dönemi Türkiye Ansiklopedisi, 13, İstanbul, İletişim, pp. 782-801.

Joppien, Charlotte (2018). Municipal Politics in Turkey – Local Government and Party Organisation, London, Routledge.

Kaya, Ayhan (2019). Populism and Heritage in Europe: Lost in Diversity and Unity, London, Routledge.

Kemahlıoğlu, Özge (2012). Agents or Bosses? Patronage and Intra-Party Politics in Argentina and Turkey, Colchester, ECPR.

Kemahlıoğlu, Özge; Özdemir, Elif (2018). ‘Municipal Control as Incumbency Advantage,’ in Sayarı, Sabri et al. (eds.), Party Politics in Turkey: A Comparative Perspective, London, Routledge, pp. 116-135.

Kılıçdaroğlu, Kerem (2019). Gelişmekte Olan Ülkelerde Siyasi Partiler ve Parti Stratejileri (trans. K. Turanlı), Ankara, Tekin Yayınevi.

Kıray, Mübeccel B. (2006). Toplumsal Yapı Toplumsal Değişme, İstanbul, Bağlam.

Kocaaga, Osman (2019). Siyasal Vekalet Eylemi Üzerinden Tahakküm İlişkilerini Düşünmek: CHP’de Aday Belirleme Süreçleri, Unpublished PhD dissertation, Galatasaray Üniversitesi, İstanbul.

Kurtoğlu, Ayça (2004). Hemşehrilik ve Şehirde Siyaset - Keçiören Örneği, İstanbul, İletişim.

Levitsky, Steven; Way, Lucan A. (2002). ‘Elections without Democracy: The Rise of Competitive Authoritarianism,’ Journal of Democracy 13 (2), pp. 51-65.

Livny, Avital (2020). Trust and the Islamic Advantage, Cambridge, Cambridge University Press.

Lord, Ceren (2018). Religious Politics in Turkey, Cambridge, Cambridge University Press.

Lust, Ellen (2009). ‘Democratization by Elections? Competitive Clientelism in the Middle East,’ Journal of Democracy 20 (3), pp. 122-135.

Lührmann, Anna; Lindberg, Staffan I. (2019). ‘A Third Wave of Autocratization Is Here: What Is New About It?,’ Democratization 26 (7), pp. 1095-113.

Machado, Kit G. (1974). ‘From Traditional Faction to Machine: Changing Patterns of Political Leadership and Organization in the Rural Philippines,’ The Journal of Asian Studies 33 (4), pp. 523-547.

Magaloni, Beatriz (2006). Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico, Cambridge, Cambridge University Press.

Mardin, Şerif (1973). ‘Centre-Periphery Relations: A Key to Turkish Politics?,’ Daedalus 102 (1), pp. 169-190.

Marschall, Melissa; Aydoğan, Abdullah; Bulut, Alper (2016). ‘Does Housing Create Votes? Explaining the Electoral Success of the AKP in Turkey,’ Electoral Studies 42, pp. 201-212.

Massicard, Elise (2013). ‘The Uses of Team Rivalry: Reconsidering Party Factionalism in Turkey,’ in Massicard, Elise; Watts, Nicole F. (eds.), Negotiating Political Power in Turkey: Breaking up the Party, London-New York, Routledge, pp. 55-76.

Massicard, Elise; Watts, Nicole F. (2013). ‘Introduction: Reconsidering Parties, Power and Social Forces,’ in Massicard, Elise; Watts, Nicole F. (eds.), Negotiating Political Power in Turkey: Breaking up the Party, London-New York, Routledge, pp. 1-16.

Meeker, Michael (2001). A Nation of Empire: The Ottoman Legacy of Turkish Modernity, Berkeley, University of California Press.

Mosca, Gaetano (1939). The Ruling Class, Karn, H.D. (transl.), New York-London, McGraw-Hill Book Company.

Navaro-Yashin, Yael (2002). Faces of the State: Secularism and Public Life in Turkey, Princeton, Princeton University Press.

Ocaklı, Feyraz (2015). ‘Notable Networks: Elite Recruitment, Organizational Cohesiveness, and Islamist Electoral Success in Turkey,’ Politics & Society 43 (3), pp. 385-413.

O’Donnell, Guillermo (1973). Modernization and Bureaucratic-Authoritarianism: Studies in South American Politics, Berkeley, University of California Press.

Özbudun, Ergun (2001). ‘The Institutional Decline of Parties in Turkey,’ in Diamond, Larry; Gunther, Richard (eds.), Political Parties and Democracy, Baltimore, Johns Hopkins University Press, pp. 238-265.

Pfeffer, Jeffrey; Salancik, Gerald R. (2003). The External Control of Organizations – A Resource Dependence Perspective, Stanford, Stanford University Press.

Sartori, Giovanni (2005). Parties and Party Systems, Colchester, ECPR Press.

Savaşkan, Osman (2019). ‘Türkiye’de Yerel Yönetimler ve AKP belediyeciliği,’ Birikim 358-359, pp. 88-99.

Sayarı, Sabri (1975). ‘Some Notes on the Beginnings of Mass Political Participation,’ in Akarlı, Engin; Ben-Dor, Gabriel (eds.), Political Participation in Turkey, İstanbul, Boğaziçi University Publications, pp. 121-133.

Sayarı, Sabri (1976). Aspects of Party Organization in Turkey, Middle East Journal 30 (2), pp. 187-199.

Sayarı, Sabri (2014). ‘Interdisciplinary Approaches to Political Clientelism and Patronage in Turkey,’ Turkish Studies 15 (4), pp. 655-670.

Schatz, Edward (ed.) (2009). Political Ethnography, Chicago-London, The University of Chicago Press.

Schedler, Andreas (2002). ‘The Menu of Manipulation,’ Journal of Democracy 13 (2), pp. 36-50.

Schüler, Harald (1999). Türkiye’de Sosyal Demokrasi-Particilik, Hemşehrilik, Alevilik (trans.: Y. Tonbul), İstanbul, İletişim.

Scott, James C. (1969). ‘Corruption, Machine Politics, and Political Change,’ American Political Science Review 63 (4), pp. 1142-1158.

Scott, W. Richard (2003). Organizations – Rational, Natural and Open Systems, New Jersey, Prentice Hall.

Sidel, John T. (1997). ‘Philippine Politics in Town, District, and Province: Bossism in Cavite and Cebu,’ The Journal of Asian Studies 56 (4), pp. 947-966.

Somer, Murat (2014). ‘Moderation of Religious and Secular Politics, a Country’s “Centre” and Democratization,’ Democratization 21 (2), pp. 244-267.

Somer, Murat (2016). ‘Understanding Turkey’s Democratic Breakdown: Old vs. New and Indigenous vs. Global Authoritarianism,’ Southeast European and Black Sea Studies 16 (4), pp. 481-503.

Somer, Murat (2019). ‘Turkey: The Slippery Slope from Reformist to Revolutionary Polarization and Democratic Breakdown,’ The ANNALS of the American Academy of Political and Social Science 681 (1), pp. 42-61.

Somer, Murat (2021). ‘Siyasal Rejim Nedir Ne Değildir?: Kavram Karmaşası Kademeli Otokratikleşmenin İçsel Bir Mekanizması Olabilir mi? (What is, and is not a political regime? Can Conceptual Confusion Be an Endogenous Mechanism of Incremental Autocratization?),’ Toplum ve Bilim 158: pp. 6-26

Sözen, Yunus (2010). Politics of the People: Hegemonic Ideology and Regime Oscillation in Turkey and Argentina, PhD Dissertation, New York University,

Sözen, Yunus (2019a). ‘Competition in a Populist Authoritarian Regime: The June 2018 Dual Elections in Turkey,’ South European Society and Politics 24 (3), pp. 287-315.

Sözen, Yunus (2019b) ‘Populist Peril to Democracy: The Sacralization and Singularization of Competitive Elections,’ Political Studies Review 17 (3), pp. 267-283.

Sözen, Yunus (2020). ‘Studying Autocratization in Turkey: Political Institutions, Populism, and Neoliberalism,’ New Perspectives on Turkey 63, pp. 209-235.

Sunar, İlkay (1985). ‘Demokrat Parti ve Popülizm,’ in Belge, Murst (ed), Cumhuriyet Dönemi Türkiye Ansiklopedisi -Cilt 8, İstanbul, İletişim, pp. 2076–2086.

Şaşmaz, Aytuğ (2022). ‘CHP’nin Dönüşümünde ve Seçim Başarısında Parti Örgütlerinin Rolü,’ İstanbul, TÜSES, accessed: 19.7.2022.

Tekeli, İlhan; Gülöksüz, Y. (1983). ‘Kentleşme, Kentlileşme ve Türkiye Deneyimi,’ in Belge, Murat (ed.), Cumhuriyet Dönemi Türkiye Ansiklopedisi – Cilt 5, İstanbul, İletişim, pp. 1224-1238.

Tuğal, Cihan (2009). Passive Revolution: Absorbing the Islamic Challenge to Capitalism, Stanford, Stanford University Press.

Turam, Berna (ed.) (2012). Secular State and Religious Society, New York, Palgrave Macmillan.

Unbehaun, Horst (2005). Türkiye Kırsalında Kliyentalizm ve Siyasal Katılım-Datça Örneği: 1923-1992 (trans.: M. Öztürk), Ankara, Ütopya.

Varol, Muharrem (1989). Yerel Siyasetin Demokratikleşmesi, Ankara, Gündoğan.

Von Beyme, K. (1996). ‘The concept of political class: A new dimension of research on elites?,’ West European Politics 9(1), pp. 68-87.

White, Jenny B. (2002). Islamist Mobilization in Turkey: A Study in Vernacular Politics, Seattle, University of Washington Press.

White, Jenny B. (2017). ‘Spindle Autocracy in the New Turkey,’ The Brown Journal of World Affairs 24 (1), pp. 23-37.

Yabanci, Bilge (2016). ‘Populism as the Problem Child of Democracy: The AKP’s Enduring Appeal and the Use of Meso-Level Actors,’ Southeast European and Black Sea Studies 16 (4), pp. 591-617.

Yaycıoğlu, Ali (2016). Partners of the Empire, Stanford, California University Press.

Yıldırım, Kerem (2020). ‘Clientelism and Dominant Incumbent Parties: Party Competition in an Urban Turkish Neighbourhood,’ Democratization 27 (1), pp. 81-99.

Yılmaz, Zafer (2015). ‘“Strengthening the Family” Policies in Turkey: Managing the Social Question and Armoring Conservative-Neoliberal Populism,’ Turkish Studies 16 (3), pp. 371-390.

Top of page


1 We acknowledge that this research was funded by TÜBİTAK (Turkish Scientific and Technical Research Council, project no: 118K279). The field research started in December 2018 and focuses on party-voter linkages, local and national party systems, and intraparty operations of four parties represented in the parliament: AKP, CHP, İYİ Parti, and the HDP (The MHP did not provide the permissions required by TÜBİTAK). We have so far visited eight different local contexts (socioeconomically and socioculturally different four sub-provinces of İstanbul and four provinces across Turkey) and conducted 206 in-depth interviews (involving more than 270 people), mostly local party activists and members, but also other relevant actors of the local political fields such as journalists and representatives of non-governmental organizations. Besides socio-economic and socio-cultural variation, the local contexts that we chose for field research reflect variance in the electoral success of parties under investigation in this research. We conducted field research in four different districts of İstanbul and four different provinces across Turkey, each representing a stronghold for individual parties under examination. Thus, we had a chance to observe every party covered in this research in their strongholds in two different local contexts. Our interviews were based on open-ended questions and usually conducted in party offices. They lasted around 45 to 60 minutes, but at times we had interviews that were longer than three hours. This provided us with invaluable opportunities for on-site observations of local functioning and activism in Turkey’s political parties. Hence, we tried to conduct the entire research process with “ethnographic sensibility”, as Schatz (2009) termed it.

2 We explain, and to a certain extent reinterpret, this term proposed by Ellen Lust (2009) in the relevant section below.

3 In this paper, we use the term “notable” in a loose fashion for referring to powerful political figures with considerable economic and symbolic resources at their disposal, which, when necessary, renders these figures capable of embracing defiant positions against the leadership of parties and other central authorities for protecting their autonomy. Our concept combines two forms of powerful networks of local politics. First, “traditional notable politics” that depends on traditional social elites (such as landlords/ağalar, notable families/eşraf aileleri, sheikhs and tribal chiefs/aşiret reisleri) and their symbolic and rural economic resources. Second, a more modern “machine politics” (Scott 1969) depending on groups of “new man,” who are combining political brokerage with their recently acquired economic resources in rural and urban contexts. See Machado (1974) and Sidel (1997) for these two types of notable politics that can be seen in the Philippines. In Turkey, too, traditional notable politics increasingly turned into “rural and urban machine politics” after the 1970s. See Ayata (2010) for a comparative analysis of the transformations of the CHP which demonstrates increasing reliance of the party in the 1970s on clientelistic brokerage run by new well-educated figures instead of patronage politics under the control of traditional local social elites. Kıray’s (2006) classic study of changing patronage structures in rural Turkey, too, vividly demonstrates the demise of the traditional notable politics in Taşköprü and the rise of a new stratum of local notables in this particular local setting relying on their recently acquired economic resources and their connections with commercial urban centers.

4 By this term we refer to the most dedicated and active elite core of party political activity including local elites at the lower ranks of party politics. Although this term is inspired by the approach of elite theorists, and most notably Gaetano Mosca (1939), we use it here in a specific way to describe particular layers of political elite profoundly –and, in most cases officially- engaged with party politics. On the one hand, the term “political elite” has a broader connotation than “political class” including traditional power holders, non-party professionals with relevant skills such as media and marketing experts as well as politicized state elites. On the other hand, the term “party elite”, in its conventional uses, may sound too restrictive by excluding lower rank local political elites, such as the examples we discuss in this paper. This is why we refrained from using both terms. Also see Von Beyme (1996) for a similar take on the concept of “political class”.

5 Here, we would like to draw attention to a classic study by James C. Scott in which he describes a “machine” as a “non-ideological organization interested less in political principle than in securing and holding office for its leaders and distributing income to those who run it and work for it. It relies on what it accomplishes in a concrete way for its supporters, not on what it stands for. A machine may in fact be likened to a business in which all members are stockholders and dividends are paid in accordance with what has been invested” (1969: 1144). Yet, we also would like to differentiate what we call “national machine” from what Scott describes as “urban political machines” run by powerful local bosses/brokers. “National machine” is much more centralized than “urban political machine”. It is a form of “mass clientelistic party” (Hopkin 2004: 632) that is highly coherent and well-coordinated. An example of a classic “machine” may be what is observed in the Philippines during the 1970s, just after the fall from political significance of rural landholding elites due to economic and social modernization: it is a powerful local network of clientelistic relations built around powerful “new men” who gained control over economic resources through their political cunning. Although “machine politics” is less autonomous than “traditional notable politics” since it relies on coalitions with national party elites, “broker-patrons” in the middle of classical machine politics have much larger room than a broker in, what we call, a “national machine”. See Machado (1974) and Sidel (1997) for personalized –and relatively more autonomous- forms of local political dominance in the Philippines.

6 Having said that and to contextualize this point, we also note that “dissent” has a distinct internal understanding among AKP members different from the understanding especially within the CHP, whose members interpret “dissent” quite differently in terms of what constitutes it and what are its normative consequences.

7 For the power of local elites in the late Ottoman Empire see Yaycıoğlu (2016).

8 One should also note, however, that the Republic with its greater coercive capacity diminished the region’s local autonomy to some extent (Meeker 2001).

9 , accessed: 10.12.2020. According to data by the Turkish Statistical Institute, which also takes small sub-provincial areas into consideration, this ratio is %92.8 in 2019. See , accessed: 11.12.2020.

10 We would like to thank Elise Massicard for drawing our attention to this important factor.

11 Züğürt Ağa, a brilliant Turkish comedy movie by director Nesli Çölgeçen, first screened in 1985, narrates the tragicomic story of a traditional landlord (toprak ağası) who migrates to İstanbul and loses his authority and wealth under the circumstances of a neoliberal urban context. The movie is emblematic of the gradual and partial fall of traditional notable politics in Turkey in the 1980s.

12 See interview 4C7.

13 In our field research, we encountered many similar figures among the AKP local elites who came from very modest backgrounds and owed their recently gained power and influence almost entirely to the party. See interviews 1C1, 2C2, 2C4, 2C5, 2C7, 2C8, 3C1, 3C2, 3C3, 4C6, 6C5, 7C1, 7C2, 7C3, 8C4, 9C4.

14 Two other interviewees -from the party’s executive board in two remarkably different provincial cities- also mentioned incidents of party members embracing dominant attitudes towards bureaucrats where higher party officials intervened in favour of the bureaucrats. Interviewees 6C3, 7C1. This also implies that such problems are not rare and may not always be easy to resolve.

15 Certainly, this may also have something to do with “in-group trust” in Islamic organizations. See Avital Livny (2020). Also see Ayan-Musil (2010).

16 In Turkish, using the second person plural (siz) is the proper and polite way to formally address someone one does not know well – particularly in an interaction with a person who occupies a position of perceived prestige and status. The second person singular (sen) is generally used either for a person one knows well or for persons with an inferior status.

17 For a few such takes on the issue, see the following commentaries arguing that we now have a kind of party-state in Turkey: , accessed: 13.4.2021. , accessed: 13.4.2021. , accessed: 13.4.2021.

18 Ceren Ark-Yıldırım drew our attention to this point. Even people with the capacity to act as local notables become subordinate in the AKP’s organizational arrangements. This is not only because the exit cost is usually too high, but the party can also always find positions in other localities for its “loyal functionaries” like a large institutionalized corporation moving its personnel to different local branches. This is especially valid in a city like İstanbul.

19 See interview 1C2. This particular interviewee gave us details on how the AKP manages clientelistic relations when it comes to providing party clientele with public or private sector jobs. It seems that there is an established system – and perhaps an online system – in the AKP for managing such relations, in a surprisingly impersonal way. It is important to note that this interview occurred in a sub-province in İstanbul controlled by a CHP municipality. Nevertheless, the AKP branch in this sub-province was very busy during our visits that took place before the March 2019 local elections, and apparently, the AKP’s higher level of coordination was used to channel the resources provided by AKP municipalities in other sub-provinces of the city to the party clientele in this particular sub-province, which was a CHP stronghold.

20 These people could be “warden administration board members” (“nöbetçi yönetim kurulu üyeleri”) or in large peripheral sub-provinces the AKP keep an entire “solution room” (“çözüm odası”), staffed with two to three party workers, open for responding particularistic demands.

21 Here we should also note that our observations in the field in eight different local contexts across Turkey are in congruence with a series of work on the clientelistic functioning of Turkish parties, and most notably the AKP, in specific neighbourhoods, mostly in İstanbul. See Arıkan-Akdağ (2014); Ark-Yıldırım (2017); Arslantaş, Arslantaş (2020) and Yıldırım (2020). All of these studies confirm the centralized and bureaucratic arrangements behind the clientelistic operations of the AKP.

22 Because the AKP distributes market cards to its clientele that could be used only in certain big retailers.

23 As Arslantaş, Arslantaş (2020) relates, construction businesses close to the party recruit workers referred by the party.

24 See Lord (2018) for the clientelistic linkages between the AKP, pro-AKP religious civil society organizations and poor constituencies.

25 Again we would like to thank Ceren Ark-Yıldırım for drawing our attention to this important point.

26 It is difficult to define the sequence between the rise of the popular broker-state and the fall of the traditional veto players (in army, judiciary and bureaucracy) in Turkey. But it is safe to argue that the AKP had this proclivity to turn into a national clientelistic machine from the very beginning. Such proclivity was further corroborated by the destruction of Kemalist establishment elites during the period between 2007 and 2011 through dubious methods deployed by the alliance between the AKP and the Gülen Community. Afterwards, the popular broker-state started to unfold fully.

27 We return to this theme of “competitive clientelism” in the section below.

28 For an important analysis of factionalism in a provincial CHP organization, see Massicard (2013).

29 In contrast, see Kemahlıoğlu, Özdemir (2018) for how the AKP center exerts a tight grip over AKP municipalities and mayors. Also see Savaşkan (2019) for the tight AKP grip on municipalities and mayors.

30 A similar observation regarding the divisive role of competitive formation of local ruling bodies in the CHP was made by Şaşmaz (2022).

31 Ayan-Musil (2010) underlines the lack of resources – that may be extracted from national office – in the hands of the central executive elite as one of the fundamental determinants of an oligarchic type of authoritarianism in the CHP.

32 For a detailed qualitative and quantitative historical analysis of the CHP combined with a local ethnography around candidate selection struggles, see Kocaaga (2019).

33 Most of the party chairs we interviewed from the HDP had experiences of custody and imprisonment and, at some point, we had this impression that such experiences are “symbolic resources” that pave the “road of honour” to ruling positions within the HDP.

34 This was evident in our field work in a South Eastern province where the İYİ Parti branch fell under the personal influence of a powerful notable family. See interview 9A1.

35 This paper uses the term “competitive clientelism” in a markedly broader way than Lust (2009). She deploys this term to explain the survival of overwhelmingly authoritarian systems with internal consensus or collusion between ruling parties and loyal oppositions in the Middle East. In our case, the semi-democratic legacy and competitive dimension of authoritarianism are much more pronounced, and, as a consequence, resources of distribution are remarkably different.

36 Our older interviewees with knowledge of party politics in the 1970s, 1980s and 1990s usually contended that the volume of such transactional relations between parties and voters dramatically increased in the AKP era. See interview 3A3 and 4A1. An expert interview with a senior colleague with fieldwork experience on political parties in the 1970s also approved that the volume and diversity of clientelistic exchange between parties and voters today is stunning compared to previous eras. See interview F7. Also see Çeviker-Gürakar’s (2016) important work that compellingly demonstrates the increase in the volume of non-competitive public procurements benefitting AKP supporters. This could be considered a reliable proxy indicator demonstrating the increasing volume of clientelism in the AKP era.

37 While there is a consensus on the authoritarian aspects of the AKP-led system in Turkey, at least since the coup d’etat in 2016, its precise nature has been a point of heated debate. Turkey is evidently not a full-scale autocracy but at times throughout the last decade of AKP rule, the political system in Turkey -at certain moments and localities- could be less competitive than a “competitive authoritarian regime”, as formulated by Levitsky, Way (2002). However, this authoritarian system has never been as stable as an ideal typical “hegemonic party regime”, examples of which are demonstrated by scholars like Sartori (2005), Magaloni (2006) and Greene (2007). Thus, here we identify the political system created by the AKP as a “fragile hegemonic party autocracy” in which the authoritarian incumbent could lose critical local offices and had to rely on a minor nationalist party to sustain its hegemonic position. Alternatively, the system in Turkey could be seen as an “indecisive state” instead of a clearly discernible and consolidated kind of an autocratic “regime” (Somer 2021). Although we do not want to delve into this discussion regarding the precise nature of the political system in Turkey, we do think that some conceptual generalizations (democracy, autocracy, electoral democracy, competitive authoritarianism etc.) are sine qua non for comparative macro-political analysis as long as researchers are aware of such concepts’ restricted analytical function and the irreducible gap between complexities of the empirical reality and the theoretical-conceptual generalizations.

38 See for example Baykan (2018), Gürsoy (2021), Yabanci (2016), Kaya (2019) and Yılmaz (2015).

39 Here it is important to note that “bureaucratic modes of authoritarianism” in the 20th century, as O’Donnell (1973) reveals in a now classic work, is anathema to populism.

Top of page


Electronic reference

Toygar Sinan Baykan and Murat Somer, Politics of Notables versus National Machine: Social, Political and State Transformations, Party Organizations and Clientelism during AKP GovernmentsEuropean Journal of Turkish Studies [Online], 34 | 2022, Online since 30 June 2023, connection on 15 July 2024. URL:; DOI:

Top of page

About the authors

Toygar Sinan Baykan

Kırklareli University

Murat Somer

Professor of Political Science and International Relations, Özyeğin University, İstanbul

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search