Justices ‘en Garde’: Ideological determinants of the dissolution of anti-establishment parties

Abstract

Examining justice-level determinants of party dissolution decisions can reveal how high Courts may influence the public choice by constraining the representation of political ideologies. We argue that Constitutional Court justices strategically engage in politics through party dissolution cases, and justices ‘en garde’ act to guard the regime against anti-establishment ideologies. As a graveyard of political parties, Turkey is an appropriate case to study this claim. By introducing a unique dataset, we demonstrate that communist, religious and ethnic parties in Turkey with considerable public support are more likely to be dissolved by justices having an activist and pro-status quo ideological stance.

Keywords

Party dissolution anti-establishment parties judicialization representation Turkish politics

Introduction

In March 2008, the Constitutional Court of Turkey (hereinafter CCT) announced its intention to review the case for dissolving the Justice and Development Party (AKP) – the incumbent conservative party. Only four months after the AKP’s first re-election with 47% of the popular votes, the Court decided that the party was a “focal point of non-secular activities”¹ that undermined the principle of secularism. The results of the voting were 6 votes for and 5 votes against the dissolution. However, the AKP managed to escape dissolution due to the lack of a qualified majority. The Court nevertheless cut the state funding to the AKP in half.

The AKP case is only one of many in which a political party with high levels of electoral support has been accused of violating the secularism principle in recent history. Ten years earlier, the CCT dissolved the Welfare Party, the AKP’s predecessor, and banned its leaders, including the former prime minister, Necmettin Erbakan, from politics. The Court has dissolved a total of 27 political parties to date. This figure is very high in comparison to the 18 party dissolutions in a total of 11 European states (Bourne, 2012), and for this reason alone merits the attention of comparative judicial scholars. Why such a high number of party dissolutions in Turkey?

Not all the parties dissolved by the CCT represent the same ideology, but in almost all cases the parties were accused of challenging the territorial integrity or the secular nature of the regime. Numerous studies on the judicialization of Turkish politics, however, claim that the Court’s decisions are often of a political character (Belge, 2006; Celep, 2012; Kogacioglu, 2003, 2004; Özbudun, 2010; Shambayati and Kirdis, 2009; Tezcur, 2009). Can we argue, then, that the Court strategically engages in politics through party dissolution cases to maintain the founding principles of the Turkish political regime?

We argue that when the anti-establishment political parties constitute a prominent threat to the regime, the judiciary takes a pro-status quo stance in favor of the state ideology. We test the viability of this claim focusing on Turkey, a country with civil law tradition. In the literature, party dissolution decisions are considered to be a “reaction” of the regime against anti-establishment ideologies (Algan, 2011: 809). Further, some scholars suggest that the Turkish political regime has come close to a government of judges due to the Court’s expansionist approach (Hazama, 2012; Özbudun, 2007; Shambayati and Kirdis, 2009; Tezcur, 2009). Examining the role of justice ideology in decision-making, hence, is crucial, as “explanations of politics are incomplete, unless they incorporate courts” (Epstein et al., 2001: 120), especially if the Courts are unconstrained but constraining actors. With this in mind, our primary purpose in this article is to investigate whether, and if so, to what extent justices’ and defendant parties’ ideological positions influence the dissolution decisions of the CCT. This differs from much of the previous literature on comparative courts, which often pertains to judicial review cases (Corkin, 2010; Couso et al., 2010; Dressel, 2012; Shapiro and Stone, 1994; Shapiro and Stone Sweet, 2002).

The justices of the CCT are argued to embrace Kemalism, which is the founding ideology of the regime and the dominant perspective of political elites, based on Turkish nationalism and secularism principles (Belge, 2006; Celep, 2012; Guney and Baskan, 2008; Kogacioglu, 2003; Shambayati and Kirdis, 2009). In fact, prior studies neither assess the ideological stances of individual justices, nor empirically examine the variation in their attitudes, and their effects on justices’ decisions. However, party dissolution decisions are by no means consensus driven. Rather, individual dissenting opinions signal that the Court’s decisions do not merely arise from legal norms but also from the ideological dispositions of justices.

We suggest that examining the justice and party-level determinants of party dissolution decisions can provide scholars with an understanding of how, and to what extent, high Courts can indirectly overrule public choice by limiting the representation of diverse political ideologies. This article is the first empirical assessment of how the ideological dispositions of the justices of the CCT influence their decisions in dissolution cases against political parties representing anti-establishment ideologies. In this regard, we both assess the validity of the attitudinal model (Segal and Spaeth, 2002) and demonstrate that the judiciary, often considered the least dangerous branch, plays a decisive role in shaping the political sphere. Given that some other European high courts have also attempted to curb anti-establishment ideologies, our findings have broader implications for the comparative judicial politics literature.

We begin by examining the party dissolution phenomenon, and discuss why this is a matter of importance. Next, we briefly summarize the attitudinal model of judicial decision-making used herein to explain the CCT’s dissolution decisions. We conclude by summarizing our findings and offering a roadmap for future research.

The dissolution of anti-establishment political parties

The ruling elite perceive anti-establishment political parties that represent traditional cleavage groups as a threat to the regime. European history reveals such examples in Italy and Germany before the Second World War (Celep, 2012). In fact, some European democracies continue to face challenges of this nature. More recent examples include the Basque nationalist Batasuna party in 2003, and the dissolution of the Belgian Vlaams Belang in 2004. Some other European countries, such as Latvia, Lithuania, Bulgaria, the Czech Republic, Spain and Greece, have also prohibited anti-establishment parties (Bale, 2007; Bourne, 2012; Molenaar, 2010). Depending on the saliency of the social cleavages politicized by anti-establishment parties, parties facing dissolution do not necessarily have small electoral constituencies. In fact, some anti-establishment parties subjected to dissolution, including the People’s Democracy, Welfare, and Virtue parties in Turkey and the Batasuna in Spain, had secured important electoral victories (Bourne, 2012).

The burgeoning literature on party dissolutions underlines the dilemma that democracies with salient social cleavages must address in regard to party dissolution. On the one hand, most democratic regimes do not make it a practice because democratic ideals recognize the importance of representing diverse ideologies and preserving civil liberties. On the other hand, excluding anti-establishment parties from the political space may help democracies protect their regimes from competitive pressures (Bourne, 2012; Celep, 2012).

Uncertainty, political polarization, and the threat of past legacies have important consequences for bolstering regimes’ reactionary stances against anti-establishment parties. To the extent that political extremism might threaten their stability, some democracies might prohibit the representation of anti-establishment ideologies. The CCT, too, is equipped with imperative powers including the power to dissolve political parties. In fact, the Court has used this power more frequently than any other democracy has, and has relied on this mechanism more than other mechanisms, e.g. denying state funding or preventing a party from running in an election.

The role of justice ideology in party dissolution cases

The legal model of judicial decision-making contends that the plain meaning of statutes, the intent of lawmakers, and precedents determine justices’ decisions (Segal and Spaeth, 2002: 48). In this view, justices are mere functionaries who simply render verdicts in accordance with the law. One problem with this model, however, is that legal texts are imprecise, and open to interpretation. Moreover, legal norms may conflict with one another. Justices may also consider the intent of constitution makers. However, constitutions only offer a general framework. Therefore, referring to intent is problematic, as different justices may reasonably draw different conclusions (Howard and Segal, 2002). Further, justices can refer to precedents for making decisions (Kassow et al., 2012; Lax and Rader, 2010 ; Lindquist and Klein, 2006 ). On this point, however, one drawback is the existence of possibly conflicting precedents relating to the same issue. As each case is unique in some respects, the decision as to which precedent to follow still involves some discretion (Segal and Spaeth, 1993).

Given the lack of precision in legal texts, the difficulty of identifying intent, and the existence of multiple precedents pertinent to a single issue, justices can be expected to base their decisions on their own interpretations of legal norms and cases, which by nature are subjective. Hence, we concur with Gibson (1983: 9) when he writes that justices’ “decisions are a function of what they prefer to do, tempered by what they think they ought to do, but constrained by what they perceive feasible to do.” This proposition does not imply that the legal model cannot explain the decisions of the Court, but rather that other factors are also influential in judicial decision-making.

We assume that justices’ attitudes (e.g. ideologies and goals) are crucial in explaining their decisions (Blake, 2012; Segal and Cover, 1989; Segal and Spaeth, 1993, 2002; Tate, 1981; Yates et al., 2013a). In many countries, there is no formal mechanism for reviewing the decisions of high courts, which implies that justices’ decisions are not monitored in any formal way, and that the justices are unconstrained in their decision-making (Epstein et al., 2001; Stone, 1992). Moreover, as high court justices are appointed for long periods or even for lifetime terms, they often do not have any concerns relating to electoral accountability or re-election (Segal and Spaeth, 2002).

The attitudinal model has been frequently tested in the American context and more recently in other established democracies with a common law tradition (Ostberg and Wetstein, 2007). It is nevertheless important to scrutinize the ideological determinants of justices’ decisions in a range of different legal and political contexts. Turkey provides an appropriate example with an activist court as an unconstrained constraining actor, where the Court’s decisions are not open to challenge and the political domain is highly judicialized. In the next section, we offer a brief examination of the CCT from historical, legal, and political perspectives in order to explain why we select the CCT in considering the role of courts as political regulators in party dissolution cases.

The Constitutional Court of Turkey

The Constitutional Court of Turkey in historical context

Established in 1961, the CCT has a legal responsibility for preserving the founding principles of the regime, and the civil liberties afforded by the constitution. This responsibility includes determining whether the views and actions of political parties pose a threat to the constitution. The Court has the right to review the constitutionality of government actions – and even to declare them void. With the enactment of Turkey’s 1982 Constitution, the scope of the Court’s judicial review power was extended. In addition to the judicial review, financial control, formal notice, stay of execution, and party dissolution are all within the Court’s discretion under the 1982 Constitution.

Until 2010, there were eleven justices serving on the Court all of whom had served until their retirement, and who were appointed by Presidents from a diverse set of candidates nominated by civil and military high courts and institutions. Since then, the number of justices serving has been set at seventeen, and all serve for twelve years with no re-election. This appointment procedure has long provided Presidents with direct influence on the judiciary and indirect influence on the political domain through the decisions of the Court.

Based on the Roman law tradition and the French–Swiss civil code, the Turkish law system dictates that justices must abide by the positive legal commands of the law and the constitution-makers (Shapiro and Stone, 1994). Yet, this lack of discretion does not impede their willingness to play an active political role. Along with the jurisdiction of the Court extending to formal notice, financial auditing, and judicial review, the right to dissolve political parties paves the way for the justices of the Court to intervene in the political domain.

The legal bases of the dissolution of Turkish political parties

Turkey has been labeled a “graveyard of political parties” (Celep, 2012: 3). Particularly through its decisions in party dissolution cases, the CCT indirectly – but actively and strategically – engages in politics to maintain the very foundations of the regime. Its constitutional responsibility to protect the integrity of the regime does not necessarily render the CCT politically legitimate, especially for the supporters of the dissolved political parties. Likewise, although the Court has the authority to dissolve political parties under the rule of law, this does not imply that such decisions are free from political considerations.

Along with the preamble of the 1982 Constitution, Article 68 of the Constitution and the law on political parties are the legal norms cited most often in dissolution cases (Algan, 2011; Hakyemez and Akgun, 2002; Karvonen, 2007). Article 68 states that the statutes, programs, and activities of political parties cannot be in conflict with the independence of the state, its indivisible integrity, human rights, the rule of law, the sovereignty of the nation, democracy, and the principle of secularism. Of these provisions, perhaps the most restrictive for the representation of cleavage groups are those regarding the indivisible integrity and secularism of the Turkish Republic. These “vague” (Özbudun, 2007: 259) restrictions are intended to consolidate the founding principles of the Turkish Republic, especially with respect to the separatist claims of the Kurdish minority, and to the religious camp (Hakyemez and Akgun, 2002; Kogacioglu, 2003).

Thus, the Constitution itself is biased in favor of party dissolutions; it falls to the justices of the CCT to enforce the founding principles of the Turkish regime. The provisions in Article 68 were amended in 1995 and 2001 in instances in which political actors have tried to curb the Court’s activism. In 1995, in line with the European jurisdiction, a constitutional change hindered efforts to dissolve political parties by requiring sufficient proof that the party subject to dissolution had become the focal point of anti-regime activities (Algan, 2011; Hakyemez and Akgun, 2002). Additional clauses limiting the power of the justices came into effect with the constitutional changes in 2001. Since then, the Court can only dissolve a political party if it violates constitutional provisions in its governing documents, by its activities, or by receiving financial aid from foreign states, persons, or corporations. In addition to the dissolution, the amendment recognized a new mechanism for checking political parties, the power to ask for a cut in the state funding for a party. Additionally, the required majority for dissolution decisions was increased from a simple to a qualified majority of at least three-fifths of the members of the Court.

Three political parties, the FP (Virtue Party), the HADEP (People’s Democracy Party), and the DTP (Democratic Society Party), have been dissolved since 2001 despite such comprehensive changes in the Constitution that have made dissolving political parties more difficult. Moreover, the lack of unanimity in the controversial FP and AKP cases in this period shows that there is still space for the justices to arrive at different interpretations of legal norms, and to render correspondingly different verdicts.

In fact, the replacement of retired justices and the appointment of new members by the former prime minister of the conservative AKP, President Gül, have radically changed the composition of the Court since his election in 2007. According to Celep (2012), the pro-status quo stance of the Court favoring the unitary state and secularism principles before Gül’s election gave way to an anti-status quo stance from 2007 onwards. In these respects, as Özbudun (2007) and Algan (2011) also argue, perhaps the main problem is not related to law or to constitutional amendments limiting the power of the Court, but to the justices’ ideologies and activism. Although this does not decrease the significance of the Constitution’s bias against anti-establishment ideologies, a lack of accord among the justices and the existence of dissenting opinions encourage us to focus on justice-level determinants of party dissolutions.

The Constitutional Court of Turkey as a political regulator

Given the judicialization of politics over time, the literature is divided in terms of the democratic legitimacy of non-elected constitutional bodies. Some authors contend that constitutional politics pose a danger to democracy, while others see it as acting as a positive check and balance (Caldeira, 1987; Hazama, 2012; Hirschl, 2004). Indeed, to the extent that judicial activism reaches certain levels, the legitimacy of such courts is a serious concern. Activist high courts, in this regard, have been accused of becoming a “government of judges” (Hirschl, 2004; Shapiro and Stone, 1994: 400). Considering its power to dissolve political parties by referring to the vague principles set forth in the Constitution, the CCT has dissolved 19 political parties since 1982. We argue that the decisions of the Court in the party dissolution cases reflect the ideological positions of its justices.

As a last resort against regime changes, the Court is intended to be a non-political and legitimate institution that has the power to constrain the representation of anti-establishment ideologies. We suggest that, although public choice cannot be restricted in democratic regimes, one way to suppress anti-establishment demands can be through manipulating the supply of anti-establishment ideologies. First, party dissolution as a way to do this is legitimate on legal grounds for a majority of the public compared to putting barriers in front of all new entrant parties. In addition, although electoral institutions are easier to change by the legislature, constitutional changes preventing activism on the part of the Court require a qualified majority in the Parliament, and often a constitutional referendum, which is difficult to achieve in fragmented political systems such as that of Turkey. Thus, limiting the supply of anti-establishment ideologies through the Court’s intervention is more legitimate in the eyes of the public, who would otherwise see such legislative actions as illegitimate. Second, although new political parties with extreme ideologies may succeed in overcoming financial and organizational restrictions, regulating the supply of political parties through the Court offers an unchallengeable means to restrict the supply of anti-establishment ideologies.² Thus, we hypothesize that the CCT functions as both a preventive and a pre-emptive instrument in shaping the political domain.

H1: Pro-status quo justices are more likely to vote for the dissolution of political parties.

So far, we suggest that justices’ ideological stances play an important role in the Court’s party dissolution decisions. In fact, not all the parties facing the dissolution threat represent anti-establishment ideologies. Compared with mainstream parties, ethnic, sectarian, and religious parties that represent traditional cleavage groups in Turkey are more likely to be targeted by political elites, as they challenge the very foundations of the regime. We expect justices to react proactively and dissolve these parties based on the alleged threat of the existence of such political parties represents.

H2: Pro-status quo justices are more likely to vote for the dissolution of anti-establishment parties that represent traditional cleavage groups in Turkey.

Despite the lack of electoral accountability, the public choice can serve as a constraint on the justices as long as they want their decisions to be considered legitimate (Caldeira, 1987; Epstein et al., 2001; Segal and Spaeth, 2002). However, empowering the judiciary might encourage activist justices who share the views of the political elite to take action against challengers with strong electoral support (Dressel, 2012; Hirschl, 2004). To the extent that those anti-establishment parties garner electoral support, and therefore threaten the status quo, the Court can intervene in the political sphere as a regulator. In such circumstances, we expect the justices who share the ideological stance of the founding and ruling elites to act in concert to constrain public choice. We, therefore, suggest that political parties with considerable electoral support representing traditional cleavage groups are more likely to be dissolved by justices whose political views are in accord with those of the ruling elite.

H3: High electoral support for anti-establishment parties makes pro-status quo justices more likely to vote to dissolve such parties.

Research design and data

Previous empirical studies on the CCT scrutinized the political divide between old and new elites by examining the decisions of the Court in judicial review cases (Belge, 2006; Hazama, 2012). Unlike those studies, which examine case-level variation in the exercise of judicial review cases, the present article is the first to examine justice-level variation. In addition, it is the first in the comparative judicial politics literature to examine party dissolution cases as another example of the judiciary influencing the political domain. Although we follow previous studies by including independent variables coded from judicial review cases as measures of the ideological preferences of justices, we employ these variables to examine party dissolution cases, which we think are likely to prove a clear indicator of the judicialization of politics.

We compiled a comprehensive dataset consisting of 2,870 cases in different domains in addition to the educational and professional backgrounds of 85 justices³ who served during the period from the enactment of the Constitution of 1982 to the end of 2011.^4,5 To measure the justices’ ideological positions, which might affect their decisions in dissolution cases, we coded the justices’ decisions and dissents in 2,074 judicial review cases in which the CCT reviewed the constitutionality of laws. In the 329 instances for which we were able to identify the ideological stances of appellants, we also coded whether the political party taking the case to the Court was right-wing or left-wing based on the Comparative Manifesto Project (CMP) dataset, which provides a reliable indicator of the ideological positions of legislative parties on the traditional left–right scale (Volkens et al., 2013).⁶ We coded our dependent variable, vote for dissolution or prohibition, from party dissolution cases that are also retrieved from the official website of the CCT.⁷

Dependent variable and model specification

Our dependent variable is the dissolution vote for a political party or the vote to ban an individual actor from politics in a given case. Because justices of the CCT cannot abstain, the variable is binary. All 41 dissolution cases between 1983 and 2011 are listed in Section A in the online Supplementary Appendix file.⁸ In addition to the decisions regarding the dissolution of political parties, we also include ten votes in these cases on whether to prohibit party leaders from politics. Moreover, we inversely coded 19 cases such that our dependent variable accounts for whether a justice votes for the dissolution of political parties or the prohibition of individual party leaders.⁹

Following the previous literature on the party dissolution phenomenon in Turkey (Celep, 2012), we also identified procedural cases in which fringe parties without any electoral support were the defendants. We estimate separate logistic regression models on the subsample composed of non-procedural cases as a robustness check. The model specification we employ in empirical analyses in the next section is as follows:

\Pr (Vote for Dissolution) = G (β_{0} + β_{1} Justice Ideology + β_{2} Status + β_{3} Experience + β_{4} Education + β_{5} Parliamentary Party + β_{6} Parliament ’ s Ideology + p ’ β + ε)

where Justice Ideology indicates the respective primary independent variable employed in the model (two-dimensional w-nominate ideal point estimates or unidimensional mean left–right score), and p’ indicates the dichotomous variables accounting for the ideological dispositions of defendant political parties in dissolution cases.

Independent variables

To test our primary hypothesis that the justices’ ideologies affect their decisions in dissolution cases, we employ two independent variables that account for justice ideology. As our first measure, we introduce two-dimensional w-nominate ideal point estimates in Model 1. Given the minimum vote requirement in the algorithm (Poole et al., 2011) and unanimous decisions in our dataset, we used 726 judicial review decisions to estimate the ideological dispositions of 54 of 58 justices who voted in a dissolution case. We correctly predict 82.26% of “yes” and 90.9% of “no” votes and classify 82.26% and 85.71% of the justices’ ideal points. Therefore, w-nominate estimates constitute a reliable account of the underlying ideological dimensions of the political space and the ideological dispositions of the CCT justices.¹⁰

The w-nominate estimates suggest that the Turkish political space, unlike the issue space of the US Supreme Court (Epstein et al., 2007; Martin et al., 2005), is likely to be two-dimensional. We refer to the first and main ideological dimension as the liberal–conservative dimension (Poole and Rosenthal, 1985, 2007). The second dimension, referred to as the activist (anti-status quo)–restrainist (pro-status quo) dimension, accounts for whether justices, regardless of their political views, adopt an interventionist stance. We expect both conservative and activist (pro-status quo) justices who score higher than liberal and restrainist justices on either or both of the first and second dimensions to be more likely to vote for the dissolution of political parties.

We introduce the unidimensional left–right score in Model 2 as an alternative measure to the two-dimensional ideal-point estimates. Our second independent variable is the mean score of whether a justice casts a leftist or a rightist vote in 329 judicial review cases in which the appellant is a parliamentary party.¹¹ We expect the effect of the mean left–right score on the probability of voting for dissolution to be positive.¹²

To test our second hypothesis, we include a set of dichotomous variables to differentiate between the party families. In addition to ethnic parties pursuing the interests of the Kurdish minority and religious parties¹³ we include three variables to control for socialist/communist parties, centrist parties, and nationalist parties. The base category in our models is fringe parties without an identifiable ideological stance. In contrast to the previous literature, we do not expect to find significant differences in regard to the justices’ attitudes toward diverse cleavage groups represented by socialist, ethnic, and religious parties. Yet, we expect to find a negative effect for parties representing the dominant rightist stance in Turkey compared to those groups.

In terms of the legitimacy of the justices in the eyes of the public, some scholars argue that public opinion has little, if any, effect on judicial behavior because the justices do not have re-election goals or electoral accountability (McGuire and Stimson, 2004; Segal and Spaeth, 2002). On the other hand, some scholars offer empirical evidence suggesting that the decisions of the Supreme Court justices in the US are affected by public opinion (Caldeira, 1987; Langer, 2003; McGuire and Stimson, 2004; Yates et al., 2013b). To test our third hypothesis, we include a measure that controls for the ideological composition of Parliament when the Court hears the case. In instances where the dominant center–right ideology of the Turkish public is translated into the Parliament, we expect it to have a positive effect on the probability of voting for dissolution to the extent that the justices take the Parliament’s ideological stance into consideration.

We introduce another measure to account for the effect of the public choice on the justices’ decisions based on the electoral support of defendant parties. If they have legitimacy concerns, justices may vote against the dissolution of parliamentary political parties that have a high level of electoral support. Conversely, we expect this variable to have a positive effect because high electoral support for extreme parties might provoke a reactionary stance in justices who wish to preserve the regime in line with the “hegemonic preservation thesis” (Hirschl, 2004).

The justices’ experience, status, and educational level are also included in empirical analyses as control variables. As chief justices might influence others’ decisions, we control for whether the President or Vice President of the Court differ in regard to how likely they are to vote in favor of dissolution. We also control for the level of education of justices given the normative debates in the literature about the juristocracy in Turkey. The experience variable is a simple counter for the difference between the years a justice was appointed and the year the decision was made.

Empirical findings and discussion

In testing our hypotheses, we employ two logistic regression models that differ in terms of the measurement of justice ideology. In discussing our findings, we report the predicted probability of voting for dissolution and the change in probability for a one-unit increase in the independent variable. The model specification is non-linear in nature; therefore, we also plot the predicted probability of voting for the dissolution of ethnic (Kurdish nationalist) parties in comparison to our base category, fringe parties. This is achieved by allowing the variables that account for justice ideology to vary and by setting other variables to their sample means or modes.

H1: the effect of justice ideology on dissolution decisions

Our first model relies on two-dimensional w-nominate ideal-point estimates on the liberal–conservative and activist (anti-status quo)–restrainist (pro-status quo) dimensions. Both measures are statistically significant and in the expected directions in Model 1. When each of the other variables is set to the sample mean or mode, the estimated probability that a hypothetical justice scoring 0 on both dimensions will vote for dissolution is 75%. Ceteris paribus, the probability for a highly liberal justice, who scores -1 on the first and 0 on the second dimension, is 84%. On the other hand, a pro-status quo justice scoring 1 on the activist (anti-status quo)–restrainist (pro-status quo) dimension and 0 on the liberal–conservative dimension has an 87% probability of voting for dissolution. When set to their minimum and maximum values, the most conservative and pro-status quo justice has a probability of 92%, which is more than twice that of a liberal and activist justice (43%). Because both measures are continuous and range between -1 and 1, we plot our estimates for nine hypothetical cases representing the minimum, mean and maximum values of respective ideological dimensions in Figure 1. As the confidence intervals around the predicted probabilities show, our predictions for the modal category in dissolution cases and ethnic parties are statistically distinguishable from the base category (i.e. fringe parties). Moreover, a conservative and pro-status quo justice has an approximately 20% higher probability of voting for the dissolution of an anti-establishment party than for a fringe party of no electoral significance.

Figure 1.

Predicted probabilities of voting for dissolution (Model 1).

Our primary independent variable in Model 2 is the mean left–right score of justices in judicial review cases with appellants as parliamentary parties. A left–right score of -1 corresponds to the far left and a score of 1 corresponds to the far right position. Like Model 1, Model 2 provides empirical support for our first hypothesis, according to which justices who have a rightist stance are more likely to vote for the dissolution of political parties. A centrist justice has an estimated probability of 71% of voting for the dissolution; it is 86% for a rightist and 49% for a leftist justice. On average, the estimated difference is 25.7% for one-standard-deviation increase in the mean left–right score.

H2: the effect of party ideology on dissolution decisions

To test our second hypothesis, we introduce a set of dichotomous variables accounting for different party families that have faced dissolution trials. Although extreme leftist ideology represents a salient divide in Turkish politics especially in the 1970s, the literature largely disregards dissolution cases against parties from the socialist and communist party family. Our results indicate that, though substantively less salient in comparison to the ethnic camps in Turkish politics, this cleavage exerts more influence on the justices’ decisions in comparison to the effect of the division between secular and religious camps.

In Model 1, for cases in which an ethnic party is facing trial, the probability of voting for dissolution increases from 36% in the base category (fringe parties) to 76%, whereas the estimated probability for socialist/communist parties is about 71%. Albeit substantively less significant than the probability of voting for the dissolution of ethnic and communist parties, the probability of voting for the dissolution of a religious party is 56.5% when each of the other variables is set to its sample mean or mode. Note that the coefficient of our independent variable controlling for nationalist parties is both negative and significant. Given the dominant Kemalist stance of the political elite in Turkey, this finding is not surprising. Indeed, it supports our expectation that conservative and restrainist justices are hesitant to dissolve political parties that share similar ideologies, but are willing to vote for the dissolution of others representing anti-establishment ideologies.

We plot the predicted probabilities of voting for the dissolution of an ethnic, and a fringe party in Figure 2. As shown in Figure 2, the predicted probabilities are statistically distinguishable for all justices in the range of the mean left–right score in our sample. Providing empirical evidence in support of both the first and second hypotheses, Figure 2 also shows that the predicted probabilities for rightist justices are not only higher than for leftist justices, but are also contingent on the political ideology of the defendant party.¹⁴

Figure 2.

Predicted probabilities of voting for dissolution (Model 2).

H3: the effects of public opinion and electoral support on party dissolution decisions

To test our third hypothesis, we introduced two measures of public choice approximating public opinion and electoral support for political parties at a given time. As noted above, one should have expected to find a decrease in the probability of voting for dissolution if the Court takes a stance in favor of diverse choices available to the public. Yet, the effect is statistically insignificant and in the negative direction in both models, even though the mean score for the ideological stance of the Parliament is 5.67, which indicates a center–right stance.

In fact, the variable that accounts for the electoral support of parties facing a dissolution trial is statistically and substantively significant in both models in Table 1. Unlike the sign of the coefficient for the Parliament’s ideology, the sign of the coefficient of the second variable, referred to as Parliamentary Party, is positive. This finding contradicts the claims in the literature regarding the moderating effect of public opinion on justices’ decisions. Ceteris paribus, justices are almost twice as likely to vote for dissolution if a party is represented in Parliament, and thus constitutes a more prominent threat to the status quo than do parties that are not represented in Parliament.

Table 1.

Logistic regressions on vote for dissolution.

	Model 1	Model 2
1st dimension (W-nominate)	−0.573***
	(0.206)
2nd dimension (W-nominate)	0.787***
	(0.229)
Mean left–right score		0.923**
		(0.443)
Socialist/Communist Party	1.487***	1.319***
	(0.317)	(0.311)
Kurdish/Alawite Party	1.724***	1.483***
	(0.365)	(0.354)
Center Party	0.076	0.321
	(0.328)	(0.358)
Nationalist Party	−1.238**	−0.949*
	(0.497)	(0.571)
Religious Party	0.852**	0.748**
	(0.422)	(0.381)
Parliamentary Party	1.040***	0.844***
	(0.340)	(0.321)
Parliament’s ideology	−0.005	−0.004
	(0.014)	(0.015)
Status of the Justice	0.250*	0.290*
	(0.136)	(0.169)
Years of experience	−0.102***	−0.097***
	(0.030)	(0.035)
Education level	−0.205	−0.379***
	(0.149)	(0.135)
Constant	−0.377	−0.280
	(0.345)	(0.436)
N	567	567
Pseudo R²	0.149	0.111
Log likelihood	−330.320	−344.903

Robust standard errors clustered by justices in parentheses.

Two-tailed tests: * p<0.1; ** p<0.05; *** p<0.01.

Our control variables also are substantively and statistically significant. On average, a year increase in our experience variable decreases the probability by 9.3% and 9.7% in Models 1 and 2 respectively. A justice with a doctorate in law is about 37% to 63% less likely to vote for dissolution in comparison to a justice with a bachelor’s degree. As expected, whereas the President of the Court is most likely to vote for dissolution, junior justices are about 19.4% to 34.3% less likely compared to Vice Presidents in the empirical analyses reported in Table 1.

Conclusion

Checks and balances are sine qua non for democracies. There are numerous examples in which political elites have advanced their interests by using political institutions. By examining party dissolution cases in Turkey, a country with a long but frequently interrupted history of democracy, we examine whether the justices of the CCT have acted in ways that overuse their powers and constrain the public choice. We show that the judiciary uses its power to dissolve political parties representing traditional cleavage groups at the expense of representative democracy.

We found that the ideologies of the justices of the CCT and the defendant political parties play an important role in party dissolution cases. Activist and pro-status quo justices act to guard the regime against anti-establishment ideologies. Moreover, anti-establishment parties are more likely to be dissolved by the Court than pro-status quo parties are. We relax the implicit assumption of uniformity across the justices in the previous literature on party dissolution cases in Turkey, and take into account the ideological stances of individual justices and defendant political parties. By modeling the effects of justices’ ideologies conditional on party ideologies, we were able to determine whether justices’ ideologies influence their decisions in dissolution cases against political parties representing salient cleavage groups in Turkey. Moreover, we demonstrated that anti-establishment political parties with considerable electoral support and that are represented in the parliament are more likely to be dissolved by justices who have conservative and pro-status quo stances. Along with these, we introduced diverse ideology measures accounting for the mean left–right scores of appellant parliamentary parties in judicial review cases, and two-dimensional w-nominate ideal-point estimates computed from judicial review cases. We showed that the justices’ ideologies are significant determinants of the Court’s decisions in dissolution cases.

We present our findings as worthy of further discussion. First, we call on scholars to test the external validity of our findings in other civil law countries. Although this article focuses on party dissolution cases in Turkey, our empirical results are mostly in line with the theoretical claims in previous comparative research on the judicialization of politics. Moreover, given the comprehensive constitutional changes in last years concerning the composition, jurisdiction, and structure of the CCT, further research could also test whether justices’ ideologies still play an important role in the justices’ decisions, and assess whether they are an artifact of Turkey’s institutional structure and frequently interrupted democracy, or only applicable to a particular era in Turkish politics.

We also call attention to party dissolution cases as an important means of judicial influence on politics. Albeit a rare phenomenon, party dissolution is not limited to Turkey. In some other European democracies that also follow the civil law tradition, High Courts limit the representation of salient cleavage groups through party dissolutions and other means, such as denying state funding to political parties. If our findings do not hold for similar regimes facing anti-establishment pressures, and other case types through which High Courts can intervene in the political domain, it could be that the exercise of judicial influence on politics is limited to regimes with similar levels of activism or those in which the High Court has a similarly high level of legitimacy in the eyes of the public.

Despite being robust to the employment of different measures of justice ideology, limiting the sample to non-procedural cases, and addressing the heteroskedasticity in the data by clustering the standard errors by cases, our empirical analyses are insensitive to temporal changes in justice ideology. Despite our best efforts to control for the changing composition of the Parliament and the Court over time, it would still be beneficial to consider the structural changes over time. In this regard, we hope that the comprehensive dataset we present will provide researchers with a foundation to account for changing dynamics of the strategic interaction between the judiciary, the legislature, and the executive in the future.

Footnotes

Acknowledgements

An earlier version of this article was awarded the Richard I Hofferbert Best Graduate Student Paper Award. The authors thank the Award Committee, Wendy L Martinek, Jeffrey L Yates, Michael D McDonald, three anonymous reviewers, as well as the editor for helpful comments and suggestions. Any remaining errors are the authors’ sole responsibility.

Funding

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

Notes

Author biographies

Mert Moral is PhD Candidate at Binghamton University (SUNY), USA. His research interests include political polarization, voting behavior, political parties and party systems, representation, and political methodology. He has previously published his research in Electoral Studies and International Journal of Electronic Governance.

Efe Tokdemir is PhD Candidate at Binghamton University (SUNY), USA. His research interests include foreign policy and public opinion, international relations and domestic politics, terrorism, and politics in the Middle East. He has previously published his research in Electoral Studies.

References

Algan

Bulent

(2011) Dissolution of Political Parties by the Constitutional Court in Turkey: An everlasting conflict between the Court and the Parliament? Ankara Universitesi Siyasal Bilgiler Fakultesi Dergisi 60(4): 809–836.

Armstrong

David A

II Bakker

Ryan

Carroll

Royce

Hare

Christopher

Poole

Keith T

Rosenthal

Howard

(2014) Analyzing Spatial Models of Choice and Judgment with R. Boca Raton, FL: CRC Press.

Bale

Tim

(2007) Are Bans on Political Parties Bound to Turn Out Badly? A Comparative Investigation of Three ‘Intolerant’ Democracies: Turkey, Spain, and Belgium. Comparative European Politics 5(2): 141–157.

Belge

Ceren

(2006) Friends of the Court: The Republican alliance and selective activism of the Constitutional Court of Turkey. Law & Society Review 40(3): 653–692.

Blake

William

(2012) God Save This Honorable Court: Religion as a Source of Judicial Policy Preferences. Political Research Quarterly 65(4): 814–826.

Bourne

Angela K

(2012) Democratization and the Illegalization of Political Parties in Europe. Democratization 19(6): 1065–1085.

Caldeira

Gregory A

(1987) Public Opinion and the US Supreme Court: FDR’s court-packing plan. American Political Science Review 81(4): 1139–1153.

Çarkoglu

Ali

Hinich

Melvin J

(2006) A Spatial Analysis of Turkish Party Preferences. Electoral Studies 25(2): 369–392.

Carroll

Royce

Lewis

Jeffrey B

James

Poole

Keith T

Rosenthal

Howard

(2009a) Comparing Nominate and Ideal: Points of difference and Monte Carlo tests. Legislative Studies Quarterly 34(4): 555–91.

10.

Carroll

Royce

Lewis

Jeffrey B

Keith

Poole

Keith T

Rosenthal

Howard

(2009b) Measuring Bias and Uncertainty in Dw-Nominate Ideal Point Estimates Via the Parametric Bootstrap. Political Analysis 17(3): 261–275.

11.

Celep

Ödül

(2012) The Political Causes of Party Closures in Turkey. Parliamentary Affairs 67(2): 371–390.

12.

Corkin

Nicola C

(2010) Developments in Abstract Judicial Review in Germany, Austria and Italy. PhD Dissertation, University of Birmingham, UK.

13.

Couso

Javier

Huneeus

Alexandra

Sieder

Rachel

(eds) (2010) Cultures of Legality: Judicialization and Political Activism in Latin America. New York, NY: Cambridge University Press.

14.

Dressel

Bjorn

(ed) (2012) The Judicialization of Politics in Asia. Abingdon, UK: Routledge.

15.

Epstein

Lee

Knight

Jack

Shvetsova

Olga

(2001) The Role of Constitutional Courts in the Establishment and Maintenance of Democratic Systems of Government. Law & Society Review 35(2001): 117–164.

16.

Epstein

Lee

Martin

Andrew D

Segal

Jeffrey A

Westerland

Chad

(2007) The Judicial Common Space. Journal of Law, Economics & Organization 23(2): 303–325.

17.

Gibson

James L

(1983) From Simplicity to Complexity: The development of theory in the study of judicial behavior. Political Behavior 5(1): 7–49.

18.

Guney

Aylin

Baskan

Filiz

(2008) Party Dissolutions and Democratic Consolidation: The Turkish case. South European Society & Politics 13(3): 263–281.

19.

Hakyemez

Yusuf Sevki

Akgun

Birol

(2002) Limitations on the Freedom of Political Parties in Turkey and the Jurisdiction of the European Court of Human Rights. Mediterranean Politics 7(2): 54–78.

20.

Hazama

Yasushi

(2012) Hegemonic Preservation or Horizontal Accountability: Constitutional review in Turkey. International Political Science Review 33(4): 421–440.

21.

Hirschl

Ran

(2004) Towards Juristocracy: The Origins and Consequences of the New Constitutionalism. Cambridge, MA: Harvard University Press.

22.

Howard

Robert M

Segal

Jeffrey A

(2002) An Original Look at Originalism. Law & Society Review 36(1): 113–137.

23.

Karvonen

Lauri

(2007) Legislation on Political Parties: A global comparison. Party Politics 13(4): 437–455.

24.

Kassow

Benjamin J

Songer

Donald R

Fix

Michael P

(2012) The Influence of Precedent on State Supreme Courts. Political Research Quarterly 65(2): 372–384.

25.

Kogacioglu

Dicle

(2003) Dissolution of Political Parties by the Constitutional Court in Turkey: Judicial delimitation of the political domain. International Sociology 18(1): 258–276.

26.

Kogacioglu

Dicle

(2004) Progress, Unity and Democracy: Dissolving political parties in Turkey. Law & Society Review 38(3): 433–462.

27.

Langer

Laura

(2003) Strategic Considerations and Judicial Review: The case of workers’ compensation laws in the American states. Public Choice 116(1): 55–78.

28.

Lax

Jeffrey R

Rader

Kelly T

(2010) Legal Constraints on Supreme Court Decision Making: Do Jurisprudential Regimes Exist? Journal of Politics 72(2): 273–284.

29.

Lindquist

Stefanie A

Klein

David E

(2006) The Influence of Jurisprudential Considerations on Supreme Court Decision Making: A Study of Conflict Cases. Law & Society Review 40(1): 135–162.

30.

Martin

Andrew D

Quinn

Kevin M

Epstein

Lee

(2005) The Median Justice on the United States Supreme Court. North Carolina Law Review 83(2005): 1275–322.

31.

McGuire

Kevin T

Stimson

James A

(2004) The Least Dangerous Branch Revisited: New evidence on Supreme Court responsiveness to public preferences. Journal of Politics 66(4): 1018–1035.

32.

Molenaar

Fransje

(2010) The Development of European Standards on Political Parties and Their Regulation. Working Paper Series on the Legal Regulation of Political Parties, No. 4. Available at: http://www.partylaw.leidenuniv.nl/uploads/wp0410.pdf

33.

Ostberg

Cynthia L

Wetstein

Matthew E

(2007) Attitudinal Decision Making in the Supreme Court of Canada. Vancouver, BC: University of British Columbia Press.

34.

Özbudun

Ergun

(2007) Turk Anayasa Mahkemesinin Yargisal Aktivizmi Ve Siyasal Elitlerin Tepkisi [Judicial Activism of the Constitutional Court of Turkey and the Reaction of Political Elites]. Ankara Universitesi Siyasal Bilgiler Fakultesi Dergisi 62(3): 258–268.

35.

Özbudun

Ergun

(2010) Party Prohibition Cases: Different approaches by the Turkish Constitutional Court and the European Court of Human Rights. Democratization 17(1): 125–142.

36.

Poole

Keith

Lewis

Jeffrey

James

Carroll

Royce

(2011) Scaling Roll Call Votes with wnominate in R. Journal of Statistical Software 42(14): 1–21.

37.

Poole

Keith T

Rosenthal

Howard

(1985) A Spatial Model for Legislative Roll Call Analysis. American Journal of Political Science 29(2): 357–384.

38.

Poole

Keith T

Rosenthal

Howard

(2007) Ideology and Congress. New Brunswick, NJ: Transaction Publishers.

39.

Schofield

Norman

Gallego

Maria

Ozdemir

Ugur

Zakharov

Ugur

(2011) Competition for Popular Support: A valence model of elections in Turkey. Social Choice Welfare 36(3): 451–482.

40.

Segal

Jeffrey A

Cover

Albert D

(1989) Ideological Values and the Votes of US Supreme Court Justices. American Political Science Review 83(2): 557–565.

41.

Segal

Jeffrey A

Spaeth

Harold J

(1993) The Supreme Court and the Attitudinal Model. New York, NY: Cambridge University Press.

42.

Segal

Jeffrey A

Spaeth

Harold J

(2002) The Supreme Court and the Attitudinal Model Revisited. New York, NY: Cambridge University Press.

43.

Shambayati

Hootan

Kirdis

Esen

(2009) In Pursuit of ‘Contemporary Civilization’: Judicial empowerment in Turkey. Political Research Quarterly 62(4): 767–780.

44.

Shapiro

Martin

Stone

Alec

(1994) The New Constitutional Politics of Europe. Comparative Political Studies 26(4): 397–420.

45.

Shapiro

Martin

Stone Sweet

Alec

(2002) On Law, Politics, and Judicialization. New York, NY: Oxford University Press.

46.

Stone

Alec

(1992) The Birth of Judicial Politics in France: The Constitutional Council in Comparative Perspective. New York, NY: Oxford University Press.

47.

Tate

Neal C

(1981) Personal Attribute Models of the Voting Behavior of US Supreme Court Justices: Liberalism in civil liberties and economics decisions, 1946–1978. American Political Science Review 75(2): 355–367.

48.

Tezcur

Gunes Murat

(2009) Judicial Activism in Perilous Times: The Turkish case. Law & Society Review 43(2): 305–336.

49.

Volkens

Andrea

Lehmann

Pola

Merz

Nicolas

Regel

Sven

Werner

Annika

Lacewell

Onawa Promise

Schultze

Henrike

(2013) The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR). Available at: https://manifesto-project.wzb.eu/.

50.

Yates

Jeff

Cann

Damon M

Boyea

Brent D

. (2013a) Judicial Ideology and the Selection of Disputes for US Supreme Court Adjudication. Journal of Empirical Legal Studies 10(4): 847–865.

51.

Yates

Jeff

Moeller

Justin

Levey

Brian

(2013b) ‘For the Times They Are a-Changin’: Explaining voting patterns of US Supreme Court Justices through identification of micro-publics. BYU Journal of Public Law 28(1): 117–143.