Types of outcomes in factional rivalries: Lessons from non-democratic parties in Turkey

Abstract

In non-democratic parties, oppositional factions have difficulty making inroads to the top executive party organs. There are two consequences for these groups: party split or leadership removal. In the former case, the oppositional faction exits and establishes its own party. In the latter, the opposition succeeds in altering the balance of power by removing the leader and the party goes through change. This article suggests that the level of power concentration within the dominant faction matters for the type of outcome in factional rivalries. If the power is concentrated in the hands of a small elite, the divisions within the elite can help the oppositional faction remove the party leader. If the power resides only with a single leader, the oppositional faction is likely to lose the struggle against the dominant faction and decide to exit. This study explores the causal mechanisms involved by comparing six non-democratic parties from Turkey.

Keywords

Turkey political parties factionalism party split leadership removal intra-party democracy

Introduction

The conflicts between the dominant and the oppositional factions in a party are the reason for their existence. Such conflicts play a vital role in leading to party change or the establishment of new parties. While party change usually occurs when oppositional factions succeed in displacing the power of the dominant faction, a split from the party may take place when the opposition loses the factional fight. Yet, what determines the type of outcome in the case of such conflicts? In other words, under what circumstances do oppositional factions succeed in party change? Or when do they lose the fight, split and establish a new party? To answer these questions, our study explores a new causal mechanism—the level of power concentration within the dominant faction—to find whether it leads to different types of outcomes in factional conflicts.

There are extensive studies of factional conflicts in the literature of party change. The conditions under which an opposition can pave the way for party change (i.e. a modification of the party policy, removal of the party leader, an electoral reform or amendments in intra-party rules) have been understood in two ways. First, it has been argued that environmental challenges matter, among which are primarily electoral defeats, external shocks or changes in the nature of party competition (Harmel and Janda, 1994: 264–7; Panebianco, 1988: 250). Such environmental challenges relating to performance of a party may cause the party’s decision-makers to undertake a fundamental reevaluation of the party’s effectiveness (Harmel and Janda, 1994: 268; Janda, 1990). A second argument focuses more on internal factors and the role of agents: An oppositional faction, without having to wait for an external window of opportunity, may achieve reforms through changing the balance of power in the party (Harmel and Tan, 2003; Harmel et al., 1995; Hellmann, 2011: 466; Panebianco, 1988: 32). Strategies may vary from open collective mobilization against the dominant faction to implicitly amending the original rules that structure behavior. Whatever their strategy is, “change agents do not make their strategic decisions in a vacuum, but they take into consideration the fact that the context favors certain strategies over others” (Hellman, 2011: 473).

Building upon this point and utilizing Hirschman’s (1970) concepts of exit and voice, we distinguish non-democratic parties from democratic ones, since we assume different levels of power concentration in a party would shape the strategies of opposition differently. In democratic party structures, despite the unbalanced nature of the power relationship, the opposition has voice and can change the motivations of the dominant faction. Within this context, party change may occur at the ideological, organizational, policy or leadership level, depending on the kind of factionalism (Janda, 1980). Examples are the mechanisms of the Western European parties in which the balance of power is manifested in the proportion of votes that each faction receives (Ceron, 2012; Levy, 2004).

In non-democratic structures, power is concentrated to the greatest extent: The dominant faction is controlled by dictatorial leaders who exclusively own the decision-making processes, which leaves no voice—i.e. no effective voting mechanism—for oppositional factions to change the policy, strategy or leadership of the party. ‘Personalistic parties’ (Gunther and Diamond, 2003: 187) and parties dominated by a small and self-selected group of elites (Scarrow, 2005: 16) are examples, since candidate selection and policy formulation processes exclude the opposition. Co-existence with the dominant faction often becomes intolerable for the opposition, which may choose the exit option and establish its own party (Boucek, 2002: 466; Boucek, 2009: 474; Lucardie, 2000)—which we call party split. Party change is also possible in non-democratic structures, but it requires intensive resistance by the oppositional faction to alter the balance of power. We assume that party change in non-democratic structures can only take place through leadership removal. Figure 1 summarizes the potential outcomes.

Figure 1.

Alternative outcomes of factional rivalry*.

Our focus in this study is on non-democratic party structures. A detailed analysis of non-democratic party structures is useful since it is an undertheorized phenomenon and one can anticipate more distinctive factors to be at play than is suggested in the existing literature about the outcome of factional rivalries. We argue that the level of power concentration within the dominant faction is likely to matter for the type of outcome in non-democratic parties: When a small elite controls the party organization, factionalism is more likely to result in leadership removal because there is always a possibility of elite division. On the contrary, when a single leader controls the party organization, factionalism more likely results in party split because power is too concentrated for the opposition to change the internal balance.

This article is structured in the following way. In the first two sections, we review alternative explanations, external and internal to the parties, concerning the outcomes of factional fights, and then present our main focus—the level of power concentration within the dominant faction. Next, we justify our methodology and case selection criteria, which include three cases from Turkey of party split and three cases of leadership removal. Following a brief description of the cases, we provide comparative analysis of the findings. We conclude with some generalizations.

Existing explanations of the outcomes of factional conflicts

The literature on party change posits that the likelihood of an opposition faction winning may vary with electoral performance, intra-bloc competition in the party system, the type of factionalism and external events.

Electoral performance

Poor performance diminishes the legitimacy of the dominant faction in the party organization (Bille, 1997; Harmel and Tan, 2003: 420; Janda, 1990). Thus, an electoral defeat and relegation to an opposition role may put the opposition in an advantageous position in its struggle for altering the balance of power through leadership removal. On the contrary, if the electoral performance of the party is stable or in progress, the dominant faction would have a stronger authority in the party. It is then plausible to expect that the opposition would give up the fight and split from the party. In short, the degree of ‘poorness’ in performance matters:

H1a: The poorer the electoral performance, the more likely leadership removal.

H1b: The better the electoral performance, the more likely a party split.

We consider the electoral performance to be poorer if the party activists view the election as calamitous or disappointing. According to Harmel et al. (1995), a ‘calamitous election’ refers to the party’s negative performance. A ‘disappointing election,’ on the other hand, implies a moderate loss of seats and votes in a single election, by its rival’s superior showing in the election, or by loss of a leading role in government (Harmel et al., 1995: 27n). Thus, relying on the data provided by the Turkish Statistical Institute, we measure electoral performance primarily on the basis of the net change in the vote and seat shares of the party and its chief competitors in the latest two consecutive elections before the intensification of the factional rivalry. Finally, we trace the perceptions of party activists of the overall election results through journalists’ reports and newspaper articles.

Intra-bloc competition in the party system

Sartori (2005: 76) argues that the lesser the competition [between political parties], the higher the intra-party fractionalism. It is reasonable to expect that a high level of party competition motivates the parties to establish a united front, lessens the tension between the rival factions and eventually works for the benefit of the status-quo seeking dominant faction. New parties are then formed mostly because the expected benefits of entering the party system are considered to be higher than the expected costs (Boucek, 2009: 474; Lago and Martínez, 2011: 4; Lucardie, 2000: 183). In this regard, we suggest looking at intra-bloc volatility. During periods of high intra-bloc volatility, oppositional factions may expect electoral benefits—counting on the floating voters within the same ideological bloc. Hence:

H2a: The higher the intra-bloc competitiveness, the more likely a party split.

H2b: The lower the intra-bloc competitiveness, the more likely leadership removal.

Based on Bartolini and Mair (1990), we measure intra-bloc competition through the change in the distribution of the vote shares between the parties from the same ideological family in the latest two consecutive elections before the intensification of the factional rivalry.

Type of factionalism

In personalistic factions, activists cluster around the personality of a great leader, whereas in ideological factions activists share a programmatic value (Belloni and Beller, 1976: 544). We presume that in non-democratic parties, an opposition with a personalized character is more likely to lose members and fall under the consent of the dominant faction, as was the case with Greece’s PASOK (Pappas, 2009). Therefore, we expect an ideological opposition to be stronger in its fight vis-à-vis the dominant faction being able to mobilize higher number of party members, create a vigorous resistance and therefore remove the dominant faction’s leader. For instance, within Kim Dae-jung’s party in South Korea, internal agents showed an interest in organizational reform only when programmatic forces started to emerge in the party (Hellman, 2011: 477).

H3a: When factionalism is ideological, leadership removal is more likely.

H3b: When factionalism is personalist, a party split is more likely.

We review the content of the intra-party conflicts through newspaper articles and biographies to identify the type of factionalism in our cases.

External events

Depending on their type and timing, external events can work either for the benefit of the dominant faction or the opposition. We hypothesize that:

H4a: If external events delegitimize the dominant faction, leadership removal is more likely.

H4b: If external events enhance the dominant faction, a party split is more likely.

While the external event may be a specific happening that occurs at a particular time (or over a defined period) and that is publicly recorded (such as a constitutional reform, provision of public funding, a new emerging party), it may also come in the form of a ‘shock’ such as the Watergate Scandal or the collapse of the Berlin Wall (Harmel and Janda, 1994: 276). We trace newspaper articles and scholars’ analyses in order to identify the type of external events (if any) prior to or at the time of the factional rivalries.

The level of power concentration within the dominant faction

We acknowledge that the four alternative explanations reviewed above are essential in understanding the power alterations between the dominant and oppositional factions. But when it comes to explaining the outcomes of factional rivalries in non-democratic parties, the party politics literature is missing another potential variable, that is, the level of power concentration within the dominant faction. We argue that the type of outcome is dependent on whether it is one leader or multiple leaders that control the dominant faction in non-democratic parties:

H5a: When a single leader controls the party organization, factionalism more likely results in a party split.

H5b: When a small elite controls the party organization, factionalism more likely results in leadership removal.

Given that establishing a new party is more costly compared with remaining in the existing party, we assume that the opposition factions will first try to remove the leader. Yet, if the leader shares his power with the rest of the party elite, it is possible to expect elite divisions that may further weaken the power of the dominant faction. As Brownlee (2007: 2) puts it, authoritarian institutions are not “inherently fragile; they weaken when their leaders drive dissatisfied elites into the opposition’s ranks.” Likewise, Gandhi and Przeworski (2007: 1280) argue that the threat to the rule of autocrats come not only from outsiders but also from those that emerge from within the ruling elite itself. Visible splits within a power bloc indicate to change actors that political space may have been opened for autonomous organization (Przeworski, 1991: 108). The distinction between personalist and elitist institutions has not really been applied to factional rivalries, even though personalistic party and cadre or elite party categories are well established. Thus, in theorizing non-democratic party structures, we believe this hypothesis should be given due attention.

In order to understand the level of power concentration in our cases, we look at the formal and informal rules of the parties. Formal party rules such as statutes and internal regulations “tend to reflect the existing balance of power within the party as a political system, and hence any shifts in that balance are likely to be reflected, at least eventually, in discernible modifications in the rules” (Katz and Mair, 1992: 6). Yet, the party rules can be of informal kind as well, since the non-democratic party organization usually lacks a certain level of institutionalization or ‘systemness’ (Randall and Svåsand, 2002: 13). There may instead be unwritten conventions or informal rules concerning the party leader or other bodies of the party organization. We review the party statutes (formal rules) and newspaper articles (to determine the informal rules) for assessing the nature of authoritarianism in selected parties.

Case selection and methodology

The research on factional politics is increasingly dominated by quantitative cross-national studies. These studies mainly ‘predict’ the causes rather than explain the causal mechanisms. Hence, they provide limited information as to how a factor interacts with the other characteristics of a political system and affects the factional rivalries.

Our study aims at uncovering the role of a potential cause that explains the variation in an outcome and the mechanism through which it operates. Thus we believe a deeper analysis of a class of cases within a country may improve conceptual validity and enable us to explore the causal mechanisms (George and Bennett, 2005: 19-21). Such analysis may still retain comparative merit by relying on concepts which are applicable in the other countries (King et al., 1994: 43; Ragin, 1989: 34). In this way, it may also contribute to testing the theory in other contexts. As aptly put, “If we are to understand the rapidly changing social world, we will need toinclude information that cannot be easily quantified as well as that which can” (King et al., 1994: 5).

Six cases of factional rivalry—three representing party split and three representing leadership removal—are selected from Turkey for two reasons: First, the Turkish context is a suitable setting to study non-democratic parties. Scholarship on Turkish parties acknowledges that the internal decision-making mechanisms of parties exclude oppositional factions (Ayan, 2010; Özbudun, 2000: 83; Sayarı, 1976; Turan, 1988: 65). Second, the focus on one single country makes the analysis suitable for a so-called most similar systems design: Coming from the same political setting, selected parties share a number of similar characteristics—they are subject to the same political history, electoral rules, parliamentary system and a political culture in which military interventions are common to observe as a distinctive type of external event on the party system. Therefore, the puzzle becomes more interesting: Why do new parties emerge in some cases while in other similar cases, parties manage to survive factional infighting by removing the leader?

Case studies from Turkey

The origin of the Turkish party system lies in what Mardin (1973) calls the center-periphery cleavage. The cleavage expresses the long-lasting conflict between a ‘nationalist, centralist, laicist, cohesive state elite’ (the left) and a ‘culturally heterogenous, complex, and even hostile periphery with religious and anti-statist overtones’ (the right) (Kalaycıoğlu, 1994: 403). The established parties had to choose between these two positions in addition to their programmatic identity. This was especially the case when the military, which was a part of the laicist and nationalist center, intervened in politics at the moments of what it considered an internal threat against the Republic such as ‘separatism’ or ‘religious fundamentalism’ (Heper and Güney, 2000: 637).

Turkish parties organizationally display similar characteristics because the Turkish Laws on Political Parties of 1965 and 1983 imposed a more-or-less standard organizational model that consists of party conventions and executive committees at the local and national level (Özbudun, 2000: 83). The law adopted in 1983 during the interim military government further strengthened the oligarchic tendencies of the parties.

Cases of party split

MHP/MÇP-1993: MÇP (Milliyetci Çalışma Partisi/Nationalist Labor Party) was established in 1985 to replace MHP (Milliyetçi Hareket Partisi/Nationalist Action Party), which was closed down in the 1980 coup along with other parties. Factional rivalry started in 1970s between the pan-Turkist dominant faction and the oppositional pan-Islamists, which became more intense in 1992 when the pan-Turkists distanced themselves from the Islamist ideology (Çınar and Arıkan, 2002: 29). Pan-Islamists split and formed BBP (Büyük Birlik Partisi/Grand Unity Party) in 1993.

RP/FP-2001: FP (Fazilet Partisi/Virtue Party) was established in 1998 as the successor of RP (Refah Partisi/Welfare Party). It belonged to the pro-Islamist movement called National Outlook (Milli Görüş) aimed at forming an Islamic front together with other Muslim societies against the degeneration brought about by Westernization (Dikici Bilgin, 2008: 415). Conflict within the RP appeared after a pseudo-coup in 28 February 1997, which brought the end of the party. An emerging reformist group criticized the radical policies of the dominant faction, which clashed with the staunch secularists of the state. Shortly after the Constitutional Court’s decision to close FP, the reformists split and established AKP (Adalet ve Kalkınma Partisi/Justice and Development Party).

SP-2010: SP (Saadet Partisi/Felicity Party) was formed by the dominant faction within FP after the Constitutional Court closed down FP in 2001. SP gained a minuscule vote share in the first elections. An opposition faction was formed against the leader based on an egalitarian re-interpretation of Turkish Islamism with an emphasis on freedom and equality.¹ The party congress of SP in July 2010 turned into a fight; the faction split and established the HAS Party (Halkın Sesi Partisi/People’s Voice Party) in November 2010.

Cases of leadership removal

CHP-1972: CHP (Cumhuriyet Halk Partisi/Republican People’s Party) was the main opposition center-left party in parliament when conflict within the party appeared in 1971. The CHP leader allied with the armed forces, which intervened in civilian politics and desired to give the control to an ‘above-party’ government. The CHP leader supported the change. The secretary general of the party was, on the other hand, against establishing any ties with a government backed by the military (Nye, 1977: 214). The party divided along this line into two factions, which eventually led to the removal of the incumbent leader.

ANAP-1991: ANAP (Anavatan Partisi/Motherland Party) was a center-right party founded by Turgut Özal in 1983 and served in the government as a single party until 1991. Özal was the prime minister and the party leader until 1989, but had to leave both posts when he declared his candidacy for presidency. He left his position to Yıldırım Akbulut, who acted as Özal’s custodian. Akbulut was challenged by a new rising face in the party, Mesut Yılmaz who ran as a candidate for the leadership post in the party. Yılmaz eventually removed Akbulut and Özal’s influence from the party.

CHP-2010: CHP was led by Deniz Baykal from 1992 and articulated a center-left ideology as an opposition in parliament. In 2010, a videotape reflecting Baykal’s improper relationship with a member of the parliament leaked to the media, which ended with his resignation. What was left behind was a group of Baykal loyalists in the central executive board.² Yet the secretary general of the CHP deviated from this line and supported the candidacy of a high-ranking CHP parliamentarian for party leadership, who was exterior to the group of loyalists. This situation created a factional rivalry in the party and led to the removal of Baykal’s influence from party leadership.

A comparative analysis of the alternative explanations

Electoral performance

Our findings show a decline in electoral performance in two party split and all three leadership removal cases (Table 1). It is difficult to trace electoral performance in the MÇP/MHP-1993 case; the party could not run in 1983 elections as only three parties were sanctioned by the military regime to compete. MÇP barely got 2.93 per cent of the vote in the 1987 elections and its members ran on the RP list in 1991 elections.

Table 1.

Electoral performance of the parties before the factional rivalry.

	Net vote change %	Net seat change	Overall performance compared with the previous election
Party Split Cases
MÇP/MHP-1993	N/A	19	Indeterminate
RP/FP-2001	−5.97	−47	Worse
SP-2010	−0.15	0	Slightly worse
Leadership Removal Cases
CHP-1972	−1.38	9	Worse
ANAP-1991	−8.83	81	Worse
CHP-2010	1.47	−65	Worse

In terms of the party split cases, the influence of external events on the party system, namely military interventions, replace electoral considerations within the parties: Pro-Islamist RP lost its electoral support after its closure. Criticisms of the policies of the party leader intensified after 1997. FP, which was established to replace RP, experienced a decrease in its vote share, which was considered as ‘calamitous’ by the reformist faction.³ As for MÇP/MHP, we did not find any reference to electoral performance in our review of the factional rivalry. The nationalists gained 19 seats;⁴ however, it is difficult to decide whether this result may be seen as an electoral success, as the party had already lost prominent members such as Yaşar Okuyan and Namık Kemal Zeybek to other parties. Finally, with regards to the division within SP, the party did not get more than 3 per cent of the vote in either the 2002 or 2007 elections. Hence, findings within these cases on the effect of electoral performance on factional rivalry appear to be indeterminate.

In the leadership removal cases, all three parties performed poorly in the latest elections prior to the leadership removal. As for the CHP-1972 case, CHP had been in opposition since the 1950 elections vis-à-vis the center-right parties in government.⁵ After losing the 1965 elections, the party started discussing the causes of the disappointing election results more openly (Ayata and Ayata, 2007: 228). These discussions started criticism of the party leader, who had been in power for 33 years. As for the ANAP-1991 case, ANAP had also suffered from a decline of vote share prior to the removal of the party leader. In addition, its incumbency as the single party in government was being challenged by the rise of a new center-right rival in the party system, DYP; in the upcoming 1991 elections DYP was expecting to gain a significant number of votes, possibly surpassing ANAP votes. As for CHP-2010, even though the party had a higher vote share in 2007 compared with the 2002 elections, it lost 65 seats in the parliament, which was publicly interpreted as ‘calamitous’. CHP was criticized as lacking the ability to transform into a strong opposition party against AKP, which increased its vote share from 34.3 to 46.6 per cent in the same consecutive elections.⁶

Thus, our findings show that poor electoral performance does lead to the intensification of factional rivalries (except in the MÇP/MHP-1993 case), yet does not have a conclusive effect on the type of outcome.

Intra-bloc competitiveness in the party system

The overall analysis of the intra-bloc volatility for the time period of each case indicates that it was higher during the factional rivalry of the party split cases, but relatively lower during the leadership removal cases (Table 2).

Table 2.

Intra-bloc competition.

	Intra-bloc volatility*	Interpretation
Party Split Cases
MÇP/MHP-1993	16.2	The right bloc is more fragmented and more volatile than the left.
RP/FP-2001	11.8	Both blocs are fragmented and very volatile.
SP-2010	15.44	Consolidation of AKP, the right bloc is more volatile than the left.
Leadership removal cases
CHP-1972	4.2	The right bloc is more fragmented and more volatile than the left.
ANAP-1991	7.3	Both blocs are fragmented, moderate volatility.
CHP-2010	15.44	Consolidation of AKP, the right bloc is more volatile than the left.

Calculated by authors based on Bartolini and Mair (1990).

In all three cases of party split, the opposition factions not only left the party, but they preferred to establish a new party rather than joining an existing political group. In the case of MÇP-1993, BBP emerged; in the case of FP-2001, AKP emerged; and in the case of SP-2010, dissidents established the HAS Party. In two cases, MÇP-1993 and FP-2001, we observe that the voter preferences have been consistently changing towards far right parties since the early 1990s (Öniş, 1997). Thus the oppositional factions might have calculated the benefits of party split when they could not remove the leader. Yet, intra-bloc volatility came to a halt after the 2002 elections due to the consolidation of AKP power, thus it is difficult to argue that decision of the dissidents in the SP to establish a new party was grounded in such calculations. However, the party program of the HAS Party states that the party is based on equalitarian values which combine religious ethic and mercy with a leftist discourse.⁷ In this case, it seems the distinct ideological orientation of the dissidents seems to make it impossible to seek refuge in any existing party.

As far as the leadership removal cases concerned, the intra-bloc volatility was weak in all three cases. In the case of the CHP-1972, there is no evidence showing that party split was ever a part of the calculations of the opposition faction. On the contrary, the opposition leader presented an image of “changing Turkey through changing CHP.”⁸ In the case of ANAP-1991, the main interest of the opposition was to acquire party leadership rather than initiating a split from the party.⁹ In the case of CHP-2010, there is also no evidence showing that the opposition faction calculated the option of party split. The main motivation was to bring a new leader to the party, who would not fall under the influence of former leader Baykal.

Thus, even though intra-bloc volatility was higher in the cases of party split than in the cases of leadership removal as we hypothesized, it is only when split becomes an option for the opposition, that is, when leadership removal cannot be achieved, that intra-bloc volatility plays a role in the outcome. Rather than explaining our puzzle, the analysis of intra-bloc volatility led us to explore when leadership removal can or cannot be achieved.

Type of factionalism

We observe that in all cases of party split (MÇP/MHP-1993, RP/FP-2001, SP-2010) and one case of leadership removal (CHP-1972), the opposition factions had an ideological character. In contrast, in two of the three leadership removal cases (CHP-2010 and ANAP-1991), the disagreements between the dominant and opposition factions were personalistic and represented conflicts over the leadership post (Table 3).

Table 3.

Type of factionalism.

	Type of factionalism	Description
Party split cases
MÇP/MHP-1993	Ideological	Conflict between non-statist pan-Islamists and statist nationalists.
RP/FP-2001	Ideological	Conflict between pro-Islamists and neo-liberal Muslim reformists.
SP-2010	Ideological	Conflict between pro-Islamists and left-wing-oriented Muslim democrats.
Leadership removal cases
CHP-1972	Ideological	Conflict between the left-of-center (party of the people) and the center-left (party of the state).
ANAP-1991	Personalistic	Conflict between a new candidate running for leadership and the leader of the party.
CHP-2010	Personalistic	Conflict between the former leader and the secretary general of the party on the new leadership post.

In three cases of party split, the parties were in opposition and had weak capacity for pork barrel distribution. The criticisms of the opposition factions were largely grounded in disagreements over party policies and ideological divergence. This is most salient in the case of MÇP/MHP-1993. Evolution of the party ideology from communitarian nationalism towards Islam has become the main source of conflict within the party since early 1970s (Çınar and Arıkan, 2002: 28). The pan-Islamist opposition was strong in the local youth organizations, yet they were underrepresented on the party’s executive board (Yanardağ, 2002: 195). The military intervention in 1980 intensified the ideological orientation of the pan-Islamist faction: The dominant nationalist faction intended to maintain its alliance with the military and thus treated the pan-Islamist faction harshly (Çınar and Arıkan, 2002: 29). Within FP, although the party’s radical pro-Islamist discourse moderated from the 1970s to the 1990s, the reformists wanted to form a more center-right party that could face the global neo-liberal challenges (Tezcür, 2010: 18). In the case of SP, which continued with the pro-Islamist ideology, the opposition faction moved towards the left of the political spectrum upholding Islamic egalitarian values. Thus all cases prove to be more of ideological rather than personalistic.

Among the leadership removal cases, it is observed that only in CHP-1972 did the opposition faction have an ideological character. The discourses of the 1967, 1968 and 1970 party congresses show that the opposition leader was already a strong advocate of the left-of-center ideology.¹⁰ He opposed the dominant faction’s policies of supporting the military government, stating that ‘left-of-the-center means that the CHP should be the party of the people, not the party of the state’.¹¹ In the case of ANAP-1991, there is not much evidence that Yılmaz’s opposition faction had an ideological character. Yılmaz had stated his motivation to run for party leadership in a way that did not include any ideological interest.¹² In the case of CHP-2010, there is also no evidence that the faction had an ideological character. The conflict emerged between the loyalists of the former leader and an emerging opposition over the question of who would be the next leader.

To sum up, we find that the ideological type of factionalism does cause the intensification of factional fights but the consequence is quite the contrary to our hypothesis (H3b): None of the party split cases were marred by personalistic conflicts.¹³ Therefore, this variable fails to explain why leaders cannot be removed in these cases.

External events

We find that external events do play an essential role in factional rivalries: they may generate the rivalry (as in the cases of CHP-1972, ANAP-1991, CHP-2010); change the type of factionalism (FP-2001); or intensify already existing divisions (MÇP-1993). Table 4 summarizes the external events and their impact on our cases.

Table 4.

The role of external events in factional rivalries.

	Type of external event	Description	Did the external event delegitimize the dominant faction?
Party split cases
MÇP/MHP-1993	1980Military coup	Intensified the already existing ideological division within MÇP/MHP	Yes
RP/FP-2001	1997Military intervention	Changed the type of factionalism—i.e. from personalistic to ideological	Yes
SP-2010	None	−	−
Leadership removal cases
CHP-1972	1971Military intervention	Generated the factional rivalry	Yes
ANAP-1991	1989Upcoming presidential elections, the party leader’s candidacy to presidency and his resignation	Generated the factional rivalry	No
CHP-2010	2010Video-tape scandal based on the party leader’s personal life	Generated the factional rivalry	Yes

During military interventions, parties usually find themselves with the choice of either allying with or resisting the armed forces, which in turn triggers divisions in the party. In two cases of party split (MÇP/MHP-1993, FP-2001) and in one case of leadership removal (CHP-1972), the dominant faction leader decided to ally with the military whereas the opposition wanted to resist. Thus in all three cases, the dominant faction lost its legitimacy in the eyes of the opposition. Among these, we notice that the 1980 coup had a quite long-lasting effect on the MÇP/MHP case. The cruelty of the coup intensified the already existing ideological division between the statists and anti-statists. The intervention in 1971 generated the ideological factional rivalry in CHP-1972, which was based on whether the CHP should be a party of the masses (oppositional faction) or a party of the state (leader’s faction). On the other hand, the military intervention in 1997 (known as the February 28 process) changed the type of division within RP/FP. While the previous rivalries primarily centered around personalistic factions within RP/FP, this time the young generation in FP did not want to be punished by the state and adopted a neo-liberal, pro-European stance as opposed to the dominant faction’s traditionalist and anti-Western attitude (Tezcür, 2010: 18).

In the case of the CHP-2010, the external event came in the form of a videotape scandal, which started an open criticism of Baykal’s policies in the party.¹⁴ In the case of the ANAP-1991, the external event was the upcoming presidential election of 1989, which led to the resignation of Turgut Özal as the party leader to run as a candidate for presidency. Özal’s departure left a huge power vacuum in the party, leading to the rise of young charismatic candidates such as Mesut Yılmaz who could challenge the authority of the former leader’s heir, Yıldırım Akbulut.

Thus, our findings do not support our hypotheses (H4a and H4b). The external events delegitimized the dominant faction in two cases of party split and two cases of leadership removal, but do not really shed light on understanding the type of outcome of factional rivalries.

The power concentration level in the dominant faction

We finally shift our attention to assessing the validity of our main variable, i.e. the level of power concentration in the dominant faction. Our main hypothesis (H5a, H5b) finds empirical support in our cases, with the exception of the ANAP-1991 case (See Table 5).

Table 5.

Power concentration level in the parties.

	Party leader	Opposition leader(s)	Is the power shared between the leader and other party elites?
Party split cases
MÇP/MHP-1993	Türkeş	Yazıcıoğlu	No
RP/FP-2001	Kutan (Erbakan)*	Erdoğan and Gül	No
SP-2010	Kurtulmuş (Erbakan)*	Kurtulmuş	No
Leadership removal cases
CHP-1972	İnönü	Ecevit	Yes
ANAP-1991	Akbulut (Özal)*	Yılmaz	No
CHP-2010	None (Baykal)*	Sav	Yes

In these cases, the preferences of the former party leaders in parentheses constitute the informal rules of the party. Their choices have superiority over the formal party statutes.

In party split cases, it is observed that the party leader was dominant and had sole control over the party organization. The MÇP/MHP had the most hierarchical structure of all, since Türkeş was declared as the unchallengeable and unquestionable, almost mythical party leader at the 1969 Adana Congress (Çınar and Arıkan, 2002: 27). The party was ruled by a group composed of the presidency council, the general executive boards and the secretary-general and his/her assistants: all swore loyalty to the leader (Yanardağ, 2002: 291). Furthermore, according to Article 30 of the statute, the party leader is the leader of all branches other than the discipline committee, and Article 32 does not give substantial power to the secretary-general in the administration of the party organization.¹⁵ Given that the pan-Islamist opposition could not win the leadership race under these circumstances, the opposition leader Yazıcıoğlu, together with five deputies split from the party and formed BBP.

In the RP/FP structure, it is possible to see a similar hierarchy. Erbakan, starting from the 1960s to the late 1990s, developed an extensive clientelistic network in the party, not only by offering its members material benefits such as fuel, food, and various commodities but also creating a personal atmosphere of closeness, affection, and companionship (Ayata, 1996: 52). The increased electoral support for RP in 1995 fostered Erbakan’s confidence and authority. We could not access the party statute of the FP; however, the media indicates the reformists were very critical of the statute in terms of intra-party democracy.¹⁶ In 2000, Abdullah Gül narrowly lost a bid to replace Erbakan’s loyalist, Recai Kutan, as FP’s next leader and the closure of the party in 2001 led the reformists to establish their own party.

On the other side, the party statute of the SP reveals the sole authority of the leader within the party organization.¹⁷ Even though Recai Kutan was far from being a charismatic leader, his strength was derived from Erbakan’s full support of his leadership, which constituted the informal rules in both the FP and the SP cases. Later on, Kurtulmuş gained the support of Erbakan as well and was elected to party leadership in 2008. Yet, soon after, a conflict developed between him and the followers of Erbakan. Signatures of a majority of the delegates asked for renewal of the congress and the sacking of Kurtulmuş. The refusal ended up in court and the court ruled that party administration should be transferred from his leadership to the board of trustees on which Erbakan’s influence was again paramount. Thus, Kurtulmuş split from the party with his followers in 2010.

In leadership removal cases, the CHP-1972 case shows that, despite the leader İnönü’s personalistic style, the formal party rules restricted the leader’s power. According to the party statutes, the secretary-general was the head of the central executive board and was responsible for maintaining communication with the whole party organization.¹⁸ Besides, the central executive board was responsible for “organizing the party, developing and controlling the organization.”¹⁹ The party leader, on the other hand, was only the representative of the party in external affairs and the main director of the party’s policy and program.²⁰ Thus, when the majority of the party congress in 1972 supported Ecevit vis-à-vis İnönü, it eventually led to the resignation of İnönü from his post.

The CHP-2010 case shows that Baykal pursued a strongly personalistic leadership style in candidate selection, programmatic or ideological debates (Ayan, 2010; Ayata and Ayata, 2007). CHP statutes granted a high degree of power to the party leader, stating that ‘party leader is the head of the whole party organization’ and ‘the central executive board’.²¹ Yet, what is extraordinary in the case of CHP-2010 was that even though the party rules gave sole authority to the leader, the videotape scandal that led to Baykal’s resignation from the party shifted the power balance in favor of the secretary general, Önder Sav. The event showed that Sav had certain influence over the party organization, although the position of the secretary-general had formally been made subordinate to that of the leader in 2007.²² Sav’s influence was visible when the party organization sided with his decision to support the candidacy of a parliamentarian, Kılıçdaroğlu, to the leadership post, without waiting for Baykal’s preferences after he resigned. Sav apparently withdrew his support from Baykal when in an interview he said, ‘From now on, we have politically separate ways with Baykal’.²³ Thus, the videotape event revealed that power was not that concentrated in the leader. It was through this scandal that Baykal’s policies could, for the first time, be criticized in the party: such criticisms were highly restricted in the party before his resignation.²⁴

In the case of ANAP-1991, the party founder Özal resigned from the party due to his candidacy for presidential elections and eventually became president in 1989. Behind the scenes, he was still the leader, whose preferences constituted the informal party rules for ANAP.²⁵ In 1989, he had picked Akbulut as his successor. Yet, when Yılmaz challenged Akbulut in the party convention in 1991, Özal seemed to take a neutral position, as the president of the country legally should.²⁶ An ANAP parliamentarian commented that ‘the neutrality of Özal with regard to who should be the new leader signals that he does not support Akbulut anymore’.²⁷ In other words, ANAP-1991 constitutes a deviant case in which a leader with sole authority willingly accepted the removal of his influence over the party by keeping neutral, since his power was restricted by his new institutional identity.

To sum up, the level of power concentration has primary influence in determining the type of outcome in factional rivalries. Where multiple leaders control the organization, it is more likely that a division among them weakens their power. Our cases show that where some elite members side with the opposition vis-à-vis the party leader, leadership removal becomes possible (CHP-1972 and CHP-2010). Where the leader has sole control over the party, it is less likely that the opposition can remove the leader (MÇP-1993, FP-2001, SP-2010). Even though one of the leadership removal cases (ANAP-1991) does not meet our expectations, we believe this case does not significantly challenge our main argument. After all, what we observe here—if not a dispersion of power at the elite level—is another type of limitation on the power of the single leader, originating from his new institutional affiliation as the president of the country.

Concluding remarks

The more that power is concentrated in a party, the harder it can be to change. In democratic parties, power is concentrated to a lesser degree than in non-democratic ones. The change agents can more easily be mobilized and justifiably demand an alteration in the leadership, policy or ideology of the party in the event of an electoral defeat, external event or a transformation of the intra-bloc competition. This study showed that such factors do not play a decisive role in such change in non-democratic parties, because the dominant faction does not tolerate the opposition’s demands. We argued that in the case of a factional rivalry, there are two possible types of outcomes for non-democratic parties: leadership removal or party split. The comparison of the six cases in Turkey showed that, with the exception of one deviant case (ANAP-1991), the type of outcome is best explained by the level of power concentration in the dominant faction. If the power is concentrated in the hands of a small elite, the divisions within the elite can help the oppositional faction remove the party leader (CHP-1972; CHP-2010). If the power resides only with a single leader, the oppositional faction is likely to lose and decide to exit (MÇP-1993; FP-2001; and SP-2010).

In addition, in each of the six cases, we observe that the opposition faction first aims at removing the party leader. Only when this fails does it follow the party split path. This signals that factional rivalry goes through certain stages, which we delineate in Figure 2. In the first stage, external events (i.e. military interventions, presidential elections as observed in Turkish cases) can generate the factional rivalries. In the second stage, poor electoral performance and types of factionalism again triggered by some external events might intensify the factional rivalries. Yet it is the power concentration in the dominant faction that determines the type of outcome. If the opposition faction fails to remove the party leader, it calculates the costs and benefits of establishing its own party or joining another party by looking at intra-bloc volatility.

Figure 2.

Causal mechanisms contributing to the outcome of factional rivalry.

Case selection based on one single country helped us to control for macro-level factors such as the government and electoral system, political culture and history of party development. However, we also restricted the generalizability of our claims by arguing only on the basis of one context. For instance, oppositional factions may prefer the option of party split under different circumstances in two-party or dominant party systems, since these systems do not provide the same opportunities for small parties as in multi-party systems. Similarly, we are not sure how our claims would stand up in countries where military interventions are not common external events. However, we believe explaining the outcomes of factional rivalries would in any political setting require us to distinguish different levels of power concentration within parties, i.e. democratic, non-democratic and the ones in which the dominant faction is led by a single leader or multiple leaders.

Footnotes

Acknowledgements

An earlier version of this study was presented at the 70th Annual Convention of the Midwest Political Science Association, 12–15 April 2012, Chicago, IL, USA. The authors thank Kenneth Janda and the three anonymous reviewers for their helpful comments.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biographies

Pelin Ayan Musil is a faculty member at the School of International Relations and Diplomacy at Anglo-American University and a research fellow at Jan Evangelista Purkyne University in Czech Republic. Her research deals with democratization, political parties, Turkish politics and EU enlargement.

Hasret Dikici Bilgin is an Assistant Professor of International Relations at Okan University in Istanbul, Turkey. Her main research interests are political elites, political parties and elections, and foreign policy analysis.

References

Ayan

Pelin

(2010) Authoritarian Party Structures in Turkey: A Comparison of the Republican People’s Party and the Justice and Development Party. Turkish Studies 11: 197–215.

Ayata

Sencer

(1996) Patronage, Party and State: the Politicization of Islam in Turkey. Middle East Journal 50: 40–56.

Ayata

Sencer

Ayata

Ayşe Güneş

(2007) The Center-Left Parties in Turkey. Turkish Studies 8: 211–232.

Bartolini

Stefano

Mair

Peter

(1990) Identity, Competition, and Electoral Availability: The Stabilisation of European Electorates 1885–1985. Cambridge: Cambridge University Press.

Belloni

Frank P

Beller

Dennis C

(1976) The Study of Party Factions as Competitive Political Organizations. The Western Political Quarterly 29: 531–554.

Bille

Lars

(1997) Leadership Change and Party Change: The Case of the Danish Social Democratic Party, 1960–95. Party Politics 3: 379–390.

Boucek

Françoise

(2002) The Structure and Dynamics of Intra-Party Politics in Europe. Perspectives on European Politics and Society 3: 453–493.

Boucek

Françoise

(2009) Rethinking Factionalism: Typologies, Intra-Party Dynamics and Three Faces of Factionalism. Party Politics 15: 45–485.

Brownlee

Jason

(2007) Authoritarianism in an Age of Democratization. Cambridge: Cambridge University Press.

10.

Ceron

Andrea

(2012) Bounded Oligarchy: How and When Factions Constrain Leaders in Party Position-taking. Electoral Studies 31: 689–701.

11.

Çınar

Alev

Arıkan

Burak

(2002) The Nationalist Action Party: representing the state, the nation or the nationalists? Turkish Studies 3: 25–40.

12.

Dikici Bilgin

Hasret

(2008) Foreign Policy Orientation of Turkey’s Pro-Islamist Parties: A Comparative Study of the AKP and Refah. Turkish Studies 9: 407–421.

13.

Gandhi

Jennifer

Przeworski

Adam

(2007) Authoritarian Institutions and the Survival of Autocrats. Comparative Political Studies 40: 1279–1301

14.

George

Alexander

Bennett

Andrew

(2005) Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press.

15.

Gunther

Richard

Diamond

Larry

(2003) Species of Political Parties: A New Typology. Party Politics 9: 167-199.

16.

Harmel

Robert

Janda

Kenneth

(1994) An Integrated Theory of Party Goals and Party Change. Journal of Theoretical Politics 6: 259–287.

17.

Harmel

Robert

Tan

Alexander

(2003) Party Actors and Party Change: Does Factional Dominance Matter? European Journal of Political Research 42: 409–424.

18.

Harmel

Robert

Heo

Tan

Alexander

Janda

Kenneth

(1995) Performance, Leadership, Factions and Party Change: An Empirical Analysis. West European Politics 18: 1–33.

19.

Hellmann

Olli

(2011) A Historical Institutionalist Approach to Political Party Organization: The Case of South Korea. Government and Opposition 46: 464–484.

20.

Heper

Metin

Aylin

Güney

(2000) The Military and the Consolidation of Democracy: The Recent Turkish Experience. Armed Forces and Society 26: 635–657.

21.

Hirschman

Albert O

(1970) Exit, Voice and Loyalty: Responses of Decline in Firms, Organizations and States. Cambridge: Harvard University Press.

22.

Janda

Kenneth

(1980) Political Parties: A Cross National Survey. New York: The Free Press.

23.

Janda

Kenneth

(1990) Toward a Performance Theory of Change in Political Parties. Paper for 12th World Congress of the International Sociological Association. Madrid, 9–13 July.

24.

Kalaycıoğlu

Ersin

(1994) Elections and Party Preferences in Turkey: Changes and Continuities in the 1990s. Comparative Political Studies 27: 402–424.

25.

Katz

Richard S

Mair

Peter

(1992) Introduction: The Cross-National Study of Party Organizations. In Katz

Richard S

Mair

Peter

(eds) Party Organizations: A Data Handbook. London: Sage, 1–20.

26.

King

Gary

Keohane

Robert

Verba

Sidney

(1994) Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton: Princeton University Press.

27.

Lago

Ignacio

Martínez

Ferran

(2011) Why new parties? Party Politics 17: 3–20.

28.

Levy

Gilat

(2004) A Model of Political Parties. Journal of Economic Theory 115: 250–277.

29.

Lucardie

Paul

(2000) Emergence of New Parties Prophets, Purifiers and Prolocutors: Towards a Theory on the Emergence of New Parties. Party Politics 6: 175–185.

30.

Mardin

Şerif

(1973) Center-Periphery Relations: A key to Turkish Politics? Daedalus, vol. 102, pp. 169–190.

31.

Nye

Roger P

(1977) Civil-Military Confrontation in Turkey: The 1973 Presidential Election. International Journal of Middle East Studies 8: 209–228.

32.

Öniş

Ziya

(1997) The Political Economy of Islamic Resurgence in Turkey: The Rise of the Welfare Party in Perspective. Third World Quarterly 18: 743–766.

33.

Özbudun

Ergun

(2000) Contemporary Turkish Politics: Challenges to Democratic Consolidation. London: Lynne Rienner Publishers.

34.

Pappas

Takis S

(2009) Patrons Against Partisans: The Politics of Patronage in Mass Ideological Parties. Party Politics 15: 315.

35.

Panebianco

Angelo

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

36.

Przeworski

Adam

(1991) Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press.

37.

Ragin

Charles C

(1989) The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies. Berkeley and Los Angeles: University of California Press.

38.

Randall

Vicky

Svåsand

Lars

(2002) Party Institutionalization in New Democracies. Party Politics 8: 5–29.

39.

Sartori

Giovanni

(2005) Parties and Party Systems: A Framework for Analysis. ECPR Press Classics.

40.

Sayarı

Sabri

(1976) Aspects of Party Organization in Turkey. The Middle East Journal 30: 187–199.

41.

Scarrow

Susan

(2005) Implementing Intra-Party Democracy. New York: NDI.

42.

Tezcür

Güneş M

. (2010) The Paradox of Moderation: Muslim Reformers in Iran and Turkey. Austin: University of Texas Press.

43.

Turan

İlter

(1988) Political Parties and the Party System in Post-1983 Turkey. In Heper

Metin

Evin

Ahmet

(eds) State, Democracy and the Military: Turkey in the 1980s. Berlin: W. de Gruyter, 63–80.

44.

Yanardağ

Merdan

(2002) MHP Değişti Mi? [Did MHP Change?] İstanbul: Gendaş Kültür.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB