The rational-institutionalist approach

The starting point for rational-institutionalist theories of democratic consolidation is the minimalist or procedural definition of democracy, first put forward by Schumpeter (Reference Schumpeter1943). Dahl’s concept of ‘polyarchy’ – a regime type that guarantees opportunities for all citizens to participate in the political process, and to contest public decision making through a series of institutional arrangements – is a widely used standard in this literature (Dahl, Reference Dahl1971). The contention here is that more extensive definitions lead to conceptual confusion and measurement problems. Additional, non-procedural, characteristics one could attach to a democratic ideal-type – high levels of education, a vibrant civil society, mass political engagement, a reasonable level of social equality, etc. – are argued to matter only for the quality of democracy, rather than for the definition of democracy itself.

To satisfy the requirement of conceptual clarity, rational-institutionalist approaches also adopt a minimal definition of democratic consolidation, concerned solely with the integrity and survival of this ‘procedural minimum’ (Schedler, Reference Schedler1998: 103). These approaches focus on elite behaviour – rather than their attitudes – within this institutional framework. Democracy is considered consolidated when rational elites prefer to take part in an institutionalized form of political competition rather than attempt to subvert the system as a whole. Consolidation thus ultimately depends on whether the cost-benefit analysis of key actors is favourable to democracy: if risks associated with non-compliance are greater than those associated with compliance, democracy will reach a self-enforced equilibrium (see Przeworski, Reference Przeworski1991; Clark, Reference Clark2002; Alevizakos, Reference Alevizakos2008).

Institutions are essential as they provide incentive structures that constrain self-serving actors – they reward compliant behaviour and sanction non-compliance. Parties first establish these institutions to jugulate the threats of unregulated competition. The initial constitutional design is thus seen to result from an ‘elite pact’ between the authoritarian regimes’ hardliners, reformers, and a newly emerging counter-elite, as all have an interest in accepting free and fair elections rather than facing potential destruction (O’Donnell et al., Reference O’Donnell, Schmitter and Whitehead1986; Di Palma, Reference Di Palma1990). Similarly, the subsequent phase of state building results from the uncertainty created by elite competition. This is the case not only because the majority in place is less likely to abuse its own power if it is checked by a strong and vigilant opposition, but also because the weakening of state institutions is likely to play in its disfavour in the plausible event of electoral defeat (Grzymala-Busse, Reference Grzymala-Busse2007). Grzymala-Busse’s reliance on the image of Hobbes’ Leviathan is a most adequate metaphor: as in the original social contract theory, elites accept the constraints of a given institutional framework because the alternative, an environment of intense competition, may imply their destruction by competing forces (Grzymala-Busse, Reference Grzymala-Busse2007). This is the logic underlying Pzeworski definition of democracy as the ‘institutionalization of uncertainty’: competition among elites creates a climate of uncertainty, one that can only be eased through a framework that institutionalizes competition, and thus a democratic framework in the minimalist sense (Przeworski, Reference Przeworski1991: Ch. 1).

Well-designed democratic institutions in turn generate elite compliance by reducing the stakes of political battle. Proportional representation and parliamentarism are thus often viewed as ‘low-stakes’ institutional design, as they offer losers means to influence the policy process, and a real prospect of obtaining power in upcoming elections. Relevant actors can then calculate that the cost of future defeat within this institutional framework is preferable to the risks associated with regime overthrow (Przeworski, Reference Przeworski1991: Ch. 1). Democratic consolidation is therefore the process by which elites get locked into an institutionalized incentive structure favourable to democracy; an incentive structure that they themselves establish to jugulate the threats of unregulated competition.

Empirical applications of the rational-institutionalist framework

Classifications of the speed and quality of democratization in the post-communist world systematically placed CEE countries at the forefront, and even the most sceptical authors considered them exceptional examples of democratic success (for instance Carothers, Reference Carothers2002: 9; Tismaneanu, Reference Tismaneanu2002: 5). Many of these positive assessments implicitly or explicitly relied on the premises of rational-institutionalism. The strong focus of post-communist studies on the constitutional frameworks that resulted from the transition period in CEE countries – including the system of checks-and-balances and the type of electoral system – typically stemmed from the assumption that these frameworks could provide incentives for actors to respect the democratic rules of the game (Zielonka, Reference Zielonka2001). The scholarly debate as to the respective merits of parliamentarism and presidentialism for the survival of democracy is most emblematic in this regard (Lijphart, Reference Lijphart1992; Linz and Valenzuela, Reference Linz and Valenzuela1994). In the second, post-transitional phase, scholars focused on the extent to which institutions protect state resources – in terms of both financial assets and employment opportunities – against elite exploitation. They took specific interest in the institutional guarantees for the transparency, openness, and fairness of state assets in CEE (Brada, Reference Brada1996; Stark and Bruszt, Reference Stark and Bruszt1998). The independent functioning of the judiciary, as well as strong institutions of oversight supervising respect for electoral procedures and fundamental liberties, were likewise prime objects of study (Grzymala-Busse, Reference Grzymala-Busse2007). Institutions were also central for the assessments of democratic progress made by international organizations. For instance, the opening of EU accession negotiations for most of these countries in the late 1990s was conditioned upon an acceptance of the Copenhagen criteria. Established in 1993, these are based not only on candidates respecting the acquis communautaire, but also on the institutional framework of democracy. This ‘political’ dimension of the Copenhagen criteria thus required CEE countries to set up an effective process of judicial review, to address the issue of clientelism and corruption, and to guarantee both political and civil freedoms, especially for minorities (Grabbe, Reference Grabbe2006).

The focus of post-communist studies on parties and party systems again results from the prevalence of rational-institutionalist understandings of democratic consolidation. Both, the development of parties as organizations of the state, and the increased regularity of patterns of party competition were taken as indicators of the compliance of mainstream actors to the rules of the democratic game. Scholarly attention was directed towards the ability of parties to develop as organizations and fulfil their functions as actors within the state, for instance to organize within the parliamentary system, formulate goals and policies, maintain a stable share of the vote, professionalize, build administrative capacity, and find sources of financing (see e.g. Krasovec, Reference Krasovec2001; Szczerbiak, Reference Szczerbiak2001; van Biezen, Reference van Biezen2003; Enyedi, Reference Enyedi2006; Kopecký, Reference Kopecký2007). Party system institutionalization – the increased regularity of patterns of competition, which ideally structure over time around two main party blocs – was also taken as a sign of democratic consolidation. In line with the rational-institutionalist framework, ‘the mark of a genuinely consolidated democracy (…) is the degree to which the alternation of parties in power is regular and accepted’ (Lewis, Reference Lewis2001a: 203). This logic is exemplified by Huntington’s ‘two-turnover test’, according to which a democratic regime can be considered consolidated when it undergoes two peaceful turnovers of ruling parties in transparent and fair elections (Huntington, Reference Huntington1991: 267).Footnote ³ The fact that elections repeatedly provoke power alternations without the results being challenges thus offers a strong indication that the institutional set-up is providing the right incentives to salient actors. Such a concern also explains the wealth of research that focused on the number of parties competing in each successive election, their size, the degree of polarization within the party system, the degree of electoral volatility, and patterns of alliances between parties (see e.g. Olson, Reference Olson1998; Toole, Reference Toole2000; Birch, Reference Birch2001; Lewis, Reference Lewis2001a, Reference Lewisb, Reference Lewis2006; Markowski, Reference Markowski2001b; Sikk, Reference Sikk2005).

The empirical studies of the early 2000s supported the hypothesis that parties and party systems were effectively institutionalizing – and thus democracy consolidating – in most CEE countries, with Hungary, the Czech Republic, Estonia, and Slovenia as frontrunners. With the exception of Vladimir Mečiar’s Movement for a Democratic Slovakia (HZDS), mainstream CEE parties preferred to follow, rather than subvert, the rules of the democratic game.Footnote ⁴ By the mid-2000s all of the future EU members had fulfilled Huntington’s two-turnover test, and most displayed party systems with increasingly regular patterns of competition. The willingness of ex-communist parties to reform was considered an essential sign of the consolidation of CEE democracies (Ishiyama, Reference Ishiyama1999; Bozóki and Ishiyama, Reference Bozóki and Ishiyama2002; Grzymala-Busse, Reference Grzymala-Busse2002; Hanley et al., Reference Hanley, Szczerbiak, Haughton and Fowler2008). While radical parties did achieve representation in the 1990s, they were excluded from governmental coalitions and lacked the broad support to be a serious democratic threat (Lewis, Reference Lewis2001a: 203; Millard, Reference Millard2004: 119–127).

The Hungarian case especially was considered a model of democratic consolidation until 2010. The country experienced one of the smoothest transitions to democracy in the region, with the development of a range of viable opposition parties as early as 1988, and the radical revision of the constitution in October 1989 (Hungarian Parliament, 1989). While the latter did not meet all the features of a ‘low-stakes’ design—because of a mixed electoral system and a fairly flexible procedure for constitutional amendment – it was nevertheless considered to ‘satisfy the needs for democracy’ (Szikinger, Reference Szikinger2001: 429). The Hungarian Communist party held its last congress during this same month, and converted to a social-democratic platform under the MSzP (Magyar Szocialista Párt) party banner. The first democratic elections followed in May 1990, during which the centre-right MDF scored 43% of the vote (Rothschild and Wingfield, Reference Rothschild and Wingfield2000: 239–245). By the mid-2000s, Hungary displayed what many scholars classified as the most institutionalized party system in CEE: strong elite competition, regular alternation between two main parties, and a rather robust institutional framework as compared with many other CEE countries (Sikk, Reference Sikk2005; Lewis, Reference Lewis2006; Grzymala-Busse, Reference Grzymala-Busse2007; Casal Bértoa and Mair, Reference Casal Bértoa and Mair2010). Further, Hungary was among the first post-communist countries to meet the Copenhagen criteria, and to start EU accession negotiations in 1998 – these went particularly smoothly, as compared with other countries such as Poland for instance (Batory, Reference Batory2008).

Challenges to democracy’s ‘procedural minimum’ in Hungary

Despite these positive assessments, democracy’s ‘procedural minimum’ has been undermined in several CEE countries following their accession to the EU. In Poland, the conservative Law and Justice party (PiS) initiated several controversial institutional reforms in 2005–07 that challenged the independence of the Polish administration (Jasiewicz, Reference Jasiewicz2007: 30–32; Michnik, Reference Michnik2007; O’Dwyer, Reference O’Dwyer2008: 1184–1187). Citizens for the European Development of Bulgaria (Gerb) – the Bulgarian conservative party that won two consecutive majorities in 2009 and 2013 – was suspected of electoral fraud in the Parliamentary elections of May 2013 (Troev and Buckley, Reference Troev and Buckley2013). Similarly in 2012, Viktor Ponta’s leftist Social Liberal Union (SUL) interfered in the proceedings of the country’s Constitutional Court to suspend Romania’s president Traian Basescu, and this shortly after having won an absolute majority in the December parliamentary elections (Buckley, Reference Buckley2012).

The case of Hungary remains the most puzzling. While it was set as a model of democratic consolidation in the EU’s post-communist space, it also experienced the most severe challenges to democratic institutions that have taken place in the region since the end of communism (Council of Europe, 2013; Dani, Reference Dani2013; European Parliament, 2013; Kornai, Reference Kornai2011; Norwegian Helsinki Commitee, 2013; United States Commission on Security and Cooperation in Europe, 2013). In the Spring of 2010, the Fidesz-KDNP coalition obtained 52.7% of the votes, a score that translated into a two-third majority in Parliament given Hungary’s mixed electoral system (Benoit, Reference Benoit1996).Footnote ⁵ Reaching such a threshold allowed Fidesz to initiate a process of constitutional re-drafting: on 18 April 2011 it adopted a new Fundamental Law, and amended it five times since (Hungarian Parliament, 2011; Kovács and Tóth, Reference Kovács and Tóth2011). Subsequently, an extensive series of Cardinal Acts were passed, including acts on the Protection of Families, on Religion and Church–State Relations, on Media Freedom, on the Rights of Nationalities, on the Judiciary, on the Constitutional Court, as well as a new Election Law (Bánkuti et al., Reference Bánkuti, Dombos, Fleck, Halmai, Rozgonyi, Majtényi and Uit2012).Footnote ⁶

It is beyond the scope of this article to offer an exhaustive analysis of these reforms – the reader is referred to the official reports and scholarly articles cited above for such analyses. Instead the following sections provide a limited number of examples, intending to demonstrate that the ‘procedural minimum’ of democracy – as defined in Dahl’s conception of polyarchy – was eroded by these reforms. This is the case for six of the eight domains in which institutional opportunities should exist for the participation in, and the contestation of, public decision making in Dahl’s model (Dahl, Reference Dahl1971: 3):

1. Freedom to form and join organizations

Hungarian Parliament passed New Cardinal Acts on the freedom of association and regulation of non-governmental organizations (NGOs) in 2011, namely Act CLXXV/2011 on the Freedom of Association, Public Benefit Status and the Operation and Support of CSOs and Act CLXXXI/2011 on the Court Registration of CSOs and the Relative Procedural Rules. Since then, Fidesz has been suspected of distributing government funds according to partisan criteria. The number of organizations receiving the funds was divided by three – in a country where in 2012 40% of NGO’ revenues originated from the state (Hungarian Central Statistical Office, 2013). Further, civil society organizations such as the Joint Civil Forum (Civil Összefogás Fórum – CÖF) were particularly favoured by the National Cooperation Fund (NEA), which is in charge of distributing these funds. In past years, the CÖF organized a number of pro-government rallies under the name of Békemenet (Peace March), and the head of the CÖF, László Csizmadia, is also the NEA council’s current chair.

In parallel, Fidesz exerted pressure on associations not supported by the government. On 19 June 2014, the Government Control Office (KEHI) gave 13 NGOs financed by the Norwegian Civil Support Fund a week to release all information concerning their financingFootnote ⁷ . On 8 September 2014, Hungarian police raided the Ökotárs foundation that was responsible for disbursing these funds. Targeted organizations, such as Transparency International, view these actions as politically motivated, and destined to intimidate independent civil society organizations (Transparency International Hungary, 2014). The existence of a legal basis for the audit of these organization was also questioned in an opinion issued on 23 July 2014 by Hungarian Ombudsman László Székely (Székely, Reference Székely2014).

2. Freedom of expression

Two Cardinal Acts that established a new Hungarian Media Authority in 2010 figured among the most controversial of Fidesz reforms.Footnote ⁸ They establish appointment procedures that compromise this authority’s independence from government, and provide it with over-extensive powers to regulate private and public media content. ‘Balanced reporting’ is now required for television and radio broadcast, and this provision is linked with a complaint system open to all citizens.Footnote ⁹ If the Media Council determines that media outlets violate this requirement, it can impose a fine, demand that the Council’s decision be published or broadcasted, as well as for the petitioner to be provided with an opportunity to publish her viewpoint.

As stated by the OSCE critical report on Hungary’s media situation, ‘a legal obligation on what content media outlets should have is not in line with free media and freedom of expression’ (OSCE, 2011). There exist concerns especially that these provisions will result in the self-censorship of media outlets (for a summary of these reforms, see Council of Europe Commissioner for Human Rights, 2011; OSCE, 2011: 10–13; Political Capital, 2011).

3. Right to vote and 4. Eligibility for public office

The Elections Act and Election Procedure Act restrict the eligibility and the voting rights of persons with mental disabilities, prisoners and ex-prisoners.Footnote ¹⁰ This resulted in the total disenfranchisement of 95,000 persons in the 2014 elections, including 26,000 individuals guilty of criminal offences that had already completed their prison sentence. These provisions were considered to ‘lack proportionality’ by the OSCE’s Office for Democratic Institutions and Human Rights, and to contravene paragraphs 7.3 and 24 of 1990 OSCE Copenhagen Document and International Covenant on Civil and Political Rights (ICCPR).Footnote ¹¹

5. Right of political leaders to compete for support

While legislation issued by the Fidesz government did not directly affect the right of leaders to compete for support, Article U of the Fundamental Law opens the door to such interference. This article stipulates that organizations related to the ex-Hungarian communist party are ‘criminal’, and mentions explicitly that the successor to this party, the MSzP, shares responsibility in these crimes. While this does not directly restrict the rights of the Hungarian Socialist Party to compete for support, the wording of this article is sufficiently vague that it could serve as a legal basis to do so in the future.Footnote ¹²

6. Alternative sources of information

Concerns exist over the restriction of media pluralism in Hungary. The government recently introduced fiscal arrangements that overwhelmingly burden the last major independent television channel in the country, the German-owned broadcaster RTL Klub. A 40% tax on media advertising revenues exceeding HUF 20 billion was adopted on 11 April 2014, while revenue below HUF 500 million is exempt from tax. Other factors curtail alternative sources of information in Hungary, such as the shrinking private advertising market, the growth of private investments by business actors supportive of Fidesz over recent years, and the pro-government bias of the public media sector (see Freedom House, 2014c).

7. Free and fair elections

Between 2010 and 2013, three new Cardinal laws regulating parliamentary elections were enacted.Footnote ¹³ A number of their provisions make the electoral system more majoritarian, and thus increase the advantage given to the leading party in an election.Footnote ¹⁴ Further, these acts include a number of provisions that favour Fidesz more specifically. For instance, the new constituency map is designed to compensate a slight left-wing lead; campaign advertising regulations advantage the government party; and acquiring electoral suffrage was facilitated for non-resident Hungarians in neighbouring countries, a population among which Fidesz voters are over-represented (for a summary of these changes and their consequences, see Bozóki, Reference Bozóki2013; Political Capital, 2013).

As demonstrated here, the reforms initiated by Fidesz eroded institutional guarantees for Hungarian citizens’ participation in, and contestation of public decision making in a majority of the eight domains stipulated by Dahl’s ‘procedural minimum’. While the history of consolidated democracies offers many examples of institutional reforms destined to favour the party in power (Alexander, Reference Alexander2001; Renwick, Reference Renwick2010) – practices of gerrymandering being a case in point – these remain limited compared with the scope of contemporary developments in Hungary. Given that, as shown earlier, most accounts relying on the rational-institutionalist framework classified Hungarian democracy as ‘consolidated’ before 2010, it appears necessary to engage in a critical reading of these theories. The following section examines two additional factors that affect democratic consolidation: the deeper commitment of party elites to the democratic process, and the capacity of parties to mobilize citizens.

Elite loyalty or the insufficiency of institutional guarantees

In the rational-institutionalist framework, well-designed institutions and sustained competition ensure that for all relevant actors the costs of undermining democratic institutions are greater than the risks of facing future defeat. The events that followed the 2010 elections in Hungary can thus be read as a reverse case, in which non-compliance became a rational course of action for Fidesz. In other words, as the elections results and the institutional design of the Hungarian state offered Fidesz the opportunity to undermine democracy’s ‘procedural minimum’ at very little cost, it seized this opportunity. This begs the question, however, whether any democratic regime could persist over time if party elites were to systematically adopt such a reasoning in comparable circumstances, that is when faced with a strong electoral outcome and relatively weak institutional constraints. Further, it raises the possibility that other, non-procedural conditions – and namely, deeper commitments of elites to the democratic process – are necessary to the long-term survival of democracy.

Alexander makes this argument explicit by applying theories of democratic consolidation to ‘first wave democracies’ – which a rational-institutionalist understanding would undeniably classify as ‘consolidated’ (Alexander, Reference Alexander2001 Reference Alexandera, Reference Alexanderb). First, established democracies count a number of ‘high-stakes’ frameworks, that increase the costs associated with electoral defeat and set weaker limitations on the power of winning majorities. These include institutional designs that favour, for instance, unitary rule over devolution, or majoritarian electoral systems over proportional representation (Alexander, Reference Alexander2001: 265).Footnote ¹⁵ The exceptionally high scores achieved by Fidesz in 2010 would thus have converted into a comparably high number of seats in many established democracies – notwithstanding the fact that Hungary’s previous mixed electoral system did not display all the features of a ‘low-stakes’ design (see Benoit, Reference Benoit1996; Szikinger, Reference Szikinger2001). Democracies with high-stakes frameworks, such as France or Britain, nevertheless survived over decades. This encourages one to consider additional characteristics of these systems, apart from institutional design, that makes them more resistant than the Hungarian framework.

Institutions are also far more malleable than posited by the rational-institutional framework, constitutions necessarily including amendment or re-drafting provisions. While requirements vary from one country to another, the parliamentary route – as opposed to revision through referendum – generally involves some form of a qualified parliamentary majority. In Hungary for instance, the threshold for a new constitutional drafting was raised in 1995 from a two-third parliamentary majority to a four-fifths one.Footnote ¹⁶ Clearly however, while high thresholds for constitutional revisions ‘might be expected to have the best chance of not generating endogenous revision impulses’, they cannot ensure that institutions will be ‘self-perpetuating or vulnerable only to exogenous shocks’ (Alexander, Reference Alexander2001: 261). In the last half-century, the history of European democracies is permeated with examples of constitutional revisions. While many of these were designed to favour the party initiating them, they did not fundamentally and durably challenge the ‘procedural minimum’ of these democratic regimes (Alexander, Reference Alexander2001: 263–264; Renwick, Reference Renwick2010). Undermining the institutional framework of democracy can, therefore, take more incremental forms than posited in the rational-institutionalist model. In other words, the choice opened to actors is not between a violent coup d’état and blind compliance to the democratic process – there is instead a wealth of intermediary possibilities, ranging from the practice of gerrymandering frequent in many established democracies, to the more radical constitutional transformations enacted by Fidesz.

To sum-up, many existing democratic frameworks allow for the emergence of strong majorities, and for these majorities to challenge the ‘procedural minimum’ of democracy through constitutional change at relatively low cost. If historically in these situations some actors, nevertheless, preferred institutionalized competition to radical constitutional reforms, this also means that additional factors to the institutional design encouraged these actors towards compliance. This implies that while well-designed institutions and high levels of political competition may set favourable conditions to democratic consolidation – by increasing the benefits of playing according to the rules and heightening the risks associated with challenging the system – they are, nevertheless, insufficient to guarantee a democratic regime against erosion. Additional conditions to those set out by the rational-institutionalist model thus require to be identified, conditions under which the ‘procedural minimum’ of a new democracy may be considered protected.

One hypothesis would be to set the deeper attachment of mainstream elites to the democratic process, or elite loyalty, as such a condition. Actors that are solely driven by the desire to acquire more power are likely to seize any available opportunity to do so. That institutions are cemented and perpetuated by the ideas and beliefs of those who act within them is an insight long recognized by cultural approaches to institutions (Gofas and Hay, Reference Gofas and Hay2007; Schmidt, Reference Schmidt2008; Bevir and Rhodes, Reference Bevir and Rhodes2010). Deeply ingrained democratic norms are thus likely to play a key role in the fact that mainstream elites in high-stakes systems, such as France or Britain, do not subvert democracy when their party obtains a strong parliamentary majority.

The question of elite loyalty to the democratic process has been given only limited empirical attention. As previously underlined, many post-communist scholars implicitly assumed that mainstream parties in CEE countries were prepared to comply with the democratic process, based on their respectful behaviour in the decade following the initial phase of transition. Problematic actors are generally identified based on pre-conceived assumptions, rather than empirical examination. Thus, many CEE democracies were declared consolidated on the premises, first, that extremist parties achieved a rather limited electoral success and were marginalized by mainstream forces; second, that reformed communist parties had accepted the democratic agenda. On the other hand, as Hanley repeatedly insisted, mainstream governmental parties from the conservative right – those that are today most problematic in both Hungary and Poland – received the least attention in the 1990s compared with those other two groups of parties, and they were also subject to the least political and academic concern (Szczerbiak and Hanley, Reference Szczerbiak and Hanley2004: 1; Hanley et al., Reference Hanley, Szczerbiak, Haughton and Fowler2008: 407–408). Fidesz was not a newcomer in Hungary: between 1998 and 2002 it led a first governmental coalition with its Christian-democratic ally, the KDNP, and lost the 2002 and 2006 elections only by a short margin to its main opponent, the MSzP (OSCE, 2002; Sitter and Batory, Reference Sitter and Batory2006). Few scholars would have then considered this party as a threat to the democratic process. The lack of attention to elite commitments to democracy in CEE thus leads to the situation where scholarship did not identify as problematic a mainstream government party that subsequently challenged the democratic process.

The overlooked role of party–citizen dynamics

While rational-institutionalist approaches over-estimated the protective role of well-designed institutions, they also under-estimated the function party–citizen dynamics may play in the success or failure of democratic consolidation. Two main assumptions structure scholarly approaches to party–citizen relations in CEE. First, in a context of widespread political disengagement, citizen mobilization by parties – their capacity to gain popular support – was seen in a positive light since the mid-1990s. Second, while important for the quality of newly found democracies, the development of party–citizen links was not considered essential to the continued integrity of democratic institutions, and thus for democratic consolidation per se. In the case of Hungary, however, the mobilization of citizens by a mainstream party was not only problematic for the democratic process, but a determinant factor in the erosion of democracy’s ‘institutional core’. The following considers these points in turn.

A large number of survey-based and electoral studies took party–citizen links in CEE as their subject of enquiry. Much of this scholarship expects the development of these links to positively affect the quality of newly formed democracies over time. One category of literature focused on correspondences between cleavages within society and political divisions within party systems. These studies identified correlations of various socio-demographic characteristics with party preferences, correlations of socio-demographic characteristics with value patterns and policy preferences, and correlations of value patterns and policy preferences with party preferences (Rose and Makkai, Reference Rose and Makkai1995; Tóka, Reference Tóka1996, Reference Tóka1998; Evans and Whitefield, Reference Evans and Whitefield1998, Reference Evans and Whitefield2000; Miller and White, Reference Miller and White1998; Tworzecki, Reference Tworzecki2003; van der Brug et al., Reference van der Brug, Franklin and Tóka2008). A second category of studies focuses on levels of political engagement in CEE. Despite the development of political cleavages, this engagement remains significantly below that of Western European citizenries. Such trends include lower and often declining voting turnouts (Millard, Reference Millard2004: 73–81), lower levels of party membership (Mair, Reference Mair1997: 185–186), higher levels of electoral volatility (Mair, Reference Mair1997: 187–192; Sikk, Reference Sikk2005), lower levels of trust in representative institutions (Misztal, Reference Misztal1996: 192; Mishler and Rose, Reference Mishler and Rose1997: 427–429; Dimitrova-Grajzl and Simon, Reference Dimitrova-Grajzl and Simon2010), and lower levels of party identification (Rose and Mishler, Reference Rose and Mishler1998: 221–223).

The CEE party studies share this positive outlook on the development of partisan ties within society. Nevertheless, they do not describe the mobilization of citizens by parties as a determinant factor in the continued integrity of democratic institutions. The weakness of partisan ties in CEE is generally considered a regrettable, yet transitory phenomenon; a problem that can be expected to regress over time, rather than a reservoir on which populist mobilization may subsequently thrive. An implicit assumption of this literature is that party system institutionalization will further citizens’ ability to identify with parties. Markowski lists the following positive consequences of increasingly regular patterns of competition: ‘“clarity of responsibility”, “decisiveness of elections”, “political representation”, (real) “opportunity for political choice” (system alternativeness) and other features of contemporary democracies (i.e. effectiveness, efficacy, political support, etc.)’ (Markowski, Reference Markowski2001a: 48).

The importance of party–citizen ties in CEE was also minimized by invoking the high levels of citizen disengagement in older democracies (on the situation of established democracies, see Dalton and Wattenberg, Reference Dalton and Wattenberg2000; Pharr and Putnam, Reference Pharr and Putnam2000; Hay, Reference Hay2007). If consolidated democracies have already gone beyond the stage of mass mobilization, then no greater engagement can be expected in newer democracies; nor is such engagement a crucial factor of democratic survival. For instance, Schmitter argues that any expectation of strong citizen mobilization in Eastern Europe ‘ignores the very substantial changes that took place in the nature and role of parties in well-established Western democracies’ (Schmitter, Reference Schmitter1992: 426–427). Similarly, Kitschelt et al. put that ‘the absence of mass party membership in Eastern Europe (…) may simply be a result of the fact that these democracies have come into existence in an era of ‘post-modern’ politics’ (Kitschelt et al., Reference Kitschelt, Markowski, Mansfeldova and Toka1999: 395; for similar stances, see Ishiyama, Reference Ishiyama2001: 35; Lewis, Reference Lewis2001a: 204–207).

The ‘economic losers’ of the early 1990s were often depicted as destabilizing forces, with the potential of being mobilized against drastic economic reforms and even the democratic process itself (Przeworski, Reference Przeworski1991; Vanhuysse, Reference Vanhuysse2006). By the early 2000s, however, few studies took such a scenario seriously. Fidesz’s strategies of citizen mobilization however played a crucial role in the subsequent erosion of democratic institutions in Hungary. The party could not have carried out its constitutional reforms without the strong majority it obtained in April 2010. The Fidesz-KDNP joint list then rallied 52.36% of expressed votes, a gain of over 10 points compared with the previous 2006 parliamentary elections. Its main opponent, the MSzP, achieved a mere 19.3% of the votes in 2010, losing 24 points compared with 4 years before. As will be discussed in the last section of this paper, this was also made possible because of Fidesz’s strategies of mobilization: a strong populist message, and the development of one of the most socially embedded party organizations in post-communist Europe (Enyedi, Reference Enyedi2015; Enyedi and Linek, Reference Enyedi and Linek2008). These developments contradict the assumptions made by the literature on the relationship between citizen mobilization by political parties and processes of democratic change. As stated above, this type of engagement has mostly been pictured as a factor that contributes positively to the quality of an emergent democracy, but is not pivotal to its institutional survival. In contrast, Fidesz’s strong form of mobilization set the conditions for the constitutional core of Hungarian democracy to be undermined. Indeed, the electoral outcome of this mobilization provided the party with both the political legitimacy and the institutional power to carry out its reforms after 2010.

From here, additional conditions to the ones set out by the rational institutionalist model can be identified, under which the ‘procedural minimum’ of a new democracy may be considered as protected from the threat of erosion. These may be spelt out in the following proposition: a democracy is consolidated when no political party that displays a lack of commitment to the democratic process also has a strong capacity for citizen mobilization in the political mainstream. The French Front National or the British National Party would be more likely to favour reforms of the type initiated in Hungary, but they are not able to mobilize the critical number of citizens to implement such reforms. This calls for an investigation of the conditions under which ‘disloyal’ parties are successful at mobilizing citizens and more generally, what makes for a good strategy of mobilization. In this regard, only a minority of studies have gone beyond the analysis of patterns of party interaction to take an interest in the substance of party competition and the content of parties’ claims to represent citizens (for exceptions, see Innes, Reference Innes2002; Grzymalala-Busse and Innes, Reference Grzymalala-Busse and Innes2003; Roberts, Reference Roberts2010). There is also little work concerning what Katz and Mair termed the ‘party on the ground’ (Katz and Mair, Reference Katz and Mair1995) – parties’ organized membership. More generally, the actual practices of citizen mobilization by parties – for instance through the organization of protests and rallies, or the development of ties with civil society organizations – have been understudied (Enyedi and Linek, Reference Enyedi and Linek2008).

The second half of this paper builds on the above-analysis to suggest some avenues of theoretical and empirical research. More specifically, it defends the need to theorize and study democratic consolidation as a process of cultural change in which party–citizen interactions play a central role. The following section describes existing culturalist approaches to democratization, and the reasons why these have been less influential than their rational-institutionalist counterparts. It then underlines ways in which party strategies of citizen mobilization can be integrated within a cultural approach to democratic consolidation. On this basis, the fourth section of the paper outlines an agenda for future empirical research.

Culturalist approaches and their critics

A number of authors emphasize the importance of mass and elite attitudinal transformations for the long-term survival of democratic regimes. In this understanding, that can be referred to as ‘culturalist’, democracy is consolidated when no significant actor or part of the population considers re-negotiating the democratic rules of the game. The stabilization of democracy thus implies that ‘democracy becomes so broadly and profoundly legitimate among citizens that it is very unlikely to break down’ (Diamond, Reference Diamond1994: 15; Plattner and Diamond, Reference Plattner and Diamond1996; Miller et al., Reference Miller, White and Heywood1997; see also Plasser et al., Reference Plasser, Ulram and Waldrauch1998). In Easton’s terminology, the democratic regime needs to generate diffuse support, an affective orientation referring ‘to evaluations of what an object is or represents – to the general meaning it has for a person’ (Easton, Reference Easton1975: 444). This also means that elite and mass support for democratic institutions does not stem from pragmatic compliance or vested interests, but from deeply entrenched values – thus scandals or economic difficulties do not fundamentally affect citizens’ regime preferences, and electoral defeat or victory do not affect elites’ regime preferences (Linz and Stepan, Reference Linz and Stepan1996: 16; Dryzek and Holmes, Reference Dryzek and Holmes2002: 9). In these approaches, cultural changes result from processes of modernization, with certain social and economic developments viewed as necessary requisites for the promotion and sustainability of the democratization process. Economic development is considered a crucial criterion, performing functions essential to the consolidation of democracy. Among others, these include allowing for the development of a middle class, encouraging the emergence of a vibrant civil society, decreasing levels of economic inequality, and accompanying the rise of education levels (Lipset, Reference Lipset1994: 2–3; Przeworski et al., Reference Przeworski, Alvarez, Cheibub and Limongi1996; Fish, Reference Fish1998).

Drawing on this academic tradition, a branch of post-communist studies emphasize the path-dependency of various CEE trajectories, tracing the effect of country-specific legacies on the social, political and economic developments of the 1990s. A share of these argue that the state and civil society structures of the pre-1989 era affected post-socialist paths of extrication from communist rule, including elite negotiations and the resulting institutional arrangements of the early 1990s (Stark and Bruszt, Reference Stark and Bruszt1998; Kitschelt et al., Reference Kitschelt, Markowski, Mansfeldova and Toka1999; Bunce, Reference Bunce2003). Subsequent developments in these democracies were also associated with the legacies of communist rule. For example, scholars examined the influence of communist regime types on the shape and representativeness of emerging patterns of party competition in Poland, Hungary, and the Czech Republic (Kitschelt et al., Reference Kitschelt, Markowski, Mansfeldova and Toka1999). Wittenberg analysed the instrumental role of church institutions in socialist Hungary for the survival of pre-communist conservative political identities in post-communist times (Wittenberg, Reference Wittenberg2006). Others examined how networks of economic ties under socialism conditioned democratic governance and economic policy making after 1989 (Stark and Bruszt, Reference Stark and Bruszt1998).

These approaches to democratic consolidation were criticized for not specifying the mechanisms by which political culture is transformed, and risks of democratic erosion subsequently reduced (Schedler, Reference Schedler1998: 104). Especially compared with the rational-institutionalist model, culturalist approaches lack analytical clarity. Certainly, many of the processes discussed by culturalist approaches – economic growth, the institutionalization of party systems, the development of mass education, the growth of civil society, etc. – are likely to play a role in the transformation of mass and elite attitudes. However, as summarized by Hanson, ‘modernization theory contains no compelling causal mechanism that might satisfactorily explain the cultural correlations it emphasizes’ (Hanson, Reference Hanson2001: 132). In other words, while correlations may be established between given economic, social or political trends and the length of survival of democratic regimes, the processes that effectively link these macro-phenomena to the diffusion of mass support for democratic institutions remain both under-theorized and empirically under-investigated (Dryzek and Holmes, Reference Dryzek and Holmes2002: 16; Kubik, Reference Kubik2003: 318–322). As a result, how culture can itself evolve or integrate different elements, for instance, by becoming more democratic, is also difficult to apprehend. This opens the door to deterministic arguments, where culture ‘descends from heaven to influence the course of history’ (Haughton, Reference Haughton2005: 6), and where phenomena are explained with reference to a country’s inalterable traditions. For instance, the idea that nationalistic and populist tendencies in CEE symbolize the ‘return of the repressed’ is widespread, these traits often being presented as long-lasting features of immature political cultures (Ignatieff, Reference Ignatieff1993; Baer, Reference Baer2001: 109; Minkenberg, Reference Minkenberg2002: 358).

The performative role of partisan mobilization

The first task of a culturalist understanding of democratic consolidation would thus be to integrate the role of agency in processes of democratic change. This section identifies parties in their functions of citizen mobilization as key agents that contribute to such transformations. These organizations mediate between citizens and the state in a democratic polity; they channel popular demands and account for their execution. The strategies they use to mobilize citizens will likely shape citizens’ perceptions of the democratic process, and more widely contribute to promote or undermine democratic norms within society at large.

Democratic theorists recently took an interest in this socializing role of institutions of representation in general, and of parties in particular (see especially Young, Reference Young2000; Mansbridge, Reference Mansbridge2003; Muirhead, Reference Muirhead2006; Urbinati, Reference Urbinati2006; Rosenblum, Reference Rosenblum2008; Saward, Reference Saward2010; White and Ypi, Reference White and Ypi2010, Reference White and Ypi2011). Despite their differences, these scholars assume that representatives have agency in how they chose to represent citizens. Because constituents never offer themselves as a homogenous, pre-defined entity with a clear and encompassing set of interests, parties need to interpret what these interests are, and make choices as to how they should be represented (Young, Reference Young2000: 126; Saward, Reference Saward2010). As emphasized by Saward, representation as an activity ‘centrally involves offering constructions or images of constituents to constituents and audiences’, and is thus primarily about ‘the active making of symbols or images of what is to be represented’ (Saward, Reference Saward2010: 14–15).

This is consequential for citizens themselves, and their self-identification. Citizen identities are conceived here not as fixed and pre-existing to representation, but as malleable and influenced by its processes. Crucially, ‘(p)arties do more than organize beliefs, interests, attitudes for political purposes. They discover and define politically relevant differences (…)’ (Muirhead and Rosenblum, Reference Muirhead and Rosenblum2006: 103). There is then a strong performative dimension to a successful act of representation. In other words, when constituents are effectively mobilized by parties, they not only recognize themselves in the images that representatives offer, but are also constituted and influenced by these images. This conception runs counter a general tendency in electoral and party studies that views political cleavages as the direct translation of social divisions. As expressed by Disch, representation ‘aims, then, not to reproduce a state of affairs but to produce an effect: to call forth a constituency by depicting it as a collective with a shared aim’ (Disch, Reference Disch2011: 107–108; see also Sartori, Reference Sartori1968).

It seems especially important to consider this creative dimension of partisan mobilization in relatively young democracies. In early 20th century Western Europe, party systems structured how generations of citizens engaged with and understood politics, and this by forging strong normative projects around existing social cleavages (Campbell et al., Reference Campbell, Converse, Miller and Stokes1960; Rokkan and Lipset, Reference Rokkan and Lipset1967). Similarly, the post-communist context is one in which new cleavages were defined and new political identities emerged. In this process, the attempts of parties to represent citizens and mobilize them around competing platforms will have had a strong influence on the stabilization of political identities. In the wake of regime changes, however, it is not only the definition of constituencies and partisan affiliations that is at stake: there also exists a broader struggle over re-defining the identity of the community as a whole. As emphasized by Jowitt, the end of communism created a ‘genesis environment’, characterized by ‘the dissolution of existing boundaries and related identities and the corresponding potential to generate novel ways of life’ (Jowitt, Reference Jowitt1992: 266).Footnote ¹⁷ With the end of decades of socialist rhetoric and the economic turmoil of the post-1989 years, leaders were given a golden opportunity to satisfy the yearning of these transitioning polities for narratives and myths (Tismaneanu, Reference Tismaneanu1998).

If parties socialize citizens, then democratic consolidation as a radical form of cultural change is also dependent on the commitment of mainstream elites to democratic norms, and on the strategies parties deploy to mobilize citizens on the basis of these commitments. As Jowitt continues in his diagnosis of the post-1989 condition, ‘for a new way of life to assert itself, a social minority must completely identify with and accept an imperative task (…) for a critical period of time during which new elites, practices, organizations institutionally coagulate’ (Jowitt, Reference Jowitt1992: 267–268). Jackson provides an example of such processes of elite legitimation in his analysis of the civilizational discourse of American and German elites in post-war Germany (Jackson, Reference Jackson2006). Demonstrating the importance of elite rhetoric in times of radical political change, he insists that ‘enormous flows of resources and reconfigurations of political practice require justification, and absent (elites’) rhetorical deployments (…) it is unlikely that the resulting policies and institutions would have taken the form that they ultimately did’ (Jackson, Reference Jackson2006: ix). Establishing the superiority of democracy as a ‘novel way of life’ in CEE will similarly require that elites legitimate the new regime through the public deployment of rhetoric.

Building on cultural theories of democratization and new theories of political representation, this section argued that party strategies of citizen mobilization have a central role in encouraging or deterring democratic forms of cultural change. On this basis, the last part of the paper outlines a three-fold research agenda for studying the influence of such party–citizen interactions on processes of democratic consolidation.

Elite commitments to the democratic process

As stressed above, a deeper commitment of mainstream elites to the democratic process ensures that these exercise restraint when faced with an opportunity to undermine democracy to their own benefit. Gathering empirical data on this issue is necessarily difficult. Given the widespread international consensus on the positive value of democracy, elites are unlikely to be explicit about these commitments.Footnote ¹⁹ Most autocratic leaders hold token elections and claim to act in the name of popular sovereignty to maintain a form of domestic and international legitimacy (Zakaria, Reference Zakaria1997). In CEE, the existence of a widespread elite consensus on the objective of EU accession in the 1990s precluded any explicit advocacy of alternatives to democracy (Batory, Reference Batory2008; Vachudova, Reference Vachudova2008). Surveys that rely on respondents’ approval or disapproval of very general categories such as ‘democracy’ and ‘authoritarianism’, would be, therefore, typically inappropriate for studying commitments to democratic norms. Not only would elites be likely to answer these surveys in a way they deem socially desirable, but these answers would say little about what individuals understand by these categories (Dryzek and Holmes, Reference Dryzek and Holmes2002). As developed below, elite public discourse may provide implicit expressions of loyalty or disloyalty to democratic norms (Linz, Reference Linz1978: 28–38), but, given the weight of EU constraints these will only offer limited evidence. The same logic applies to one-on-one elite interviews, in which respondents will not be transparent about the nature of their commitments to the democratic process (Steiner, Bächtiger et al., Reference Steiner, Bächtiger, Spörndli and Steenbergen2004: 54; White, Reference White2011: 45).

If elite discourse can be expected to vary depending on the context and audience that is being addressed, studying such variations would provide a first indication of the norms they uphold. One would then need to compare instances of elite discourse that take place in environments that are more or less constrained by the international context, or in other words, observe whether elites are more explicit about undemocratic commitments where costs of doing so are lower. Comparing discourse destined for an international public to one destined to a domestic public may provide such indications, especially when the national language is not widely spoken abroad. Similarly, one could compare the public declarations of elites to the ones they use in more private settings, or among their peers. Participant observation, a method defined as ‘the process of learning through exposure to or involvement in the day-to-day or routine activities of participants in the researcher setting’, may be one way of getting closer to such private forms of discourse (Schensul and LeCompte, Reference Schensul and LeCompte1999). Immersion in the internal life of a given party, for instance spending an extended period of time within a party headquarter or a ministry, would typically allow access to group discussions among elites where more explicit expressions of democratic or undemocratic commitments would be evident (see Marková, Reference Marková2007; Bevir and Rhodes, Reference Bevir and Rhodes2010; White, Reference White2011: 40, 45; Belzile and Oberg, Reference Belzile and Oberg2012: 467). Given the absence of a common standard in the literature for evaluating such commitments, country comparisons are also warranted in this domain. Comparing the nature and level of democratic commitments in new and old European democracies would be especially crucial, as this would provide information on the degree to which a history of democracy matters to the value-systems of party leaderships.

Discursive and organizational strategies of citizen mobilization

As previously emphasized, to undermine democracy’s ‘procedural minimum’ mainstream parties with a disloyal leadership need to win a large parliamentary majority, and thus to rally a substantial part of the electorate. Empirical work is thus warranted on how such parties can successfully mobilize within the confines of a democratic constitutional framework, and in the context of a constrained international environment. Two dimensions of these strategies would deserve specific attention. First, the ideational component, or in other words, the platforms on which mainstream parties mobilize; and second, the organizational component, that is the networks, events, and resources that parties rely on to increase their societal reach.

As mainstream parties with a disloyal leadership cannot be explicit about their lack of democratic commitment, alternative discursive strategies can be expected that will render their claim to power more acceptable in the domestic and international arena. Populism is a likely candidate, precisely because it carries an ambivalent democratic message. The populist speaker claims to act in the name of the ‘People’, and yet denies the legitimacy of alternative claims to citizen representation.Footnote ²⁰ One of the distinguishing features of CEE populism is that it frequently affects the political mainstream, in contrast to Western European party systems where such rhetoric tends to be contained at the fringes of party systems (Mudde, Reference Mudde2002). Concerning Fidesz, existing studies show that the party started deploying such rhetoric before the 1998 election campaign and radicalized continuously until 2010. Fidesz’s discourse repeatedly depicts the nation as a unitary actor that can only legitimately be represented by the Fidesz, and their opposition as a corrupt and elitist clique that tramples on the nation’s interests (Palonen, Reference Palonen2006; Reference Palonen2009: 322–324; Bozóki, Reference Bozóki2008; Bozóki and Kriza, Reference Bozóki and Kriza2008: 217; Centre for Fair Political Analysis, 2013; Enyedi, Reference Enyedi2015).Footnote ²¹ More systematic comparative work across the region would be required to map the type of discursive strategies mainstream parties deploy, and the degree to which they can be labelled as populist (for an operationalization, see Deegan-Krause and Haughton, Reference Deegan-Krause and Haughton2009). Attention should also be given to the evolution of these discursive strategies over time, and the circumstances under which they prove successful for mobilizing citizens. In Hungary for instance, the deep political and economic crisis under MSzP’s 2006–10 mandate was pivotal in giving traction to Fidesz’s appeals.Footnote ²²

Citizen mobilization also has an organizational dimension. Beyond the platforms parties defend, they deploy means to communicate their message and achieve public visibility. In established democracies, parties progressively ceased to rely on membership to gather funds and diffuse their ideas from the 1960s onwards, and now depend far more heavily on state resources and the mass media for these purposes (Katz and Mair, Reference Katz and Mair1995, Reference Katz and Mair2009). Given the weak ties CEE parties had with civil society in the early 1990s, they also established statuses that minimized their dependence on private donations and membership fees, gave a strong role to state funding, and tightly regulated campaign advertisement (Kopecký, Reference Kopecký2006). Despite the steep decline in the membership numbers of Western European parties, these numbers still remain far higher than in post-communist Europe (Biezen van et al., Reference Biezen van, Poguntke and Mair2012). At the same time, studies show that CEE parties that achieved societal anchorage – namely through the development of an extensive network of local branches and their membership base – have a significant advantage over other parties in mobilizing popular support during elections (Tavits, Reference Tavits2012).

More empirical research is thus warranted to explain variations in the organizational models of parties across the region, and explore the reasons why some were more successful than others at developing a strong social basis. The Hungarian party system is a case in point. While the ageing membership of MSzP declined steadily since the early 1990s, Fidesz party members increased from less than 5000 in 1990 to over 40,000 in 2011 (Saltman, Reference Saltman2014: 105–106). In parallel, the number of local party branches of Fidesz also increased between 2001 and 2005 from around 400 to 1050 (Enyedi and Linek, Reference Enyedi and Linek2008: 462–463).Footnote ²³ The electoral advantages that such a member-based organization provides to parties have been under-researched in the case of CEE. Fidesz achieved high visibility in the public space of major cities thanks to its dense network of supporters, organizing regular town-hall meetings, public discussions, and demonstrations. Spectacular actions included the coordination of a number of mass rallies that united over 100,000 participants in 2006, and the organization of four large-scale petitions that collected close to a million signatures (Enyedi and Linek, Reference Enyedi and Linek2008: 464). It remains to be established how these strategies served its 2010 landslide electoral victory.

The socializing role of parties

If the long-term consolidation of democracy is dependent on parties promoting democratic norms among the broader citizenry, disloyal parties will also seek to diffuse to the broader public their lack of democratic commitment. This process serves them in a number of ways. It may expand the party’s capacity for mobilizing citizens, first through the radicalization of its existing basis of support, and second through expanding its reach to new constituencies. The diffusion of non-democratic norms will also make institutional reforms more acceptable to the broader population, and limit the risks of resistance to such changes. Socialization may thus increase the chances of disloyal actors remaining in power once the rules of the political game are modified to their advantage.

Empirical research should examine how disloyal leaderships draw on the existing preferences of citizens, and the extent to which they also socialize citizens into non-democratic norms. Pre-existing political identities – values, interests, and group membership – necessarily limit the creative dimension of partisan representation (Enyedi, Reference Enyedi2005). More broadly, the ability of parties to mobilize around a project is dependent not only on the leader’s agency, but also on what Gamson terms the ‘cultural resonance’ of the party’s broader message (Gamson, Reference Gamson1992: 135). In this sense parties are constrained by a pre-existing context, a context that may not be favourable to the emergence of pluralistic norms of political competition. This opens up a wealth of other relevant questions, for instance whether certain types of mobilization strategies are more ‘creative’ than others over time; whether certain groups in a given population are more influenced by these strategies than others; or whether socializing effects are stronger in new rather than in established democracies.

A number of disciplines relying on a variety of different methods explore these types of questions, and may provide inspiration for future research. The recent ‘contextualist’ turn in public opinion and political psychology studies is providing mounting evidence that citizens opinions do shift according to how parties frame issues (Manza and Cook, Reference Manza and Cook2002; Druckman, Reference Druckman2004; Sniderman and Theriault, Reference Sniderman and Theriault2004; Chong and Druckman, Reference Chong and Druckman2007). For example Chong and Druckman demonstrate, using experimental data, that party frames can moderate ideological extremes and that the ‘relative strength’ of these frames does affect citizens’ engagement with them (Chong and Druckman, Reference Chong and Druckman2007). While these types of dynamics are under-researched in CEE, there is also evidence that they strongly operate in newly found democracies. Relying on in-depth interviews and participant observation methods to study youth activism in contemporary Hungary, Saltman shows that parties are core institutions of political socialization among younger Hungarian cohorts (Saltman, Reference Saltman2014). But quantitative methods may also be appropriate for such an endeavour. For instance, using surveys and voting data, Enyedi defended the relevance of an approach to cleavage formation in CEE that starts with the ‘strategic calculations of the political actors and not with the distribution of the preferences in the society’ (Enyedi, Reference Enyedi2005: 699). According to him, the ideological trajectory of Fidesz towards a form of national-populism since the mid-1990s contributed to attitudinal changes within Hungarian society, encouraging especially the development of authoritarian values among younger generations.