Party (de)institutionalization in times of political uncertainty: The case of the Justice and Development Party in Turkey

Abstract

Recently many polities around the world as different as Hungary, Turkey, Venezuela, Thailand suffer from autocratization. This has led to a growing scholarly interest in the process of autocratization. Yet, despite this emerging generation of studies on democratic setbacks, we still do not know much about the changing nature of party politics in the process of autocratization. We argue, in this article, that during autocratization, the incumbent party follows the path of internal and external party deinstitutionalization in response to the changing nature and intensity of political uncertainties. Using the case of the Justice and Development Party in Turkey, we address three questions: (1) How can the concept of party de-institutionalization be revised and used to understand party transformation during autocratization? (2) What explains party deinstitutionalization in transitional contexts? (3) What is the relationship between party de-institutionalization and autocratization? In doing so, this article increases our understanding of party transformation in transitional contexts and more specifically incumbent party change in the process of autocratization by providing a causal theory of party deinstitutionalization.

Keywords

autocratization party change party deinstitutionalization the Justice and Development Party Turkey electoral authoritarianism

Introduction

Recently many polities around the world as different as Hungary, Turkey, Venezuela, Thailand suffer from autocratization which is defined as “a process of regime change towards autocracy” (Cassani and Tomini, 2019). Coming to power through free and fair elections, populist leaders in these countries have begun to play with the democratic rules of the political game—impairing institutional checks and balances, eroding the freedom of media, violating political rights and freedoms, and intimidating the opposition forces (see Levitsky and Way, 2010). As such, move from democracies toward electoral democracies, electoral authoritarian regimes in the hands of populist leaders has become a salient global challenge affecting several countries (see Lührmann and Lindberg, 2019). This has led to a growing scholarly interest in the process of autocratization (see Cassani and Tomini, 2019; Diamond, 2015; Haggard and Kaufmann, 2016; Levitsky and Ziblatt, 2018). Yet, despite this emerging generation of studies on democratic setbacks, we still do not know much about the changing nature of party politics in the process of autocratization. Put differently, the puzzle remains as to what happens to an incumbent party which emerges in a democratic context and then becomes the major agent of autocratization. We argue, in this article, that the incumbent party changes significantly during the process of autocratization in order to respond the needs of new political regime. We find the concept of party de-institutionalization useful to explain these changes experienced by the incumbent party. Hitherto, the concept of party de-institutionalization has been mostly used to understand the evolution of far-right radical parties (see, Harmel et al., 2018). However, party de-institutionalization also accurately captures the dynamics of party transformation in the process of autocratization which changes the nature of political uncertainties forcing the incumbent party to restructure its organization and reform its relationship with voters.

In this article, we suggest an analytical understanding of the transformation of an incumbent party during autocratization by revisiting the concept of party de-institutionalization. Our contention is that rising political uncertainties in a changing regime context forces the dominant group within the party leadership to prioritize party de-institutionalization to hold on to power. More specifically, during autocratization, the incumbent party follows the path of internal and external party de-institutionalization in response to the changing nature and intensity of political uncertainties. Accordingly, we address three questions: (1) How can the concept of party de-institutionalization be revised and used to understand party transformation during autocratization? (2) What explains party de-institutionalization in transitional contexts? (3) What is the relationship between party de-institutionalization and autocratization? In doing so, this article increases our understanding of party transformation in transitional contexts and more specifically incumbent party change in the process of autocratization by providing a causal theory of party de-institutionalization.

The case of the Justice and Development Party (Adalet ve Kalkınma Partisi, AKP) in Turkey stands out as an important case for theorizing for the process of party (de)institutionalization during autocratization for several reasons. First, the party has remained in office uninterruptedly, which offers a natural control for the effects of loss of incumbency on party (de)institutionalization. Secondly, under the AKP rule, Turkey has experienced democratization followed by a sharp shift to democratic backsliding, which culminated in autocratization and the emergence of an electoral authoritarian regime (see, Esen and Gumuscu, 2016; Ozbudun, 2015). Such an evolving regime trajectory under the uninterrupted incumbency of one party makes Turkey an interesting case to study the effects of changing regime-level characteristics on party (de)institutionalization. Finally, we assert that previously proposed indicators of party (de)institutionalization need to be revised to account for mainstream parties. The AKP’s trajectory of party (de)institutionalization shows us ways to do so.

As follows, the first section provides a theory of party de-institutionalization in transitional contexts, which begins with developing the concept of party de-institutionalization, and then offers a detailed causal theory explaining why and under what circumstances party de-institutionalization might occur. In the following section, the dimensions of de-institutionalization are applied to the case of the AKP in Turkey, and the final section, using process tracing method, accounts for what explains de-institutionalization of the AKP in Turkey. Concluding part discusses the prospects of electoral authoritarian regime in Turkey with reference to the AKP’s decay.

A theory of party deinstitutionalization under changing regime context

The concept of party de-institutionalization: De-routinization and de-alignment

Party institutionalization remains one of the most contested concepts in comparative politics given that the literature provides many definitions, with different dimensions and indicators offered by various scholars (for a recent review, see Casal Bértoa, 2017; Harmel et al., 2018). Nevertheless, there is still a broad consensus on the minimal definition of party institutionalization as a process by which political parties acquire value and stability. Following this track, we also define party institutionalization as a process by which political parties reproduce stable and consistent patterns of intra-organizational structure and/or popular support and mass mobilization (Casal Bértoa, 2017: 9). While the internal dimension involves routinization, which entails increasing stabilization through reoccurring patterns of behavior based on formal/informal rules and procedures, the external dimension is related to the stabilization of the party-voter linkage, whereby the core electorate of the party maintains its support for the party.

We argue, however, that institutionalization is not a constant condition; rather, the dynamism of institutions—in our case political parties—also embodies process of de-institutionalization. Interestingly, the concept of party de-institutionalization remains weakly theorized, being mainly used for understanding the evolution of far-right marginal political parties (e.g. Harmel et al., 2018).

De-institutionalization of political parties can be conceptualized as the mirror image of the process of institutionalization (Table 1) referring to a process of deterioration in the organizational features that enhance the stability of organization and declining attachment of its core electorate to the party (Harmel et al., 2018).¹

Table 1.

Dimensions of party (de)institutionalization.

	Internal	External
Party Institutionalization	Routinization	Strong Party-Voter Linkage
Party De-Institutionalization	De-Routinization	Weak Party-Voter Linkage

Source: Author’s own compilation based on Harmel et al. (2018)

Harmel et al. (2018) conceptualize the internal dimension of de-institutionalization as de-routinization, which refers to a process in which established patterns of behavior inside the party are replaced by ad hoc decision-making (p. 113). Secondly, following Panebianco (1988), Harmel et al. (2018) suggest that another sign of de-routinization is the personalization of the party under a strong leader who violates the organization’s formal and informal rules (p. 113). Although personalization has various connotations (see Cross et al., 2018, among others), personalization of political parties includes an increasing emphasis on the party leader at the expense of the party. Externally, the signs of de-institutionalization are changing perceptions of other parties and destabilization of parties’ core electorates (Harmel et al., 2018: 113–114). From their empirical study on “non-mainstream parties,” Harmel et al. (2018) assert that a sign of party institutionalization is parties’ level of blackmail or coalition potential and other parties’ perceptions of that party. While this observation seems valid for cases of radical right and left parties portrayed as anti-establishment parties, for mainstream parties, the level of legitimacy and organizational relevance can be taken for granted. We, therefore, define external party de-institutionalization as de-alignment between the party and its core electorate. While sudden shifts in the electoral performance of political parties and increasing single party electoral volatility rates are important indicators of external party de-institutionalization, we suggest that de-alignment also entails increasing leadership-based voting at the expense of party identification, which indicates a weakening of the party-voter linkage. Table 2 summarizes our approach for analyzing de-institutionalization.

Table 2.

Dimensions and indicators of party de-institutionalization.

Party De-institutionalization	Dimensions	Symptoms	Indicators
Internal	De-routinization	Increasing instance of ad-hoc decision-making even in the presence of formal rules and procedures. Intensifying personalization and over-centralization which increases the prospects of ad-hoc decision-making inside the party.	Organizational changes that leads to further personalization. Instances of ad-hoc decision-making outside the rules, norms, and procedures.
External	De-alignment	Increasing voter attachment to the existing party leader at the expense of the party itself. Significant vote changes from election to election.	Survey-based research results that measure the rationale for voting among party’s core electorate. Campaign process and the role of party leader in campaigns.

Source: Authors’ own compilation based on Harmel et al. (2018).

A causal model for explaining party de-institutionalization

The question remains why an incumbent party who controls power de-institutionalizes. In other words, why would a party in government for more than a decade change and restructure its internal decision-making procedures and its relations with society even when it continues to attract significant amount of support? Harmel et al. (2018) offer some possible factors that may trigger (de)institutionalization such as the level of organizational complexity, availability of financial resources, sudden leadership changes and/or electoral decline (pp. 45, 114). Although these endogenous and exogenous factors might trigger de-institutionalization, our case shows significant stability in terms of organizational strength/complexity, intra-party cohesion, and electoral stability. Moreover, the AKP neither experienced a sudden leadership change caused by severe factional rivalry or death of its leader, nor a sudden disintegration of its core electorate. We suggest that although the party continues to attract significant electoral support, experiences no leadership shocks, owns its issue arena, and has strong organizational capabilities, it can still be forced to restructure itself and change its organization and relations with society. This happens in an effort to hold on power by centralizing power in the hands of party leader in response to the lack of predictability in a transitional context defined by “extraordinary uncertainty” (O’Donnell and Schmitter, 1986). Accordingly, we suggest that this process is not triggered by anticipated sources of deinstitutionalization in democratic regimes but rather by the fragile and uncertain conditions emerged in the process of autocratization.

The question of who initiates the deinstitutionalization is also central to comprehend the complex process. Instead of treating incumbent parties under autocratization as monolithic entities, our theory, building on Panebianco (1988), treats incumbent parties as organizations comprised of different tendencies which engages in organizational power struggles. Such an approach enables us to unpack the party, to delineate intra-party power struggles, and to understand how party leadership and other actors inside the party react to twin uncertainties inherent to the process of autocratization. In the process of autocratization, “politics of uncertainty” (Schedler, 2013) becomes the defining feature of political regime and inflicts pressure on incumbent party goals. Building on Lupu and Riedl (2013), uncertainty, here, is defined as the imprecision with which political actors can predict future interactions and developments (pp. 1341–1343). While uncertainty is inherent in both democratic and authoritarian regimes, the difference is that democratic elections have substantive uncertainty but provides procedural certainty (Przeworski, 1986). In authoritarian elections, on the other hand, procedural uncertainties are combined with substantive certainties in which “autocrats enjoy the prerogative of meddling with procedures and of determining outcomes in advance” (Schedler, 2013: 26). Although uncertainty is a constant feature of both democracies and authoritarian regimes, in transitional periods “extraordinary uncertainty” emerges which refers to the total lack of “the relative stability and predictability” (O’Donnell and Schmitter, 1986).

We argue that two uncertainties define the transitional politics of autocratization: horizontal and vertical (e.g. Svolik, 2012). Horizontal uncertainties refer to elite-level threats which intensify when prospects of intra-party and/or extra-party elite-level struggles increase and become more imminent. As autocratization deepens, the genuine opposition may begin to emerge within the party. Controlling intraparty politics and preventing the emergence of any alternative sources of power inside the party becomes a major priority for the aspiring autocrat. Extra-party horizontal uncertainties entail any elite-level threat outside the incumbent party such as threats from other state institutions, NGO’s, and business.

Vertical uncertainties include rising electoral uncertainty and/or risk of societal backlash. Electoral uncertainty refers to the growing prospect of opposition gains or victories in future elections. Previous election results may increase the incumbent party’s uncertainty regarding upcoming elections. The establishment of a new party or the increasing party systemic relevance of existing opposition parties can also increase electoral uncertainty for incumbents in times of transition in this case from defective democracy into electoral autocracy. When the regime transition completed, electoral uncertainty might decline as the playing field becomes uneven decreasing the opposition’s odds of winning. However, during the process of autocratization the incumbent must maintain its democratic-procedural legitimacy (Dukalskis and Gerschewski, 2017) and doing so requires minimizing repression/authoritarian electoral manipulation and maximizing its vote. Thus, as in democracies, the prospects of oppositional mobilization and electoral gain disrupt this delicate balance that incumbent needs to sustain between legitimation, cooptation, and repression (Gerschewski, 2013) during the transitional phase of autocratization. Moreover, electoral uncertainty here does not only refer to decline in votes but also increasing mobilizational capacity of opposition parties, their collaboration, or formation of new parties that have the potential to disrupt the balance between opposition and incumbent.

The prospect of a societal backlash constitutes another form of vertical uncertainty. Increasing social unrest in the form of protests challenge the incumbent party’s strength at three different levels. First, it reveals the degree of discontent from the incumbent which is an expected outcome in any regime. More importantly, it raises electoral and horizontal uncertainties. Electorally, if the protests are widespread, the credibility of incumbent declines, creating an electoral insecurity. Moreover, especially internationally supported widespread protests might strengthen intra-party opposition when some factions side with the movements and started to question the government’s way of dealing with protests. As such, these factors in our typology of uncertainties prevalent in regime transition might interact and mutually reinforce each other. This amplifies their impact on the decisions of the incumbent party leaders and elites around them to alter party structure (see Figure 1).

Figure 1.

Casual model for incumbent party de-institutionalization under autocratization. Source: Authors’ own construction.

Accordingly, incumbent parties in transitional contexts perform in an extraordinary uncertain political platform that forces certain factions within the party leadership to restructure the party by de-routinizing the decision making in party organization and de-aligning voter-party linkage. In other words, so as to hold on power in times of “disorder and indeterminacy” resulted from autocratization, a group of party elites might opt for (further) centralizing decision-making in the hands of party leader and changing party-society relations from strong voter-party identification to voter-party leader identification both of which lead to party de-institutionalization. As such, the incumbent party leader and faction aligned with him/her is the agent of autocratization while the party as an organization becomes the object of de-institutionalization.

The process of the AKP’s de-institutionalization: De-routinization and de-alignment

The AKP was established in 2001 following factional rivalry inside the pro-Islamic Virtue Party. Shortly after its establishment, the AKP won general elections to become the governing party. Meanwhile, the party began a process of institutionalization in a hostile domestic political environment due to strong skepticism among state institutions, civil society organizations, and the media regarding the party’s Islamist background. Several scholars suggested that the party attained a certain level of institutionalization in this domestic political context (Hale and Ozbudun, 2010; Kumbaracıbaşı, 2009; Lancester, 2014; Yardımcı-Geyikçi, 2015).

When tested against the classical indicators of party institutionalization, the AKP appears to have high levels of institutionalization. For example, the party’s electoral performance, membership strength, and party identification rates indicate strong external party institutionalization (see Tables 3, 4, and 5, respectively). Similarly, the AKP’s complex organizational structure, its local presence in all provinces and sub-provinces, membership density levels (Table 4), and stability of party finances (Table 6) all indicate strong internal institutionalization (for the AKP’s organizational features, see Kumbaracıbaşı, 2009; Hale and Ozbudun, 2010). Nevertheless, we contend that these figures are misleading for assessing the AKP’s level of institutionalization in the context of a changing regime trajectory from democratization to autocratization.

Table 3.

AKP’s number of votes, vote shares and single party volatility rates in general elections (2002–2018).

General Elections	2002	2007	2011	2015 (June)	2015 (November)	2018
Number of votes received	10,808,229	16,327,291	21,399,082	18,867,411	23,681,926	21,338,693
Vote share %	34.3	46.6	49.8	40.9	49.5	42.6
Volatility	—	6.2	1.6	4.5	4.3	3.5

Source: Turkish Statistical Institute (TUIK) www.tuik.gov.tr.

Table 4.

Membership strength and membership density figures of the AKP and the CHP (2007–2018).

Year	Parties	Numb of members	Party Votes*	Membership Density (M/Party Voters)	Membership Strength (M/Total Electorate)
2007	AKP	1,834,520	16,327,291(07)	11.2%	5.2%
2007	CHP	529,703	7,317,81(07)	7.2%	1.5%
2011	AKP	7,844,15	21,399,082(11)	36.7%	18.3%
2011	CHP	953,416	11,155,972(11)	8.2%	2.1%
2018	AKP	9,949,451	21,338,690(18)	46.6%	19.8%
2018	CHP	1,219,793	11,354,191(18)	10.7%	2.4%

* The numbers in parentheses are the years of general elections. Sources: Parties’ total number of voters and total number of voters were gathered from Turkish Statistical Institute (TUIK). Party membership figures for 2007 were gathered from Hürriyet (2007). For 2011 membership figures, see Hürriyet (2013a). For 2018 membership figures, see T24 (2018a).

Table 5.

Individual party identification figures.

Political Parties	Individual Party Identification (%)
AKP	61.2
CHP	68.7
MHP	67
BDP	63.8

Source: European Social Survey (ESS) Data Round (2008).

Table 6.

Percentage splits of party income.

Sources of Income (% of total income)	AKP (2004)	AKP (2006)	AKP (2011)	AKP (2016)
State Subsidies	85.2	89.8	85.3	83.7
Donations	8.1	3.2	5.2	14.4
Other	6.7	7.0	9.5	1.9
Total	100	100	100	100

Source: Justice and Development Party official website.

For instance, such high levels of membership density, party identification, and electoral stability should be scrutinized against the party’s politics of patronage and clientelism—both of which have become prevalent features of electoral politics in Turkey (see, Sayarı, 2014). Researchers argue that party-voter linkages based on material benefits weaken external party institutionalization by discouraging voters from forming emotional attachments to the party (see Panebianco, 1988). Clientelist party-voter linkages, vote buying, and patronage are all suggested as important factors explaining the AKP’s electoral dominance (see Ark-Yıldırım, 2017; Çarkoğlu and Aytaç, 2015; Gümüşcü, 2013).

Furthermore, the AKP has raised clientelism and patronage to new heights by establishing a complex network of socialization and benefit distribution that involves central government, local party organizations, municipalities, charitable foundations, religious brotherhoods, and highly active party activists (Ark-Yıldırım, 2017; Sayarı, 2014). Such a network has created a “habitus” that enables the party to socialize its voters into the party through regular meetings, benefit distributions, and daily problem solving in which party activists play a major role alongside religious brotherhoods and charitable foundations (Ark-Yıldırım, 2017). Thus, rather than building short-term materialistic party-voter relationships, the AKP has prioritized a long-term relationship to achieve continuous support through successive elections. The party’s membership has been an important mechanism to achieve an effective long-term relationship between the party and its voters.

As a result, strong party membership figures do not indicate party institutionalization because the AKP increasingly uses party membership to control and direct benefits to potential voters. Several studies also suggest that, rather than following a “catch-all” strategy, the party has selectively targeted voters with long-term party affiliation for benefit distribution (Çarkoğlu and Erdem, 2015). Moreover, the AKP voters have steadily become party members to receive benefits both during and between election campaigns. Therefore, high membership density, which normally indicates organizational strength, becomes invalid and unreliable as an indicator of party institutionalization.

In parallel with such a materialistic party-voter linkage, survey-based indicators of party identification also lose much of their validity. Several studies conclude that, in cases with high levels of clientelism and patronage, survey-based studies measuring party identification figures are prone to respondent bias regarding clientelism and patronage. This may lead to invalid inferences about external party institutionalization (Çarkoğlu and Erdem, 2015: 548). Similarly, the pervasiveness of clientelism and patronage, and the presence of an uneven electoral playing field in authoritarian contexts, significantly reduce the validity of single-party volatility as an indicator of party (de)institutionalization.

When tested against the classical indicators of party institutionalization, the AKP shows high levels of internal and external institutionalization. Nevertheless, when scrutinized against the changing regime context and the party’s cycle of dominance, these indicators lose their analytical utility to delineate the degree of its (de)institutionalization. We therefore suggest that the party’s trajectory of de-institutionalization can be more accurately observed by a deeper analysis of intensifying personalization and ad hoc decision-making (de-routinization), and a party-voter linkage characterized by increasing voter attachment to the party leader rather than the party (de-alignment). We argue that, since beginning its third term, the AKP has been de-institutionalizing itself, as manifested by increasing de-routinization and de-alignment.

De-routinization

The AKP has experienced an intensified process of over-centralization and personalization starting from its third term which increased the frequency of ad hoc decision-making inside the party. These two processes, we argue, are symptomatic of de-routinization.

Following Erdoğan’s election as president and resignation from the party in 2014, his external interference in the party’s internal decision-making processes has significantly intensified. For example, he started to chair cabinet meetings himself, which were previously headed by the prime minister (as de jure party leader) (Habertürk, 2015). Although this might indicate deinstitutionalization of parliamentarian system, this move certainly has lessened the authority of de-jure party leader as he lost his key role in the government which indirectly de-routinized the structure of party decreasing the power of party leader. Beside this, there were also several instances of de-routinization through Erdoğan’s external interference in the internal workings of the party organization, such as changing the nomination process of the Central Decision-Making and Executive Council (CDEC), changing rules regarding the intra-organizational hierarchy, and ad hoc changes to intra-party operations. All have further centralized power in the hands of Erdoğan.

A clash between the then de jure party leader Davutoğlu, and Erdoğan indicates an increasing prevalence of ad hoc decision-making inside the party. Prior to the Party Congress before the November 2015 elections, Erdoğan pressurized Davutoğlu to change the CDEC’s composition by packing it with Erdoğan loyalists (T24, 2015). The formal procedure for electing the CDEC members is through a secret ballot by the Party Congress (AK Parti Tüzüğü, 2019). However, despite this formal routinized procedure, due to the dominance of the party leadership over central and local party organizations, the informally routinized procedure required the approval of a list prepared by the party leadership through a procedural quasi-electoral process. The subsequent so-called “the CDEC crisis” between Erdoğan and Davutoğlu provides a perfect instance of deviation from formal to informal routinized intra-party procedures.

Following Erdoğan’s intervention, the CDEC—packed with his loyalists—took decision-making power over provincial and sub-provincial party organizations from the de jure party leader Davutoğlu (T24, 2016). This instance of ad hoc decision-making was made to increase the hold of de facto party leader Erdoğan over the de jure leader Davutoğlu. Similarly, after the June 2015 general elections, when the AKP lost its absolute parliamentary majority, Erdoğan’s reluctance to form a coalition with opposition parties, despite Davutoğlu’s enthusiasm, once again demonstrated Erdoğan’s power over the party organization and party leadership (Öniş, 2016).² Such instances of ad hoc decision-making, coupled with several other conflicts between Erdoğan and Davutoğlu, forced the latter to resign. He was replaced by a firm Erdoğan loyalist, Binali Yıldırım (Hurriyet, 2016).

The de-routinization indicated by increasing ad hoc decision-making and over-centralization was further intensified after Erdoğan was appointed as a party leader for the second time. Since then, through several amendments to the powers of the CDEC, the party has become much more personalized. For instance, the powers of the CDEC, which is dominated by Erdoğan, vis-à-vis local party organizations were significantly increased (Hürriyet, 2017). The party leadership also acquired sole authority regarding the application of the “three terms rule,” which was previously introduced as a routinized procedure to decrease intra-organizational personalization in the party (Hürriyet, 2017). Such over centralization and personalization significantly increased the prospects of ad hoc decision-making by a strong party leader.

Another symptom of ad hoc decision-making subject to the will of the leader was the forced resignation of various mayors and local party administrators. Following the aforementioned amendments, the party leadership announced a major purge of local party organizations and municipalities based on the claim that there was significant “metal fatigue” in local organizations (Haberturk, 2017).³ Following the announcement, several heads of provincial party organizations were forced to resign alongside mayors of key cities, including Istanbul and Ankara. All this occurred at the discretion of a certain faction within the party headed by the party leader (Reuters, 2017). During the selection of new candidates for local party organizations, the party leadership used its undisputable authority to personally interview local party staff (CNNTürk, 2018). In short, particularly after Erdoğan’s move to the presidency, the AKP operations as an organization have been significantly de-routinized through increasing instances of ad hoc decision-making accompanied and intensified by the personalization of the party under the sole control of Erdoğan and his loyalists.

De-alignment

The classic indicators of external party institutionalization, such as the party’s electoral performance and party identification, suggest that the AKP enjoys strong external institutionalization. Nevertheless, we argue that the AKP’s party-voter linkage has deteriorated due to stronger attachment to the party leader than the party as an organization. This has occurred on both the supply and demand sides of the linkage. By describing these processes, we can observe that the conditions of external party de-institutionalization have resulted from an alteration of the electoral mobilization prioritized by the party and its leader.

Supply-side of party-voter de-alignment

Regarding the supply-side of party identification, although the party leader has always been an important determinant of support for the AKP, this has not weakened party identification among the AKP voters. Instead, during the initial phase of institutionalization, a strong leadership encouraged strong identification with a party whose message was effectively delivered to the electorate by the party leader (Harmel and Svåsand, 1993). For instance, during first two terms, the party leadership made extensive investments on promoting its proposed party ideology framed as conservative democracy (see, Simsek, 2013).⁴ In fact, it was through the proposed ideology of conservative democracy that the party held together its ideologically eclectic voter base. During the AKP’s third term, however, this amplificatory function of the party leadership was transformed into a relationship in which the party organization became the transmitter of the leader’s message. In turn, this message started to identify the party in the image of its leader deemphasizing party ideology. Such a symbiosis between the leader and party identity has made it almost impossible to disentangle party identification from voter attachment to the party leadership.

For instance, during the June 2015 election campaign, a dual and unbalanced process emerged in which the party-level campaign proceeded in parallel with Erdoğan’s individual campaign as President. The party’s message thus became the message of the party leader, conceptualized as an empty signifier called the “cause”-dava.⁵ Such a message increasingly included nationalistic and statist tones, combined with an emphasis on service, which showed clear signs of detachment from the party’s initial identity of “conservative democracy” (Bashirov and Lancester, 2018). This new identity was made concrete by Erdoğan’s repeated use of his famous slogan: “One nation, one flag, one homeland, one state.” This was later incorporated into the party’s bylaws to become the party’s official principle (see Ak Parti Tüzüğü, 2019). The AKP elites also amplified this leadership-based identification at the expense of the party by Erdoğan’s statements and conflating the party’s identity with the party leader. For example, during the 2016 Party Congress, one prominent party figure and staunch Erdoğan ally, Bekir Bozdağ, stated that “as our people say, the AKP is Tayyip’s party…It is impossible to think of the party apart from our President and our President apart from the party” (Sözcü, 2016). Such a leadership-centered campaign strategy was coupled with an increasingly polarizing populist discourse initiated by the party leadership, as evident in the intensification of Erdoğan’s populist discourse (Figure 2). This further diluted the party brand and increased voter identification with the leadership at the expense of the party (on brand dilution of political parties, see Lupu, 2014). Increasing populist discourse of the party leadership during the campaign processes here stands as a proxy for dilution of party’s initial ideological brand indicating party investment on personal appeal of the leader at the expense of party ideology. So, populism here becomes a political strategy that a party leader uses in order to establish an unmediated personalized relationship with his/her audience independent of party identity (for a similar conceptualization of populism see, Weyland, 2001).

Figure 2.

Populist discourse of the AKP leader. Source. Hawkins et al. (2019). Note. 0 = non populist 1.5–2 = very populist.

Demand-side of party-voter de-alignment

Turning to the demand-side of party identification, starting from the AKP’s third term, voter attachment to the party leadership significantly increased at the expense of the party as an organization. Recent survey-based research conducted among the AKP electorate found that 71% of the AKP supporters stated that they would never give up voting for their party (Konda, 2018). While this figure indicates strong party identification, the same study also found that only 9% of respondents agreed that the AKP (as a party) should govern the country whereas 74% agreed that Erdoğan should govern (Konda, 2018). Compared to leadership-based voting during the 2011 elections, surveys found declining leadership-based voting in both the June and November 2015 elections, which was mainly due to Davutoğlu’s leadership at that time (Figure 3). Similar survey-based research indicates a 5% increase in leadership-based voting following Erdoğan’s return as a party leader (Figures 4 and 5). Overall, the significantly lower percentage of respondents that prioritized the party over the party leadership clearly demonstrates the AKP’s external de-institutionalization. Even in the 2014 local elections, when the party leadership’s role should be weaker, the AKP’s electorate strongly emphasized the party leader as the main reason why they supported the AKP.⁶ Another symptom of external party de-institutionalization was the vote difference between Erdoğan and the AKP. In the 2018 general elections, the AKP’s vote share was less than Erdoğan’s while the party continued to lose votes to its far-right ally, Nationalist Action Party (Milliyetci Hareket Partisi, MHP). This became so clear that, prior to the 2018 general elections, Erdoğan found it necessary to accuse voters planning to vote for him but not the party of hypocrisy.

Figure 3.

Rationale for voting among the AKP electorate. Source: Konda (2018).

Figure 4.

Rationale for voting among the AKP electorate. Sources: IPSOS: Post-Ballot Box Research (2015a, 2015b, 2018).

Figure 5.

Process tracing: AKP’s route to de-institutionalization. Source: Authors’ own compilation.

Accounting for agency-based de-institutionalization under autocratization: The case of AKP

The preceding analysis empirically shows that the AKP’s initial fragile institutionalization reverted to de-institutionalization during its third term in office. The question of why the party regressed in this way requires further explanation. Using the process tracing method, we contend that increasing and changing political uncertainties during the process of transition from defective democracy into electoral autocracy forced the party leadership and a group of elites aligned with the party leader to invest in party de-institutionalization instead of further stabilization and value infusion of the party as an organization.

The AKP’s route to de-institutionalization

Following its third landslide victory in the 2011 general elections, many scholars argued that the AKP had become a pre-dominant party by maintaining its electoral dominance for a third time while even increasing its votes (see Gümüşcü, 2013; Müftüler-Bac and Keyman, 2012; Sayarı, 2016). Apart from this electoral consolidation, the party also significantly enhanced its power vis-à-vis the secular state establishment, civil society, and the media. Consequently, by the 2011 general elections, the party had already alleviated the domestic political uncertainties that had defined the first two terms of its incumbency to achieve a “cycle of dominance” (Gümüşcü, 2013). During its third term, however, the “democratization” process initiated in the AKP’s first term was transformed into autocratization, which culminated in a electoral authoritarian regime (see, Esen and Gümüşcü, 2016; Somer, 2016, among others). Concomitant with this process, the types of uncertainties the party faced during its first two terms altered significantly. Now, confronted by horizontal and vertical uncertainties within the regime-level context of autocratization, the party leader and a faction aligned with the leader preferred to centralize control in the hands of Erdoğan himself at the expense of the AKP as an organization, thereby leading to de-institutionalization.

Firstly, in 2015, the party’s electoral dominance was significantly weakened. In the June 2015 general elections, the party lost its absolute parliamentary majority for the first time. Although it regained its lost votes in the second elections in November 2015 after terminating coalition talks, its electoral dominance reduced. Due to autocratization during the same period, the appearance of a regime-level cleavage (see Selçuk and Hekimci, 2020) decreased the ideological distance among opposition parties. This within-bloc de-polarization among the opposition and regime-level polarization (Somer, 2019) further increased the AKP’s electoral uncertainty. In addition to intensifying coordination among opposition parties, the party-systemic influence of the pro-Kurdish People’s Democracy Party- Halkların Demokrasi Partisi (HDP) compounded the electoral uncertainty. Such dynamics became more prominent following the constitutional referendum of 2017, which moved Turkey to an executive presidential system (Esen and Gümüşcü, 2017, 2018). The new majority voting system further triggered opposition coordination, forcing the AKP to ally with the far-right MHP. Finally, the close margin of the victory despite the biased electoral playing field, further intensified the AKP’s electoral uncertainty.⁷ The opposition’s efforts to increase societal mobilization to counter the government’s undemocratic measures and a societal backlash against government policies, particularly the Gezi Park protests (see Yardımcı-Geyikçi, 2014) and the Justice March-Adalet Yürüyüşü, also amplified the vertical threats against the regime and jeopardized the position of the incumbent party.

Alongside these vertical uncertainties, intra-party horizontal uncertainties against the AKP leadership snowballed, coming from the founding party members such as Ali Babacan, Abdullah Gül, and prominent party figures such as Ahmet Davutoğlu. In addition to intra-party horizontal uncertainties, extra-party horizontal uncertainties also intensified starting from the AKP’s third term. The most prominent of these was the clash between the Gülenist movement and the government which culminated into a failed coup attempt in 2016, planned by the latter group.

The AKP’s initial fragile institutionalization process was sharply reversed by these intensifying horizontal and vertical uncertainties (Figure 5). As to the intra-party horizontal uncertainties, some prominent party figures’ differing stances on prominent issues have indicated intra-party resistance to autocratization initiated by Erdoğan and his loyalists. For instance, during the Gezi Park protests compared to the repressive stance of Erdoğan and his close circle against the protesters, the then President Abdullah Gül took a more conciliatory stance emphasizing the importance of an open and inclusive society (Hürriyet, 2013b). Later the influence of Gül inside the party was rapidly eradicated through direct interference of Erdoğan (BBC Türkçe, 2013; Hürriyet, 2013b). Similarly, another prominent party figure Ali Babacan also clashed with Erdoğan and his close circle on economic policies and pushed for some economic policies that should counter the clientelist and patronage-based links between the party and pro-government business (T24, 2020). Intra-party clash mentioned already between the former party leader Davutoğlu and Erdoğan over intra-party matters and important points of dissensus on critical policy areas are also indicative of increasing horizontal uncertainty against the dynamics of the autocratization process within the party.

These figures and others inside the AKP-while cannot be portrayed as a monolithic faction against Erdoğan- had the potential to increase their influence in the party and to disrupt the autocratization process initiated by Erdoğan and his allies. Thus, such a presence of potential intra-party rivalry significantly increased the horizontal uncertainty for Erdoğan and his close circle. As a reactive response to such a rising uncertainty, Erdoğan and his close allies initiated a process of centralization and personalization, leading to de-routinization of the party which significantly increased the prospects of ad-hoc decision-making inside the party at the discretion of party leader. The party leadership’s reflexes of over-centralization and personalization have been further intensified by rising electoral uncertainty alongside increasing prospects of intra-party rivalry following the 2017 constitutional referendum. From this point on, many of the symptoms of internal de-institutionalization were justified by the party’s elites and leadership as a way to increase the flexibility of the party organization in its struggle against both elite and mass-level attacks (e.g. CNNTürk, 2017; Habertürk, 2017; Hürriyet, 2017).

This internal party de-institutionalization took place in tandem with external party de-institutionalization, as seen by increasing party elite-level investments in highly personalized election campaigns. Increasing vertical threats in the form of both electoral uncertainties and mass-level counter mobilization required significant ideological flexibility to maintain the party’s winning electoral margin under a majoritarian electoral system. In response to these uncertainties, and relying on Erdoğan’s lasting charisma, the AKP’s party leadership has preferred to personalize their campaigning by deprioritizing the party brand. This party-level investment in brand dilution has increased the party’s adaptive capacity against rising electoral uncertainties. For instance, against rising electoral uncertainty due to the HDP’s growing party-systemic influence, the AKP initiated a highly personalized campaign centered on polarizing nationalistic themes that enabled the party to keep its core electorate loyal to the party leader rather than the party (on the dynamics of polarization in Turkey see, Somer, 2019). Later such a radical shift to the right-end of the political spectrum with apparent policy shifts such as termination of Kurdish opening process is culminated into an alliance with the ultra-nationalist MHP. Counterfactually, such a rapid shift would not be possible for a party which has strong party-based ideological links with its core electorate. Thus, the prevalence of leadership-based linkage enabled the AKP to hold on to its core electorate following such a rapid ideological shift. However, the cost was the de-institutionalization of the AKP as a political party.

Conclusion

Our detailed analysis of the AKP’s trajectory in Turkey has demonstrated that an incumbent party can be significantly transformed in the process of autocratization in response to political uncertainties enhanced in transitional contexts. The party elites may opt for de-institutionalization both internally and externally at the expense of “the party as an organization” to hold on to power. In reaching this conclusion, we first revealed that the classic indicators of internal and external party institutionalization lose their validity in non-democratic contexts because they indicated that the AKP remains institutionalized. However, our detailed analysis and proposed indicators of internal and external party de-institutionalization empirically show that the AKP has in fact entered a process of de-institutionalization despite performing strong according to classic indicators of party institutionalization. This shows the urgent need to expand our conceptual framework and find more appropriate indicators of party (de)institutionalization in non-democratic contexts.

Using process tracing, we also explain why the AKP has de-institutionalized, arguing that this process is agency-driven, initiated by the party leadership and party elites to increase the party’s adaptive capacity within the context of rising political uncertainties, which have been intensified by autocratization of the regime. Our analysis also reveals the relationship between de-institutionalization and autocratization by demonstrating how they are mutually driving processes. That is, the more that the AKP has expedited the autocratization, the more political uncertainties increased. This in turn has forced the party leader and his close circle to prioritize holding onto its power over the party as an organization. This has led to further de-institutionalization of the party.

The literature on electoral authoritarian regimes posit that organizational strength of the incumbent party is an important condition for longevity of these regimes (e.g. Levitsky and Way, 2010). While being related, organizational strength is not identical with the process of party institutionalization (Harmel et al., 2018). The case of AKP demonstrates that while withholding its organizational strength a party can revert to an agency driven process of de-institutionalization. Such a process of de-institutionalization increases the adaptive capacity of the incumbent party within the transitional context of autocratization. Nevertheless, such a process of agency-driven de-institutionalization can easily revert into an organizational decay.

For the time being, by engaging in agency-driven de-institutionalization, we argue that, the AKP has experienced an organizational decay. The symptoms have become even more apparent following the opposition parties’ recent local election victories, when the AKP lost nearly all the major metropolitan municipalities (see Esen and Gumuscu, 2019). Two key figures of the party Ali Babacan and Ahmet Davutoglu have established new political parties targeting the AKP’s core electorate. Moreover, the loss of the major metropolitan municipalities, such as Istanbul and Ankara, has critical repercussions for the AKP’s politics of patronage and clientelism, which has been an important determinant of party membership and identification since the AKP’s third term in government. While the AKP may outlive its current party leader and re-institutionalize itself, the prospects of organizational decay remain high. Such party-level organizational decay will also have important implications for the longevity of Turkey’s electoral autocracy given that the literature on such regimes suggests that political parties are key organizational determinants of authoritarian endurance.

Footnotes

Acknowledgements

The authors would like to thank Sarah Birch, Susanne M. Klausen and two anonymous reviewers for their valuable feedback on earlier drafts of this article.

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Sebnem Yardimci-Geyikci

Hakan Yavuzyilmaz

Notes

References

Ark-Yıldırım

(2017) Political parties and grassroots clientelist strategies in urban Turkey: one neighborhood at a time. South European Society and Politics 22(4): 479–490.

Bashirov

Lancester

(2018) End of moderation: radicalization of AKP. Democratization 25(7): 1210–1230.

BBC Türkçe (2013) Gezi Parkı: Gül ve Erdoğan’dan farklı yorumlar [Gezi Park: different comments by Gül and Erdoğan]. BBC Türkçe. Available at: https://www.bbc.com/turkce/haberler/2013/06/130603_erdogan_gul (accessed 1 March 2020).

Çarkoğlu

Aytaç

(2015) Who gets targeted for vote-buying? Evidence from an augmented list experiment in Turkey. European Political Science Review 7(4): 547–566.

Casal Bértoa

(2017) Political parties or party systems? Assessing the “Myth” of institutionalization and democracy. West European Politics 40(2): 402–429.

Cassani

Tomini

(2019) Autocratization in Post-Cold War Regimes. London: Palgrave Macmillan.

CNNTurk (2017) AK Parti’de tüzük değişikliği: işte maddeler…[Constitutional change in AK Party: here are the articles]. CNNTurk. Available at: https://www.cnnturk.com/turkiye/ak-partide-tuzuk-degisikligi-iste-maddeler (accessed 1 March 2020).

CNNTurk (2018) AK Parti’de mülakatlar tamamlandı, 600 ismi Erdoğan belirleyecek [The interviews are finalized in AK Party, Erdoğan will designate 600 names]. CNNTurk. Available at: https://www.cnnturk.com/turkiye/ak-partide-mulakatlar-tamamlandi-600-ismi-erdogan-belirleyecek (accessed 15 January 2020).

Cross

Katz

Proyers

(eds) (2018) The Personalization of Politics and the Challenge for Political Parties. London: Rowman & Littlefield.

10.

Diamond

(2015) Facing up to the democratic recession. Journal of Democracy 26(1): 141–155.

11.

Dukalskis

Gerschewski

(2017) What autocracies says (and what citizens hear): proposing four mechanisms of authoritarian legitimation. Contemporary Politics 23(3): 251–268.

12.

Eligür

(2009) Turkey’s March 2009 elections. Turkish Studies 10(3): 469–496.

13.

Esen

Gümüşcü

(2016) Rising competitive authoritarianism in Turkey. Third World Quarterly 37(9): 59–73.

14.

Esen

Gümüşcü

(2017) A small yes for presidentialism: the Turkish constitutional referendum of April 2017. South European Society and Politics 22(3): 303–326.

15.

Esen

Gümüşcü

(2018) Perils of “Turkish presidentialism.” Review of Middle East Studies 52(1): 43–53.

16.

Esen

Gümüşcü

(2019) Killing competitive authoritarianism softly: the 2019 local elections in Turkey. South European Society and Politics 24(3): 317–342.

17.

ESS (2008) European Social Survey Round 4 Data: Norway: Data File Edition 4.0. Norwegian Social Science Data Services-Data Archive and distributor of ESS data.

18.

Gerschewski

(2013) The three pillars of stability: legitimation, repression, and cooptation in autocratic regimes. Democratization 20(1): 13–38.

19.

Gümüşcü

(2013) The emerging predominant system in Turkey. Government and Opposition 48(2): 223–244.

20.

Haberturk (2015) Cumhurbaşkanı Erdoğan’ın başkanlık ettiği toplantı Cumhurbaşkanlığı Sarayı’nda yaklaşık 8.5 saat sürdü [The meeting chaired by President Erdoğan took around 8.5 hours in the Presidential Palace]. Haberturk. Available at: https://www.haberturk.com/gundem/haber/1032261-iste-ilk-kare (accessed 15 January 2020).

21.

Haberturk (2017) Ak Parti’nin “Metal yorgunluğu” raporu Erdoğan’a sunulacak [AK Party’s “Metal fatigue” report will be presented to Erdoğan]. Haberturk. Available at: https://www.haberturk.com/gundem/haber/1593147-ak-parti-nin-metal-yorgunlugu-raporu-cumhurbaskani-recep-tayyip-erdogan-a-sunulacak (accessed 1 February 2020).

22.

Haggard

Kaufman

(2016) Dictators and Democrats: Masses, Elites, and Regime Change. Princeton, NJ: Princeton University Press.

23.

Hale

Özbudun

(2010) Islamism Democracy and Liberalism in Turkey: The Case of AKP. New York, NY: Routledge.

24.

Harmel

Svåsand

(1993) Party leadership and party institutionalization: three phases of development. West European Politics 16(2): 67–88.

25.

Harmel

Svåsand

Mjelde

(2018) Institutionalization (and De-institutionalization) of Right-Wing Protest Parties: The Progress Parties in Denmark and Norway. London: ECPR Press/Rowman & Littlefield.

26.

Hawkins

Rosario

Erin

, et al. (2019) Global Populism Database: Populism Dataset for Leaders 1.0. Available at: https://populism.byu.edu/Pages/Data (accessed 15 January 2020).

27.

Hurriyet (2006) Başbakan Davutoğlu istifasını sundu, Binalı Yıldırım görevi aldı [Prime Minister Davutoğlu submitted his resignation, Binali Yıldırım become Prime Minister]. Hurriyet. Available at: http://www.hurriyet.com.tr/gundem/basbakan-ahmet-davutoglu-istifasini-sundu-binali-yildirim-gorevi-aldi-40107477 (accessed 15 January 2020).

28.

Hurriyet (2007) Her 100 seçmenden 14ü bir parti üyesi [Every 14 out of 100 electorate are members of a party]. Hurriyet. Available at: http://www.hurriyet.com.tr/gundem/her-100-secmenden-14u-bir-parti-uyesi-5867722 (accessed 1 March 2020).

29.

Hurriyet (2013a) AKP üye sayısında fark attı [AKP outscored in terms of number of members]. Hurriyet. Available at: http://www.hurriyet.com.tr/gundem/ak-parti-uye-sayisinda-fark-atti-22588205 (accessed 12 February 2020).

30.

Hurriyet (2013b) Cumhurbaşkanı Gül’den Gezi Parkı açıklaması [A statement on Gezi Park by President Gül]. Hurriyet. Available at: https://www.hurriyet.com.tr/gundem/cumhurbaskani-gulden-gezi-parki-aciklamasi-23424869 (accessed 21 February 2020).

31.

Hurriyet (2016) Binali Yıldırım Başbakan olarak parti grubunda konuştu [Binali Yıldırım gave a speech as Prime Minister in party group]. Hurriyet. Available at: http://www.hurriyet.com.tr/gundem/binali-yildirim-basbakan-olarak-parti-grubunda-konustu-40108212 (accessed 1 March 2020).

32.

Hurriyet (2017) Ak Parti’de Erdoğan yeniden Genel Başkan [In the AK Party again Erdoğan is the leader]. Hurriyet. Available at: http://www.hurriyet.com.tr/gundem/son-dakika-ak-partide-kongre-gunu-40464627 (accessed 1 March 2020).

33.

Internethaber (2014) Bu da seçim sonrası anketi: Neden AK Parti? [This is a post-ballot box election survey: Why AK Party?]. Internethaber. Available at: https://www.internethaber.com/bu-da-secim-sonrasi-anketi-neden-ak-parti-655700h.htm (accessed 21 February 2020).

34.

Ipsos (2015a) 2015 Genel Seçim Sandık Sonrası Araştırması [Post-ballot box Research of 2015 General Elections]. Ipsos. Available at: http://www.arastirmakutuphanesi.com/wp-content/uploads/2015/06/Ipsos_SandikSonrasi_CNNTURK-web2.pdf (accessed 1 March 2020).

35.

Ipsos (2015b) 1 Kasım 2015 Genel Seçim Sandık Sonrası Araştırması [1 November 2015 General Elections Post-Ballot Box Research]. Ipsos. Available at: http://www.arastirmakutuphanesi.com/wp-content/uploads/2015/11/Ipsos_Sand%C4%B1kSonras%C4%B1_Bulgular_Gorseller_04112015_CNN_FinalDosya.pdf (accessed 1 March 2020).

36.

Ipsos (2018) 24 Haziran 2018 Cumhurbaşkanı Seçimi ve Milletvekili Seçimi sandık sonrası araştırması [Post-ballot box research of 24 June Presidential and Parliamentary elections]. Ipsos. Available at: https://www.ipsos.com/sites/default/files/ct/news/documents/2018-07/Ipsos-SandikSonrasiArastirmasi-CnnTurk-2Temmuz2018.pdf (accessed 21 February 2020).

37.

Konda (2018) Seçmen kümeleri-AKP [Sets of Electorate-AKP]. Konda. Available at: https://konda.com.tr/wpcontent/uploads/2018/05/KONDA_SecmenKumeleri_AkParti_Secmenleri_Mayis2018.pdf (accessed 12 August 2019).

38.

Kumbaracıbaşı

(2009) Turkish Politics and the Rise of the AKP: Dilemmas of Institutionalization and Leadership Strategy. New York, NY: Routledge.

39.

Lancester

(2014) The iron law of Erdogan: the decay from intra-party democracy to personalistic rule. Third World Quarterly 35(9): 1672–1690.

40.

Levitsky

Way

(2010) Competitive Authoritarianism: Hybrid Regimes After the Cold War. Cambridge: Cambridge University Press.

41.

Levitsky

Ziblatt

(2018) How Democracies Die. New York, NY: Broadway Books.

42.

Lührmann

Lindberg

(2019) A third wave of autocratization is here: what is new about it?. Democratization 26(7): 1095–1113.

43.

Lupu

(2014) Brand dilution and the breakdown of political parties in Latin America. World Politics 66(4): 561–602.

44.

Lupu

Riedl

(2013) Political parties and uncertainty in developing democracies. Comparative Political Studies 46(11): 1339–1365.

45.

Milliyet (2014) Erdoğan: Muhafazakar Demokrasi anlayışına önem atfediyoruz [Erdogan: We attach importance to conception of Conservative Democracy]. Milliyet. Available at: https://www.milliyet.com.tr/gundem/erdogan-muhafazakar-demokrasi-anlayisina-onem-atfediyoruz-5127721 (accessed 1 March 2020).

46.

Müftüler-Bac

Keyman

(2012) Turkey under the AKP: the era of dominant party politics. Journal of Democracy 23(1): 85–99.

47.

O’Donnell

Philippe

(1986) Transitions from Authoritarian Rule: Tentative Conclusions About Uncertain Democracies. Baltimore, MD: Johns Hopkins University Press.

48.

Öniş

(2016) Turkey’s two elections: the AKP comes back. Journal of Democracy 27(2): 141–154.

49.

Ozbudun

(2015) Turkey’s judiciary and the drift toward competitive authoritarianism. The International Spectator 50(2): 42–55.

50.

Panebianco

(1988) Political Parties: Organization and Power. Cambridge: Cambridge University Press.

51.

Parti

(2019) AK Parti Tüzük [AK Party Bylaw]. Available at: https://www.akparti.org.tr/media/279929/cep-boy.pdf (accessed 15 January 2020).

52.

Przeworski

(1986) Some problems in the study of the transition to democracy. In: O’Donnell

Philippe

Laurence

(eds) Transitions from Authoritarian Rule: Comparative Perspectives. Baltimore, MD: Johns Hopkins University Press.

53.

Reuters (2017) Turkish mayor says forced to quit in Erdogan purge of local government. Reuters. Available at: https://www.reuters.com/article/us-turkey-politics/turkish-mayor-says-forced-to-quit-in-erdogan-purge-of-local-government-idUSKBN1CZ21G (accessed 15 January 2020).

54.

Sabah (2017) Cumhurbaşkanı erdoğan: istifa davada sorun olarak görülmemeli [President Erdoğan: resignation should not be seen as a problem]. Sabah. Available at: https://www.sabah.com.tr/gundem/2017/10/05/erdogan-istifa-davada-sorun-olarak-gorulmemeli (accessed 15 January 2020).

55.

Sayarı

(2014) Interdisciplinary approaches to political clientelism in Turkey. Turkish Studies 15(4): 655–670.

56.

Sayarı

(2016) Back to a predominant party system: the November 2015 snap election in Turkey. South European Society and Politics 21(2): 263–280.

57.

Schedler

(2013) The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism. Oxford: Oxford University Press.

58.

Selçuk

Hekimci

(2020) The rise of the democracy–authoritarianism cleavage and opposition coordination in Turkey (2014–2019). Democratization. Epub ahead of print 10 August 2020. DOI: 10.1080/13510347.2020.1803841

59.

Şimşek

(2013) Conservative Democracy as a cosmetic image in Turkish politics: the semiology of AKP’s political identity. Turkish Studies 14(3): 434.

60.

Somer

(2016) Understanding Turkey’s democratic breakdown: old versus new and indigenous versus global authoritarianism. Southeast European and Black Sea Studies 16(4): 481–503.

61.

Somer

(2019) Turkey: the slippery slope from reformist to revolutionary polarization and democratic breakdown. Annals AAPSS 681(1): 42–61.

62.

Sozcu (2016) Bozdağ: AK Parti, Tayyip’in partisidir [Bozdağ: AK Party is Tayyip’s Party]. Available at: https://www.sozcu.com.tr/2016/gundem/bozdag-ak-parti-tayyipin-partisidir-1241001/ (accessed 1 March 2020).

63.

Svolik

(2012) The Politics of Authoritarian Rule. Cambridge: Cambridge University Press.

64.

T24 (2015) ‘Saray kadrosunun damga vuracağı AKP’nin MKYK listesinde Gül’ün ekibi tasviye edilecek’ [Within the AKP’s CDEC which will bear the mark of the Palace cadre, the Gül’s cadre will be discharged]. T24. Available at: https://t24.com.tr/haber/saray-kadrosunun-damga-vuracagi-akpnin-mkyk-listesinde-gulun-ekibi-tasfiye-edilecek,309415 (accessed 15 January 2020).

65.

T24 (2016) Davutoğlu’nu kopuşa götüren 20 Kriz [20 crisis that lead Davutoğlu to detachment]. T24. Available at: https://t24.com.tr/haber/davutoglunu-kopusa-goturen-20-kriz,339220 (accessed 15 January 2020).

66.

T24 (2018a) MHP’nin üye sayısında 4.5 ayda 14 bin 341 düşüş [A 14.341 decline in MHP’s number of members within 4.5 months]. T24. Available at: https://t24.com.tr/haber/mhpnin-uye-sayisinda-45-ayda-14-bin-341-dusus,567043 (accessed 15 January 2020).

67.

T24 (2018b) Eski AKP MKYK üyesi Osman Can: 2015’te CHP ile koalisyona çok yaklaştık, Erdoğan engelledi [Old AKP CDEC member Osman Can: In 2015 we were too close to a coalition with RPP, Erdoğan prevented it]. T24. Available at:https://t24.com.tr/haber/eski-akp-mkyk-uyesi-osman-can-2015te-chp-ile-koalisyona-cok-yaklastik-erdogan-engelledi,638029 (accessed 1 March 2020).

68.

T24 (2020) Ali Babacan Erdoğan’la yaşadığı ilk krizi açıkladı: Mali kurallara bağlı maliye politikalarını reddetti [Ali Babacan explained the first crisis he experienced with Erdoğan: He rejected fiscal policies that are bound to fiscal regulations]. T24. Available at: https://t24.com.tr/haber/ali-babacan-erdogan-la-yasadigi-ilk-krizi-acikladi-mali-kurallara-bagli-maliye-politikalarini-reddetti,884343 (accessed 15 January 2020).

69.

Weyland

(2001) Clarifying a contested concept: populism in the study of Latin American politics. Comparative Politics 3(1): 1–22.

70.

Yardımcı-Geyikçi

(2014) Gezi Park Protests in Turkey: A party politics view. Political Quarterly 85(4): 445–453.

71.

Yardımcı-Geyikçi

(2015) Party institutionalization and democratic consolidation: Turkey and Southern Europe in comparative perspective. Party Politics 21(4): 527–538.