SEVENTH FRAMEWORK PROGRAMME THE PEOPLE PROGRAMME MARIE CURIE ACTIONS – NETWORKS FOR INITIAL TRAINING (ITN)
ELECDEM TRAINING NETWORK IN ELECTORAL DEMOCRACY
GRANT AGREEMENT NUMBER: 238607
Deliverable D5.1 – Institutional Structures and Partisan Attachments Final Report
Early Stage Research fellow (ESR)
Steven Van Hauwaert
Host Institution University of Vienna, Austria
The ELECDEM project was funded by the FP7 People Programme ELECDEM 238607
A. ABSTRACT The academic literature proposes a wide variety of factors that contribute to the explanation of far right party development. However, these constructs are typically structural in nature, rather variable oriented and are not necessarily able to explain far right party development as a whole. Much too often, the existing literature assumes far right parties develop independently from one another, even though processes such as globalisation make this highly unlikely. Therefore, this study refutes this assumption and claims far right party development is much more interdependent than the literature describes. To do so, this study proposes to complement existing explanatory frameworks by shifting its principal focus and emphasising more dynamic variables and processes. This innovative study’s main objective is to bring time and agency back into the analysis, thereby complementing existing frameworks. In other words, the timing and the pace of far right party development should be considered when explaining this phenomenon, just like it should include the far right party itself. Largely based on social movement and policy diffusion literature, this study identifies, describes and analyses the different facets and the importance of diffusion dynamics in the development of West European far right parties. The focus on the similarities and differences of diffusion patterns and the ensuing consequences for far right party development, allows this study to explore the nature, the role and the extent of diffusion dynamics in the development of West European far right parties. In order to successfully explore this diversity of diffusion patterns, the study relies on the comparative method, which combines the study case familiarity, much like case studies, and the study of differences, much like large n studies. This approach allows for the rejection of uniformity and generality, which is often the case in large n studies. Rather, it places diffusion patterns (and far right party development) in a socio cultural and historical context, and it allows for this approach to add to the existing theory. Within this comparative framework, and following the logic of triangulation, this study uses three complementing methodologies. Based on the CSES dataset, it uses multilevel modelling to analyse far right party identification and form an image of volatile voters. This helps determine far right parties’ degrees of similarity and the subsequent possibilities for diffusion. In addition, the extensive analysis of primary, secondary and tertiary literature, help form a theoretical framework for diffusion dynamics and, most importantly, serve as a foundation for the last methodology. Elite and expert interviews allow for the assessment and improvement of the theoretical framework, as well as the creation of an empirical framework. Regardless of the empirical nature of this study’s last section, following its innovative nature, most findings must be interpreted with some caution and one must not make generalisations beyond those provided in the study. After an extensive overview of the existing literature and the methodological approach this study takes, the first chapter provides an analysis of partisan attachments, which indicates that volatile voters differ considerably in different political systems. These observed differences indicate different diffusion opportunities but also the necessity of a case specific approach to a certain extent. Therefore, in the second chapter, the study designs a three step algebraic model, which confirms the presence and the importance of the master frame and diffusion dynamics in far right party development. A third chapter identifies four diffusion mechanisms, and it describes how these can influence far right party development differently. The following chapters take a more empirical approach and can be divided in three sections. A first section makes a clear distinction between a myopic and a deliberate state of diffusion. Directly following this, a second section makes a distinction between direct and indirect diffusion patterns. The examination of these two questions creates a set of diffusion characteristics that help describe how, when and why diffusion can become successful. Ultimately, a simulation model is used to illustrate the validity of the above claims.
2
B. TABLE OF CONTENT
A. Abstract ...... 2
B. Table of Content ...... 3
C. Executive Summary...... 4
D. Full Report ...... 8
1. Objectives and Aims ...... 9
2. Situating the Research in the Literature ...... 15
2.1 Literature Review ...... 15
2.2 Party identification ...... 64
3. Measurement, Data and Methods ...... 69
3.1 Political Process Model ...... 69
3.2 Methodology ...... 91
3.3 Three step model for diffusion ...... 127
3.4 Four mechanisms of diffusion ...... 149
4. Results and Analysis ...... 170
4.1 Outline ...... 170
E. References ...... 175
3
C. EXECUTIVE SUMMARY In the past three decades, the literature on socio political and electoral factors that influence and explain far right party development has grown significantly. In this discussion, however, the role trans national diffusion patterns play and the extent of their influence on far right party development have been severely understudied. Subsequently, this study identifies, describes and analyses the nature and scope of diffusion patterns between different West European far right parties. In addition, the study explores the diversity of such dynamics between West European far right parties, and throughout different developmental stages. This allows for the identification and examination of diffusion as an explanatory dynamic for far right party diffusion, and a possible causal connection between the two. Chapter 2.1 situates this study’s research question in the existing far right party literature. After providing an extensive overview of the most relevant scholarship, the following chapter indicates that overall party identification has decreased, while far right parties have managed to increase their partisan attachment. This observation, together with an increasing volatility and the absence of one cross national volatile voter, indicates there are both similarities and differences between far right parties. In order to explain this, one needs a general theoretical framework that allows for a case specific approach when necessary. One of the more important objectives of this study is to introduce two fundamental concepts the existing literature understudies: master frame and diffusion. The former is done via the application of the political process model to contribute to the explanatory framework of far right party emergence (see chapter 3.1). The latter is done via the conception of an algebraic model that indicates the presence and the weight of diffusion in far right party development (see chapters 3.1 and 3.3). One of the primary claims this study makes is that the diffusion of a new master frame can be partially responsible for successful far right party development (see chapter 3.4). In the final, more empirical sections, the study provides a clear distinction between myopic and purposeful diffusion, the values and limitations of direct and indirect diffusion, and it denotes the explanatory (and causal) value of diffusion by using a simulation model to confirm the accuracy of diffusion claims (see chapter 4.1). Much of the literature does not properly distinguish between different developmental stages, only takes national explanatory variables into consideration or neglects trans national dynamics. It is important, however, to also take into account processes, rather than only variables, dynamics, rather than only structures, and international factors, rather than only national factors, especially when analysing a European wide phenomenon. In other words, the study does not intend to provide an alternative and competing explanatory model for far
4
right party development, but instead it intends to add to the existing scholarship and complement the existing models by suggesting the inclusion of trans national dynamics and emphasising the explanatory value of more dynamic factors. Far right parties are the most significant and important challenge to post war electoral democracy. In general, they accept the notion of electoral democracy, however, they wish to not only change its operationalisation, but also improve representation and thereby its overall quality. To uphold their development, far right parties need to continue to increase and/or stabilise their vote share. Party consolidation can help increase the party identification between voters and far right parties, most notably compared to more traditional parties and their electorates. In general, the decreasing party identification with traditional parties in Western Europe benefits the appeal of far right parties. This increased appeal (which often leads to an increased vote share and possibly emergence) can spark interest from other far right parties as to how the former party emerged. This could start different kinds of trans national diffusion dynamics that, eventually, could increase far right party identification. This study wishes to examine the validity of such claims and conceives the following objectives: o Most scholars study far right parties as independent entities. However, socio political phenomena, such as globalisation and internationalisation, make this highly unlikely. The study introduces trans national diffusion, which has been facilitated by the growing importance of globalisation, and analyses how this dynamic influences far right party development and partisan attachment. o The literature’s assumption that parties have developed independently has also led to a failure to include communication patterns into the explanation. This study examines the interaction between far right parties on a trans national level, and analyses its nature, extent and diversity throughout the different developmental phases. o Following certain societal evolutions, such as the personalisation of politics, and the often complex access to far right parties, previous scholarship has neglected the party itself in its analyses of far right party development. Therefore, this study reemphasises the far right party as an important unit of analysis and as an institution, which influence their development and the development of other far right parties. Most of this study’s data are the result of extensive data gathering and data creating processes. However, due to its innovative character, this does not lead to a typical large n dataset, which has an impact on the methodologies applied throughout this study. In general, the study uses a comparative approach, which makes it possible to analyse the diversity in diffusion patterns, to construe the significance of diffusion patterns in far right party development and to
5
complement and advance existing theoretical frameworks. Within this comparative framework, triangulation of methods and sources contributes to the causal validity of this study. Multilevel modelling (with CSES data) is used to analyse differences in partisan attachment. Primary, secondary and tertiary literature, together with participant observation, is used to create a support mechanism and generate an initial empirical foundation for a theoretical framework. Expert and elite interviews are used to generate data, and verify and empirically test the theoretical framework The analysis of diffusion patterns is fairly new in party politics, and more specifically in far right politics. On one hand, one of the primary objectives of this study is to serve as a foundation for further, more elaborate research. The empirical framework this study uses should be expanded and form the foundation for a larger n analysis. On the other hand, the study’s primary ambition is to complement the explanatory frameworks so as to include both variables and processes. This ambition has already led to a set of important findings: o Far right party identification is much higher and volatility much lower as compared to more traditional parties. This is a structural observation one can find throughout Western Europe. Additionally, there is no one volatile voters, which means that master frame implementation cannot be the same in different political systems. o The extensive analyses of demand side and supply side variables have led to the negligence of more dynamic factors, such as organisational structures or the master frame, and processes, such as diffusion. Therefore, the study proposes the political process model as a more complete explanatory framework for far right party development. o The study introduces trans national diffusion to far right party research by proposing a theoretical model that complements the existing, more variable oriented explanatory models. Within this framework, the study highlighted why and how a new master frame diffuses between far right parties. In this process, the four primary dimensions of diffusion are analysed as well: the transmitting and adopting agents, the object of diffusion and the channel(s) of diffusion. o The study brings these different diffusion dimensions together, and formulates four diffusion mechanisms. Based on their different characteristics, different mechanisms operate differently and have different outcomes. The illustrations show that a learning mechanism is more likely to be successful than, for example, an imitation mechanism. o The study proposes a master frame change model that provides an algebraic explanation for master frame change. Based on a rational choice approach and
6
expected utility functions, the study acknowledges the existence of diffusion patterns and a far right party’s rationale to choose one master frame over the other. o The study provides an indication of how a far right party’s master frame decision can be influenced by other far right parties, i.e. how adoption decisions are interdependent. Additionally, it discusses the different adoption dynamics and how/if it is possible to aggregate these individual adoption decisions. o As diffusion is not always a voluntary or a purposeful dynamic, the study differentiates a myopic state of affairs (where diffusion can be accidental) from a more intentional dynamic. From this analysis, it is clear that most far right parties purposely seek to diffuse, as this proves to be beneficial. o The study’s theoretical claims are not only substantiated by the empirical illustrations, but the researcher also designs a simulation model that rigorously confirms the previous conclusions and allows for the final rejection of the study’s null hypothesis of independence.
7
D. FULL REPORT
8
1. Objectives and Aims
9
Originally, party identification and partisan attachment were stable and strong predictors for vote share. However, since the socio political evolutions in the 1970s and 1980s, this has changed significantly. Academic literature provides an extensive account of how the strength and importance of both party identification and partisan attachment decreased, thereby giving room to a more ad hoc approach to politics and voting. These evolutions provided quite a suitable environment for new parties to enter the political arena and challenge electoral democracy, which was especially true for far right parties. The literature on the societal, socio political and electoral factors that influence and help explain the development of such parties is extensive (see chapter 2.1). Nonetheless, as will be indicated throughout this study, the role and the extent of trans national diffusion dynamics have been severely understudied. Therefore, this study’s ambition is to identify, describe and analyse the nature and scope of trans national diffusion patterns between West European far right parties. It will analyse the variation in these dynamics throughout the different developmental stages of far right parties so as to establish and support a causal relationship between the two. In the contemporary political structure, far right parties are the most significant challenge to electoral democracy. They consider the quality of electoral democracy ambiguous, and they present clear critiques on the implemented representative system and the increasing politics of cartelisation of traditional parties. Contrary to some of the early far right party literature, far right parties operate within the boundaries of electoral democracy and can be considered a “pathological normalcy” rather than an anomaly of electoral democracy. In other words, far right parties criticise the system in place, and wish to change and improve it, not abolished it in favour of an alternative. Rather than less representation, they advocate for better representation so as to improve the quality of (electoral) democracy. One of the core ambitions of this study, which is often neglected in contemporary scholarship, is to provide an accurate and objective description of far right parties and how they can operate within the boundaries or even as components of electoral democracy. Typically, such an evaluation is based on the following criteria: representation, legitimacy and accountability. Usually, far right parties propose to decrease the distance between representatives and its electorate, not necessarily by disregarding the notion of representation, but by improving it. They do believe in the value of democratic institutions, but systematically question its proceedings and consistency, mostly as a result of traditional parties’ influencing them. In other words, they question the legitimacy of those institutions but more as an opposition to traditional parties rather than an opposition to its democratic or representative nature.
10
Directly following this, far right parties also seek to increase accountability of representatives and give the true power back to the electorate.
Objective 1: Evaluation of the effects of globalisation on the quality of electoral democracy In the far right party literature, globalisation is often described as one of the probable explanatory factors that provide a fertile breeding ground for far right parties to emerge (see chapter 2.1). At the same time, globalisation has also contributed to the decreasing distance between the electorate and its representatives, thereby changing representation and increasing accountability. Since far right parties seek to change the content and the format of electoral democracy, globalisation certainly plays an important role in this process. Several far right party scholars take a case study approach, which focuses on one or a few far right parties and is frequently accompanied by ad hoc theorisation and incorrect generalisations. Trans national dynamics such as globalisation make it implausible and unlikely that far right parties developed independently from one another. Other scholars prefer to take a large n approach, which often puts the primary focus on specific explanatory variables, and forgets to include more dynamic factors or far right parties themselves (i.e. institutions). Whereas other fields do take interdependence and processes into account, this is still highly underdeveloped in far right party research (see chapter 2.1). Dynamics such as globalisation allow far right parties to observe, imitate, and learn from others. Simultaneously, media and propaganda approaches can also help them to successfully diffuse objects themselves. All together, such trans national mechanisms are the core of this study and need to be given a place in the explanatory framework for far right party development. Combined with their Sartorian characteristics, such dynamics allow far right parties to influence the overall quality and implementation of electoral democracy. In short, this study complements existing variable oriented explanatory frameworks by introducing and emphasising the importance of more dynamic factors such as globalisation.
Objective 2: Evaluation of the effects of communication on the quality of electoral democracy When scholars study communication, they typically study rhetoric, meaning or ideology. In today’s electoral arena, communication exceeds this description and its role is much more diverse. Communication patterns and the subsequent dynamics have been neglected throughout the study of party politics as a direct consequence of the assumption that parties developed independently from one another. The researcher refutes such claims and includes inter party and trans national communication related dynamics, and how they
11
influence (the quality of) electoral democracy. More specifically, this study emphasises the communication dynamics between partisan entities on an international level. It observes and describes the diffusion that occurs between different far right parties, and it will analyse this throughout different developmental phases in order to indicate and analyse diversity. The four diffusion mechanisms the study identifies each consist of specific characteristics and operate under certain conditions. It is important to note that more (or less) communication, in whichever form, between far right parties does not necessarily lead to more or better electoral democracy. Here, outside factors, such as mainstream strategies, the electoral system, etc., are also important and influence the possible effects communication patterns can have on the nature and the quality of representation (see chapter 2.1). Therefore, communication patterns should be seen as a part of a puzzle that helps explain the overall characteristics of electoral democracy. In addition, far right parties can identify un/successful communication patterns, learn from other parties abroad, observe different media strategies, or other specific forms of trans national diffusion, to influence electoral democracy and its constructs.
Objective 3: Evaluation of the effects of institutions on the quality of electoral democracy The majority of far right party studies focus on demand side factors and political opportunities. By doing so, they fail to include institutional factors such as the party itself. More than its Sartorian characteristics, a far right party is also an organisation and a structure (be it successful or not). Even though a far right party only exercises such a function once it consolidates, it is nonetheless an important function. It is through this function that a far right party can directly influence electoral democracy and its quality. Therefore, this study emphasises the far right party itself and puts the focus again on agency, rather than structure. Numerous socio political and societal evolutions have contributed to a decreasing importance of institutions since the 1970s (see chapter 2.2). In general, the emphasis has gradually shifted from institutions to individuals, and can be summarised by the so called ‘personalisation of politics’. As a consequence, scholarship has also shifted its emphasis away from the party itself. Despite the increasing importance of individuals (especially for far right parties), this study attempts to reemphasise the importance of the party itself, as this is still the most important unit of analysis that can influence electoral democracy. In addition to this, many studies neglect when something happens or happened, i.e. the time of an incidence. This study reemphasises this because when something happens can help explain why and how something happens.
12
How to reach these objectives: The comparative method The study uses the comparative method to describe and analyse the diversity of trans national diffusion patterns between far right parties, and their relationship to the parties’ developmental patterns. This set of far right parties is specified at the outset of the study because, much like case study research, this approach seeks to establish a certain level of familiarity with each of its cases (see chapter 3.2). At the same time, contrary to case studies, where the emphasis lies on similarities between cases or through time, the comparative method focuses on a specific phenomenon (i.e. trans national diffusion) and the principal goal is to explain diversity between different far right parties. Even though large n studies and comparative studies both focus on differences, the former seeks to explain covariations based on many cases, whereas comparative research focuses on diversity. The ambition of this study, together with the possibilities of the comparative approach, allows for three basic goals. A first goal is the analysis of the diversity in diffusion patterns. Since it is important not to assume uniformity or generality, the study seeks to uncover and analyse causal mechanisms (i.e. diffusion dynamics) that separate far right parties in different categories. A second goal is the interpretation of socio cultural and historical significance of trans national diffusion dynamics. For this, the classification as far right parties, and within the far right party family is important. A third goal is to complement and advance existing theory by putting the focus on diffusion as a process. The comparative method emphasises diversity and its principal focus lies on similarities within the far right party family, on parties with the same diffusion outcome so as to differentiate that category from other categories and to explain the differences in diffusion outcome. In this case, the study of diffusion patterns is a study of both similarities and differences (i.e. diversity) between West European far right parties. Its process is fairly straightforward. A more detailed description of the use and the objectives of this method, together with the rationale for the case selection, the analytical frames this study proposes and a comprehensive analysis of diversity, can be found in the methodological chapter.
Methodological constructs As a part of the comparative approach, this study gives great importance to the logic of triangulation. This means methodologies and data sources are numerous, each sharing their benefits and allowing others methodologies or sources to complement their limitations. This study mostly draws on three principal methodologies. At first, the study takes a rather quantitative approach to analyse party identification, volatility, and the political opportunities
13
these give to far right parties. These are measured by analysing the party vote in the previous two elections and will eventually be regressed by using multilevel logistic regression. The CSES (third wave) dataset provides all the necessary variables and it contains most of the countries this study includes in its further analysis. Since this study analyses various countries, a separate model will be created and tested for each of the countries in order to properly take into account possible contextual factors. More specifically, a step wise regression will help determine which variables to include as independent variables, and which not to. Using this procedure, the study is able to verify whether the independent variables responsible for volatility are the same in all countries considered, or whether the independent variables are partially or completely different. In other words, this analysis determines whether volatile voters share characteristics trans nationally or whether their characteristics are contextually determined. This will help determine how similar far right parties are, and to which extent diffusion is possible. The analysis of diffusion patterns largely consists of the study of instances of influence between far right parties. In order to study these, a researcher needs to be aware of instances of diffusion, or so called incidents. Various sources serve as the basis for this: party materials, academic literature, popular literature, communication, third party analyses, participatory behaviour, etc. The main objective is to gather enough constructs so as to design a theoretical framework that can eventually make inferences about the constructs, the characteristics and the effects of communication. In short, it is the analysis of who (i.e. the transmitter) said what (i.e. the object) to whom (i.e. the receiver), how (i.e. the channel), why (i.e. the rationale) and with what effect. In this study, the effect and the different reactions of other far right parties are also of crucial interest. Ultimately, interviewing is the most important methodology used throughout this study. Elite interviews provide a necessary inside perspective, whereas expert interviews provide an important outside counterbalance. The study chose to conduct only semi structured interviews. On one hand, as this is an extremely specific study, that asks very detailed questions, one must be sure to obtain such information during an interview that is almost always restricted in time. On the other hand, a fully structured, almost survey style, interview would not be beneficial since most of the inquiries are not necessarily categorical variables (but rather incidents). It allows the interviewee some freedom to answer how he or she wishes, while the interviewer keeps control and can steer the interview in any desired direction. In addition, the interviewer must remain flexible to respond and inquire into some of the responses by the interviewees.
14
2.1 Situating the research in the literature
15
1. Conceptualisation and classification A. Introduction Since the 1980s, an extensive literature has been created that discusses numerous facets of far right parties (FRPs). Within this literature, however, arguments are diversified and consensus is exceptional. One of the primary areas where there is a structural lack of standardisation is the conceptualisation and typology of the phenomenon under study. The vast majority of scholars refer to the same phenomenon (with some occasional differences); however, they fail to refer to it consistently. Therefore, at the offset of a substantial study, it is important to specify certain aspects of the study. What are some of the ideological constructs used throughout the literature of conceptualise this phenomenon? Why the term FRPs, rather than extreme right parties (ERPs)? How do delineate between the two? Which parties are classified at FRPs (and why) and which are not? The primary unit of analysis of this study is the agent, i.e. the FRP. It forms the core of the study and is the most important unit of analysis. Therefore its proposed classification scheme does not include social movement, fringe movements, extreme right movements or other groupuscules. Such entities operate under different dynamics and should be subjects of a separate analysis (e.g. Klandermans and Mayer, 2006). The first part of this section provides an overview of the most important conceptualisations in the literature. It puts the emphasis on nationalism, exclusionism, authoritarianism, (liberal) democracy and populism. It uses these ideological constructs to build a proper conceptualisation and terminology of FRPs. It provides a justification as to why these specific ideological constructs are used throughout this study and why the term FRPs has been chosen over more common terms such as extreme right or populist right. The parts that follow describe a classification and provide a rationale as to why certain parties are included.
B. Conceptualisation in the literature. One of the core problems in the field of far right politics is the lack of scholarly consensus about how to unambiguously conceptualise FRPs. This includes its differentiation from other parties, and its categorisations within the party family. In addition, even when scholars do agree on the theoretical constructs of the conceptualisation, the empirical observations are often problematic. This section discusses the most common constructs used in the literature to conceptualise FRPs and concludes with the conceptual framework that serves as the foundation of this research project’s notion of FRPs. Nationalism
16
Both philosophy and political science have an extensive and influential literature on nationalism (e.g. Gellner, 1983; Hobsbawn, 1990). This literature redefined nationalism from a political attitude (on the individual level) to a more aggregated doctrine or paradigm. As a concept, nationalism has been discussed extensively. Nonetheless, one definition stands out due to the combination of its simplicity and its comprehensiveness. According to Mudde (2000, p.187) nationalism entails “the belief that the political unit (the state) and the ethnic unit (the nation) should be congruent”. Nationalism comprises two fundamental components: the state and the nation. The state is the centre and principal of political authority, and the only entity that can legitimately use violence (Weber, 1978). Without the state, there cannot be nationalism. The nation can be defined based on culture or will. The former describes a nation based on ideas, signs, associations, behaviour, etc. The key problem is there would be too many nations compared to states. The latter describes a nation based on mutual recognition of belonging, which comes with responsibilities and opportunities. This foundation, however, risks including groups that do not constitute nations. Rather than relying on these Weberian ideal types, it would be more pragmatic to assume people form a nation because they have a will to be politically united with all those who share their culture. Inherently, this definition has an internal and an external component (Koch, 1991; Gelner, 1997). On one hand, nationalism seeks internal homogenisation, which means the state should only comprise members of the nation. This can be achieved by separatism, assimilation, expulsion, genocide or any combination of those strategies (Mudde, 2000). On the other hand, nationalism also seeks external exclusiveness, which means it attempts to unite all the members of the nation into one state. A population transfer, territorial expansion, or a combination of those two strategies can result in this (Mudde, 2000). Following such a dualistic nature of nationalism, foes are often more numerous than friends. Even though nationalism is not universal (Gellner, 1997), in a modern socio political context it is so common and almost generic due to extensive (or even excessive) conceptual stretching. In order for it to still be something significant and unambiguous, it becomes almost necessary to specify nationalism even further. One of the most relevant and most acknowledged distinctions is made between state and ethnic nationalism (Spencer and Wolman, 1998; Greenfeld, 2001). These are two different strands of nationalism, each parting from a different foundation. For state nationalism, the state is of primary importance. It refers to the nationalism that can be found throughout history, up until the 18 th century when the world was divided
17
into nation states. It has been fundamental and crucial in the successful development of numerous countries (e.g. France, Austria) and international organisations (e.g. NATO), to such an extent that it is not something proclaimed by one specific party family (Billig, 1995). The nation’s core identity is that of a political entity. In other words, it is not necessarily based on a shared ethnic history. For example, the FN upholds strong state nationalistic values, in line with French political culture. This differs from ethnic nationalism, which puts the primary focus on the ethnic unit (e.g. Fennema, 1997). In this case, membership to the state (i.e. citizenship) is limited to one culturally constructed (e.g. ethnic) group. This group is based on a shared ethnic identity, which (unlike citizenship) can remain throughout generations. An example of a party that upholds ethnic (rather than state) nationalist values is the VB. For them the Flemish identity is more important than the Belgian citizenship. For several scholars, nationalism is an indispensible feature of an FRP’s ideological core (Mudde, 1996; 2007; Ignazi, 2003). Therefore, if one aims to conceptualise FRPs, drawing on nationalism as one of its principal indicators, the above differentiation may cause some additional problems. Both of these nationalisms are Weberian ideal types, however, empirically, a combination of both is often found. This means that, despite the usual presence (in varying degrees) of both forms of nationalism, a general conceptualisation of nationalism does not require either strand to its fullest extent. Therefore, a conceptualisation based on nationalism is not necessarily as problematic as originally assumed.
Exclusionism In popular literature and contemporary media, FRPs are often mentioned together with rather negative notions such as racism, discrimination, etc (e.g. Hunter, 1998a; 1998b). However, most often these references are too abstracted, incomplete or even incorrect. Throughout the academic literature in the past three decades, several forms of exclusionism have been discussed and attributed to FRPs. This section provides an overview of those different forms of exclusionism, and will indicate the accuracy of each of those claims. The oldest, and most prevalent form of exclusionism in popular literature and media is biological racism. Racism is an ideology that refers to the belief in hereditary (i.e. natural) and permanent differences between groups of individuals, with the centrality of differences in race (Mudde, 2000, p.187). This includes the additional belief there is a hierarchical order between the different races, which justifies discrimination based on race. Since it is based on specific ideas about humanity and civilisation, racism is often seen as an ideology. In
18
addition, it has the belief these ideas should have political consequences. During World War II, racism was a prominent part of Nazism’s ideological core, but it remained absent from fascism’s ideological core (Traverso, 2001). Nonetheless, throughout the literature, FRPs are often seen as the contemporary successors of these parties because many of the original FRP scholars have their origin in the fields of right wing extremism and historical fascism. Therefore, in the older literature, racism is frequently seen as a part of the ideological core of FRPs (e.g. Mudde, 1996; Billiet and De Witte, 1996). Upon careful study and analysis of party literature and party rhetoric, one can be certain this is not the case (e.g. Mudde, 2000; Rydgren, 2005a). More precisely, the adhesion to racism would be one of the primary reasons not to include a party in the FRP party family (see below for the difference with ERPs). Then, how does such confusion come about? As is very often the case, this is a consequence of a conceptualisation problem. The UN’s definition of racial discrimination (and many others) does not limit itself to discrimination based on race or heritage, it also includes discrimination based on cultural factors, such as ethnicity. Such conceptual imprecision is one of the principal reasons why there remains discussion on whether FRP promote racism. Therefore, it benefits overall clarity when biological racism is differentiated from so called cultural racism or ethno pluralism. These latter terms emphasise differences of a cultural nature, which are well defined and bound entities with specific and strict geographical boundaries between them. According to ethno pluralism, cultures are of equal value but they are different (i.e. there is no hierarchy between different cultural groups) and should therefore not be mixed (Taguieff, 1990; Betz, 1999; Rydgren, 2005a). Alain de Benoist, founder of the Nouvelle Droite and the French think thank GRECE, introduced the concept of ethno pluralism in the 1970s (de Benoist, 1974; 1994). He opposes Islamisation and American style multiculturalism, and supports a civilisation in which ethnicities, cultures and nations live and develop in separation of each other. He claims European cultures are disintegrating and sees ethno pluralism as the solution for this. Even though many of his followers adhere to the FN and Jean Marie Le Pen, he clearly opposes the latter due to his occasional racist and anti Semitic rhetoric (DeClair, 1999; Shields, 2007). Both racism and ethno pluralism are closely related to the concept of xenophobia. This refers to an unquestionable or heightened fear of someone foreign or strange (Mudde, 1999; Andersen, 2000; Ignazi, 2003). It is not necessarily ideological, but mostly psychological. In other words, it is an irrational fear of certain groups of individuals and it is the difference with them one fears. This may or may not refer to ethnic groups; it can also be
19
much broader and refer to anyone outside the own personal identity. In the case of FRPs, the feared out group usually consists of all kinds of minorities, going from Jews and Muslims, to asylum seekers and refugees, to immigrants in general (von Beyme, 1988; Betz, 1994; Fennema, 1997; Van der Brug, Fennema and Tillie, 2000; Van der Brug and Fennema, 2003).
Authoritarianism Authoritarianism is also surrounded by conceptual ambiguity. It is used in a variety of disciplines, all of which define authoritarianism differently. Like most scholars in political science, this study follows the Frankfurt school’s socio psychological conceptualisation. Adorno e.a. (1982, p.157) describe authoritarianism as “a general disposition to glorify, to be subservient to and remain uncritical toward authoritative figures of the in group and to take the attitude of punishing out group figures in the name of some moral authority”. From this definition it is clear this approach is easily combined with other ideologies, such as nationalism and ethno pluralism (Christia and Jahoda, 1954; Kirscht and Dillehay, 1967). Adorno e.a. created the F scale, which indirectly measured the degree of authoritarianism of a personality. The F scale consisted of nine separate components that are added up to a total score, which could then be correlated to various socio demographic variables. Altemeyer (1981) later specifies authoritarianism as the belief in a strictly ordered society where punishment is severe, and refers to it as right wing authoritarianism. He also proposes an improvement of the original F scale. According to him, only three of its components are relevant. Authoritarian submission measures the degree of submissiveness to the authorities that are perceived as established and legitimate in a society. Authoritarian aggression measures the degree of aggression directed against various out groups that are perceived to be targets according to the established authorities. Finally, conventionalism measures the degree of adherence to society’s traditions and social norms, and the belief that others in one’s society should also be required to adhere to above mentioned norms (Altemeyer, 1981, p.147 48). In the past, the F scale has been considered a good indicator for voters’ attitudes and behaviour, as it indicated ‘extremist potential’. However, in recent literature, it has received criticism from various angles. Most commonly, scholars have pointed out the severe empirical issues related to the measurement of the F scale (Ray, 1984; Robinson, Shaver and Rightsman, 1991), the political bias (Kreml, 1977), the proposed correlation between societal processes and individual personality (Boehnke and Hadjar, 2004), and the confusion of psychological variables with political criteria (Jost, Glaser, Kruglanski and Sullloway, 2003).
20
In the end, there appears to be somewhat of a consensus on the three dimensional F scale as proposed by Altemeyer (1981). FRPs are often seen as the natural successors of fascism and Nazism because several of them have (had) roots in fascism or Nazism (e.g. some of the original FN and VB factions) and because both adhere to authoritarianism. At the same time, one has to make two important observations. First, contemporary research indicates that such shared history does not contribute to FRP development in any way (Zaslove, 2004; Coffé, 2004; Art, 2006), which moderates the validity of the successor thesis. Second, one of the crucial dissimilarities is the former’s totalitarian nature, which differs from authoritarianism in its unconditional adherence to order and authority (Altemeyer, 1981, p.152). In addition, authoritarianism does not necessarily imply or preclude anti democratic attitudes, which totalitarianism does. Therefore, both anti system and traditional parties can have an authoritarian core. In contemporary politics, authoritarianism plays an increasingly important role. More specifically, the defreezing of the original cleavages in the 1960s and several post materialist evolutions in the 1970s led to the recognition of a cultural socio political axis, ranging from libertarianism to authoritarianism. Several scholars see the emergence of FRPs, or the shift from libertarianism to authoritarianism, as a direct reaction to those evolutions. Very often, this is summarised as the silent counter revolution (e.g. Ignazi, 1992) or the post materialism thesis (Betz, 1994; Kitschelt and McGann, 1995). Both of these will be extensively discussed in the section on political opportunities (see below).
Democracy and liberal democracy Democracy has been defined by many different disciplines. Its conceptualisation alone can be the subject of an academic study. In this case, however, the thoughts behind democracy and the distinction between democracy and liberal democracy are more important than the debate on the concept itself. Therefore, in order to be concise, the conceptualisation of (liberal) democracy will not be given excessive attention. Generally, democracy refers to a political system that includes judicial equality, due judicial process, civil liberties, human rights and the acceptance of civil society (Dahl, 1989). Whereas this used to be somewhat debated, contemporary literature agrees that parties unwilling to accept these principles are at the fringes of the multidimensional political spectrum and do not play any substantial role in today’s politics (e.g. NPD, EPEN, CP’86, NF, etc.) (e.g. Carter, 2005; Mudde, 2007). In other words, in order to participate in the post war political arena, a party has to accept the rules of the game, i.e. democracy. For example, in 1999 Haider stated, “there
21
exists no alternative to democracy, but there very well exists alternatives to the ruling parties” (Probst, 2003, p.120), indicating he accepts democracy, but not its operationalisation. Within this democratic context, however, FRPs oppose liberal democracy and any evolution they believe will weaken the nation (the in group). Liberal democracy constitutes a democratic context that also provides constitutional rights and liberties for minorities, promotes political pluralism and protects minorities (Eatwell, 2000; Mudde, 2004). Some of the earlier research claims FRPs have anti democratic and anti system tendencies and they refuse to acknowledge the moral principle of equality among men (e.g. Backes and Jesse, 1993; Mudde, 1996; Backes, 2001). However, most contemporary literature agrees that the parties labelled in this study as FRPs oppose liberal democracy, yet they do not proclaim anti democratic sentiments (Carter, 2002; 2005; Mudde, 2007). Hence, it is important to make a clear distinction between these two constructs. To even further dissect the concept of democracy, FRPs also favour the increase of participatory democracy (e.g. direct democracy) and want to limit representative democracy (Manin, 1995; Mudde, 2004). This will be discussed extensively in the following section on populism.
Populism In the literature, the discussion of populism is often referred to as “defining the indefinable” because it is almost impossible to provide an unambiguous definition of populism. Nonetheless, this study provides a comprehensive overview of the most important strands of research in the literature, and which of those are most appropriate for this study. In general, the literature provides three possible approaches to define populism, each of them with a set of advantages and disadvantages. Following the work of Ionescu and Gellner (1969), a first approach is bases the definition of populism on the specific circumstances in which it occurs. This is an appropriate approach for case studies; however, it has little regard for further generalisation. A second approach, based on the work of Canovan (1981), provides an overall taxonomy. It abandons the general concept of populism, and emphasises the variety on the subject. A third approach, based on Taggart (2000), uses a Weberian ideal type (based on common features) to which no populist party is identical, yet they will all resemble. Populism, as a widely discussed topic, is not always part of the discussions on political ideologies. In the party literature, and more specifically in the FRP literature, one can observe a debate surrounding populism. On one hand, populism can be perceived as a political style. It is not a belief one seeks to implement; rather it is a technique or a method that dictates how
22
to successfully engage in politics (Betz, 1993; 1994; Hainsworth, 2008). On the other hand, populism can be analysed as an ideology. It is a set of belief, values and norms a party upholds and attempts to implement (Mudde, 2004; Mudde and Kaltwasser, 2012). This latter approach is fairly recent, however, its value must not be underestimated. Regardless of the approach, most conceptualisations agree on the key juxtaposition of ‘the elite’ versus ‘the people’ (or ‘the heartland’ as Taggart (2000) refers to it). Mudde (2004) provides an all inclusive, yet succinct and unambiguous, definition of populism: “An ideology that considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ versus ‘the corrupt elite’, and which argues that politics should be an expression of the volonté générale (general will) of the people” (Mudde, 2004, p.543) At its core, populism relies on the rejection of two distinct principles: elitism, which is seen as the mirror image of populism, and pluralism, i.e. societal heterogeneity. In addition, populism is defined as more than just a style. Taggart (2000) and Mudde (2004; 2012) see it as a ‘thin centred’ ideology. In other words, populism has a restricted core, attached to a narrower range of political concepts, which means it can easily be combined with other ideologies. Its emphasis on friends and foes and rejection make it moralistic rather than programmatic (Wiles, 1969, p.167). Very often, populism is defined based on its charismatic leadership, however, this is a facilitating, not a defining factor (Van der Brug and Mughan, 2007). Populism aims to change the players and the rules, not the game (i.e. democracy), therefore, it is a reformist, not a revolutionary ideology (Mudde, 2004). In other words, it seeks to change the status of ‘the people’, not ‘the people’ as such.
C. The study’s conceptualisation For Betz (2003) and Mudde (2007), nativism is one of the most central ideological constructs of this party family. Following a concise literature review, Mudde defines nativism as “an ideology, which holds that states should be inhabited exclusively by members of the native group (“the nation”) and that non native elements (persons and ideas) are fundamentally threatening to the homogeneous nation state” (Mudde, 2007, p.22). He sees nativism as a combination of nationalism and xenophobia. Nonetheless, this study does not completely follow the terminology, but it adheres to the underlying premise. More specifically, the study adheres to the hypothesis that FRPs have nationalism and xenophobia at their cores, but it does not adhere to the Europeanisation of the concept nativism. Authoritarianism is defined as an ideological feature, rather than as an attitudinal feature (as Adorno e.a. (1969) and Altemeyer (1981) suggested proposing variations of the F
23
scale). In general it refers to the desire for an ordered and strict society, where violations against the established authority are severely punished. Concretely, authoritarianism is mostly expressed by emphasising the following traits: law and order, increasing police resources, increasing power for the judiciary, longer prison sentences, conservative morality (e.g. abortion, drug legalisation, same sex marriage, etc.) and occasionally also militarism and a preference for the death penalty. This study also adheres to populism as an ideology. It acknowledges the presence of a populist style of politics and a populist rhetoric; however, for FRPs, the study asserts populism as more than that. At the same time, it recognises that populism can never be more than a thin centred ideology (along the lines of Taggart and Mudde). It divides society in two homogenous groups, ‘the pure people’ and ‘the corrupt elite’, and proposes that politics should be an expression of the will of those ‘pure people’. Concretely, pure populism is usually expressed by the demand for better representative power, less intermediate power and more direct democracy. However, most often, populism is combined with other ideologies. Based on extensive first hand analysis of the literature, the party programmes, the rhetoric, the party publications, the debates, etc. (see chapter 3.2) this relies on nationalism, xenophobia, authoritarianism and ideological populism as the ideological core of FRPs. Therefore, this study upholds a sizeable part of the conceptual framework proposed by Mudde (2007). Nonetheless, for reasons that are discussed in the following sections, this study does not adhere to the notion of nativism or the terminology of populist radical right parties.
D. This study’s terminology: far right parties There is no conformity and no consistency in the literature when it comes to the terminology used to refer to this party family. As the literature grows, it becomes problematic that some literature uses multiple terminologies to refer to this party family within the scope of one publication. Therefore, it is important to remember that terminology and precise conceptualisation are important; nonetheless, the consistency of one’s own definition and terminology is even more important. In other words, as long as terminology is properly argued for and used consistently, most criticism will be normative in nature. One of the few general consistencies in the literature is that this party family is referred to as ‘right’, either implicitly or explicitly. Here, the common misconception is to interpret this term in socio economic terms (e.g. Betz, 1994; Kitschelt and McGann, 1995). In the 1980s and early 1990s, this might have been the case for most FRPs; however, currently their socio economic ideology lies closer to a Third way (e.g. Freeden, 1999) or a
24
Christian Democratic (i.e. fairly centrist) ideology (Spicker, 2000; Huntington and Bale, 2002). More appropriately, in this case, right should be interpreted in cultural terms. Specifically, this refers to hostility to modernity, i.e. morally conservative (Eatwell, 1989) and the recognition that individual inequality falls outside of the limits of the state (Bobbio, 1994). For an excellent overview on the history of the so called right and its terminology, see Ignazi (2003, chapters 1 2). The most common term used in the literature (and the media) to refer to this party family is ‘extreme right’. However, this term remains rather ambiguous. Originally, this terminology stems from the German notion Rechtsextremismus , which the German Bundesamt für Verfassungsschutz (Office of the Protection of the Constitution) defines as anti democratic, anti liberal and anti constitutional (Backes and Jesse 1993; Minkenberg, 1998; Backes, 2001). Such a term refers to a particular ideology, which is usually held by those parties that explicitly oppose democracy. Since it indicates extremeness in ideological and political space, it also has a spatial component. Therefore, rechtsextremismus and extreme right have often been deemed interchangeable (Weinberg 1993). On the contrary, however, most parties referred to as extreme right do not fit this definition seeing as they oppose liberal democracy but accept democracy as such. The term ‘radical right’ is also often used, be it by itself or in combination with other ideological references. Much like extreme right, this term is rather ambiguous. Originally, radical right referred to non partisan movements such as the Birch Society and McCarthyism that were active in a specific geographical and time sensitive context, i.e. the late 1940s, early 1950s in the United States (Bell, 1963; Lipset and Raab, 1978). They are characterised by conservative morality and anti communism, rather than anti democracy (Himmelstein, 1990). The literature identifies this phenomenon through individual level personality traits, and not through political values (like in the case of the ideologies discussed above). All this together, makes the term ‘extreme conservatism’ more appropriate for this phenomenon. Furthermore, the literature also uses ‘radical right’ to refer to a counter revolutionary anti modern tradition of thought (Evola, 1969; 1973), that took on violent proportions on occasion. At best, this only refers to a small segment of our intended concept. Historically, the term ‘radical’ was also used to refer to the supporters of the French Revolution, i.e. the ‘left’ (Schwartz 1993; Mudde 2007). Similarly, in German, radikalismus is described as disapproval and criticism on the constitutional order, which can have its origin on the left or on the right (Scheuch and Klingeman, 1967). Today it is still an often used term with respect to left wing and/or progressive parties such as Parti Radical de Gauche and Parti
25
Radical in France, Politieke Partij Radikalen in the Netherlands, and Partido Radicale Italiano in Italy. Therefore, in general, the term ‘radical right’ is not only too ideographic, but also much too broad to describe this phenomenon. Therefore, this study draws on the term FRPs when it references to the party family that serves as its core object. It is a term that, unlike the often used extreme right and radical right, does not have internal ideological contradiction or inherent history that might change its connotation. In addition, the term FRP represents simplicity and is relatively uncomplicated, which is often not a primary concern in the contemporary literature. This terminology also has a minor drawback; more specifically the proposed relativism included in the term ‘far’ runs the risk of not properly delimitating its borders. This would add to the already ambiguous conceptualisation of this concept in the existing literature. However, one must consider that, considering the state of the field, the obtained clarity of this terminology far outweighs the possibly ambiguity. In addition, the primary intention of this chapter has been to provide the reader with a rationale as to why and how this terminology is preferred over others, and how it can bring more precision and transparency to the field.
E. Classification issues An explicit and solid classification scheme is particularly pertinent for those borderline cases that, without such a scheme, could be considered both. The literature describes a wide variety of criteria that can serve as the basis for a classification of FRPs. Generally, these are all based on four possible selection criteria. The first two criteria, party name and international alliance, are designed by the FRP itself. In other words, the foundation of the classification is FRP self identification. The latter two criteria, ideology/policy and origin/sociology, are reflections by external agents, such as journalists, researchers, etc.
Classification based on self-identification If, for the sake of argument, one starts from the assumption that FRPs know themselves better than anyone else, one could assume FRPs are the appropriate agents to identify and clarify their own characteristics. Generally, in the political party literature, two core approaches are most frequently applied to establish a relevant categorisation (Mair and Mudde, 1998; Gallagher e.a., 2005). A first approach is based on a party’s name, whereas a second approach is based on the party’s membership to possible international alliances. The party name approach is classic, especially to indicate party families traditionally on the left. Originally, socio democratic, socialist, communist and Green parties could be
26
easily classified using their names as a reference category because most of them had the name of the party family (or some form thereof) in their party name. However, since the end of communism in 1989, this has become more problematic. Left wing parties still emerged; however, they would often not have any ideological reference in their name due to the stigmatised nature of communism (e.g. the traditional Italian left wing Partito Democratico ). The right also suffered from post war stigmatisation of Nazism and fascism, thereby making name recognition of right wing extremist parties such as Fiamma Tricolore (Tricolour Flame) in Italy, and Nationaldemokratische Partei Deutschlands (National Democrats) in Germany, etc. rather difficult. On the traditional right, party names have always been more difficult to recognise. The label ‘liberal’ or ‘people’ has been used for various ideologies. The label ‘centre’ serves both agrarian parties (e.g. the Centre Parties in Scandinavia) and extreme right parties (e.g. Centre Democrats in the Netherlands). In addition, the right is characterised by so called sui generis labels, such as Fianna Fail (Soldiers of Destiny) in Ireland, the Rassemblement pour la République (Rally for the Republic) in France, Forza Italia (Forward Italy) in Italy, Volksunie (Union of the People) in Belgium, etc. For FRPs, a combination of stigmatisation due to its spatial proximity to extreme right parties, and locally embedded and significant names, have resulted in a difficult recognition of the party family based on their names. A few examples are Vlaams Blok/Belang (Flemish Bloc/Interest) in Belgium, Die Republikaner (The Republicans) in Germany, the Freiheitliche Partei Österreichs (Freedom Party in Austria) and the Dansk Folkeparti (Danish People’s Party) in Denmark. More difficulties emerge when the same name refers to two different party families: the National Front in the UK is an extreme right party, whereas the French Front National is a typical FRP. Therefore, party name might not be the most appropriate criterion to categorise FRPs. An alternative self identification approach is that of membership to trans national organisations. One might assume that, based on ideology, parties that belong to the same party family (and share similar ideas and characteristics) will participate in the same trans national organisations. Such international party links are considered a serious and widely accepted indicator, most importantly because it is based on a party’s own choice. One can find such organisational entities both in a structured (e.g. the European Parliament) and an unstructured environment (e.g. informal international alliances). Such an approach is also subject to several strands of criticism. Most importantly, not all political parties are members of an international alliance or have international links. This is especially the case for smaller parties, like FRPs are often. In addition, the presence of
27
international links and alliances is often dependent on the willingness and the size of a political party. More specifically, when a political party becomes successful and grows in size, it becomes more likely to participate in an international alliance, both from an individual perspective and a group perspective. This is an excellent theoretical classification criterion, however, much like the party name, this poses some empirical problems. More specifically, some parties have multiple international alliances. For example, the Austrian ÖVP is a Christian democratic party, yet it is a member of both the Centrist Democratic International (CDI) and the more conservative International Democratic Union (IDU). Another criticism is that there is little consistency in affiliation, since certain parties have contradictory affiliations through time. For example, the Italian LN had been a member of the regionalist Rainbow Group (1989 1994) before it became a member of the Euroliberals (neither one indicating an FRP affiliation). On several occasions, FRPs have attempted to form a group in the EP; however, despite their continuous representation since the early 1980s this has proven to be problematic due to their nationalistic character (e.g. Fieschi, 2000). Following a mainstreaming strategy, certain FRPs refuse to be associated with other (trans national) FRPs due to their stigmatised nature (e.g. Haider refused any sort of contact with or recognition of Jean Marie Le Pen and the FN). Some of the FRPs, however, were able to form two groups between 1984 and 1994. Ever since, FRPs only managed to create a non ideological (i.e. technical) group from 1999 to 2001 and the extremely short lived ITS Group in 2007. Most of the alliances have been formed outside the European Parliament and often focus on the rejection of the contemporary evolution of European integration (e.g. EuroNat, Alliance of European National Movements, European Alliance for Freedom, etc.). The above criticisms make clear that a classification based on international alliances or party name is not always that evident. For both selection criteria, the case of the FRPs complicates such a classification even more. Based on the above discussion, one can make a claim that summarises possible categorisations based on party name and/or international alliance. When forming a party family, or more specifically a group (e.g. in the European Parliament), political and electoral relevance are sometimes more important than ideological similarities (Bardi, 1994; Andeweg, 1995, p.64). Therefore, in general, self identification might not be the best tool to successfully categorise FRPs.
28
Classification based on external agents Rather than the above described self identification approach, one can also rely on external agents to provide the classification criteria. A common approach for this is the perceived ideology and/or policy FRPs (claim to) propose (Mair and Mudde, 1998). The ideological approach is rare in comparative politics (Janda, 1993), and especially in cross national party politics research. The principal reason for this is the recurring importance of the end of ideology thesis in the late 1980s (Budge e.a., 1987; Ignazi, 1993). The policy approach is difficult to operate as a classification criterion because only a number of FRPs have been in a position to propose policy. Research has been done into FRP influence on traditional parties and their proposed policies (cfr. Verrechtsing ), but this influence is often implicit and happens behind closed doors. In other words, it is difficult to categorise FRPs based on influence if one does not know when and where the influence begins or ends. Regardless of their operational difficulties, if one can observe congruence between the ideologies/proposed policies by different FRPs, this would be a valid argument to consider them part of the same party family. Such an observation can be based on several sources, such as for example expert judgements, legislative behaviour, mass opinion survey, formal policy statements, etc. The analysis of these observations can lead to datasets that are well suited for large n cross country quantitative studies, such as the Comparative Manifesto Project (in general, Budge e.a., 1987; on FRPs, Cole, 2005). This categorisation approach also has several drawbacks. First, there are questions of inter coder reliability (Mikhaylov, Laver and Benoit, 2008). Secondly, the codebooks one uses in large n quantitative studies suffer from conceptual rigidity and restrictions through time. More specifically, this significantly favours short term over long term analysis, and can even become problematic in long term analyses. Thirdly, similar restrictions can be found in the cross unit comparisons. In other words, the comparability resulting from such a quantitative analysis can be an artefact of the method of analysis. Even though more and more FRPs are gaining representation, support minority governments and join governmental coalitions, only a minority has been able to directly contribute to a country’s policy outcomes (Heinisch, 2003; Bale, 2003; Luther, 2009). Therefore it would be preferable and most appropriate to use an ideological approach. Not many studies in party research emphasise the primary importance of ideology and take an ideological approach. Since ideology identifies a rather normative projection, it is important to diversify one’s sources, i.e. use a triangulation approach. It is also important to analyse
29
both internal (e.g. party papers) and external sources (e.g. party manifestos) in order to get a complete impression of the ideology an FRP proposes (Sainsbury, 1980). Nonetheless, one principal problem remains: Does the FRP or an external agent (such as a journalist, researcher, etc.) determine the ideology? Most scholars choose for the latter. Backes and Jesse (1993) describe anti democratic dispositions as the core of parties positioned at the far end of the left right political spectrum. Backes (2001) adds the refusal of equality between individuals as an important indicator. Betz (1994) describes the rejection of socio cultural and socio political systems in addition to the rejection of equality between individuals. According to Fennema (1997), ethnic nationalism, anti parliamentarism, anti materialism and conspiracy theories form their core. Kitschelt and McGann (1995) describe ‘the new radical right’ based on three dimensions, and even provide a winning formula for their success (see below). Originally, Mudde (1996) counted 58 different features among 28 different authors, and later he indicated the vast majority of scholars found nationalism, racism, xenophobia, anti democracy and a strong state (Mudde, 2000). In his seminal work in 2007, Mudde specified FRPs’ main ideological constructs as nativism, authoritarianism and populism. However, much like the terminology, there are many more. An alternative approach is to base the categorisation on the origin and/or the sociology of the FRP and its electorate. This approach parts from the assumption that party family members will have mobilised under similar historical circumstances and with intentions to represent similar interests. Based on cleavage theory by Lipset and Rokkan (1967), parties are part of the same family if they represent similar socio demographic groups, i.e. electorates (e.g. Seiler, 1980; von Beyme, 1985; Mudde, 1998). Such an approach has one fundamental advantage, which is that the categorisation is embedded in broader theoretical patterns of party development theory (e.a. Strøm, 1990). Within this framework, it becomes difficult to observe the evolution of a party family, since electorates constantly evolve. Contemporary electorates do not necessarily fit into the original four cleavage model anymore due to socio political developments, the defreezing of old cleavages (Mair, 1993; 1997) and the development of possible new cleavages (e.g. globalisation, gender, immigration, etc.). A possible solution would be to construct different categorisations for different developmental phases of the FRP. Therefore, this approach might be optimal for an extensive long term study..
30
F. Delineating the borders: Far right parties Regardless of the ambiguity surrounding their conceptualisation and their terminology, most scholars do agree on the FRPs that form the core of the party family. This section provides a brief overview and justification as to why this study considers these parties FRPs. However, the classification of certain parties, and the decision whether to include them or not remains ambiguous and scholars do not necessarily find a consensus. Therefore, it is important to be explicit and transparent about the choices this research project makes and why it makes them.
The most obvious choices The French Front National is widely considered a prototype for other FRPs. Most of the literature agrees on its status within the FRP family. For an extensive overview of the FN and its history, this study refers to DeClair (1999) and Shields (2007). In short, the FN emerged in 1972 and can be considered an FRP since the end of the 1970, when it first adopted the successful core of ideologies discussed above. Nationalism has always had deep historical roots in French history (e.g. Jeanne d’Arc, French Revolution, Napoleon, the Algerian War, etc.) and, on a more individual level, it has been systematically embedded in French Republicanism (e.g. laicité ). The FN’s xenophobia originally stems from the Nouvelle Droite (which was adopted from the French PC, and the British Torries), one of its core factions. Its authoritarianism is expressed through social traditionalism, extensive law and order policies, centralisation, no federalisation and hostilities towards civil liberties. Populism also stems from one of its core factions, Poujadism , which protected ‘the man in the street’ from ‘the elites’ (Hoffman, 1956). In 1999, one of its factions split, formed its own party (the Mouvement National Républicain ), but also adheres to the core FRP ideologies. In general, the Flemish Vlaams Blok/Belang is also considered a prototype FRP (with Ignazi’s original 1992 categorisation as the most notable exception). For an extensive overview of the VB and its history, the researcher refer to Coffé (2005). It originated in 1978 as a nationalist party, which adopted similar ideologies to the FN in the early 1980s. Like many Flemish (right wing) parties, its nationalism takes a rather regional approach. Since its origin, Flemish independence has been one of the VB’s core ideologies. Led by de Winter, this has been combined with xenophobia and opposition to multi culturalism. It proposes similar authoritarian values than the FN. Its constant exclusion from the political game facilitates its dichotomous (i.e. populist) approach to politics. Belgium hosts (or hosted) two FRPs, one in Flanders and one ‘significant’ FRP in Wallonia, namely the Front National (belge) . For an extensive overview of the FNb, the
31
researcher also refers to Coffé (2005). It originated in 1985 as a close reproduction of the French FN (e.g. logo, rhetoric, name, structure, etc.). However, due to the absence of a charismatic leader and a solid organisation (among others), its similarity to the FN was far from complete. Its ideological constructs are similar to those of the VB, however, rather than emphasising Flemish independence, it emphasises state nationalism (i.e. federalism). Generally, it also tends to focus more on xenophobia than the VB (although this can be debated since the 2000s). The Austrian Freiheitliche Partei Österreichs is also generally considered an FRP. Even though it was founded in 1956 and it has always had a nationalist and a liberal faction, one has to specify that the FPÖ only became an FRP since Jörg Haider became its party leader in 1986. For a more elaborate overview of the FPÖ and its transition from neo liberal party to FRP, the study refers to Luther (2008). Since 1986, the FPÖ has been dominated by its nationalist faction, and has taken a strong xenophobic course, emphasising the concept of the Heimat . An inter party conflict led to a split in 2005, and the emergence of the Bündnis Zukunft Österreichs , which can also be considered an FRP. Similarly to the FN and the MNR, there are few important ideological differences between the FPÖ and the BZÖ. Founded in 1995 as a split party from the Danish Fremskridtspartiet , the Dansk Folkeparti adheres to the same ideological constructs as the other FRPs. For an elaborate overview of the DFP and its history, the study refers to Rydgren (2005b). Despite the general consensus on its FRP status, the DFP differs from some of the other FRPs. Contrary to most other FRPs, it has not (yet) experienced troublesome factionalisation. Since its emergence in 1995, a female party president, Pia Kjærsgaard, has led the DFP. It has supported minority governments in the past. Much like the FPÖ in the past, it prefers to keep its public distance from some of the other FRPs. A far less successful or obvious, yet generally accepted FRP, is the German Die Republikaner . A schism from the CSU, the REP becomes an FRP in 1985 under the leadership of Schönhuber. A more detailed overview of the REP and its history can be found in Jaschke (1994) and Minkenberg (1998). Its success has remained mostly regional and local, only exceeding the five percent national electoral threshold once. However, since 1985 it has always adhered to the FRP ideologies. With two parties to its right, the DVU and the NPD 1, the REP is the only German FRP.
1 The DVU is generally considered less extreme than the NPD. However, since their non competition pact in 2004 and their 2011 attempt to form a pre electoral alliance and present one electoral list, it would be fair to say that both the DVU and the NPD are extreme right parties.
32
Some less obvious choices One of the most controversial and problematic parties as far as its classification goes, is the Italian Lega Nord . Several notable scholars refuse to consider the LN an FRP (e.g. Ignazi, 1992; Fieschi e.a., 1996; Gallagher e.a., 2005), whereas other scholars do accept this premise (e.g. Mudde, 2007). Some scholars include the LN as a FRP but base their rationale on a flawed classification scheme (e.g. Betz, 1994; Van der Brug e.a., 2000; Van der Brug and Fennema, 2003; Carter, 2005), whereas others exclude the LN based on a flawed classification scheme (e.g. Kitschelt and McGann, 1995). Upon closer analysis of its manifestos and its literature, and more specifically the changes thereof through time, one can observe an evolution of the LN’s ideological core. Upon its creation in 1991, the LN was a populist, regionalist and relatively neo liberal party. However, in the mid to late 1990s, the LN’s priorities changed significantly: It took a harder stance on immigration, the EU, South Italians and civil liberties. This evolution toward xenophobia and authoritarianism is now complete, and the LN’s core is similar to those of other FRPs. A difference with other FRP is the more neo liberal economic programme the LN proposes due to its relative larger appeal from the middle class. However, most authors agree the economy is not part of FRPs’ ideological core, nor is it a primary reason for voters to identify with or vote for the party (see above). Therefore, this cannot be considered a reason not to include the LN in the FRP family. Another contested FRP is the Schweizerische Volkspartei in Switzerland. Due to its normalisation and the extensive cantonal character of Swiss politics, certain scholars have difficulty labelling the SVP. Like this is the case for the SVP, a party can emphasise different ideological aspects in different cantons, thereby making it extremely difficult to categorise the national party as a member of the FRP family. On one hand, the SVP has a rather populist and culturally traditional character in the Zürich canton. On the other, it has a much more liberal character in the Bern canton. Subsequently, certain scholars use the former faction as the basis for their classification (e.g. Husbands, 2000; Betz, 2004), whereas other scholars use the latter faction (Hennecke, 2003). Since the (implicit) leadership of Blocher, the Zürich faction has gradually increased its influence over the Bern faction. It succeeded in expanding its ideology, its adherence to Blocher, and it significantly increased its influence in the national SVP. On average, the SVP scores 17.9 on Benoit and Laver’s (2006) scale of party positions, thereby putting it on the far right side of the political spectrum. Since the 1990s, the Zürich wing has been dominating a transformation into a ‘new’ SVP, which resulted in the radicalisation of the SVP. Therefore, it would be justified to include the SVP as an FRP
33
from the 1990s on, and especially from 2005 on, when Blocher entered the Swiss government as an SVP representative (Skenderovich, 2005). More often than not, the British National Party is not included in the list of FRPs (e.g. Ignazi, 2003) for two reasons. On one hand, the BNP’s history is not favourable to its perception as a democratic party. More specifically, because of its split in 1982 from the neo fascist British National Front and the leadership by Tyndall until 1999, the BNP is often perceived as a neo fascist or even a neo Nazi party. However, under the leadership of Griffin, the BNP has systematically distanced itself from racism, excessive nationalism, anti Semitism and revisionism. He moderated the BNP’s ideology to that of a democratic, yet still extremely stigmatised FRP (Eatwell, 2000). On the other hand, the BNP is not always considered an FRP or included in any sort of analysis because of its limited (national) electoral success. Much like the REP, the BNP has various local fiefs (e.g. Stoke on Trent) and has successfully challenged for various local and regional seats. The BNP also obtained two seats in the 2009 EP elections. Therefore, this is usually a practical reason not to include the BNP, not an ideological one. The exact same argument can be made for the Centrumdemokraten in the Netherlands, which is generally not included in the FRP family either. In 1984 it emerged as a split from the more extreme and less democratic Centrumpartij en gained minimal success (Lucardie, 1998; Mudde, 2000). The primary difference is the CD’s dissolution in 2002, whereas the BNP is consolidating. Throughout Western Europe several other smaller FRPs have emerged throughout the past three and a half decades. However, not all of them have been able to surpass the stage of single issue or flash party into a consolidated FRP. Especially since the European wide success of FRPs in the early 2000s, several new FRPs have attempted to emerge or have emerged. Unfortunately, academic literature on these parties has been scarce. Most of these parties can be found in Scandinavia. The Sverigedemokraterna exist since 1988. However, it has not been since the election of Åkesson as party president in 2005 that it obtained 20 seats in parliament. This led to a drastic moderation in its ideology in 2011, which know qualifies it as an FRP. The Finnish Perussuomalaiset is a fairly young party (1995) that also adheres to the FRP ideology. In a very short time it has managed to obtain 39 seats in parliament. The Norwegian Fremskrittspartiet developed from an anti tax party to an FRP. They have been steadily consolidating their electoral success since the early 1990, leading to a record high 41 seats in the 2009 parliamentary elections. Its inclusion in the FRP family is, however, debatable since nationalism does not take up a central place in its ideological framework. This is a discussion that remains open for debate.
34
In Spain, there is one regional party that adheres to the FRP framework and also maintains relations with some of the previously discussed FRPs, more specifically Plataforma por Cataluña. Italy knows several smaller parties on the right that are often included in the FRP framework. However, none of them fit the definition of an FRP: Forza Nuova (neo fascist), Fiamma Tricolore (neo fascist), Fronte Sociale Nazionale (neo fascist), and La Destra (national conservative, fitting Kitschelt and McGann’s (1995) winning formula). The Danish Fremskrittspartiet is an anti immigration, anti taxation party. The Dutch Lijst Pim Fortuyn is also often confused with an FRP because it had a strong anti immigration stance, however, it expressed libertarianism rather than authoritarianism. The Swedish Ny Demokrati is also often seen as an FRP, however, it did not uphold nationalism as a core characteristic. Obviously there are several extreme right parties that do not posses the democratic component necessary to be an FRP (e.g. Austrian NDP, German NPD, British NF and Greek GD).
G. Classification used in this study Regardless of these previous classifications and the merits they provide for the study of FRPs, this study uses its own classification scheme. It is based on three broad criteria: a spatial, an attitudinal and a systemic criterion. The spatial criterion refers to the party’s positioning to the right of the traditional parties, but still to the left of extreme right parties (which are undemocratic, anti system parties). One can ask how far a party needs to be from a traditional right wing party, or if there is an exact distance between traditional right and FRP or FRP and extreme right that serves as delineation? However, there is no standardised answer to such criticism. After all, the position of each party is dependent on the demand side factors, the supply side factors and the interaction with other parties (see below). Therefore, it is necessary to complement this relative criterion with two additional criteria. The attitudinal criterion refers to the political values an FRP proposes, i.e. its attitudinal or partisan ideologies. The primary focus here lies on nationalism, xenophobia, and authoritarianism, all as they have been described in the previous sections. It is important here to notice there is no reference to fascism or Nazism, which are both undemocratic constructs. The presence of such ideologies would indicate a party is extreme right, rather than far right, and should be excluded from the party family. This is the principal reason why, for example, the Italian MSI is not considered an FRP in this research project. The systemic criterion refers to the party’s attitude vis à vis the societal system as a whole. According to Kirchheimer (1966, p.237) there are two sorts of system attitudes. One can oppose to the constitutional constraints of a system and express “ a belief system that does
35
not share the values of the political order within which it operates” (Sartori, 1976, p.133), which refers to an anti democratic and anti system party. Examples of such parties have already been given in the form of Nazi and/or fascist parties. Or, one can remain loyal to the system, and oppose the ambition of the system (parties). This is referred to as goal differentiation. FRPs take this latter approach and use populism to criticise the existing system (see above) (Mudde, 2007). Since one of this research project’s main focuses is the interaction between FRPs, and the diffusion between one another, time is a crucial dynamic. The following chapters will clearly show that when an FRP emerges, when it consolidates, and when it adopts the new master frame, is crucial for this study. The use of time as a criteria for classification, will lead to three main categories. The first category is referred to as the innovators. It contains those parties who constructed the new master frame and served as examples for others (e.g. FN). The second category can be referred to as the adopters. It indicates relatively risk taking FRPs that adopt the new master frame following the example of an innovator (e.g. VB, LN, FPÖ). The final category contains the laggards. These are the FRPs that have waited until the last moment to adopt the new master frame, and can be considered rather risk averse (e.g. SD, Finnish PS). They need the success of others before they engage in master frame adoption themselves.
H. Conclusion This section identifies the most important ideological constructs the literature brings forward and uses them to construct its own ideological framework. This framework, based on nationalism and xenophobia, complemented by authoritarianism and populism, serves as the foundation of this entire study. In the literature, the term ERP is most often used to refer to this phenomenon. However, regardless of that term’s popularity, it refers to an anti system and often even fascist ideology, which is incompatible with democracy. This study therefore chooses the term FRP, which refers to a populist approach to democracy and an opposition to liberal democracy. Like for conceptualisation, the design of an appropriate classification and the importance of its implications for the rest of the study should not be underestimated. It is a principal component of any comprehensive analytical framework. Before describing the actual classification, this section identifies an ideological rationale that emphasises party ideology as the rationale behind the classification. FRPs are differentiated from other parties based on a spatial, an attitudinal and a systemic dimension. Furthermore, this study also
36
differentiates within the FRP family, based on a temporal criterion that indicates the developmental progress of an FRP. This results in three categories: innovators, adopters, and laggards. By also taking time into account as the foundation of a classification, a party can be a traditional party at one point in time and an FRP at another. It is important to recognise the intention, and therefore the limitations, of this chapter. Its main purpose is to provide an overview of the existing literature on FRP conceptualisation and classification, complemented with a justification for the conceptualisation, the typology and the classifications used throughout this study. It is not this section’s intention to provide new analyses. It merely serves as a clarification and a specification of some of the key notions of this study. However, this does not negate its importance, because without these sections, this study would be ambiguous, unclear, not delineated, vague and indecisive.
3. Demand-side factors A. Introduction Scholars indicate numerous factors (i.e. variables) that can describe and explain the current development of FRPs, and serve as the basis for explanatory models. Modelled after Eatwell (2003) and Mudde (2007), and in order to design a theoretical framework that is transparent and comprehensive, the factors comprising such explanatory models can be divided into demand and supply side factors. One also has to differentiate between developmental phases. Distinct variables might not be pertinent in all developmental stages or might have a different impact depending on the developmental stage they are active in (Coffé, 2004). This chapter exclusively discusses the literature on demand side factors, sometimes also referred to as breeding ground factors. Such variables explain why and which political systems have the potential for an FRP favourable electorate. Such variables are absolutely necessary for an FRP to successfully develop, however, they are far from sufficient. It does not necessarily result in consolidation and/or success at the party level, making it only part of the equation. It is the supply side factors that provide an explanation for the cross national differences in development and success between FRPs. Especially for demand side factors it is beneficial to differentiate between macro , meso and micro level variables. Most demand side factors are macro factors, which typically refer to broad economic, historical and social processes that take place at an international or supra national level. Nonetheless, meso and micro level variables are also essential. This chapter provides an overview of the scholarship and different some of more important hypotheses the literature proposes as (partial) explanation for FRP development.
37
B. Macro level Hypotheses The majority of demand side factors are macro level variables. The lowest possible level they can have is that of the national level. However, it is not out of the ordinary for demand side factors to be present on the European, international or even global level. Therefore, macro level demand side factors are only rarely FRP specific, which means other parties and movements also suffer or benefit from them (Husbands, 2002). Their biggest advantage is that they can explain “similar developments in very different settings” (Mudde, 2007).
Social Breakdown Thesis Between the 1920s and the 1960s, Lipset and Rokkan’s (1967) original four cleavages were able to describe both party and electoral structures. Since then, several socio political developments led to the so called ‘defreezing’ of the classic socio political cleavages (Mair, 1993; 1997). These demand side evolutions either changed the existing cleavages or helped create new cleavages, i.e. electoral niches (Rydgren, 2005b). New parties, such as various Green parties and FRPs, took advantage of such niches in the 1970s and 1980s. One of the more important demand side evolutions is dealignment, which refers to large portions of the electorate that abandon its partisan affiliation between elections. Typically this can be observed by a decrease in partisanship (e.g. weaker party loyalty), an increase in electoral volatility (e.g. an increase in floating voters) and a decrease in electoral participation (e.g. decreasing turnout) (Franklin e.a., 1992; Putnam, 2000). This is generally combined with an overall loss of faith in traditional parties and national institutions, which by and large indicates a growing interest in more issue based and protest politics (such as FRPs). Another significant demand side evolution is realignment, which refers to drastic changes in the political system or electoral behaviour (read preferences ). Key (1955, 1959) defines this as a change in voters' political allegiances that culminates in the creation of a new majority party that subsequently dominates elections. In the 1970s one can observe such change as a result of the decreasing importance of the economic dimension (especially traditional class interests) and a greater interest for lifestyle (or so called cultural) issues. These socio political developments are often referred to as post materialism or post industrialisation (Inglehart, 1977; Ignazi, 1992). A direct consequence of these two socio political evolutions is usually the increase of individualisation (Bauman, 2001). The public interest (in politics) decreases together with the political participation as a result of people’s disinterest with the general good and class interest. This is visible not only on a political level, but also on a societal level with the
38
reduction of collective bargaining, trade union membership, and the overall collectivisation that defined society in the four decades before that. Individual mobility increased, hence autonomy and loneliness increased as well, resulting in individuals becoming more conscious about their personal situation and less about the overall conditions. Similarly, secularism increased as a result of the defreezing of the religion cleavage and the increasing individualisation of people’s religious experiences. The social breakdown thesis indicates this renders the climate favourable for FRPs because they give increasing attention to cultural factors such as ethnic nationalism, family and other morally conservative values. An FRP that represents such ideologies and deals with such issues and values should therefore appeal to individuals that do not feel represented by traditional parties, and are frustrated with the political situation at that time. Relatively speaking, younger voters are expected to find more appeal in such parties because their partisan attachments have not fully formed yet. Its internal logic is solid, however, one can observe that many of the contemporary FRP voters do not suffer from the assumed social isolation (Norris, 2005). In the 1980s, this thesis was more accurate than it has been since the 2000s. Also, it has been more accurate for some FRPs (e.g. FN) than for others (e.g. LN). This thesis often claims that the impact of an FRP is more substantial in a consociational society (Luther and Deschouwer, 1999). The internal logic of this claim can be justified since the above described evolutions would result in relatively more disenfranchised voters in consociational societies. In practice, FRPs are not always more successful in consociational societies (e.g. the CD in the Netherlands) and not all party systems with successful FRPs are consociational systems (e.g. FN in France).
Modernisation/Economic Interest Thesis One of the principal explanatory factors for the evolution from materialism to post materialism was the increasing wealth people/countries experienced in the aftermath of the Marshal Plan. The modernisation thesis describes how the evolutionary advances in science, technology, production, etc. lead to an increased standard of living and a previously unseen development of society. This increased standard of living resulted in urbanisation, increasing levels of education, rationalisation, increasing individualisation, etc. (e.g. Betz, 1994; Koopmans and Statham, 2000; Norris, 2002). Most scholars (partially) attribute the emergence of FRPs to the modernisation thesis. More specifically, FRPs are strong opponents of modernisation, and attract so called ‘losers of modernisation’, i.e. those who lose in the economic competition over scarce resources
39
and/or those who suffered from some form of relative deprivation (Betz, 1994; Minkenberg, 1998; Decker, 2004). This theoretical claim is confirmed by several empirical studies (e.g. Kriesi, 1999, Fetzer, 2000), while other scholars claim the FRP electorate contains both ‘winners’ and ‘losers’ of modernisation (Betz, 1993; Flecker, 2007). In other words, the empirical evidence for this claim is rather contradictory. The internal validity of this thesis is rigorous, however, it is theoretically too general and empirically too flawed to exist as the sole explanatory thesis for FRP development. The exact linkage between this thesis and successful FRP development remains unclear. Several scholars even claim that, based on this thesis, the most likely FRP voters, do not vote for FRPs (Mayer, 2002; Norris, 2005). Even though, in the 1980s and the early 1990s, considerable shares of FRP voters were modernisation losers, the electoral group of modernisation losers was much larger than only those who casted an FRP vote (Minkenberg, 2000). In addition, the modernisation process was similar in the entire Western democratic world, yet only few democracies see the successful development of an FRP, thereby indicating this can only be a contribution to the overall explanatory model
Globalisation Thesis Originally, the modernisation thesis was developed in and for the 1980s; however, several scholars have reformulated this thesis so as to adjust it to more contemporary socio political evolutions (e.g. Minkenberg, 1998; Holmes, 2000). One such reformulation can be referred to as the globalisation thesis (Swank and Betz, 2003). It claims FRPs will successfully develop and obtain votes from those who fear or suffer from economic change, i.e. globalisation. Economic and cultural globalisation lead to unemployment and McDonaldisation in those industries that are vulnerable to foreign competition (e.g. blue collar workers) and public sector employees that are likely to be hit by tax and state expenditure cuts (Kitschelt and McGann, 1995; Minkenberg, 2000; Kriesi e.a., 2005). In other words, the so called losers of globalisation are the most likely FRP voters. However, just like the modernisation thesis, there are some possible problems with the explanatory value of the globalisation thesis. Most importantly, there exist several difficulties (mostly methodological, but also theoretical) trying to connect a macro level phenomenon like globalisation to individual level votes for an FRP. The use of multi level models can accommodate such dangers for an ecological fallacy, however only few scholars have attempted this (cfr. Amsterdam school ). In addition, the empirical evidence for this claim is
40
rather limited. Does globalisation account for a new cleavage or is this the evolution of an existing cleavage? Either or, when did globalisation start? Regardless of these theoretical and empirical limitations, the globalisation thesis has the possibility to explain differences between geographical regions since different regions might experience or react to globalisation differently. This is an additional strength, compared to the social breakdown thesis, which was a West European evolution. However, by itself, neither the globalisation nor the modernisation thesis can explain the differences within a certain region (e.g. why is there a strong FRP presence in Italy, but not in Spain?).
(Reverse) Post-Materialism Thesis By the end of the 1990s, post materialism lost some of its significance as an explanatory voting mechanism. The progressive freedoms that had been advocated during post materialism became threats to traditional values, and mainstream elites are blamed for such social liberalisation. This evolution shifts the focus from the left libertarian side of the cultural axis, to the right authoritarian side of the axis. At the same time, the increasing importance of globalisation and job insecurity reverted some of the concerns (and voting mechanisms) back to the original Downsian dimension. The (reverse) post materialism thesis describes FRP development to be more successful (and significant) there where post material voting is persistent. After all, there where post materialism is most present is where socio political evolutions like globalisation will create most discontent and rejection. Following this thesis, Kitschelt and McGann (1995) claim a society needs to obtain or even surpass a post industrial state in order to experience the successful development of an FRP. More specifically, some scholars claim there is a strong relationship between an open economy and the potential for successful FRP development (Swank and Betz, 2003; Veugelers and Magnan, 2005) This thesis is subject to the same lines of criticism as the modernisation thesis. However, there are two important differences. First, this thesis has much more explanatory value in Eastern Europe as a result of the often local nature of the FRP vote in Western Europe (e.g. Southeast of France, Antwerp in Belgium, Carinthia in Austria, etc.). Second, contrary to the modernisation thesis and very similar to the globalisation thesis, this thesis has some explanatory value for differences in FRP development between regions (e.g. South versus North, East versus West).
41
(Economic/Political) Crisis Thesis The literature emphasises the advantageous role a crisis can play in the successful development of an FRP (e.g. Taggart, 2000). However, the term “crisis” has to deal with severe conceptualisation and operationalisation problems. Contrary to the previously described theses, it is the theoretical foundation and its selected proxy indicator that are often weak and problematic, and it is the empirical framework that is quite solid. Unfortunately, most scholars ignore the conceptualisation issues and focus on the empirical evidence. The economic crisis thesis claims a causal connection between the state of the economy and FRP electoral success. Most often, unemployment is used as a proxy indicator of the economy. The literature, however, does not find a consensus as to whether this connection exists, and if so, in which direction it exists. Such differences can partially be explained by the lack of a proper conceptualisation and the use of a wide variety of different data sources and methodologies. Some scholars claim there is a positive correlation between the two (e.g. Jackman and Volpert, 1996), i.e. the higher unemployment is, the higher the FRP vote share will be. Some scholars find no correlations whatsoever between the two (e.g. Givens, 2002; 2005). Some scholars also find a mediated effect of certain welfare provisions (e.g. Swank and Betz, 2003) or immigration (e.g. Golder, 2003). Most scholars, however, find a negative connection between the two (e.g. Knigge, 1998; Lubbers, 2001; Wendt, 2003; Arzheimer and Carter, 2006), i.e. the lower the unemployment, the higher the vote share for an FRP. This is not surprising since FRPs do not posses issue ownership over socio economic issues. Therefore, the decline in issue salience of the economy (i.e. the decline of unemployment as part of a crisis) should logically increase the electoral appeal of an FRP. The political crisis claims a causal connection between the socio political state of society and FRP electoral success. Most of the studies that analyse this thesis use political dissatisfaction or political discontent as a proxy indicator for the socio political state of society because surveys and opinion polls have shown record highs for political distrust all over Europe since the 1990s (Norris, 2002; with the exception of Denmark: see Anderson, 2002). Here as well, the literature does not find consensus on whether the correlation is positive (Knigge, 1998; Van der Brug, 2003) or whether a correlation as such even exists (Norris, 2005). This thesis has one serious problem. The correlation on the micro level, i.e. voter dissatisfaction, is plausible, however, this is not the case on a macro level, i.e. political crisis. In order to credibly make such a claim, it is necessary to find another proxy indicator for
42
political crisis. In the literature, two important macro level alternatives can be found. Some scholars claim patronage encourages medium support (and the vote) for right wing authoritarian and populist anti state parties (Helms, 1997; Kitschelt, 2002). This claim is empirically tested and confirmed by using clientelism and corruption as a proxy for political crisis (Veugelers and Magnan, 2005). Others scholars claim political crisis can be linked to the consociational nature of a party system (Kriesi, 1995; Andeweg, 2001; Evans and Ivaldi, 2002; Papadopoulos, 2005). According to them, consociational societies are more prone to populism since the degree of cartelisation is high and the distinction between left and right low. This claim regarding populism is correct, however, this does not automatically mean consociational systems are a breeding ground for FRPs.
Immigration thesis In the literature, this thesis has been referred to under different names, most notably the single issue thesis (e.g. Mudde, 1999) and the ethnic backlash thesis (e.g. Mudde, 2007). The immigration thesis claims that FRPs will do especially well at times when there are major concerns about new immigration. It sees FRPs as a response of the majority to a perceived threat, be it ethnic, cultural or even economic, by a minority (e.g. Veugelers and Chiarini, 2002; Wendt, 2003). Here more than for other theses, the diversity of proxy indicators used for the “threat by the minority” has much to do with this. Studies have used foreign born citizens (e.g. Golder, 2003), new immigrants (e.g. Lubbers, 2001; Swank and Betz, 2003), immigration (e.g. Husbands, 2001), asylum seekers (e.g. Wendt, 2003), ethnic polarisation (e.g. Evans and Need, 2002), etc. The majority of the empirical studies find a positive correlation between their respective proxy indicators for “threat by a minority” and the successful development of an FRP (Knigge, 1998; Lubbers, 2001; Golder, 2003; Swank and Betz, 2003). Few studies find no relationship between the two (e.g. Kriesi, 1995; Dülmer and Klein, 2005), whereas certain studies even find a negative correlation between the two (e.g. Wendt, 2003). So, in short, empirical analyses are rather inconclusive, mostly as a result of the differences in proxy indicators, and the different dataset and methodologies used in the analyses. Most FRPs do not only emphasise immigration, but they also have much broader ideological issues that attract voters. On some occasions, immigration has not been an issue at all (e.g. LN in the early 1990s). In addition, there appears to be no correlation between FRP success and the current waves of immigration, indicating FRPs might not be a reaction to such waves. Thus, the immigration thesis has difficulties to explain temporal and spatial
43
differences in FRP development (Mudde, 1999). This can be observed by the propensity to engage in case studies to analyse the immigration thesis, rather than rely on the advantages of comparative studies (with the foremost exception of Norris’ (2005) seminal study).
C. Meso level Hypotheses The second demand side dimension comprises meso level factors that could help provide a fertile breeding ground for FRPs to successfully develop. However, this dimension is extremely underdeveloped and understudied. Due to the extensive time and cost requirements of meso level research, few empirical studies have been done thus far. Unfortunately, the absence of structural datasets or analytical possibilities sometimes leads to shortcuts, such as the study of macro level data at the meso level. When this is not done appropriately, this can result in an ecological fallacy or a fallacy of composition. The few studies that have used meso level explanations are worth noticing. Based on several survey studies, Eatwell (2000) claims a significant negative relationship between the level of education and the FRP vote. In other words, the lower one’s education level is, the more likely one is to vote for an FRP. The autocorrelation of the education variable with income and possibly even socialisation (or a more extreme claim, genetics) can become problematic. Adorno e.a. (1969) claim that those raised by an authoritarian father are predisposed for authoritarian attitudes. In their study this was used as a proxy indicator for a fascist vote. Since FRPs also have authoritarian constructs at their core, one could also use this as a proxy indicator for an FRP vote. Despite its many reformulations (see above), this correlation has been empirically and theoretically discredited (e.g. Stone e.a., 1993; Martin, 2001). Klandermans and Mayer (2006) have done the most elaborate study that uses a meso level approach to the study of FRPs. They claim the principal FRP attitudes are the result of a socialisation process by families that are favourable to FRPs. In other words, this study claims there is positive correlation between families that are favourable to FRPs (and cast a vote) and successful FRP development. However, contemporary individualisation processes, combined with a decreasing birth rate, make it difficult to generalise this claim to an entire electorate. Due to its limited number of cases it is also difficult to generalise from it. In addition, the proposed explanatory factors are rather structural and are unable to explain the FRP success in the 2000s.
44
D. Micro level Hypotheses Most actual empirical tests take place on the individual level. Such tests attempt to identify and analyse why certain groups of people vote for an FRP by establishing a correlation between individual level characteristics and voting behaviour. In order to avoid any of the previously discussed fallacies, it is import to analyse these appropriately. In this section, the researcher provides an overview of the three most prominent individual level theses.
Attitudinal Thesis Due to its inherent logic, the attitudinal thesis is most likely the most researched thesis in FRP research (especially on the micro level). The attitudinal thesis claims that successful FRP development is a direct consequence of the prevalent FRP ideology among the individual members of the electorate. In this case, FRP views, attitudes or values (usually measured with survey data) are used as a proxy indicator of FRP ideology. Here also, this is done with a variety of data sources and methodologies. Nonetheless, the vast majority of the research upholds this thesis (e.g. Van der Brug e.a., 2000; 2005; Van der Brug and Fennema, 2003). More FRP voters are nationalist compared to the traditional electorates and they are relatively more nationalist than traditional electorates. Similarly for authoritarianism, more FRP voters are authoritarian and they are more authoritarian than the traditional electorates. The analysis of populism on a micro level is slightly more complicated and encounters some operational problems, although Rooduijn and Pauwels (2012) have indicated it is possible to provide a measurement that is reliable and valid in a comparative context. Generally, protest voting and/or abstentionism are used as proxies for populism. The empirical foundation of this thesis is usually very reliable due to the widespread availability of survey data. Unlike most of the previous theses, its problems are usually of a more theoretical nature. Most notably, as a result of the abundance of data, one has to be cautious not to confuse correlation with causality. In order to claim causality, one must not only have correlation (i.e. non spuriousness), but also cause before effect, longitudinal data and a solid theoretical foundation. Also, the relative character of a conceptualisation (most notably its spatial criteria) makes any cross national comparison difficult. Perhaps most problematic is the widespread nature of FRP attitudes in the general electorate. The percentage of the electorate with FRP attitudes far outweighs an FRP’s vote share (Norris, 2005). Simultaneously, empirical studies indicate the absence of FRP attitudes in FRP voters and the non FRP vote of the majority of those voters with FRP attitudes, which fits within the larger debate of rational voting (Lubbers, Gijsberts and Scheepers, 2002). It is
45
very challenging to study whether an individual casted a rational vote, rather than a strategic, emotional or protest vote. Specifically, the literature does not necessarily confirm that people vote for an FRP because of their FRP attitudes (Mudde, 1999).
Insecurity Thesis Since the defreezing of the original cleavage structure, most countries have experienced a significant decrease in trust and confidence (e.g. Almond and Verba, 1989). The insecurity thesis claims that due to the before mentioned macro level developments, certain electorates have become insecure about certain aspects of their lives. FRPs provide a collective identity (see above discussion on in and out groups) and propose security for the future. In other words, similar to the macro level post materialism thesis, successful FRP development is a materialist counter revolution against the economic insecurities of today’s society (Minkenberg, 2000; Christofferson, 2003). Concretely, welfare chauvinism is often used as a proxy variable for insecurity because it combines the fear of an out group (i.e. xenophobia) and the lack of identity or economic discomfort of an in group (i.e. nationalism). Even though this is often thought as the principal rationale to cast an FRP vote, there is no strong empirical evidence that supports this thesis. The FRP electorate does not necessarily hold welfare chauvinistic attitudes (compared to the general electorate), nor are welfare chauvinistic attitudes central to the FRP vote choice. Even more, welfare chauvinistic attitudes only play a marginal role in motivating the electorate to vote for an FRP (Coffé, 2005).
Protest Thesis Rational voting, and its validity as a post Lipset and Rokkan justification for vote choice, is often very debated (Green and Shapiro, 1994). The literature has described several alternatives, and with regard to the development of FRPs, one of those is particularly relevant: the protest vote. In a post industrial era, the electorate is becoming increasingly dissatisfied, resulting in an increasing number of protest votes. The protest thesis claims FRPs are the main beneficiary of these protest votes, which explains their successful development. In other words, this thesis explicitly rejects a vote in support of the FRP ideology. Once again, due to the extensive availability of survey data, one can find numerous proxy indicators for a protest vote throughout the literature: decreasing turnout, decreasing vote share for traditional parties, decreasing political activism, etc. All of these indicate that ‘traditional’ politics is becoming less prevalent, whereas unstructured and transient voting is
46
becoming more customary. In other words, following the dual nature of the protest thesis, people are voting less based on ideology (or support) and they are predominantly voting based on political resentment (Ignazi, 2003; Mayer, 2005). In addition to the increase of protest voting, one can also observe the relative stability of FRP electorates (compared to those of traditional parties). Following this incongruity, most of the literature specifies that it is important not to interpret the protest and the ideological vote as mutually exclusive (Swyngedouw, 2001) but to acknowledge and accept they can appear simultaneously. Concretely, during its emergence, the majority of FRP votes are protest votes (predominantly in the 1980s and early to mid 1990s) (Arzheimer, 2009). However, if an FRP wishes to persist electorally and consolidate its support, it needs to transform these protest votes into support votes, which is what the majority of the ‘older’ FRPs have done in the 2000s (Betz, 2002; Schmidt, 2003).
E. Conclusion The study of the demand side factors takes up a significant and important part of the FRP literature, most notably in the 1980s and 1990s. One can find a wide variety of research designs, data sets and methodologies throughout this literature. However, unlike for the conceptualisation issues, there is no dominant approach and there is a genuine lack of consensus, which rather often leads to important contradictions (for almost each and every thesis). In addition, the lack of proper data and the lack of originality often lead to ecological fallacies and fallacies of composition (mostly on the meso level). The above chapter gives a brief overview of some of the most important hypotheses that contribute to the demand side framework for FRP development, together with some of the literature that discusses these hypotheses. As has become clear from this chapter, the contradictory results do not make the analysis of the demand side factors evident. Therefore, one must make two remarks. First, only few studies take into account each thesis discussed in the above chapter, which often results in studies being rather ad hoc. Second, in order to be complete, an explanatory framework should also include supply side factors. Regardless of their extensive empirical problems, the importance of demand side factors should not be underestimated. They can explain why FRPs are likely to emerge under certain circumstances and which electorates might potentially vote for FRPs. These factors need to be beneficial for an FRP to be able to emerge. In other words, in order to develop successfully, an FRP’s breeding ground must be fertile. This is an absolute, yet not sufficient requirement. To complement these factors, supply side factors explain why FRPs emerge and
47
who votes for FRPs. Their inclusion in an explanatory framework allows for the explanation of geographical differences.
4. External Supply Side A. Introduction A fertile breeding ground is necessary for the emergence of an FRP, however, by itself, it is not sufficient to explain FRP development. If an FRP wishes to surpass the emergence stage, favourable demand side factors alone are not enough. More specifically, once an FRP has emerged, the demand side factors become less significant. A demand for an FRP does not necessarily result in long term success and consolidation. In order to achieve this, one has to analyse internal and external supply side factors as well because it is these factors that translate the demand side factors into actual specific party politics. The external supply side factors, or political opportunities, can explain why FRPs have successfully developed in some political systems and failed in others. They can help explain the temporal and spatial differences demand side factors cannot. They can facilitate or impede the development of an FRP; they do not determine its development. Once an FRP has successfully emerged, it has the possibility to influence these external supply side factors and change them in a more favourable direction. Since the 2000s, the literature has given this concept, its definition and its operationalisation, increasing attention. This section extensively discusses the terminologies and the conceptualisations used throughout the literature. It justifies the choice made in this study to refer to this concept as political opportunities, rather than the more common political opportunity structure. The literature does not provide a unilateral operationalisation that would allow this study to describe and measure external supply side factors. Therefore, this section also includes an overview of the operationalisation schemes used throughout the literature. In a final section, specific attention is given to the facilitating and impeding nature media can have.
B. Terminology and conceptualisation The concept of political opportunities originally comes form the social movement literature, where it is usually referred to as political opportunity structure (POS). Since significant parts of the POS are not necessarily structural, but rather perceived, situational and dynamic, its definition is somewhat of an oxymoron (Eisinger, 1973). The structural sources of FRP development are mostly represented by socio economic and cultural factors, which have been
48
discussed above in the section on demand side factors (see also chapter 3.1). Therefore, it would be more appropriate to refer to them as political opportunities. Despite the general conformity on the terminology, there is no consensus on the broader conceptualisation of political opportunities. Since each factor that has the possibility to facilitate or impede FRP development could technically be labelled a political opportunity, its conceptualisation becomes particularly delicate. On one hand, if it is defined too narrowly, the concept becomes increasingly inadequate and implausible, resulting in a loss in explanatory power (McAdam, 1996). On the other hand, if the concept is defined too broadly, it becomes trivial and possibly tautological, also resulting in a loss in explanatory power. Then, it “is in danger of becoming a sponge that soaks up every aspect of an organisation’s environment” (Gamson and Meyer, 1996, p.275). This dilemma indicates why there is a lack of scholarly consensus regarding the conceptualisation of political opportunities (e.g. McAdam, McCarthy and Zald, 1996; Tarrow, 1998). Some of the more notable scholars have provided precise accounts of political opportunities. Kitschelt (1986, p.58) describes political opportunities as “specific configurations of resources, institutional arrangements and historical precedents for social mobilisation, which facilitate the development of protest movements in some instances and constrain them in others”. Meyer and Minkoff (2004) define political opportunities as factors exogenous to the FRP enhance or inhibit prospects for mobilisation, for specific claims to be advanced rather than others, for particular strategies of influence to be exercised and for movements to affect mainstream institutional politics and policy. Like McAdam e.a. (1996), these descriptions emphasise the relatively dynamic character of the relationship between a political system and movement mobilisation. Such conceptualisations, however, remain rather general. Therefore, this study relies on the conceptualisation provided by Tarrow (1994, p.85): ”consistent – but not necessary formal, permanent, dimensions of the political environment that provide incentives for people to undertake collective action by affecting their expectations for success of failure”. It is more detailed and it explicitly emphasises resources outside the movement (or in this case FRP) that can either open or close the political space for (further) mobilisation. The inclusion of political opportunities in the theoretical framework for FRP development has only increased the multiplicity of its conceptualisations (e.g. Kitchelt and McGann, 1995; Minkenberg, 1998; Decker, 2004; Rydgren, 2005b). Mudde (2007) reviews the literature on political opportunities in an FRP context, and provides his own classification based on four overlapping contexts: an institutional, a political, a cultural and a media
49
context. His suggestion, however, is too restricted since political opportunities go further than the theoretical framework he proposes. This study proposes a more elaborate approach. First, more conceptual attention to the mechanisms that drive political opportunities and the analytical road to collective action (i.e. mobilisation) could benefit its general understanding. Second, more attention to the importance of issue or constituency specific aspects of political opportunities, and their impact on the overall process of FRP developmental, as opposed to the emphasis on general political opportunities. Third, too often studies try to explain different dependent variables with the same political opportunities. This is something that should be avoided, and different studies should have different political opportunity frameworks. These challenges will be incorporated in chapter 3.1 when political opportunities are discussed in more detail.
C. Operationalisation: The multi dimensionality of a concept Since the conceptualisation and delineation of political opportunities are quite complicated, its study is forced to deal with all sorts of uncertainty and ambiguity. Therefore, this section provides a concise overview of some of the operationalisations used throughout the literature, together with a more extensive discussion of this study’s operationalisation, which finds its foundations in the works by Kriesi e.a. (1992; 1995) and van der Heijden (2010). It provides a solid foundation for further use throughout this study. The main purpose of the political opportunities is to explain spatial and/or temporal variation in FRP development. Kitschelt and McGann (1995) provide a relatively comprehensive study about the role of political opportunities in FRP development. However, they do not allow the framework to precisely determine the influence of the independent variables on the FRP vote. Jackman and Volpert (1996) include the electoral system, the party system and several socio economic factors in their analysis. Conversely, they fail to include any socio economic variables. Knigge’s (1998) study takes the opposite approach. She only uses socio economic variables and fails to include (among others) electoral and party system variables. Abedi (2002) only includes party system factors, ignoring the influence of socio economic and institutional factors. Like Kitschelt and McGann, Lubbers’ (2002) study is very comprehensive. However, it suffers from severe methodological, spatial and case selection problems. Finally, Carter (2005) provides a framework that includes party ideology, party organisation, party competition and the institutional environment. According to Kriesi e.a. (1992; 1995) and van der Heijden (2010), five broad dimensions constitute the operationalisation of political opportunities for FRP development.
50
First, the status of the socio political cleavage structure is an instrumental factor for any emerging political party (Lipset and Rokkan, 1967). As previously discussed, post industrialisation and post modernisation led to the ‘defreezing’ of traditional cleavages, which was symbolised by dealignment and realignment processes. The growing discontent with the elites and the decreasing socio partisan identification (Putnam, 2000) led to a more volatile electoral arena and a more open political space (Rydgren, 2005b). Politicisation of new issues, agenda setting and mainstream parties’ strategies (Meguid, 2008) resulted in the relocation of the competition between parties toward new and salient cleavages (Duyvendak and Koopmans, 1992; Perrineau, 1997). Together with traditional parties’ failure to adjust to this development, this contributed to the issue ownership FRPs generated over ‘new’ issues like immigration, law and order, and welfare chauvinism (Rydgren, 2004a; Meguid, 2008). A second important dimension of political opportunities is the institutional state structure, i.e. the points of access to the existing political system. The levels of vertical decentralisation (i.e. federalism) need to be favourable, together with a minimal level of horizontal concentration of state power between the different state branches. The electoral system needs to be accessible to new and small parties and offer them the possibility to influence political decision making procedures. Other important access points are the availability of direct democratic structures (e.g. referenda), and the access to state subventions and state funding (see Carter, 2005). A third political opportunity dimension constitutes the (informal) strategies of traditional parties vis à vis FRPs. On one hand there are inclusive strategies by which the traditional elites seek to integrate FRPs (e.g. facilitation, cooperation, assimilation, etc.). On the other hand, there are exclusive strategies by which traditional elites seek to exclude FRPs (e.g. polarisation, repression, confrontation, etc.). The combination of these two leads to a degree of presence or absence of elite allies, which has proven to play an important role (see Meguid (2008) for an extensive overview of the different strategies and their consequences). Directly following this, the nature of a party system’s coalition format and the available political space (Minkenberg, 2001) also has a sizeable influence on FRP development. A fourth dimension consists of the power relations within the existing party structure. Social movement theory states that a majority of social movements are more reserved in their opposition in periods when traditional left wing parties are in government, and more outspoken when those parties are in opposition (e.g. Kriesi e.a., 1995). This argument finds its rationale in the post materialist values most social movements share with traditional left wing parties. Subsequently, it would be plausible to reframe this for FRPs, stating their
51
opposition will be more moderated when culturally right wing parties are in power, which means the possibilities for access to power and political influence increase. This reformulated hypothesis is intuitional, however, it has not been tested empirically yet. Nonetheless, one can make the following observation: the few instances when FRPs have participated in government have always been in coalition with culturally right wing parties. This implies, or at least indicates, an inherent ideological affinity of FRPs to form pre electoral agreements (e.g. LN), post electoral coalitions (e.g. FPÖ) or support minority governments (e.g. DFP) that include traditional right wing parties. A fifth dimension is the mass media structure. Despite its fairly novel nature (since the 1970s in Europe), mass media’s role in politics is already considerable. Gradually, but progressively, its societal importance has increased, which also increased its role in party development processes. For FRPs, the access to the media has become an important dynamic, particularly when publicising its ideology or familiarising the electorate with its ideas and personalities (Carter, 2005). Like all other political opportunities, it can have a facilitating or an impeding effect. Its most important societal functions are agenda setting, status conferral, stereotyping, priming, gatekeeping, and framing (Koopmans, 2004). In short, the media select and allocate prominence to the public discourse of their preference. They produce the first and most basic selection mechanism in the public sphere, namely visibility, which influences the further transmission and diffusion of the public discourse (see section below).
D. Mass media Due to the increasing importance of the media, and the inability to consistently include this dimension in the political opportunities framework for FRP development, this section provides a more elaborate analysis of mass media, the role it plays in FRP development, and more normatively, why this should always be included, especially in an explanatory model for FRP development. Following a brief overview of the modernisation of the media, this section will mainly focus on three fundamental dimensions of mass media: the different functions the media holds, the different paths to media access, and the content of the media. As a result of its increasing societal importance, media’s overall position in politics and its role in FRP development have unquestionably become more prevalent. One of the more important roles media occupies is that of agenda setting, which maintains that steady and cumulative media attention to an issue influences the public’s perception of that issue (Cohen, 1963; McCombs and Shaw, 1972). In other words, it allows for issues to transfer
52
from the media’s agenda to the public’s agenda, thereby giving the media the possibility to influence issue salience and issue ownership (Budge and Farlie, 1983). The media also determines to what extent, and at what time, media attention will be given to individuals, i.e. status conferral (Koopmans, 1996). This function can influence the social standing of politicians, their authority and their legitimacy. Therefore, regardless of the influence on individual politicians, status conferral can have an important influence on the appeal of a party (Lazarsfeld and Merton, 1948; Simonson, 1999). The media provides a forum or a platform for a politician to diffuse the party’s ideology and to debate with other politicians. Since FRPs often have a charismatic leader, such a personalised relationship with the media is usually considered to be favourable for FRPs (e.g. Lepenism, Haiderisation, etc.). As a public actor, media has the power to create, modify or disregard stereotypes (e.g. Dyer, 1993). Like media, a stereotype is a cognitive shortcut that one uses to simplify the complex world (Van Hauwaert, 2012). This function, however, can easily lead to prejudices, inflexibility, stigmatisation and scapegoating (Taguieff, 1984) through the spread of dualistic ideologies and rhetoric such as xenophobia and nationalism. This is facilitated through the use of populism. Therefore, the social media function can be very beneficial for FRPs and their successful development. A related, but more general concept is that of priming. Media can broadcast images, content, formats, etc. that stimulate related thoughts in cognition of consumers (Krosnick and Kinder, 1990; Miller and Krosnick, 2000). Based on such broadcasts, and by the activation of related knowledge, individuals will evaluate an issue, an individual or a party. Recently one can observe a media evolution from unintentional to intentional priming, or from influence to manipulation. Among others, this can be attributed to the increasing presence of political advisors hired by politicians to sway political opinion, or so called spin doctors. More generally, media can influence and control the access certain political parties or individuals have to the public sphere. Koopmans (2004) refers to this as the media’s gate keeping function. Media have the ability to select, shape, amplify or diminish the public messages political parties or individuals have, partially depending on what the media believes is desirable. They have the authority to allocate prominence to information as they see fit, and to select the information they consider important. Koopmans and Olzak (2004) claim this selection bias is dependent on visibility, resonance and legitimacy of the information parties or individuals seek to diffuse. Not only do media provide us with content, they also provide the public with frames. Such framing processes provide schemata of cognitive interpretations on which individuals
53
rely in order to understand and respond to social reality (Snow e.a., 1986; Benford and Snow, 2000). The media’s framing role consists of three separate functions: a diagnostic, a prognostic and a motivational function (Snow and Benford, 1988). Respectively this corresponds to the expression of an issue, the specification of possible solutions and a rationale as to why this is important. The ultimate objective of this function is to increase the salience of certain content by using primary frames. An FRP’s access to the media has become particularly important due to the media’s increasing presence, reach and influence in the socio political system (Semetko, 1996). Today, media (and more specifically television) comprise some of the more fundamental campaign instruments and it is the electorate’s primary source of information about politics and elections. However, it should be noticed there is a considerable degree of variation between, and sometimes even within, political systems as to the general importance of the media and which functions are most prevalent (Carter, 2005). Nonetheless, media is an integral part of an FRP’s political opportunities. It is a very solid theoretical argument, however, empirically there is no clear pattern that indicates a lack of access to the media impedes successful FRP development (e.g. Carter, 2005). One can observe successfully developed FRPs that have easy media access (e.g. Denmark, Belgium and Norway) and more difficult media access (France and Austria). One can also observe countries where media access is relatively easy and an FRP has not successfully developed (e.g. Germany and Spain). However, despite some empirical indications, one major questions remains: Is an FRP’s access to media the cause or the result of its successful development? Perhaps it is not necessarily the access and its facilitating nature that are important for successful FRP development, but rather the content of media. FRP discourse and rhetoric, and their degree of moderation, are crucial factors (see internal supply side factors) and can have an important impact on the success of an FRP (e.g. post 2004 VB, post 2011 FN, post 1999 FPÖ, post 2011 LN, etc.). It becomes clear that certain types of extremist rhetoric are not appropriate if an FRP desires to become successful. Often, the ideological moderation is included in a party’s internal supply side factors (see below). To conclude, it does appear that positive media coverage favours FRPs and is beneficial for their development. On the contrary, negative media coverage is not necessarily detrimental to an FRP’s successful development. It appears the saying “any publicity is good publicity” can be applied here to some extent. At the same time, it is perfectly possible for the media to be both “friend and foe” (Mudde, 2007). Which role media will fulfil largely depends on the interaction of traditional parties with the FRP (see Meguid, 2008). Regardless
54
of the direction of its relationship, it is clear media plays an important role in the development of an FRP, and this mostly during its emergence.
E. Conclusion Whereas demand side factors indicated why FRPs gain support, political opportunities can facilitate or impede the development of FRPs. By themselves, political opportunities cannot determine the faith of FRPs; rather they translate why FRP support (i.e. fertile breeding ground) can lead to successful development, or (electoral) emergence. Its inclusion in the explanatory framework for FRP development allows for the explanation of certain temporal and/or geographic changes or differences. Since political opportunities are not structural factors, it is possible for FRPs to influence the political opportunities once they have successfully emerged. As a result of this dynamic character, the political opportunities framework can change over time, both in content and in impact, and influence FRP development differently compared to an earlier state. Most scholars agree the explanatory value of a single dimension of the political opportunities is rather limited. It is the aggregate framework that influences an FRP’s development. For example, one would expect a proportional system to be beneficial for an FRP’s development, however, by itself; its explanatory power is extremely limited. In general, most scholars do agree on a certain number of claims. Following the theoretical framework provided by Meguid (2008), the stigmatisation of an FRP by traditional parties is very often detrimental to its development. At the same time, it appears that the presence of a nationalist subculture contributes to the explanatory framework for FRP success. Media, as a political opportunity, should not be underestimated. They have an important role in an FRP’s development, most specifically in its emergence phase. In other words, media plays an important role in the introduction of an FRP and its ideology to the broader electorate.
5. Internal Supply-side Factors A. Introduction The two previous sections have indicated that specific external characteristics need to be constructive for an FRP to successfully emerge. On one hand, the breeding ground needs to be fertile so as to present a political environment that favours the development of an FRP. On the other hand, political opportunities need to be favourable, i.e. they need to present an FRP with possibilities. Then, it is up to the FRP to take advantage of them. Now, it is important
55
the FRP itself is also included in the explanatory framework so as to go beyond the notion of an FRP as a non actor (Berman, 1997) and bring the agency back into the analysis. As already mentioned before, different dynamics can come into play between different developmental stages (Coffé, 2004). Generally, scholars agree that external factors are more important for an FRP to emerge, whereas internal, party specific factors play a more significant role when an FRP seeks to consolidate and become an established partisan actor. Therefore an FRP’s emergence is highly dependent on ‘others’, whereas its consolidation is mostly a matter of the ‘self’. Scholars generally describe three main variables that contribute to FRP consolidation when they describe internal factors: the moderation of its party ideology, the strength of its leadership and a structured party organisation. However, in most cross national explanatory frameworks, internal supply side factors are not included (e.g. Eatwell, 2003; Van der Brug e.a., 2005; Carter, 2005). Some of the principal reasons for this are the inherent difficulty to observe, study and analyse such factors due to FRPs’ general lack of transparency. In addition, internal supply side factors also have to deal with serious conceptualisation and operationalisation problems, which is the case for most studies that do include these factors (e.g. Lubbers, 2001; Lubbers e.a., 2002; Norris, 2005; Mudde, 2007). Therefore, these sections provide a general overview of the literature and discuss some of the problems involved in their analyses.
B. Moderate party ideology As emphasised in earlier sections, regardless of whether or not contemporary FRPs should be seen as the logical successors of fascism and Nazism, there are no FRPs with structural ideological links to either fascism or Nazism. A connection of certain factions or individuals within an FRP does not necessarily constitute a structural partisan connection to fascism or Nazism. After all, the inherent anti democratic and anti system nature of fascism and Nazism is not compatible with the FRP ideology. Some scholars even successfully argue that any links to fascism and/or Nazism lead to electoral failure (e.g. Hainsworth, 2000). The majority of the FRP literature often argues in favour of a moderate ideology thesis as a more general version of that observation. This thesis claims that ideological moderation is crucial for successful FRP development, more specifically its consolidation. In other words, if an FRP wants to sustain itself and consolidate, it is strongly recommended to moderate its ideology (e.g. Ignazi, 1992; Kitschelt and McGann, 1995; Ignazi, 2003; Cole, 2005; de Lange, 2007). However, following the numerous classifications in the literature, it is plausible the significant overlap between the category of consolidated FRPs and those with a
56
moderate ideology is the consequence of a misspecification of ideological characteristics. If one were to rely on flawed classifications to make inferences about the consolidation of FRPs, this would significantly weaken the overall validity of this hypothesis. Most scholars do agree on the differentiation between FRPs and ERPs, and conclude that the latter are extremely unsuccessfully throughout Western Europe (e.g. Carter, 2005). However, even within this group of FRPs, one can often find flawed classifications. For example, the MSI is often described as an ERP (e.g. Ignazi, 1992), and similarly the VB is often described as an ‘old’ ERP (Ignazi, 1992), a ‘traditional’ ERP (Ignazi, 2003) or the welfare chauvinist party (Kitschelt and McGann, 1995). Regardless of the differences in terminology, according to these scholars, both the MSI and the VB uphold such extremist ideologies they are no longer considered FRPs. This in its turn indicates they should fail to successfully consolidate (Mudde, 2000). The opposite is true for the CD, the MNR and the REP. Most scholars categorise them as ‘moderate’, ‘new’ or non extreme; however, regardless of their more moderate ideology, they failed to successfully develop. Therefore, it is important to either complement the above thesis with other possible explanations or to further specify it. For example, an important factor the above thesis does not take into account is the political culture. Seeing as how a moderate ideology does not always appear to suffice, perhaps the openness of a political culture to more extremist challenges plays a role in its success. Mudde (2007) argues that political extremity is more likely to damage the development of an FRP when political culture is unreceptive of such extremity. He illustrates this by referring to the openness in Flanders, Austria and Italy and the success of its FRPs. With the growing influence and importance of the media, perception has become a key concept in (electoral) politics. Studies on leadership and individual presentation propose the perpetual balance theory, which indicates that it is not necessarily a person’s characteristics that influence success, but rather its image, i.e. the perception others have of a person’s characteristics (Nimmo and Savage, 1976). Such an argument can be made for parties as well: It is not an FRP’s ideological moderation that matters; it is the perceived ideological moderation and an FRP’s general image that matter. An FRP has numerous instruments to influence the general perception of its ideology. One if its more important instruments are its propaganda. Most FRPs have created a well organised and professional organisation around its propaganda. In addition to this, FRPs can also use their presence in media to influence their image. Mudde (2007) adds that independent media are important to introduce FRPs to the electorate, whereas party
57
propaganda becomes more important when an FRP consolidates and needs to create party loyalty. This aligns with the notion that external factors are more important during the emergence, whereas internal factors are more important for an FRP’s consolidation.
C. Leadership The literature generally acknowledges the importance of leadership (e.g. Immerfall, 1998; Taggart, 2000; Probst, 2003; Mudde, 2012). King (2002) states leadership will be most influential when electoral volatility is high and ideological voting is low. Since this has increasingly been the case since the 1960s, its role in the successful development of FRPs has become more prominent (Plasser and Ulram, 1995; Mayer, 2002). Since the 1980s, one can also observe an increasing centralisation and personalisation of power (e.g. Poguntke and Webb, 2005; Karvonen, 2010), which is even more pertinent for FRPs (Husbands, 1998; Minkenberg, 1998; Van der Brug and Mughan, 2007). Generally, one can differentiate between two dimensions of leadership: external or electoral leadership, and internal or institutional leadership (which has been relatively understudied). External leadership is often associated or considered equivalent to charismatic leadership (see Machiavelli’s Il Principe and Hobbes’ Leviathan ), which can be defined in two distinct ways. On one hand, charisma can be an absolute phenomenon and refer to a set of individual characteristics (e.g. Van der Brug e.a., 2005). On the other hand, charisma can be a relative phenomenon and refer to a relationship with one’s followers. Weber (1922) defines such leadership as a specific quality of an individual personality, by virtue of which he or she is set apart from ordinary men and treated as endowed with supernatural, superhuman, or at least exceptional powers or qualities. Such powers are not accessible to the ordinary man, but are regarded as of divine origin or as exemplary, and on the basis of them the individual concerned is treated as a leader. This definition clearly indicates that charisma is determined by the perception and recognition of its followers. While the literature emphasises the importance of external leadership and its role as an explanatory factor for FRP development, three additional comments should be made in order to complete this theoretical argument. First, since few empirical comparative studies are available, it is difficult to explicitly analyse the exact effects external leadership has. Secondly, whereas a charismatic leader can be an advantage, it can also be a disadvantage. If the charismatic leader is a controversial personality, i.e. he or she polarised the electorate, this can also lead to a specific vote against the FRP. Lastly, one should not overestimate the
58
effect of external leadership. Some FRPs do not have a charismatic leader and have been able to successfully develop and even consolidate (e.g. DFP). The importance of external leadership is also subject to change throughout developmental stages. Mostly, external leaders are used to attract (new) votes and (partisan) support, which has been the principal occupation of the FRPs in the 1980s and the 1990s (Eatwell, 2005; 2006). Because FRPs are masters at transferring votes into support (which Weber (1987) refers to as the ‘routinisation of charisma’) and in keeping their electorate stable, the importance of charismatic leaders gradually declined. In general, it appears external leadership becomes less important after an FRP’s emergence. External leadership is not sufficient for successful electoral development and sustained electoral success; one also needs internal leadership. Besides its Sartorian purpose, FRPs are also organisations. In order to survive and to sustain in the political arena, such organisations must become institutionalised. Unfortunately, charismatic leadership and institutionalisation are not always compatible due to the inherent contradictions between the focus on the ego and the focus on the party (Panebianco, 1988). Recent evidence has shown, however, that not only the combination between both is possible, but also that the cooperation between both is possible (e.g. Pedazhut and Brichta, 2002).
D. Organisation An internal leader organises the party and gives it a sustainable structure. Most scholars agree that without such a structure and without an organisation, an FRP cannot sustain its initial success (e.g. MNR, CD). Therefore, this internal factor is often explicitly specified as the most crucial one (e.g. Betz, 1998; Carter, 2005). It is absolutely crucial if an FRP desires to translate electoral success into political influence. Directly following this claim, and referring back to the differences between internal and external leadership, the literature also agrees an FRP’s organisation is more important in its consolidation phase (Coffé, 2004). After all, a structured organisation increases party cohesion and stability, which is something an FRP needs to maintain its credibility and its vote share (Betz, 2002). Despite its importance, the organisational factor faces several dilemmas. Firstly, as for several other components of the explanatory framework for FRP development, party organisation is incredibly difficult to operationalise and it is nearly impossible to obtain reliable information. The few studies that have explicitly attempted to include party organisation in their explanatory framework have mostly relied on expert surveys (e.g.
59
Lubbers e.a., 2001; Norris, 2005). However, since little information is available on the internal organisation of FRPs, the term ‘expert’ can be questioned here. An additional problem, which other factors in the previous sections have also faced, is the notion of causality. Several empirical examples sustain the claim that a stable party organisation is not required for an FRP to emerge (e.g. REP). However, the question remains whether a stable party organisation is the cause or the consequence of electoral sustainability? The majority of FRPs were not successful, but they imploded only after their emergence, i.e. after their initial electoral success. From this, one can infer that further electoral success (i.e. consolidation) was not possible without an accompanying stability of the FRP’s organisation. Mudde (2007, p.266) describes such parties as “victims of their own success”. They started to become successful, but due to the absence of a proper party structure, the pressure on the party resulted in their collapse. In the case of FRPs, such pressures are usually caused by the absence of (adequate) competent personnel, a lack of supporting structures (e.g. youth organisation, social movements, etc.), a failure to contest in the same districts over time and/or heightened degrees of factionalisation. It is needless to say that organisational weakness is detrimental for governmental participation and that this can lead, not only to electoral punishment, but also to implosion (e.g. FPÖ). In order to create a solid foundation and a stable party organisation, the literature describes two key factors: the presence of grass root organisations and local strongholds (Swyngedouw, 2000; Sully 2002; Ellinas, 2008; Mudde, 2011). The former refers to social movements (e.g. liberation movements, nationalist movements, etc.) that carry out the party ideology but are non partisan. Many FRPs have a strong and successful youth organisation that plays an important role in the ideological socialisation of its youngest (newest) members. The latter refers to a local or regional set of constituencies where an FRP can elaborate its organisation and build its electoral success. Marcus (2000) refers to factionalism as “a perennial problem” for FRPs. Since the 1980s, the FRP family has experienced more splits than any other party family, most of them following factionalisation and/or electoral success (Minkenberg, 1998). Despite some of the empirical evidence, FRPs are not necessarily more prone to factionalism. Following Pederson’s (1982) argument non traditional parties in general are more prone to factionalism. Most of the FRPs that experience internal struggle are new parties with a low degree of institutionalisation. Their increased dependence on few individuals makes them more susceptible to factionalism than parties with a supporting organisation. The
60
institutionalisation thesis claims that older and better organised parties seem to overcome factionalism without implosion, which appears to be supported by FRPs.
E. Conclusion Of all the different variables the literature describes as possible contributors to the explanatory value of FRP development, none have been given less attention than the internal supply side factors. When they are included, it often remains theoretical due to severe operationalisation problems. Such problems are usually the direct consequence of the closed and often secretive character of the internal workings of a FRP. There is almost no quantitative data available and only very few researchers have been allowed to observe the internal functioning of an FRP. Naturally, this restricts the research possibilities to theoretical accounts of the factors this chapter describes. During its emergence, the literature mostly agrees that a moderate ideology and charismatic leadership can positively influence the likelihood of successful development. They also appear to largely agree that, in this developmental phase, FRPs help shape their own faith and are largely responsible for their continued existence. FRPs need a well structured organisation and an infrastructure that can sustain the party. At the same time, an FRP needs local implantation that can serve as its basis in times when electoral results are not favourable. An FRP also benefits from a professionalised propaganda machine that helps distribute and diffuse its rhetoric and ideology. This process should be led by a stable and practical internal leader, which can hold an FRP together in times of difficulty.
6. Conclusion Ever since their general emergence in the 1980s, scholarship has created an extensive body of research on FRPs. A wide variety of sources, techniques, approaches, methodologies, cases and dimensions have been studied. However, up until today, the FRP literature has significant discrepancies, which does not make it evident to rely on previous scholarship. Moreover, this results in the absence of a standardised and generally accepted framework on which to construct further research. Therefore, this chapter’s principal objective has been to map the relevant strands of research, bring forward their strengths and weaknesses, and bring them together so as to construct a theoretical foundation to which this study can add. This absence of consensus is most visible in the theoretical frameworks the literature describes, particularly the conceptualisation and terminology of FRPs. As a direct consequence of the vast diversity of conceptualisations the literature does not necessarily
61
discuss the same phenomenon and it becomes difficult to construct and apply one overarching theoretical framework throughout the field. The first sections of this chapter, therefore, provide an overview of the conceptualisations most used in the literature, and discuss their strengths and weaknesses. In general, the literature does not necessarily agree as to which method to use to conceptualise this phenomenon and which ideological constructs to use to achieve this. Through extensive analysis of both party and scientific literature, this study claims that FRPs must uphold four principal ideologies to be considered FRPs: the core ideologies of nationalism and xenophobia, accompanied with different forms of authoritarianism and populism. The ideological character of populism is still debated throughout the literature; however, as per the rationale provided in this chapter it is most appropriate to define their populist nature as an ideology rather than merely a style. In the second section of this chapter the researcher emphasises the debate and the lack of consensus surrounding the terminology of this phenomenon. Even though consensus is also absent in this field, the most prevalent and generally accepted terms are still ‘extreme right’ (e.g. Ignazi, 2003), and to a lesser extent ‘populist radical right’ (e.g. Mudde, 2007). The researcher provides an extensive rationale as to why these terminologies are not necessarily appropriate for this phenomenon. Particularly for ‘extreme right’, the researcher claims this refers to a phenomenon different from the core of this study. More precisely, whereas this study deals with parties that seek to change the democratic process and reject liberal democracy, the term ‘extreme right’ refers to parties that explicitly reject democracy as a political system. Therefore, this study proposes the less common term ‘far right parties’, which is constructed on a set of spatial, attitudinal and systemic components that inherently refer back to their ideological position. The following three sections provide an overview of what the literature brings forward as potential factors that (can) contribute to FRP development. In other words, it gives an overview of the most important theoretical frameworks that are present in today’s literature and discusses the value of its constructs. Overall, this study has chosen to follow the approach both Eatwell (2003) and Mudde (2007) take, i.e. divide the explanatory framework for FRP development into demand side, external supply side and internal supply side factors. Not only does this approach allow for the identification of patterns of diversity, but it also allows the study to differentiate between different developmental phases. During the emergence phase, demand side factors and external supply side factors are most pertinent for an FRP’s development. During further developmental phases, the role of internal supply side factors increases and FRPs are able to change their external supply side factors.
62
Demand side factors provide a similar environment for all (far right) parties, and create a potential electorate for FRPs. In other words, they indicate why and which political systems have such a potential for an electorate that is favourable to FRPs. The section reviews the most important hypotheses mentioned in the literature: the social breakdown thesis, the modernisation thesis, the globalisation thesis, the (reverse) post materialism, thesis, the crisis thesis, the immigration thesis, the attitudinal thesis, the insecurity thesis and the protest thesis. By themselves, none of them can explain FRP development, as demand side factors are not sufficient. Nonetheless, they are necessary factors. Similarly for external supply side factors, or political opportunities, they need to be complemented by other factors. Not only to they contribute to the explanation of FRP emergence, they also allow the study to analyse patterns of diversity and thereby the differentiation between successful and failed emergence. Based on a solid conceptualisation and justification for the terminology, this study relies on and adds to the operationalisation provided by Kriesi e.a. (1992, 1995) and Van der Heijden (2010). They identify five different dimensions of political opportunities that can facilitate or impede FRP emergence: the socio political cleavage structure, the institutional state structure, traditional parties’ strategies, the power relations within the existing party structure, and the mass media structure. This latter dimension is often overlooked, but its role in the introduction of an FRP and its ideology to the broader electorate should not be underestimated. Once FRPs consolidate, the importance of these factors will decrease and the significance of the party itself will increase. These factors are referred to as internal supply side factors, and usually comprise ideology, (external, but mostly internal) leadership, and organisation. In general the literature agrees that a stable organisation and infrastructure, a moderate ideology and good internal leadership are the crucial factors for FRP consolidation. However, unfortunately, these components are incredibly difficult to study due to the often closed nature of these parties, which does not necessarily allow for the large scale or quantitative analysis of internal supply side factors. Overall, this chapter has provided an extensive analysis of the contemporary scholarly literature and a scientific foundation to build further research on. Therefore, it is important to emphasise the descriptive nature of this chapter and allow this study to reject or complement the existing explanatory frameworks in the following chapters. The researcher will indicate where the current explanatory frameworks are incomplete, and how this study will contribute to and complement these frameworks.
63
2.2 Party identification and voter volatility Does THE volatile voter exist?
64
1. Situating the Research Question Until the 1960s, American literature, and most notably the Michigan model, emphasised the stability of party identification, and subsequently low volatility. Scholarship indicated that the group (in this case the party) is the generalised dimension most critical in defining the individual group relationship (Campbell e.a., 1960, p.121). From the late 1960s on, however, party identification steadily decreased (Wattenberg, 1998) and some scholars started to question the conceptualisation of party identification. Rather than necessarily seeing it as a stable, affective and identity based concept, they proposed a more volatile, rational and policy based conceptualisation (e.g. Fiorina, 1981; Saunders and Abramowitz, 2004). This proposed modification directly followed American scholars’ findings there existed a growing consistency between ideology and party identification, which indicated there was a significant rational component to it ( Abramowitz and Saunders 1998; Carsey and Layman, 2006 ). In Europe, scholars were much more sceptical and sometimes even critical of the notion of party identification. This was predominantly the result of the original application of party identification in a two party Anglo Saxon system (Budge, Crewe and Farlie, 1976; Thomassen, 1976; Baker e.a., 1981; Richardson, 1991). Most notably, certain scholars doubt it is possible to separate long term party identification from short term vote intention in European style parliamentary systems. Additionally, they claim the notion of “partisan independence” does not exist in European party system (with perhaps the exception of the UK). In general though, European scholars do recognise the differentiation between vote choice and some form of partisan attachment. The debate is not necessarily focused on the validity and the weight of party identification, but rather on its conceptualisation and its measurement. Similarly to American scholars, European scholars generally agree that party identification is decreasing, that it is still more stable than party vote, and that it only changes under extraordinary circumstances, such as realignment (e.g. Key, 1955; Rosenof, 2003). Regardless of possible differences in measurement, ever since the ‘defreezing’ of the original cleavage structures (Mair, 1993; 1997), the political environment has been much less stable than originally assumed (Dreyer, 1973; Meffert, Norpoth and Ruhil, 2001). Since the electorate’s perception of politics and its attitudes towards the political class changed in the 1970s, its political participation has decreased. This was most conspicuous by a decline in voting and campaign activity, particularly in electoral turnout (Finkel and Opp, 1991; Back and Teorell, 2005). This, together with numerous other socio political and cultural evolutions, led to a changing political environment (Franklin e.a., 1992); one where new
65
issues, new cleavages, and new niches gave opportunities to post materialist parties such as FRPs (Eatwell, 2003; Mudde, 2007). More specifically, the changing political environment has led to decreasing party identification and increasing volatility, which, most scholars agree, has been beneficial for FRPs (Ignazi, 1992; 1996). As a new political player, FRPs were likely to attract those voters who saw their party identification with other, more traditional parties decrease. Therefore, it is not a surprise that the original success (i.e. emergence) of FRPs has often been contributed to a sizeable protest vote (Mayer and Perrineau, 1992; Eatwell, 2000; Perrineau, 2002). Simultaneously, most of the literature agrees that FRPs have done particularly well in attracting voters and consolidating this support (Van der Brug, Fennema and Tillie, 2000; Norris, 2005). Party identification with FRPs appears to be stronger than with more traditional parties. Especially since the 2000s, FRPs have been incredibly successful in transferring this protest vote into a support vote, thereby significantly increasing party identification (De Witte, 1998; Betz, 2002; Schmidt, 2003). In addition to these changing levels of party identification, one can also examine the patterns of the above mentioned change. If a voter continues to identify with and vote for the same party, his or her vote is considerably stable. However, if or when a voter abandons a party tie only to return to that same choice later, the vote is considered deviated. If or when a voter does not return to the same party, this dynamic can be referred to as realignment, dealignment or volatility (among others), which is often summarised as positive change. These latter dynamics indicate that vote change becomes more frequent, and, generally, parties are less capable of creating ties with voters. The result is a substantial increase in floating voters, and a change in party strategies to attract such voters. Therefore, the increasingly volatile nature of voters changes the electoral environment. In other words, the increasingly volatile political climate has proven to be very advantageous for FRPs. However, considering Western Europe has not experienced uniform FRP success, it is important to go beyond this observation and further analyse any possible patterns of diversity one can find. Therefore, a comparative research project such as this one provides an opportunity to examine whether the West European far right party electorates (that originally voted for more traditional parties) actually share similar characteristics across different West European party systems. Such an analysis can help determine whether different strategies are necessary to attract voters, or whether the far right party family can uphold a more uniform approach to attract those voters that do not strongly identify with a specific (traditional) party. In other words, one of the principal questions examined in this
66
paper is whether one can speak of THE volatile voter or whether a volatile voter depends on the contextual and socio demographic differences between party systems?
2. Methodology For both party identification and volatility it is important to be clear as to how to conceptualise and how to measure them. The former will be discussed extensively in the first sections of the paper. As for the measurement of these concepts, several restrictions have to be taken into account. Most measurements of volatility, or even party identification, require three time points in order to determine whether the change in voting behaviour is also an indication of volatility. However, often, as is the case here, one is restricted by the available data. Therefore, this research project will both theoretically and methodologically argue that two time points are sufficient as a basis for the study of party identification and volatility. Considering the comparative nature of the research question, and the importance of panel data, which allows for the study of individual vote history rather than aggregate vote share change, this paper will rely on module two of the CSES study. Not only does it include the desired cases (i.e. countries), it also includes the appropriate variables (i.e. survey questions). When selecting the cases, only countries with multiparty systems have been included in order to avoid conceptual discussions surrounding the meaning and validity of party identification. The analysis will include Belgium (2003), France (2002) and Italy (2006) as its principal country studies; however, it will also take a closer look at Denmark (2001), Norway (2001), Sweden (2002) and Switzerland (2003) as additional illustrations. Rather than one aggregate model, a separate dataset and multilevel logistic regression model will be constructed for each individual case. This will allow for the consideration of contextual differences between different West European party systems (see Eatwell, 2003; Mudde, 2007). As a dependent variable, this research project designs a volatility value for each unit of analysis based on an individual’s vote choice in the most recent election and its vote choice in the previous election. This will result in a dummy variable on the party level. In addition, the model will also include a volatility dummy variable on the coalition level, thereby indicating whether an individual voter has shifted coalitions between elections. Based on the three classic questions used in the CSES dataset, party identification will also be incorporated for each unit of analysis: "Is there a party that you feel close to?" (direction), “If yes, how strong you feel close to it?" (intensity), and “if no, is there a party you feel closer to than the others?" (leaning). For each unit of analysis, a set of socio demographic variables (using a step wise regression procedure) will also be included. These will describe the
67
individual voters and contribute to the overall image of each electorate. If possible, a third and fourth dummy variable will also be created, which will reflect the stability in terms of casting a vote and the stability in terms of voting for the post election coalition. This differentiation between party and coalition, and the subsequent creation of two separate levels of volatility represent the principal innovative character of this paper. These constructed datasets will allow for the identification of the volatile electorates (i.e. those who are volatile and those who are not), their descriptions (i.e. their socio demographic factors) and the analysis of possible within group or between group similarities and differences (i.e. their possible diversity across party systems). These three analytical dimensions can help explain why certain individuals are more likely to be volatile than others. More specifically, such a detailed breakdown can help determine whether there are common socio demographic characteristics between cross national volatile electorates. In its turn, this can provide an indication as to how different or how similar FRP electoral strategies could be.
68
3.1 The Use of Social Movement Theory to Explain the Emergence of Far Right Parties The Political Process Model
69
1. Introduction Throughout the past three decades, several third wave FRPs have emerged successfully (von Beyme, 1988). Traditionally this has been explained by demand side factors and (external) supply side factors (Mudde, 2007). This chapter suggests a different and more comprehensive framework, and applies a social movement approach to FRP emergence. More specifically, this chapter uses the political process model (as designed and proposed by McAdam, 1982) as an explanatory model for FRP emergence. It provides a new theoretical approach to an often wrongly or carelessly studied phenomenon. The political process model (PPM) describes three distinct dimensions, which combine different aspects of the emergence process. Firstly, the PPM describes political opportunities, which refers to (more dynamic) factors outside the FRP that either facilitate or impede FRPs’ emergence and/or access to power. Secondly, the PPM describes resource mobilisation, which refers to (more structural) factors through which people can mobilise and engage in collective action. Lastly, and most notably, the PPM refers to a situationally adjusted master frame, which includes cognitive and cultural factors that influence FRP emergence. The emphasis on the latter dimension is the main innovation of this paper. A favourable combination of these three political process dimensions is both necessary and sufficient for FRPs to emergence successfully. Yet since its introduction in the 1980s, the PPM has experienced some criticism as well, especially from the more culturalist approach to (new) social movements, namely multi institutional theory. Some critiques will serve as the basis for a slight modification of the PPM. However, most critiques are a consequence of a misinterpretation or misspecification of the original PPM, its application on FRPs or an attempt to explain why movements emerged instead of how.
2. Traditional explanations of FRP emergence The literature provides a set of extensively researched explanatory factors for FRP emergence. Demand side factors often find their origin in mainstream electoral politics. They indicate “broad economic, historical, social processes that take place at the national, supranational and sometimes even global level” (Mudde, 2007, p.202). Prime examples are increasing levels of immigration (Betz, 1994; Martin, 1996; Ignazi, 1997), rising unemployment (Givens, 2002; Swank and Betz, 2003), economic crises (Betz, 1994; Knigge, 1998), protest voting (Fennema and Tillie, 1998; Van der Brug e.a., 2000), new cleavages (Ignazi, 1992), modernisation (Minkenberg, 1998; Decker, 2004), and demographic changes (Veugelers and Chiarini, 2002; Wendt, 2003).
70
By itself, these variables focus on structural phenomena and are biased towards finding one universal cause for FRP emergence (Betz, 1994; Rydgren, 2005a), which is a direct consequence of the literature’s ambition to avoid ad hoc theorising. However, much of the literature ignores the fact that (i) FRP emergence can only be explained by a set of dynamics, not by one structural factor, (ii) it lacks explanatory power for developmental and electoral differences, and (iii) the dynamics responsible for FRP emergence are very different from those sustaining its success (Coffé, 2004). This is very well represented in the works of both Eatwell (2003) and Mudde (2007), who conclude that in order for FRPs to emerge, two explanatory variables are key: (i) the fertility of the breeding ground, and (ii) the political opportunity structure (POS). The literature agrees the latter factor plays a more important role during FRP emergence than in any other phase (Jungerstam Mulder, 2003) because if its facilitating or impeding character (see below). Most importantly because the POS can explain (i) why certain FRPs emerged and why others did not, and (ii) why it benefited an FRP and not one of the traditional parties (Rydgren, 2005a; Arzheimer and Carter, 2006). Additionally, once an FRP emerges, it can change the POS both in impact and in content (e.g. the media) so as to benefit from its change (Mudde, 2007). The most important drawback of the POS is the literature’s disagreement on its conceptualisation and operationalisation (see below). This traditional model, however, has several limitations. It is too limited in its scope 2 (Rydgren, 2005a), and overemphasises structural factors (Jackman and Volpert, 1996), which are often not correctly specified and the result of strategic party political choices (Goodwin and Jasper, 1999). Simultaneously, it frequently disregards the possible explanatory value of dynamic factors and of several lower level factors like individual behaviour/attitudes or social networks. In order to effectively take into account such factors, this paper proposes to take a more comprehensive and multi dimensional approach.
3. Introducing social movement theory in political science A social movement theory can only justifiably explain FRP emergence when possible differences between FRPs and social movements prove to be non existent, or at least irrelevant. Generally, there are three essential differences between social movements and FRPs: (i) FRPs have access to state power, (ii) they have different bargaining styles, and (iii)
2 Many of the proposed theories, or dimensions of theories, do not hold in comparative contexts. This is where this paper tries to contribute, by proposing a model or an approach that can explain the emergence of FRPs throughout Western Europe.
71
they participate in elections (Tarrow, 1994). Despite their differences, however empirical research shows that the actors, fates and structures of social movements and political parties are often closely intertwined. Goodwin and Jasper state “researchers in this tradition (i.e. the PPM) view protest groups as being like political parties, except operating outside the electoral system” (2009, p.190). Common goals emphasise similar processes, especially when both are mobilising in the early developmental stages. More specifically, such goals are referred to as (i) interest articulation, (ii) interest deliberation, and (iii) interest aggregation (Aminzade, 1995; Maguire, 1995). In other words, they are different means to similar goals; they aim to translate popular preferences into policy while operating in the same political space (McAdam, e.a., 1996). FRPs often closer resemble social movements than traditional political parties. Even though political parties serve as major institutions for interest articulation and aggregation, social movements have often been forceful competitors, especially in representing post materialist values (Schmitter, 2001). In the 1980s, the majority of the European electorate showed a growing anti political attitude that regarded formal politics as dirty and corrupt, in contrast to ‘cleaner’ social movement activism (Gershman President, 2004). FRPs adopted this populist approach to politics as one of the key ideological pillars, thereby increasing their credibility and similarities to social movements (see master frame ). Given the above rationale of equivalence, social movement theory is used here to explain FRP emergence. Generally, there are four possible approaches to social movements. The classic model and resource mobilisation form the basis and the core of the PPM used to explain FRP emergence in this paper. More recently, the new institutionalist or culturalist approach serves as a possible critique to the PPM. However, some of its propositions will be incorporated in the discussion and will actually complement the PPM. The classic model claims that unconventional politics (like far right politics) are redundant and counterproductive to political goals. Like other forms of deviant and irrational behaviour, it is explained by psychological factors. McAdam (1999) started from the basic assumption that ‘structural strain’ was the primary explanatory variable. This was determined by three factors: (i) mass society alienation due to industrialisation and urbanisation, (ii) absolute deprivation and frustration as a consequence of economic and material conditions, and (iii) relative deprivation and status inconsistency as a consequence of economic and material conditions. This in turn led to a disruptive psychological state (e.g. Adorno e.a., 1950), and the increase of contentious politics (e.g. the emergence of FRPs).
72
This model was prevalent during the 1950s and 1960s, however, criticism on its irrational character and its sole emphasis on deprivation and grievances 3 increased, especially by rational choice economists (Olsen, 1965). Even though Olsen was rather extreme in his rational choice view, he recognised protest as a rational possibility. McCarth and Zald (1977) had a similar criticism, claiming organisations, just like individuals, act rational. In short, protest as process was no longer seen as contentious or dangerous but as rational and normal. As a response, resource mobilisation theory developed in the 1970s (McCarthy and Zald, 1977). It sees social movements as rational actors that emphasise the importance of resources, and have mobilisation and political action as their primary goals. Social movement organisations 4 play an important role in dividing the different resources and connecting members with each other. In this process, both cooperation (i.e. achieving common goals) and competition (i.e. resource rivalry) play an important role (Zald and McCarthy, 1979). This model works well to explain moderate and professional groups; something contentious movements like FRPs are not.
4. The political process model Resource mobilisation also experienced increasing criticism due to its failure to include political dimensions. Scholars realised the role of the state was not only to be a target for protest but also an arbitrator for grievances, and questioned the structural role and nature of external influences in this process. Therefore, a more dynamic theory – including explanations for the oft rapid changing political space – was needed. This paper draws on the more comprehensive political process model to identify and analyse how FRPs emerged (McAdam, 1982; Tarrow, 1983; Kitschelt, 1986). Besides the integration of different dimensions, it presupposes that organisational features, action repertoires and their relative impact are determined by specific socio political conditions external to the movement (van der Heijden, 1997) 5. The PPM comprises three key elements that can explain FRP emergence: (i) expanding political opportunities, (ii) indigenous organisational strength, and (iii) cognitive liberation (McAdam, 1982). The latter
3 Grievances were considered necessary factors for social movement emergence, however, they were far from sufficient. For this reason it is necessary to look beyond grievances and also take other explanatory factors into consideration. 4 Social movement organisations (SMOs) are defined as specific, organised components of a social movement. They usually share the social movement’s goals and have coordinating roles, but they do not employ or direct most of the supporters (Zald and McCarthy, 1979). 5 Van der Heijden (1997) analyses the institutionalisation of environmental movements, but when discussing the PPM, he specifically refers to the trans national differences of similar factors. In other words, similar structures, dynamics and processes can differ in their impact, depending on the national context they operate in.
73
two concepts are updated to mobilising structure s (Gamson and Meye r, 1996 ) and a master frame 6 (Snow and Benford, 1988) respectively. These factors are not only necessary but also sufficient; they explain both emergence and a lack thereof.
Additionally, three relations stand out. Firstly, the arrow from ‘expanding po litical opportunities’ to ‘cognitive liberation’ indicates that political opportunities available to FRPs influence the master frame an FRP develops. Secondly, the arrow between ‘cognitive liberation’ and ‘indigenous organisational strength’ indicates res ource mobilisation and the master frame are interdependent variables. Lastly, the arrow from ‘socio economic processes’ (i.e. demand side factors) to ‘expanding political opportunities’ and ‘indigenous organisational strength’ implies that structural chan ge directly influences political opportunities and the master frame, and only indirectly influences FRP emergence.
Expanding political opportunities Most scholars use the term political opportunity structures (POS), which Koopmans (1999) defines as (i) c ollective actions external to the movement, (ii) consequences of interactions between political actors and/or institutions, and (iii) excluding possible products of strategic interaction. Since significant parts of the POS are not structural, but rather perceived, situational and dynamic, its definition is an oxymoron (Eisinger, 1973). It would be more appropriate to refer to them as political opportunities. The structural sources of FRP emergence are mostly represented by socio economic and cultural fac tors ( see Figure 1 ). Each factor that facilitates or impedes emergence can be labelled a political opportunity. It is thus extremely delicate to define political opportunitie s (Gamson and
6 Snow and Benford (1988) originally rename cognitive liberation as cultural framings. However, since we are dealing with macro level entities (i.e. FRPs), their term of mas ter frame is considered more appropriate in this case.
74
Meyer, 1996). The more broadly defined, the more trivial and tautological political opportunities become (Gamson and Meyer, 1996). The more narrowly defined, the more inadequate and implausible the concept and its explanatory power becomes (McAdam, 1996). This dilemma illustrates the lack of scholarly consensus regarding the definition of political opportunities (Gamson and Meyer, 1996). The specification of political opportunities needs to be more emphasised, and the research more synthesised (Meyer and Minkoff, 2004). McAdam e.a. (1996) emphasise the relatively dynamic character of the relationship between a political system and movement mobilisation. Therefore, this paper prefers Tarrow’s description of political opportunities as “consistent – yet not necessarily formal or permanent – dimensions of the political environment that provide incentives for people to undertake collective action by affecting their expectations for success or failure” (1994, p.85). It is much more specific and it explicitly emphasises resources outside the FRP 7 that either open or close political space for mobilisation (Kingdon, 2010). Political opportunities can explain emergence and short term electoral changes because they are fast variables (Koopmans, 1999), i.e. they are situational and can alter quickly (Tarrow, 1998; Gunning, 2008). Their precise effect is historically and situationally contingent, and depends on structural (e.g. electoral system), strategic (e.g. elite alliances) and cultural factors (e.g. resonant slogans). Political opportunities also reconnect structure and agency; something ‘resource mobilisation’ failed to do (Koopmans, 1999; Meyer, 2004). Additionally, because people join FRPs in response to political opportunities, the “when” of FRP emergence is important. This brings back timing as an explanatory variable. Largely based on the more elaborate work of Kriesi e.a. (1992; 1995) and van der Heijden (2010), five broad dimensions constitute the operationalisation of political opportunities for FRP emergence 8. First and foremost, the nature of the socio political cleavage structure is an instrumental factor for any emerging political party (Lipset and Rokkan, 1967). In the early 1970s, the traditional cleavage structures went through a process of ‘defreezing’ 9 as a result of post industrialisation and post modernisation (Pedersen, 1979;
7 This stands opposed to the internal factors of a social movement usually emphasised in the literature, i.e. grievances and resources (Meyer and Staggenborg, 1996). 8 Since the focus is on political parties and not on contentious politics, this paper only considers those political opportunities that channel political action into legal and electoral routes. If the analysis of political opportunities would take illegal and disruptive routes into consideration, this would provide a theoretical framework for neo fascist and extreme right parties (e.g. British National Front , Italian Fiamma Tricolore , etc). 9 Since the 1970s, the four traditional cleavage dimensions, as proposed by Lipset and Rokkan in 1967, started losing their importance. At the same time, new cleavage structures appeared alongside a more cultural/post materialist dimension (e.g. globalisation, gender, modernisation, religion, etc) (e.g. Ignazi, 1992).
75
Budge, 1982; Dalton, McAllister and Wattenberg, 2002). This evolution was symbolised by dealignment and realignment processes 10 (Kriesi, e.a., 1995), which resulted in a significant decrease of the salience of traditional cleavages. The growing discontent with the elites and the decreasing socio partisan identification (Putnam, e.a., 2000) led to a more volatile electoral arena and a more open political space, i.e. new niches emerged in the electoral arena (Rydgren, 2004a). Agenda setting, politicisation of new issues and mainstream parties’ strategies toward new issues (stigmatisation or imitation) resulted in the increasing salience of certain new (culturalist) cleavages, in addition to the relocation of the competition between parties (Duyvendak and Koopmans, 1992; Perrineau, 1997; Clark and Lipset, 2001). Together with traditional parties’ failure to adjust to this evolution, this contributed to the issue ownership FRPs generated over several ‘new’ issues like immigration, law and order, and welfare chauvinism. A second important dimension of political opportunities is the institutional state structure, i.e. the points of access to the existing political system. The levels of vertical territorial decentralisation 11 (i.e. federalism) need to be favourable, together with a minimal level of horizontal concentration of state power between the different state branches 12 (Kriesi e.a., 1995). The electoral system needs to be accessible to new and small parties 13 and offer them the possibility to influence political decision making procedures. Other important access points are the availability of direct democratic structures (e.g. referenda), and the access to state subventions and state funding (Carter, 2005). A third dimension constitutes the (informal) strategies of the mainstream parties vis à vis FRPs. On one hand there are inclusive strategies by which the political elites seek to integrate FRPs, which can range from facilitation to cooperation or even assimilation. On the other hand, there are exclusive strategies by which political elites seek to exclude FRPs, which can range from polarisation to repression or even confrontation. The presence (absence) of elite allies has proved to play an important role (Meguid, 2008). A similar
10 This can be measured by political discontent, the number of scandals, corruption, political alienation, mistrust toward the elites, Goldthorpe’s class voting scheme, the number of floating voters, and the political complexity of and access to the political/electoral process. 11 This can be measured by the distribution of funds, effort, and personnel allocation alongside the different governmental levels. Additionally, the use of Lijphart’s index is also useful and convenient to measure the levels of disproportionality. 12 This can be done by a qualitative description of relative strength and facilitating character of the three different state branches. 13 This can be measured by looking at the electoral formulae, the district magnitude, the number of districts, the assembly sizes, legal and effective thresholds and the Gallagher index of disproportionality.
76
observation can be made for the available political space 14 (Minkenberg, 2001). The nature of the party system’s coalition format also has a sizeable influence on an FRP’s emergence. A fourth dimension consists of the power relations within the existing party structure. Social movement theory states that most social movements are more reserved in their opposition in periods when left wing parties are in office, and more outspoken when left wing parties are in opposition (Kriesi, 1992)15 . This argument can be reframed for FRPs, stating their opposition will be more moderated when culturally conservative parties are in power, which means the possibilities for access to power and political influence increase. Given, this hypothesis is intuitional, however, the few instances when FRPs have participated in government has been in coalition with traditional right wing parties, thereby clearly indicating their ideological affinity for the right wing over the left wing (e.g. FPÖ and LN). A fifth dimension is the mass media structure. For FRPs, the access to the broadcast media is an important facilitating or impeding factor when publicising its ideology or familiarising the electorate with its ideas and personalities (Carter, 2005). The media has different societal functions that can influence an FRP’s emergence: Agenda setting, status conferral, stereotyping, priming, gatekeeping 16 , and framing. They select and allocate prominence to the public discourse of their preference. They produce the first and most basic selection mechanism in the public sphere, namely visibility, which heavily influences the further transmission and diffusion of the public discourse (Koopmans, 2004).
Indigenous organisational strength, or mobilising structures Even though expanding political opportunities are necessary, they are not sufficient for FRP emergence, and are complemented by mobilising structures. Similar to resource mobilisation, the main focus of mobilising structures is on the resources from within used for the continuous development of FRPs (van Noort, 1988; Kitschelt, 1991). However, contrary to resource mobilisation, this factor complements others; it does not propose to be the sole decisive factor in the emergence phase (Kriesi e.a., 1992). Like political opportunities can be
14 Following Kitschelt and McGann (1995), Ignazi (2003) claims FRPs only benefit from mainstream convergence after the political system has polarised. The simple convergence thesis has also encountered serious criticism by some scholars disagreeing with the operationalisation of convergence (e.g. Veugelers, 2001; Norris, 2005). Generally most scholars agree with the simple convergence thesis, however, there are disagreements regarding which parties’ convergence facilitate FRP emergence. Some argue in favour of the mainstream right parties (e.g. Van der Brug, e.a., 2005), some argue in favour of the mainstream left parties (e.g. Van den Brink, 2005), while others claim it is a combination of the two (e.g. Carter, 2005). 15 This is primarily because most social movements advocate post modernist values and culturally left wing politics aims to advance similar values. 16 A prime example of such a gatekeeper is the Bundesverfassungsgericht in Germany.
77
impeding, most mobilising structures can also be ‘demobilising’ structures and hinder any potential mobilisation (McAdam and Paulsen, 1993; McAdam, McCarthy and Zald, 1996). Throughout the existing literature, there is no clear demarcation between open and closed structures so any possible specification can be contested. However, the danger of ambiguity and tautology denote the importance of a proper and clear conceptualisation (Goodwin and Jasper, 1999). If this is not done, the concept loses its explanatory value (see above). McAdam (1982) refers to mobilising structures as the resources of the aggrieved population that allow them to exploit the changes in the political opportunities. In later definitions, the emphasis changes from ‘resources’, which is passive, to ‘collective action’, which is more active. McAdam, McCarthy and Zald (1996) argue that mobilising structures can be broadly defined as “those collective vehicles, informal as well as formal, through which people mobilize and engage in collective action". More comprehensively, McCarthy (1996) later defines them as:
“Those agreed upon ways of engaging in collective action which include particular "tactical repertoires", particular “social movement organizational" forms, and "modular social movement repertoires". I also mean to include the range of every day life micro mobilization structural social locations that are not aimed primarily at movement mobilization, but where mobilization may be generated: these include family units, friendship networks, voluntary associations, work units, and elements of the state structure itself. (1996, p.141) Some scholars view these organisational and social networks as pre existing structures, not as dynamic creations of movement organisers. Very often, they are seen as physical structures, rather than as the information, ideas and emotions that “flow” through them (Emirbayer and Goodwin, 1994; Jasper and Poulsen, 1995), which is what mobilising structures really consist of. Based on the limited description by McAdam (1982, 1999) and the more extensive work by Goodwin and Jasper (1999), three dimensions constitute its operationalisation. A first dimension of mobilising structures is the existing solidarity network. Many individuals share prior ties or connections with others that can lead to solidarity. When this leads to a collective identity and the creation of a group, a person might have a big stake in the group’s future. Therefore, “when collective action is urgent, an individual is likely to contribute his or her share even if the impact of that share is not noticeable” (Fireman and Gamson, 1979, p.22). In short, recruitment takes place primarily through pre existing solidarities (e.g. Snow, e.a., 1980; Bonnell, 1983; McAdam and Paulsen, 1993) 17 . Secondly, FRPs attract supporters even (or especially) in the absence of prior identities or social networks (Jasper and Poulsen, 1995). Fellow participants or supporters share similar
17 Mobilising structures (i.e. identities) come from shared positions in networks of patronage (Gould, 1998), urban residence (Gould, 1995), political affiliation (Mische, 1996), submerged networks (Melucci, 1989), and other formal and informal, ideological and institutional organisations (McAdam, McCarthy and Zald, 1996).
78
demographic and/or economic traits but these common characteristics do not form any pre existing collective identities (like those previously mentioned). All that is lacking for such a collective identity is often the institutional infrastructure. In this case, FRPs first create the collective identity, which then leads to social networks and further institutionalisation (Buechler, 1990; Jasper, 1997; Klandermans, 1997). Thirdly, the process of identity attribution and reinventing collective identities may also happen independent from FRP efforts (Fantasia, 1988; Hirsch, 1990; Polletta, 1998). Factors external to FRPs or the possible adherents can be responsible for the implementation of a collective identity. Such an external event can mobilise people that do not know each other or the movement that will institutionalise their collective identity (Jasper, 1997). For example, the mediatisation of sudden political scandals (such as those in the early 1980s in Belgium) can bring together a group of protesters previously unacquainted and create a collective identity 18 based on moral grounds, which can eventually be adopted by the FRP in order to mobilise and create a social network. Therefore, since mobilising structures do not always require pre existing personal or collective identities, the adherents’ attempts to strategically and structurally frame these identities, are crucial for further mobilisation and recruitment of possible supporters. In this context, the concept of a frame refers to the interpretative schema developed by an FRP in order to mobilise and recruit potential supporters (Snow, e.a., 1986).
Cognitive liberation, or master frame The term framing was first used by Goffman and defined a “schemata of interpretation that enables individuals to locate, perceive, identify and label occurrences within their life space and the world at large” (1974, p.21). The use in social movement theory differs from the more cognitive construction of “schema” (Klandermans, e.a., 1999) 19 because collective action frames, unlike schemas, are not merely the aggregates of individual attitudes and perceptions. Their addition of the outcomes of negotiated shared meanings renders them more inclusive (Gamson 1992). Later, collective action frames were used as an interpretative function by simplifying and condensing aspects of the “world out there”, but in ways that are “intended to mobilise potential adherents and constituents, to garner bystander support, and to
18 To avoid any overextension of the concept, collective identity is defined as “an individual’s cognitive, moral and emotional connection with a broader community, category, practice or institution. It is a perception of a shared status or relation, which may be imagined rather than experienced directly, and it is distinct from personal identities, although it may form part of a personal identity” (Polletta and Jasper, 2001, p.285). 19 The concept of framing has been widely used in psychology (Tversky and Kahneman, 1981; Klandermans, 1997), and various other disciplines as well (Tannen, 1993; Scheufele, 1999; Triandafyllidou and Fotiou, 1998).
79
demobilise antagonists” (Snow and Benford, 1988, p.198). It is an “action oriented set of beliefs and meanings that inspire and legitimate the activities and campaigns of a social movement organisation” (Benford and Snow, 2000, p.614). Master frames play the same role as collective action frames do, only on a bigger scale (Snow and Benford, 1992). They are not limited to the interests of a particular group or a set of related problems. Its scope and influence reach much further, and are designed to attract a broader audience or deal with a broader set of problems. Contrary to collective action frames, a master frame influences and constrains the orientations and activities of other parties and can be considered a flexible and all inclusive framework that goes beyond a set of ideologies (Snow and Benford, 1992). Ideology and master frame are two highly interrelated concepts, and both refer to general constructs within which more specific ideas are understood. However, analytically, master frame must be clearly distinguished from ideology. Whereas master frame refers to a cluster of fairly broad and inclusive rhetorical strategies on which FRPs can draw (Swart 1995; Carroll and Ratner 1996), ideology refers to support for more specific articulations of theory and value nested within more general ones (Oliver and Johnston, 2000). In other words, a master frame does not necessarily include the socio political theory and normative value systems that characterise an ideology 20 . The success of such a master frame is contingent upon its ability to attend to the interrelated problems of “consensus mobilisation” and “action mobilisation” (Klandermans, 1984). The former indicates FRPs need to gather support for their proposed master frame. The latter indicates FRPs need to activate and mobilise those already supporting their proposed master frame. In order to do this, framing can be divided into three core tasks: diagnostic, prognostic and motivational framing (Wilson, 1973; Snow and Benford, 1988). The mobilisation of both people and resources requires identification of the current situation as unjust or problematic (diagnostic), providing people with possible changes and a normative view of society (prognostic), and giving them the incentive to either become a member or allocate some resources to the FRP (motivational). In addition to these core features of a master frame, the literature also identifies several more dynamic features, which give the master frame its specificity and its uniqueness compared to other master frames. The most important features are the problems or issues addressed by the master frame (problem identification), the direction of attribution (e.g.
20 For an excellent example of the distinction between ideology and master frame, see Aarelaid’s (1998), and Johnston and Aarelaid Tart’s (2000) studies on Estonian national movements.
80
Gerhards and Rucht, 1992), the extent of themes and ideas incorporated and articulated (flexibility and rigidity; inclusiveness and exclusivity), the scope and degree of influence of the master frame, and the effectiveness or mobilising potential of a master frame (degree of resonance). These more dynamic factors become important when a master frame travels across political systems (i.e. trans national diffusion). Few FRPs have the intellectual background and the financial resources to develop their own master frame (Rydgren, 2005b). Usually they draw on an existing (and successful) master frame, strategically alter this, and adopt it. This is especially the case for those master frame aspects that are deliberative, utilitarian and goal oriented (Snow and Benford, 2000). The strategic efforts to connect master frames to prospective constituents and resource providers, is referred to as frame alignment (Snow, e.a., 1986). There are four different processes: frame bridging, frame amplification, frame extension, and frame transformation. They are all present, be it in different degrees, and each have a specific impact on the process of trans national diffusion, and the possible success of the newly implemented master frame.
6. Fourth generation social movement theory: New social movements Several scholars criticised the PPM for its inadequate description of collective action. They claimed the weakening of traditional structures, the increasing individualisation, and a general and profound change in the transformation of conflict structures led to the rise of so called ‘new social movements’ (NSMs) (Raschke, 1985). Subsequently, the theoretical focus shifted from collective action strategies and resources to fostering new, post materialist identities (Cohen, 1985; Klandermans, 1991). This new course hardly constitutes a new theoretical approach, but rather a new emphasis (Flacks, 2004). NSM theorists see symbolic action in the civil or cultural sphere, as well as in the state or political sphere (McAdam, 1983), as an integral part of collective action (Cohen, 1985; Melucci, 1989). Simultaneously, processes that increase self determination and self sufficiency are emphasised over the traditional power maximisation strategies (Habermas, 1987; Rucht, 1988). This increases the importance of post materialist and cultural values as opposed to conflicts regarding material resources (Inglehart, 1990; Dalton, Kuechler and Burklin, 1990). An additional emphasis is put on the construction of collective identities and group interests, rather than their structural determination (Melucci, 1989; Hunt, Benford, and Snow, 1994). Similarly, NSM theorists emphasise the social construction of grievances and ideology, whereas the PPM presumes they are a consequence of a movement’s structural
81
location (Klandermans, 1992). Lastly, NSM theorists recognise a variety of temporary networks that form the basis of mobilisation processes, whereas the PPM presumes a centralised organisation as a key component. A first reflection is that (certain) social movements are ‘new’ (Kriesi, e.a., 1995) and therefore require a re evaluation of their original assumptions. However, their novelty and ties to the political sphere are often exaggerated due to their emphasis on culturalist goals (Plotke, 1990), and their commonalities neglected (Melucci, 1994). Tarrow (1991) argues many so called ‘new’ social movements have grown out of older organisations and have long histories, which NSM theory neglects. This goes back to the original argument of ‘cycles of protest’ (Tarrow, 1998), according to which these ‘new’ social movements are merely another cyclical phase of collective action (Brandt, 1990). Therefore there is more continuity between old and new than implied by NSM theorists (Johnston, 1994). This conceptual disagreement is also reflected in the second major criticism. NSM theorists claim traditional social movements have been rather progressive in nature, whereas new social movements are more reactive following the intrusion of capitalist institutions in society (Habermas, 1987). However, Rucht (1988) claims that continuous modernisation produces conflicts alongside culturalist lines, and that the identity oriented development this provokes is progressive. Simultaneously, (culturalist) modernisation provokes a reactive protest against conservative economic or political changes. This indicates that, both old and new, social movements are progressive and reactionary in their societal response to change and protest (Cohen, 1982, 1983). A third major criticism claims most new social movements are cultural in nature, whereas the old social movements are more political actors. NSM theorists claim, due to their focus on identity, they are about something larger than traditional politics. The risk of such a statement is that it creates a dichotomy between two concepts that are not necessarily mutually exclusive. McAdam (1994), therefore, claims that all movements play both a cultural role and take (explicit or implicit) political stances 21 . Brandt (1986) takes this even further and claims they combine both dimensions and still attempt to transform power relations. Therefore, this defines more the modernisation of both society and movements, rather than the nature of ‘new’ social movements.
21 He even argues that those movements that do not wish to partake in the political environment or opt out of any conventional contestation to power, still take a political stance of – what he refers to as – quietism (McAdam, 1994).
82
A fourth major criticism is that participation in NSMs cannot be predicted anymore by class location (Touraine, 1981; Melucci, 1996; Castells, 1997). They claim they are now defined by cultural dynamics like race, ethnicity, religion, gender, etc. However, the fact that NSMs have broader and more ideologically identified constituencies is overemphasised and does not constitute a significant change in how movements emerge. This merely represents a shift from group based politics (interests) to value based politics (ideology). Even for those scholars that accept a change in social base 22 , class and collective action have been decoupled in advanced capitalism and the role of culture has increased (Eder, 1993).
7. Additional critiques on the political process model An important critique on the PPM is its strong bias toward ‘structures’ (Goodwin and Jasper, 1999). The updated models (McAdam, McCarthy and Zald, 1996; Tarrow; 1994) include this in the political opportunities dimension; however, structures are inherently non dynamic (which political opportunities are). If the separation between these two is not clear, the danger exists that several non structural factors (e.g. agency, strategy, etc) are studied as structural factors. Therefore it is more appropriate to include all structural factors in the resource mobilisation dimension of the PPM. Closely related to this is a critique on the role and place of culture in the PPM. The separate influence of culture is often obscured by the focus on networks as a part of mobilising structures. Culture and mobilising structures are often analysed as affecting recruitment separately, however, mobilising structures achieve very little without “the ideas and affective bonds that keep them together” (Goodwin and Jasper, 1999, p. 46), which has been indicated in the original PPM by the interdependent relationship between the two concepts. Therefore it is pertinent the conceptualisation of both culture and mobilising structures reflect their uniqueness. Several authors also claim the PPM suffers from severe selection bias: The PPM only applies to those social movements that pursue political participation and political rights. Jasper and Poulsen (1993) claim that social movements that do not target the state as their
22 Most NSM scholars argue that social movements have indeed changed their social class base, which can now be conceptualised as a middle class base in contrast to the blue collar class base of old social movements (Wright, 1985; 1989). Kriesi (1989) argues that even the idea of a new oppositional middle class forming the basis of social movements is both too broad and too narrow. It is too broad because it is not the whole middle class that supports NSMs, but rather the younger generations. It is also too narrow because there are other groups/social classes besides the middle class that support new social movements. Therefore, the support for some movements is best described in terms other than class or social base, but rather in terms of values, ideas and goals supporters agree with.
83
main adversary do not really ft the PPM. Protestors and supporters with full citizenship rights (like those in NSM theory) do not have to wait for expanding political opportunities and reduced repression mechanisms. However, this paper does not generalise the PPM to other social movements or political parties. Its sole claim is the explanatory power of the PPM for FRP emergence, a political party that targets the state and aims for political participation.
The Front National in the early to mid 1980s To empirically verify the above theoretical model, this paper uses the French FN as an illustration. Since it is considered to be the mother party of all contemporary West European FRPs, and it seeks electoral power in a political system that is generally considered unfavourable, this is the most evident case to apply the PPM to. In the following section I will discuss the three dimensions of the PPM for the French context at the time of the FN’s successful emergence (early to mid 1980s). Political opportunities, as described by Kriesi e.a. (1995) and Van der Heijden (2010), have not been considered favourable for the FN in the early to mid 1980s. Following the “Trente Glorieuses”, France experienced a period of economic and social crisis: Inflation, unemployment, lack of integration by foreigners, increasing small delinquency, decreasing importance of culture, and a decreasing quality of public life (Bréchon, 2009). Until 1982 83, French traditional parties failed to adjust, and incorporate these societal concerns. As seen in many West European countries, France became a country of two speeds following post modernisation and post industrialisation (Wieviorka, 1992). Following this, the original cleavage structure, as described by Lipset and Rokkan (1967), decreased in importance and gave room to new identification possibilities. The original economic bipolarisation disappears and a new “ immense group central ” emerges (Giscard D’Estaing, 1976). In other words, the newly found social questions provided (new) parties focussing on the socio cultural dimension with political salience. Traditional parties’ failure to act and societal frustration opened up political space for populism and parties being critical of the (democratic) system. More specifically this was translated in a vast unhappiness with the PS (i.e. Mitterrand), and its inability to deal with the crisis, which eventually led to political dissatisfaction and disenchantment (Perrineau, 1997). Even compared to other European countries, France was characterised by a lower level of confidence in political parties and a weak partisan organisation (i.e. low party identification). The evolution from mass parties to cartel parties (Katz and Mair, 1995) has also been more important in France than in other
84
European countries (Bréchon, 2005), thereby allowing populism to flourish. Generally, these factors have led to the crumbling of the traditional left and right. With the exception of one major change, the French institutional state structure has been rather stable since the creation of the 5 th Republic. Until 1986, the French electoral system has been a two round majoritarian system (i.e. district magnitude = 1), with an assembly size of 474 and an effective threshold of about 37.5 percent (Lijphart, 1994). This means the opportunities for smaller parties, like the FN, to even advance to the second round were limited. In 1986, the PS changed the electoral system to a d’Hondt PR system to allow FN emergence, and general fragmentation of the right. Each of the 96 départements became an electoral district, with an average district size of 5.79 seats, a legal threshold of five percent at the district level, and a total assembly size of 556. The effective threshold decreased to about 11 percent and allowed the FN to obtain 35 seats in the French Assemblée Nationale . The RPR UDF 23 obtained a majority after all, but the result created a sort of political legitimacy for the FN (DeClair, 1999; Shields, 2007). Before the change back to two round majority in 1988, there were no substantial subsidies available for the FN. If a candidate obtained more than five percent of the regular vote on the first ballot, which they FN did not, they would be reimbursed for the costs of the campaign (i.e. letters, flyers, posters, propaganda, etc.). Additionally, they would be reimbursed for their deposit of FFr. 1.000 (about USD 330 in 2012), which was lost for those candidates who did not reach the five percent threshold. Generally, in France, public funding and state subventions to political parties was inexistent until 1988. Regardless of these rather limiting state structures, the FN has taken advantage of the available political space. The change to a PR system, and the coalition between RPR and UDF on the right opened up political space for a nationalist and populist right. The 1983 coalition with the RPR UDF list in the second round in Dreux (and several others on the local and municipal level in 1984 and 1985) provided the FN with increasing legitimisation and decreasing stigmatisation. A Figaro SOFRES barometer substantiates this in 1985, and showed nine percent of French citizens declaring they have a good opinion of Le Pen, and a same percentage would like him to play a bigger role in French politics 24 . Generally the mainstream had difficulties how to approach this new phenomenon. On one hand, the traditional left demonised the FN (e.g. via SOS Racisme ), and used anti racism
23 The coalition between the Rassemblement pour la République (Rally for the Republic) and the Union pour la Démocratie Française (Union for French Democracy), together comprising the traditional right in France at that time. 24 SOFRES, 1985, Opinion publique , Paris, Gallimard, pp. 182–83
85
campaigns to fill the void of its empty ideology during the early 1980s. At the same time it is perfectly content to let the FN compete with the traditional right for the ‘rightist’ vote. Following Meguid (2008), we can say the traditional left takes an ‘adversarial’ strategy toward the FN. The traditional right, on the other hand, took a more diversified approach. On the national level it opposed any kind of coalition with the FN, thereby (also) taking an ‘adversarial’ approach. On the municipal and local levels, the traditional right did get into coalitions with the FN, thereby taking a more ‘accommodating’ approach. Following Meguid’s theoretical framework, this combination (on the national level) is beneficial for the FN, and eventually leads to vote gains for the FN (which it did up until the early 2000s). In France, media access is generally very difficult and restricted for political parties. Even though paid political advertising used to be possible (until the early 1990s), media access for the FN in the early 1980s was difficult and minimal due to specific requirements and limited financial resources. In periods between elections, 20 minutes of free time was available every month for those parties represented in parliament (which the FN was not). During election campaigns, parties represented in parliament received additional airtime 25 . Up until 1982, the French (public and private) media actively refused the FN and its leader any kind of access. In May and June of that year, Le Pen addressed several letters to Mitterrand discussing the media boycott the FN suffered, after which Mitterrand instructed French media to “devote equitable coverage to the FN” (Shields, 2007, p.196). This was followed by a number of media appearances by Le Pen on both radio and television, with as milestone his prime time interview in L’heure de la vérité 26 where he presented himself as a politician rather than a social stigma, and described himself as a ‘Churchillian democrat’. In hindsight, both Le Pen and Mitterrand acknowledge the importance of this emission, and the more general presence in the media scene, for the FN’s credibility and future success. This can be seen in the Figaro SOFRES polls following the media breakthrough 27 . Up until the beginning of the 1980s, the political opportunities were rather restrictive for the FN. Even through traditional cleavage structures were decreasing in importance, it did not find a favourable state structure, advantageous mainstream strategies or any kind of access to the media structures. This changed somewhat in the mid 1980s, and political opportunities
25 Until 1986, this was about three hours of free airtime, which was equally distributed among the parties prior to the first ballot 26 Broadcasted on France’s public channel France 2 (then Antenne 2 ) on February 13 th 1984. 27 In May 1984, SOFRES findings showed 18 percent of respondents declaring themselves ‘very or somewhat sympathetic to Le Pen’, while levels of support for Le Pen’s ideas on immigration, law and order, and anti communism reached 28 percent, 26 percent and 25 percent respectively (SOFRES, 1985, Opinion publique , pp. 178 180 and 182 183).
86
become more favourable when mainstream strategies rendered the FN legitimacy and respectability, and allowed it access to media structures. From then on, we see political opportunities changing in a more favourable direction for the FN. It is more difficult to describe the mobilising structures that played a role in the emergence of the FN. Its three dimension are not necessarily present to the same extent. Since the FN was trying to establish itself in the political system, it did not yet have an existing solidarity network of voters it could rely on. Due to different factions and surrounding organisations it did, however, have a network of prior ties on which it could rely for solidarity, such as national conservatives (led by Roger Holeindre), revolutionary nationalists ( Ordre Nouveau ), Third Way solidarists (led by Jean Pierre Stirbois) nationalists (represented by Le Pen), Poujadists, Algerian War veterans, monarchists, the Nouvelle Droite (led by Alain de Benoist), Catholic fundamentalists (led by Bernard Anthony), neo fascists, and many other smaller groups. Upon its emergence, a wide variety of smaller groups were united under the leadership of Le Pen, as he created a common identity for those following and supporting the FN. As soon as the party obtained legitimacy and recognition (by forming coalitions with the traditional right), the FN obtained what it had been missing before: The institutionalisation of its collective identity. This sparked several politicians from both the traditional right and left to join the FN, and further expand its existing social networks. This, not only, improved the FN’s popular perception and support, but it also paved its path to normalcy. Demand side factors or environmental factors also contributed to the FN’s emergence (Mudde, 2007). Increasing discontent, the increasing number of scandals, political alienation, increasing mistrust towards the elites, dealignment and realignment processes, increasing complexity of electoral process, decreasing party identification, and the general failure of the traditional parties to adjust to these societal demands, have been crucial for the FN’s emergence. It mobilises (and monopolises) those who protest, and it provides the foundation for a common (populist) identity. Here, however, the FN needed to make sure it framed such external events strategically alongside its own ideologies, or master frame in order to fully take advantage of this. This brings us to the most unique and innovating aspect of the FN’s emergence: The master frame. As a concept, this is not new, however, the content of this master frame, and its application on an FRP, is. Since its original construction by the FN, it has massively spread throughout Western Europe, and it has served as the basis of the third wave of right wing extremism (von Beyme, 1988). Based on the ideological constructs proposed by Eatwell
87
(2003) and Mudde (2007), the FN’s master frame is composed of four elements: nationalism, xenophobia, authoritarianism, and populism. Contrary to Mudde (2007), and for reasons beyond this paper, I do not combine xenophobia and nationalism into nativism. Nationalism has deep historic roots in France and among French citizens (e.g. Jeanne d’Arc, the French Revolution, Napoleon, the Algerian War, de Gaulle, etc.) and has always been omni present in French political culture. It is something that is embedded in the French concepts of ‘Republicanism’ and ‘laïcité’. With the Italian MSI as an important role mode, the FN (led by the Ordre Nouveau faction) adopted this ideological framework in order to appeal to a broader right wing electorate. Since the FN aspired to attract majority support, it abandoned all reference to the past (contrary to the MSI) and focussed on its innovative character. Due to rising mass immigration, post modernisation and globalisation (Kriesi e.a., 1995), the issue of immigration became an important one. It became in conflict with the traditional French model of individual integration, which led to French citizens having increasingly the feeling immigrants invaded their country, their religion, and their neighbourhoods. Whereas traditional parties refused to include such challenges, the FN happily did so. Regardless of its original use in the 1970s by the French PC and the UK Torries (more specifically their traditionalist conservative wing), the FN easily established issue ownership as the issue became more salient in the 1980s. Within the FN, xenophobia and ethno pluralism were originally advocated by the Nouvelle Droite , which saw separate civilisations and cultures very much along the lines of Huntington’s ‘The Clash of Civilisations’. Therefore it must be emphasised this new master frame does not include biological racism, but rather incorporates more moderate forms of ethno pluralism. The authoritarian component of the master frame, following the approach of the Frankfurt School (Adorno e.a., 1969), is often connected to other attitudes and ideological features such as anti Semitism (as on occasion professed by Le Pen) and ethno centrism (exemplified by the welfare chauvinism advocated by the FN). The focus on social traditional values was on one side, a reaction to the rise of post materialism (Ignazi, 1992), and on the other side a deliberate implementation of (successful) MSI strategies. The FN’s strategies included extensive law and order policies (e.g. “sécurité et liberté” voted in 1980 28 ) and scapegoating immigrants for unemployment and insecurity with for example the slogan “Les Français d’abord” ( The French first ).
28 This law, enacted on February 2 nd 1981, reinforces security and protects personnel freedom. It was implemented by Giscard d’Estaing, but Mitterrand abolished important sections of the law in 1983.
88
Populism, as an ideology, was adopted by the FN due to its successful use in the past, and as a response to demand side processes. On one hand, Poujadism had successfully used populism in the 1950s 29 to represent the middle class and those suffering because of socio economic change. In the early 1980s, it served to attract those who were considered the ‘losers’ of globalisation, industrialisation and modernisation. It rejected the heterogeneous nature of multiculturalism and the omni present elitism in French society. By adopting populism as an ideological feature, rather than just a style, the FN became the crusader for the ‘man in the street’ against the political, economic and intellectual elites. This is largely considered to be the basis of the FN’s appeal to protest and issue voters The combination of these four ideological perspectives, and how they complement each other, provided the FN with a very successful master frame. Together with favourable political opportunities and mobilising networks, it proved to be successful and led to the emergence of the FN in the latter half of the 1980s. Additionally, the FN and its master frame served as a role model for other FRPs. After careful and specific frame alignment processes, this master frame diffused trans nationally to most West European FRPs: the Belgian VB, the Austrian FPÖ, the German REP, the Italian LN, the Danish DF, etc.
8. Concluding Comments This paper suggests the political process model as an explanatory model for FRP emergence. Although it is rather uncommon to apply a social movement theory to FRP emergence, the paper has shown that the equivalence of these two concepts (in this particular case) is undeniable. FRPs are actually more closely related to social movements when it comes to their emergence dynamics, than to traditional political parties. Additionally, most social movements combine political goals with more culturally oriented efforts, which constitute two crucial dimensions of an FRP. The PPM presents three necessary and sufficient factors responsible for FRP emergence: favourable political opportunities, open mobilising structures and a properly aligned master frame. The first dimension refers to all dynamic factors outside the FRP that facilitate FRP emergence. This includes media structures, mainstream party movement, cleavage structures, etc. The second dimension refers to a variety of organisational and social networks that support FRP emergence. Examples are formal and informal networks, collective identities, and strategic alliances. The last dimension refers to the more structural
29 Poujade’s Union de Defense Commercants et Artisans (UDCA) was rather successful toward the end of the 4 th Republic, obtaining 53 representatives in 1956, among which Jean Marie Le Pen.
89
and culturally factors that make sure individuals organise and/or mobilise around FRP ideas and proposals. Several social movement scholars claim the PPM is outdated and NSM theory provides a more complete and correct representation of social movement dynamics. Besides being empirically restricted to a limited number of ideal type social movements in North America, NSM theory is better at raising questions about the sources of identities than at answering them (Polletta and Jasper, 2001). This is mainly because NSM theory attempts to explain the “why” of social movement emergence, rather than the “how” (Melucci, 1985; Klandermans and Tarrow, 1988), which is what this paper focuses on. Exactly because NSM theory examines why social movements emerged, it blatantly neglects to include how such dynamics come to play. The same argument can be made for the “when” and the “where” of social movement formation across different, yet similar societies (Tarrow, 1994) To conclude, the use of the PPM as the explanatory model of a political phenomenon, such as FRP emergence, has been both theoretically and empirically justified throughout this paper. The paper indicates, describes and justifies the choice of this model above any other. The three key dimensions of the PPM have been conceptualised, operationalised and applied to the process of FRP emergence. There are, however, some minor comments and critiques that are relevant to the application of the PPM on the FRP emergence. Nonetheless, most of these merely indicate further specification or clarification of the model. In the end, the precise description of the three dimensions and their interactions/interdependencies serve as a fitted explanatory model for the emergence of FRPs.
90
3.2 Methodology: The Comparative Method
91
1. Introduction Following the systematic disregard of diffusion processes in FRP research, the role this study has is primarily one of conceptualisation, operationalisation and data generation, not necessarily one of extensive data transformation, hypothesis testing or large n data analysis. This requires a research design specifically tailored for these purposes. One of the core notions of the proposed research design is triangulation, which refers to the use of multiple strategies to support a study’s claims. This is not limited to the cross disciplinary nature of this study’s theoretical foundations, but also refers to the analytical and empirical sections. More specifically, this chapter describes the logic of triangulation in the description of the study’s methodology and data collection so as to overcome some of the problems commonly associated with single method or single theory approaches (Burgess, 1982). The primary strength of the logic of triangulation is its diversity, which is not only found in the variety of data collections methods, but also in the use of the comparative method as the principal data analysis tool. When proposing and describing this approach, it is also important to indicate how triangulation increases the quality of the research compared to some other approaches. Previous studies that have included diffusion mostly did so theoretically and descriptively (e.g. Rydgren, 2005a). This study can contribute, not only by providing a more comprehensive framework, but also by taking an alternative approach. The literature’s predominant emphasis on theory contributes to a structural absence of empirical frameworks and analyses that incorporate diffusion. Hence, it does not come as a surprise that the field lacks a structural foundation to build on. Part of this study’s objective is to substantiate empirical claims, which can then serve as a foundation for future studies of diffusion. Therefore, in short, this study adopts the roles of both theoretical specification and empirical initiation of trans national diffusion dynamics. This chapter describes the data collection and data analysis methods this study uses to specify this latter dynamic. A first section discusses the distinction between quantitative and qualitative data, its requirements and its possibilities, which allows the researcher to describe and justify the predominance of qualitative data throughout this study. The following sections describe the data collection methods that are used throughout this study. Following the logic of triangulation, a set of supporting or complementing data collection methods can be divided in primary, secondary and tertiary data sources. Most importantly, elite and expert interviews are the study’s most sizeable and most significant data collection tool. The following section describes and justifies how some of the methodological choices contribute to the quality of the obtained information. The researcher uses a comprehensive seven stage process to report
92
and optimise the data collection through interviews: thematising, designing, interviewing, transcribing, analysing, verifying and reporting. Each of these seven stages is specified to indicate its importance in the overall interview process and how it contributes to resolving the study’s main research questions. Following the extensive discussion of data collection methods, the final section of this chapter discusses the study’s primary data analysis method: the comparative method. Its overall focus on diversity serves as the primary justification of this method’s use and advantages for this study. Not only does it allow the study of both differences and similarities between FRPs and within the FRP family, it is also the most suitable analytical method for explorative studies such as this one that seek to contribute to existing theoretical frameworks. This section pays provides a rationale for the study’s case selection, why it prefers a case oriented approach as an alternative to a more variable oriented approach, how this approach influences the study’s conceptualisations, and how certain pitfalls of the comparative method have been taken into account when designing the study.
2. The principality of a qualitative approach Qualitative research is often wrongfully perceived as a set of unconnected and isolated methodologies that do not have a set uniform perspective. However, with a proper analytical framework, one of qualitative research’s main strengths is its ability to provide an analysis of dynamics and developments in situ (Denzin and Lincoln, 1994). Qualitative methodology is often received rather sceptically or even looked down upon by some of the more quantitative scholars. Some of their core arguments are the absence of statistical generalisation, the lack of reproduction possibilities, its small n, researcher bias and difficulty to rigorously analyse its results. In its turn, some of the quantitative research can often be questioned for its failure to provide uniform models, its difficulty to add to the general scientific knowledge, and its disregard for theoretical generalisation. This chapter discusses different approaches and methodologies that protect against such possible biases and enhance the research’s rigour (Mays and Pope, 1995). The difference between a qualitative and quantitative approach mostly lies in the principles qualitative research holds: breadth, precision and accuracy (Becker, 1996). More precisely, instead of quantitative research’s emphasis on reproduction and generalisation, qualitative research focuses on intrinsic value and accuracy. Regardless of its more limited scope and smaller case selection, qualitative research allows more analytical depth and variation. Rather
93
than focussing on and isolating individual variables, qualitative research allows for the study of the broader analytical processes and explanations. Qualitative research often goes beyond the more common testing of hypotheses and engages in a more dialectic process between the research question and the data. On one hand, qualitative research can add to or specify the more general and quantitative research. On the other hand, qualitative research can serve as an instigator for further (more quantitative) research. The more flexible nature of qualitative research allows for the constant adjustment and regeneration of information while the research is in progress. This allows for immediate and continuous improvements of theoretical and analytical shortcomings of a study. From these brief descriptions of qualitative and quantitative research, it must be clear these different approaches do not necessarily compete with each other. Rather, they have the appropriate characteristics to, under the right circumstances, serve as complementary approaches. Most of contemporary FRP research, and more specifically the analysis of the explanatory factors of its development, is quantitative in nature (see chapter 2.1). Since this study provides a comprehensive framework of specific cases, it scrutinises the diversity among FRPs and it seeks to improve existing frameworks by introducing new explanations, the data and/or knowledge generation characteristics of a more qualitative focus are more appropriate. In addition, a more qualitative approach is more suitable to incorporate the proposed shift from a variable oriented to a more process oriented approach (see below). More specifically for this study, a more qualitative approach, with its specific set of advantages and benefits, helps examine why and how certain dynamics operate under specific circumstances, and it also helps study research questions that are not easily quantifiable. It focuses its analyses on the time and context specific operation of certain broader social constructs and dynamics of the socio political environment, such as diffusion patterns, which requires the researcher to take a broader environment into consideration, while still providing an in depth analysis of each case (see section on the comparative method).
3. Data collection: The far right party landscape The original FRP studies (e.g. von Beyme, 1988; Ignazi, 1992) were great in depth, more qualitative studies, whereas contemporary studies are for the most part quantitative in nature. Due to temporal and financial restrictions, few studies are capable of collecting or creating their own data, and, therefore, most scholars choose (or are forced) to rely on existing material, be it literature, data or analyses (notable exceptions: Mudde, 1996; Klandermans and Mayer, 2006). For analytical reasons, such dependency on secondary and tertiary
94
data/literature can be problematic. Since one of the principal objectives of this study is the introduction and analysis of new and undiscovered concepts such as master frame and diffusion, the non existence of data should not come as a surprise. In addition, dynamics and processes (most notably diffusion patterns) are an important focus of this study, rather than variables, which are what most existing literature analyses. Therefore, the researcher is compelled to collect data rather than rely on an existing dataset, and construct a dataset specifically tailored for this study.
A. The supporting role of primary data sources: Far right party archives Lichtman and French (1978) proposed the most commonly described differentiation between primary, secondary and tertiary data based on timescale and intended audience criteria. They describe primary sources as evidence that was part of the event in question and was not intended for external use or circulation (e.g. governmental documents, private papers, unpublished diaries and other forms of communication). Secondary sources are available for external use and circulation either at the time of the event or shortly thereafter (e.g. surveys, manifestos, governmental publications, newspapers and other periodicals). Tertiary data is available after the event and usually constitutes a reconstruction (e.g. academic literature, books, dissertations and other interpretative publications). As a theoretical principle, primary sources are quite valuable. They are not written with the intention of being circulated or having third parties analyse and interpret them. Some scholars would even argue primary sources constitute facts, whereas secondary and tertiary sources can be considered literature. According to such logic, if the primary sources are unavailable for analysis, the research should not be continued because the empirical basis of the research will be flawed (Webb and Webb, 1975). However, a more adequate and contemporary approach is to be more flexible in one’s approach to data collection, and emphasise the importance and the merit of diversity in data sources, i.e. triangulation. The most interesting primary data sources are some of the records FRPs hold; including their affiliated groups and some of the associations they work with or belong to. When deciding which archives, libraries, documentation and/or records to select, the value of secondary and tertiary literature cannot be underestimated. In other words, some of the existing academic and journalistic literature can help the researcher identify specialists and important actors in the field that can serve as the basis for the more interactive data collection methods, such as interviews and, to a lesser extent, participant observation.
95
When selecting a primary data source, one must take into account certain criteria, such as the feasibility to obtain access. The higher the degree of FRP stigmatisation, the more unlikely it is an FRP grants access to its archives. Hence it is more likely a moderate FRP grants access to its archives. Directly related to this is the desirability of the access to a particular FRP’s archives. Following the research questions and the exploratory purpose of this study, it would also be fair to claim that if a FRP is rather small, extreme and/or relatively new compared to its fellow family members, its archives are unlikely to be beneficial for this study because those their role in trans national diffusion has been limited. It would be more beneficial to obtain access to the archives of more sizeable and more significant FRPs. Practically there are also certain restrictions that need to be taken into account. Due to a limited timeframe, the researcher has not attempted, nor gained access to every archive. Following limited financial resources and mobility, the researcher has not been able to challenge for access to each FRP’s archives. Following limitations in linguistic knowledge, the researcher would not be able to analyse each FRP archive. Because of the economic law of marginal returns and the redundancy to study a complete dataset, the researcher studied and analysed primary materials of two of the most prevalent FRPs. Practically, the researcher was only granted partial access and he was not allowed to make photocopies or take information outside. However, considering the purpose of the primary data source, they have been a great source of information, both for the further development of the constructs of diffusion as for the empirical illustrations of trans national diffusion.
B. Quality control: The value and limits of primary sources Primary sources, and their general importance, indicate that the workings of an FRP can also be a vital resource for a political analysis. A study should not always exclusively rely on academic literature and survey studies. Regardless of its many advantages, the use of primary data sources, and more specifically the use of FRP archives, also has several drawbacks. The literature describes these possible limitations alongside four different quality control criteria: authenticity, credibility, representativeness and meaning (Platt, 1981; Scott, 1990). In general, the authenticity principle refers to whether something is genuine (Scott, 1990). It includes the consideration whether this something is original and whether it is possible to confirm its identity. Specifically related to the information in the FRP archives, it is not that straightforward to authenticate the information there. They are not publicly overseen entities; hence one cannot make specific claims about the authenticity of the information. Nonetheless, it is possible to check the originality and the validity of the
96
information found in the archives by crosschecking and cross referencing this with other data sources, which is done extensively in this research project. At no moment, did the researcher find contradictory evidence between or within different data sources. The credibility criterion indicates how sincere and accurate the author of a document was. To best evaluate this one must assess the conditions under which the information was produced and the author’s interest in its production. Since the main characteristic of the primary source data is internal circulation, its underlying conditions need not be questioned too much. They can be assumed to serve either informative or informal purposes within the party. For other sources, such as party manifestos, which are created with the intention to be broadcasted and diffused, one can and should justifiably question the credibility. The third criterion is that of representativeness, which indicates if the consulted information is representative of the totality of documents. An unrepresentative sample of documents can lead to wrongful conclusions. In addition, the inclusion of too many (unnecessary) primary sources can lead to top down bias, where the principal focus of the data collection lies on the primary source. However, the danger for such biases is limited in this case as a direct consequence of the variety of different data sources (Plummer, 1983). The researcher relies on primary sources and academic/journalistic literature and he collects data himself in the form of interviews. A last criterion, meaning, refers to the actual and literal meaning of the documents. Based on their individual value sets, each researcher can have his or her own interpretative understanding of certain sources, and thereby give different meaning, or even, importance to them. Therefore, it is important the researcher has sufficient contextual and specific information, in addition to the authors’ intention and interest in the production of the information, so as to interpret the actual meaning of the document correctly. Using the interview as a complementary data collection method, which is what constitutes a large party of the data collection of this research project, can do this.
C. The introduction of secondary and tertiary data sources The analysis of primary data sources can be of great value for any study, and despite some of its limitations (see above), it is generally considered one of the most reliable data sources. Regardless, it is not a data source that is very often used in political science, with perhaps the exception of large and often publicly funded projects. Scholars like Mudde (2007) emphasise the positive contributions primary data sources can make and question whether political research can even be done properly by only relying on secondary and tertiary data sources.
97
One of the most prevalent problems with secondary and tertiary data sources is their reliability. How well do party manifestos represent the party’s ideology? How objective are newspapers? How partisan are party publications? Which medium represents the majority of the party? It would be fair to claim that governmental or parliamentary publications are most reliable, however, here they offer little representativeness and thus limit the chances to make an original contribution to the literature. The analysis of newspapers or party publications could form an alternative, however, they are spatially and temporally bound in their information and their analysis. In addition, the reliability and accuracy of some partisan publications is questionable at best, most notably when it comes to the information on FRPs. Hence, by itself, secondary data sources can never comprise the complete dataset unless the researcher engages in a full analysis of the entire publication process. Nonetheless, party publications do have an informative value. On one hand general party publications, for the partisan or even broader audience, can be written or oral. Written party publications can be newsletters and newspapers ran by the party, newsletters and newspapers favourable to the party, party logos, columns/editorials, blog entries, books authored by party members, public statements, written correspondence, etc. Oral party publications can be speeches (although those can also be considered written), conversations, discussions, interviews, debates, parliamentary questions and answers, rhetoric, etc. On the other hand, party manifestos, which are mostly used by researchers and analysts. Manifestos are the political parties’ programme that comprises its norms, values, and ideologies, or in short, that gives an indication of the constructs of its master frame. Usually a party publishes a manifesto right before an election, however, some parties publish a manifesto more often. A brief overview of these secondary sources can be found in appendix one, whereas an overview of the tertiary sources can be found in the literature review (see chapter 2.1).
D. Conclusion The combination of these three data sources serves as the basis of FRP classification, and the extensive FRP conceptualisation and operationalisation. In addition, it also serves as the basis for the overall data collection, which helps to illustrate the claims this study develops. Primary data sources found in the FRP archives serve as a first indicator of possible master frame characteristics and serve as a reference for further data sources. Secondary and tertiary data can provide a more precise description of diffusion characteristics and help confirm past events and occurrences that might contribute to the identification of diffusion dynamics. The combination of these different data sources exemplifies this study’s triangulation approach.
98
All three data sources act as selection mechanisms for further data sources and/or references, which then serve as a foundation for a further, more detailed data collection process. As part of this study’s triangulation approach, the researcher uses these preliminary data collection methods to complement other, more extensive and substantial methods. In this particular case, the researcher relies on both expert and elite interviews to serve as the principal data collection method of this study.
4. The primary data collection tool: Elite and expert interviews In this study, and following the approach of Kvale (2007), some of the interviews’ most important characteristics, possibilities and limitations are outlined following a sequence of seven stages: thematising, designing, interviewing, transcribing, analysing, verifying and reporting. The first three stages are pre interview, whereas the latter four are done post interview. In the very beginning, the purpose of the research project and the conceptualisation of the theories are formulated and thematised . Upon completion, the researcher can design the interview process, hereby also taking into account the ethical implications of the study. The interview itself is then conducted with a reflective approach to the desired knowledge and the research questions sought to answer. Soon thereafter, the interview is transcribed and the material is prepared for analysis. Next, according to the appropriate modes of analysis , the researcher evaluates the prepared data. Subsequently, the findings are verified , which means the researcher determines their reliability, their internal and their external validity by carefully scrutinising them with other interviews, the literature and other scholars. In the end, the reworked findings, together with the implemented methods and approaches, are reported to the academic world in the form of a dissertation.
A. Thematising the interview The researcher has to integrate some of the study’s more imperative questions in his preparation of the interview process: why, what and how. In other words, the thematising phase refers to the formulation and specification of the research hypotheses (what) and a theoretical clarification of the study’s purpose (why). Both of which are discussed at length in the following chapters. These questions form the basis of a correct methodological design and an appropriate analytical framework so as to engage in meaningful analysis (i.e. how). The overall rationale for this methodology can be found throughout this chapter. For a more elaborate discussion of the conceptual and theoretical foundations of FRP development, the researcher refers to the literature review. The conceptual and theoretical discussion of
99
diffusion is not as extensive and profound as the literature review on FRP development because the academic background of diffusion is so cross disciplinary, that it would be difficult and impractical to unambiguously report this (see chapter 3.3). Without an overview or analysis of existing literature, it is difficult to indicate what this study’s scientific contributions are and whether its findings are innovative. In this case, the existing research has neglected the role of diffusion patterns in the discussion of FRP development. This study claims that diffusion has been wrongfully overlooked, but also that diffusion needs to be considered a prominent factor in the discussion of FRP development. One of the main purposes of this study is thus to increase scientific understanding of diffusion pattern in FRP development. In order to do so, this study proposes a more process oriented approach and seeks to integrate this into the existing, more variable oriented literature. To gain an increasing understanding of this new focus, the researcher seeks to acquire practical knowledge of events, dynamics and socio political phenomena through interviews. Interviews usually have one of two distinct characteristics: They either part from a hypothesis testing rationale or from an exploratory (i.e. hypothesis creating) rationale. Most contemporary studies take a hypothesis testing approach, however, this study contains both approaches. The early stages of the study claim the proper alignment of the newly diffused master frame is an important explanation for successful FRP development. The researcher investigates the new master frame and examines several (theoretical) claims concerning diffusion patterns in the early stages of FRP development. During the interviews, this is dealt with in the first sections of the interview. The latter stages of the study are more exploratory and discuss specific trans national diffusion patterns with other FRPs, including their characteristics, nature and extent. It goes beyond the diffusion of the master frame, but also discusses more interdependent forms of diffusion, such as groups in the European Parliament, international meetings, structural contacts, etc. Due to the absence of data, the interview is much more open and flexible in this phase, with only a general structure to guide it. Familiarity with both the study’s theory and the methodology are necessary if a researcher wishes to be explicit, comprehensive and contribute to the scientific literature. This is not only obtained through a literature review but several studies have shown that it is beneficial for the quality of the interviews and the study to be present in the environment where the interviews are conducted (Lave and Kvale, 1995). This form of participant behaviour allows for the study of behaviour, international presence at meetings, social interaction and more generally, the environment in which FRPs and their functionaries operate. This can provide more valid knowledge than inquiries during interviews.
100
For this study it serves as a more informal complement to the two core data collection methods: elite and expert interviews. It introduces the local language, the daily routines, and can even provide insights in the power structures of FRPs. Perhaps most importantly, it creates visibility for the researcher, which, combined with support from high ranking FRP members, positively contributes to the legitimacy of the researcher and his access to potential interviewees. In practice, the researcher has engaged in a wide variety of social activities organised by FRPs, and this in various countries, both in and between election times. The content and extent of participant observation has been discussed more carefully above.
B. Designing the interview The design of the study provides a structured overview of the dilemmas the researcher faces and how their clarifications are justified in this particular study (Flick, 2007). The design of an interview has a logical temporal dimension, which functions as one of the guiding principles throughout the whole interview process. Every study starts with an outline of its goals, aims, objectives and ambitions. In a relatively non standardised study such as this one, the researcher chooses its methodology only after the subject and purpose of the study have been set and expressed. Such methodological choices must be considered carefully as they shape choices made in following stages. Simultaneously, it is important to consider that the practical implementation of these phases is more interdependent than theory describes.
The quality and quantity of potential interviewees The selection of interviewees is not random; most of them are privileged, i.e. the interviewees have high political knowledge. This study is mainly interested in those individuals who have access to privileged information and have political experience. Unlike most of the literature, the researcher makes a difference between elite and expert interviews. The former refers to interviews with (mostly high ranking) party officials, whereas the latter refers to interviews with FRP experts (e.g. journalists, scholars). Practically, the former category outweighs the latter because most FRP experts are generally rather uninformed on the subject of trans national diffusion patterns. During an interview there is always a certain power asymmetry, usually in favour of the interviewee, because the interviewee has more information. Extensive knowledge on the interview topic and meticulous preparation can (among others) increase the interviewee’s respect for the researcher and thereby shift the balance in favour of the researcher (Foucault, 1980; French and Raven, 2001). If such a power shift occurs, the researcher can contribute
101
his conceptions to the interview dialogue and possibly challenge certain statements made by the interviewee. This should be done with caution as this can either alienate the interviewee or increase the quality of the dialogue. In this particular case, such an approach proved somewhat beneficial, particularly with those subjects that with an academic background. The number of interviewees is based on the economic principle of diminishing marginal returns. Since there is little to no existing (qualitative) FRP research that includes the dynamics of trans national diffusion, the initial interviews are incredibly valuable and serve as an important starting point for this study’s analysis. The researcher relies on two unconventional sampling techniques to form the foundation of its interview pool. Firstly, based on some of the primary and secondary data sources, the researcher makes a selection of which individuals are most relevant, useful and important to interview. From this preliminary list, the researcher chooses a selected number of high ranking FRP officials that belong to the more sizeable, successful and active FRPs. They provide the study with an enormous amount of information and the researcher with credibility and legitimacy. Secondly, these few high ranking party officials serve as indicators for further interview sampling. After the debriefing, the researcher asks each interviewee if he or she knows other individuals who can be relevant for this study. This is referred to as ‘snowball’ sampling and is very common among qualitative scholars. After each interview, therefore, the list of potential interviewees grows. The researcher applies rigorous selection criteria to these potential targets to determine whether they are good candidates. Gradually, as the number of interviewees increases, the marginal contribution of each interview progressively decreases. At a certain point, the researcher reaches the maximal possible utility of one additional interview and moves from a phase of contribution to one of saturation. Since the process of interviewing is rather time intensive, the researcher wishes to avoid any such inefficiency. The question is, therefore, when this saturation point is reached. In leading studies, the average number of interviews that generally proves to be useful and efficient tends to be around 15, with a standard deviation of ten (Siegel, 1956). However, this number is influenced by all sorts of factors such as the time of the interview, the resources of the interviewer, and most importantly the length of the actual interview. So there is no uniform rule as to what number of interviews represents the inflection point one wishes to avoid. When doing interviews, simultaneously analysing previous interviews and preparing for future interviews, this decision remains an ad hoc decision. In this particular case, the first three to four interviews have contributed greatly to the empirical data this study relies on. They served as a selection mechanism for possible
102
interviewees and have added new and important subjects to the selection. In short, they carry a more sizeable weight in the study’s empirical framework. The amount of information gathered from the interviews has been considerable since most of them lasted between one and two hours. Therefore, one can observe the inflection point relatively quickly. The researcher has done a total of several dozens of interviews, however, after about ten interviews a decisive amount of information became to be repeated and it was safe to say the researcher entered the saturation phase.
Interviewee accessibility One of the foremost challenges when a researcher engages in elite interviews is the access to those elites, if this is at all likely or possible (Stedward, 1997). Typically, members of this elite group are busy, not always (geographically) in the same place, and often need to be provided with motivation why they should participate in an academic study (Goldstein, 2002). In other words, in order to interest elites, it is important to indicate why their contribution is important and how the topic can be of interest to them. There are some standard steps a researcher can go through but, in the end, it is important to anticipate and personalise the approach depending on each interview target. Throughout this process, it is possible a researcher has to deal with so called ‘gatekeepers’ or with the typical standardised ‘one type fits all answers’ (Stedward, 1997). Various plans and approaches were designed to help surpass gatekeepers and standardised answers. However, surprisingly, the access to high ranking FRP officials proved much less problematic than originally anticipated, which resulted in a response rate of more than 90 percent. In general, FRP access has only been problematic in Austria, for the FPÖ and the BZÖ. In this case, the researcher had to deal with gatekeepers, who refused access to the most trivial and common object (such as manifestos, party publications, etc.), and standardised answers to any requests or written attempts at communication. Specifically, the researcher attempted to obtain an interview with an Austrian high ranking far right politician who would have been able to contribute greatly to this study. He or she could have served as the foundation for the Austrian case in this study; however, he or she refused any and all forms of cooperation. At first, he or she responded to the written attempts of contact. He or she suggested to answer the interview questions via email. After respectfully declining this offer, the potential interviewee did not respond to any further forms of communication.
103
Individual Interviewer Bias Despite some linguistic restrictions, the study employs a single interviewer, i.e. the researcher. Since the nature of this study is rather specific, there are few researchers with the same combined thematic and substantive knowledge of FRP development, diffusion dynamics, master frames, their conceptualisations and their operationalisations. Because of financial limitations and strict confidentiality agreements (see ethical guidelines), it was not possible to hire and train an associate, or outsource certain interviews. In addition, the researcher has also singlehandedly done the transcriptions and post interview analyses. Even though this design has not been completely deliberate, the literature indicates that the combination of interview preparation, the actual interview and its analysis in one single person can lead to individual interviewer bias (Kvale, 1994). This refers to the differences in interviewee responses as a consequence of their individual reactions to the social style and personality of the interviewer, or to the presentation and/or formulation of a particular question. Such problems can be severe, however, they can also be controlled for fairly easily when the researcher (very generally) takes the context into consideration. In addition, when the researcher is the interviewer, the danger for this bias is often increased by the self fulfilling prophecy. Since the researcher also participates in the FRP environment, the risk of potentially influencing interviewees becomes more significant. However, both parties respect the necessary boundaries between them and agree to the interview conditions (see ethical guidelines), which allow the researcher to control for such possible bias. Furthermore, in this particular case, individual interviewer bias is not necessarily something that should be controlled for; rather this can also be beneficial. It can serve as a legitimation tool or as evidence of the researcher’s objective and truthful intentions.
C. Conducting the interview To maximise efficacy, the constructs of each interview study depend on its limitations and its objectives. In this specific case, the researcher chose a semi structured life world interview format to tailor its specific needs. A semi structured life world interview describes the life world of the interviewee with respect to the interpretation of the described phenomenon, i.e. trans national diffusion patterns between FRPs. They are direct, in depth, controlled, uniform, and allow for intervention if necessary and observation of secondary factors such as tone and facial expressions. More specifically, the researcher provides a solid description of the new master frame and its trans national diffusion, while also interpreting the diversity of trans national diffusion patterns throughout the development of FRPs.
104
Two of the more important benefits of this specific interview technique are its openness and its flexibility (Barnes, 1992). Its flexibility allows the researcher to make changes in the sequence and the style of the questions so as to maximise the quality of the obtained information. In this process, the researcher uses an interview guide as a general set of suggestions rather than as a set list of questions (see appendix 3 and 4). The researcher allows the interviewee to teach him what the problem, the question, or the situation is (Dexter, 1969). This gives the interviewee an opportunity to answer the question but also to expand on it. It is precisely these opportunities that allow the researcher to discover information that was not necessarily anticipated beforehand. Similarly to participant observation, expert and elite interviews that are designed according to the above described format have certain limitations. Most importantly, the entire interview process is extremely time consuming (Berry, 2002). The preparation for each interview can take up to several hours (excluding the time spent on deigning the research project, conceptualisations, interview questions, etc.), the actual interview between one and two hours, each transcription between 4 8 hours, and the analysis also several hours. In addition, privileged interviews are also geographically limited, require a sizeable amount of resources, are in danger of leading to subjective analyses and require appropriate training of the interviewer (Hertz and Imber, 1995).
Different interview forms There are different epistemological conceptions of conducting interviews depending on whether the interview is seen as a process of knowledge collection or as a process of knowledge construction. The former conception regards interviews as a data collection method that is separated from the following data analysis. This is the case in most social sciences where knowledge is already available. The latter conception sees interviews and their analyses as an intertwined phase of knowledge construction. This is more along the lines of a post modern comprehension of socially constructed knowledge. For this study, the interview is more or less divided into two separate sections. The first section, which describes the master frame and its trans national diffusion, follows a process of knowledge collection. The second section, which describes the interdependent relationship between trans national diffusion and FRP development, follow a data construction approach. Considering their general purpose and central place within the study, the interviews have two main forms or functions: factual and narrative. A factual interview goes beyond the opinions and behaviour of the interviewee (which are not necessarily at the core of this study),
105
and focuses on obtaining valid and reliable information, so as to (for example) construct a reliable dataset. The literature indicates the importance of question wording and active listening, and how that affects the obtained information (Barry, 2002). Particularly for this study, the researcher seeks to gather facts and information about instances of trans national diffusion so as to create, describe, and analyse patterns and dynamics. A narrative interview has a slightly different focus, and allows the interviewee to discuss and/or communicate events, stories and/or experiences during the interview. These sorts of interviews often emphasise the temporal, social and meaning structures of an interview (Mishler, 1986). Such a natural cognitive and linguistic interview form is closest to a genuine and everyday conversation compared to any other interview forms. Rather than ask questions to gather facts and be an active listener, the main role of the interviewer is to be a listener and abstain from interruptions. For this study, due to the absence of empirical literature or a solid dataset, such an approach allows for the interviewee to share information previously unknown to the researcher. In short, both of these are usually used within the same interview. A factual approach allows the interviewee to share information and the researcher to create a dataset. At the same time, a more narrative approach allows the researcher to obtain information on the underlying patterns and dynamics. Within this study factual and narrative approaches do not contradict but complement each other, and the researcher often changes approach within the interview, depending on the appropriate needs and requirements.
Briefing and debriefing the interviewees A general briefing precedes each interview in which the researcher explains the situation for the interviewee (i.e. length and general outline of the interview) and provides a general indication of the purpose of the interview. At the same time, the interviewee has the opportunity to make any pre interview inquiries. If the interviewee does not initiate this, however, any questions are postponed and reserved until after the interview as not to influence the interviewee. After all, following the asymmetric power relations, the interviewee often attempts to obtain an understanding of the researcher and his research before freely divulging privileged information. Usually, this is done at the very beginning of an interview. In other words, a solid rapport and understanding need to be established between the interviewee and the researcher for an interview to be successful. Key attributes to achieve this are conscientious listening, a demonstration of respect and a general focus on clarity of questions and responses.
106
In this case, the researcher made the conscious choice to use a tape recorder. This allowed for a complete record of the often long interviews, and it gives the researcher the opportunity to actively listen during the interviews and to anticipate or react to the interviewee’s statements. The researcher always asked the interviewee for permission to record the interview immediately after the briefing. Probably as a result of the interviewee’s familiarity with the media, camera and public appearance, this never posed a problem. After each interview, the researcher always debriefs the interviewee. It is possible there is some tension between the actors at the end of an interview because the interviewee feels he or she talked too openly, or he or she wonders about the later use of the interview. At this time, while emphasising the ethical guidelines, the researcher answers any additional questions about the detailed purpose and design of the study. When this tension is present and the interviewee feels uncomfortable with the imbalance of sharing information, the researcher paid extra attention to a more detailed explanation of the purpose of the interview and this study.
The interview guide One of the core preparatory requirements for an interview study is the interview guide, which can be thought of as a script that structures and organises the discourse during an interview. Following the deliberate choice to engage in a semi structured interview study, the interview guide contains both general topics and specific questions. It does not include detailed sequences of carefully worded questions like a survey does. Such an organisation allows the researcher to follow up on an interviewee’s responses and even explore new directions following these responses. Contrary to more funnel shaped interviews, where an interviewer starts with detailed questions and works toward more general questions, this interview study relies on fairly direct and unambiguous questions that allow the interviewees to structure their responses themselves. Notwithstanding the overall flexibility of semi structured interviews, there are three general principles the researcher upholds when designing this study. First, regardless the detailed nature of the questions, the interviewee should have a list of topics, not necessarily ordered, which the researcher wishes to cover (Devine, 1995). Second, the researcher must prioritise his topics and questions to make sure all the essential and necessary information is obtained within the timeframe of the interview (Stedward, 1997). Third, the interview design must be so flexible to allow the interviewee to address new topic and extensively develop
107
older topics. In general, a balance between these three principles must be found to maximise the efficacy of an interview (Bryman, 2001). Each question in the interview guide has a thematic (what) and a dynamic (how) dimension. The former refers to its contribution to knowledge production, whereas the latter refers to its contribution to a positive interpersonal relationship between the researcher and the interviewee. Specifically for this study, the former refers to the expansion of knowledge on master frames and trans national diffusion, whereas the latter refers to the creation of a trustful and intellectual relationship between the researcher and the interviewee. This latter dimension must not be underestimated as the creation of a respectful relationship can lead (and has led) to access to both additional information and other FRP interviewees. Since there are sizable differences between the academic and the political world (e.g. jargon), the interview guide consists of two separate but interconnected parts: one with the project’s hypotheses or objectives and another with the interview questions that provide the researcher with answers to these hypotheses and objectives. The academically formulated hypotheses and objectives are translated into more field specific and colloquial questions in order to be understandable and produce spontaneous and useful responses. It is very well possible (and likely) that one hypothesis is translated in multiple interview questions/topics. Following the specific role of the FN as a mother party for other FRPs and its central role in the diffusion process, the researcher provides two separate interview guides for the FN and all the other FRPs, respectively appendix 4 and 5. The questions are formulated specifically so as to engage the interviewee to provide extensive responses. At the same time, it allows the researcher the opportunity to intervene when a response is seen as incomplete or irrelevant. This means the researcher is the key to a successful and information loaded interview. The researcher’s abilities to interpret a response, to place it into context and to generate unexpected but useful information from this, are key attributes of the interview as a data collection method. Such an objective cannot be achieved without extensive and detailed knowledge of FRPs and their development.
General interview quality The overall quality of an interview is largely determined by the qualifications and characteristics of an interviewer. The researcher himself plays a key role in this and must possess extensive knowledge on the topic, master conversational skills, and be proficient in the linguistic style of the political interview. Generally, in order to do a successful interview and to produce rich knowledge, a researcher must be knowledgeable, structured, clear,
108
diplomatic, sensitive, open, assisting, critical, remembering and interpreting. In addition, all these factors also have a substantial influence on the eventual analysis, verification and reporting of the findings. Flick (2007) describes the following six criteria that help determine the quality of a semi structured interview: • The extent of spontaneous, detailed and relevant answers from the interviewee. • The shorter the question and the longer the answer, the better. • The degree to which the researcher follows up and clarifies the meanings of the significant features of the responses. • Generally, the interview should be interpreted at the time of the interview. • The researcher attempts to verify his understanding of the interviewee’s answers during the interview (i.e. prompting). • The interview itself is ‘self reported’, i.e. it is autonomous and does not need extra explanations. Since this is not a common research topic, and the researcher only exceptionally sent an advance overview of the questions 30 , one can assume that most interviewees have been spontaneous in their answers. In general, the surrounding settings can also play an important role in determining an interviewee’s comfort and spontaneity. Each interview was done in the interviewee’s native tongue to facilitate the interviewee’s response patterns. The researcher specifically chose to conduct the interviews on a one to one basis in a quiet place where the interviewee is most comfortable. Most often, this was the office of the interviewee. As shown in the interview guides, all questions have been structured in such a way as to invoke long answers, especially in the latter sections of the interview where a more narrative and open ended approach is taken. As there is no dataset to fall back on, all interview responses are carefully clarified, cross validated, analysed and interpreted so as to avoid ambiguity. The autonomous nature of each of the interviews allows the researcher to specifically design each interview depending to the country and FRP specific situation, thereby increasing the quality of each of the interviews individually and the study as a whole. To maximise the quality of an interview, it is important to balance methodological requirements and some of the more practical challenges to get the interviewee to share information or to get the interviewee to put the focus on the research topic when he or she is wandering (Harrison, 2001). During this steering process, it is generally important to not
30 This was only done upon request of the FRP official/functionary (or usually his or her secretary).
109
approve or disapprove of the information the interviewee shares. In this particular case, the researcher cannot emphasise enough how important this notion has been (and continues to be in the reporting phase). The researcher presents oneself with the clear ambition to provide a professional, neutral and correct representation of FRPs.
D. The post interview process After each interview it is necessary to go through a post interview routine in order to maximise and optimise the quality of the obtained information. In order to be efficient, it is important to do this as structured as possible, so as to refute possible criticism on the process (see below) and to increase the efficacy and comparability of the study. Once all interviews are done, it is also important to collectively process them and transfer the information into supporting material for a dissertation. This post interview process consists of three important stages: the transcription, the analysis of the data and the reporting of the data (Kvale, 1995). The transcription of an interview occurs as soon as possible after the actual interview (usually within hours). This allows the researcher to take contextual factors, impressions and undertone into account. This transcription is a translation from a spoken to a written narrative. According to Ong (1982), oral and written texts are based on different cultural and linguistic constructs. When a researcher translates one form to another, this entails a series of assumptions and assessments by the researcher. In this case specifically, the transcription loses the interviewee’s body language due to the use of a tape recorder, and it loses the tone and intonation of the interviewee’s voice following the change to a written form. Therefore, critics of interviews, and qualitative data collection in general, see interview transcripts as weakened, decontextualised reproductions of interview conversations. However, very often such critiques do not correctly reflect how the transcripts will be used for a specific study. Particularly, the formality and the level of detail of the transcript give an indication of the eventual purpose of the transcription. Here, contextual and intonation factors play less of a role than for example when one examines the appeal of FRP politicians. Highly specialised transcriptions (e.g. ten Have, 1999) are neither feasible, nor necessary, for the conceptual and contextual analysis of trans national diffusion in a far right context. Its main purpose is to report the interview so the researcher can emphasise its facts, narratives, statements and information. Following the transcripts, interviews are carefully analysed so they can be used as the empirical foundation or illustration of theoretical claims made in the study. Generally, the quality of the data analysis depends on a researcher’s analytical skills, a researcher’s
110
knowledge of the research field, the sensitivity for the linguistic medium and the expertise of tools available for analysis of the linguistic expressions. There is not one uniform analytical method that can successfully analyse interview transcriptions but useful methods can be divided into those focussing on meaning and those focussing on language. This study argues for the increased importance of trans national diffusion patterns when describing far right development, so therefore the meaning of the interview outweighs the linguistics. While the analysis emphasises the substantial content of each interview, it uses two guiding principles to facilitate the interpretation of the information: mean condensation and mean interpretation. The former refers to a sort of meanings shortcut. It translates statements and responses into shorter and relevant only formulations, thereby keeping the language and the general understanding of the interviewee in order to preserve the original intention. The latter refers to the interpretation of the interviewee’s intentions, which provides a critical and comprehensive reflection of the interview. It gives the researcher the opportunity to go beyond some of the strategic and tactical responses. These analytical tools are part of the general method this study uses to analyse its data and support its claims: the comparative method. This method examines patterns of diversity within a given set of cases. Compared to some of the other methods, its main benefits are the examination of patterns of similarities and differences, the familiarity with the cases, the inclusion of cultural and historical significance, and the opportunity to advance theoretical frameworks. The following chapter discusses and portrays this method more extensively. Finally, once the data has been collected and analysed, it has to be reported, which is done in the form of this report, and will later be done in a more extensive dissertation form. One of the most important criteria for outside reporting is the objectivity of the study’s interpretations, which have to be systematic and careful. Since this is a valid and often expressed criticism toward most of the existing literature, this study seeks to avoid that at all cost. Therefore, the researcher has cultivated a set of creative, yet very dexterous, writing skills that treat writing as a ‘method of inquiry’ and allow for the objective analysis of a stigmatised, yet often severely misunderstood phenomenon.
E. Ethics and interviewing: The study’s ethical guidelines Following the wide variety of ethical implications this study can entail, the researcher specifically chose a consequentialist approach, which is based on Mill’s constructs of utilitarianism (Mill, 1910). This stands opposed to a more idealistic approach, advocated by Kant’s deontological theory. Describing the former, Kent argues, “people should seek to act
111
in accordance with the consequences of their behaviour and minimise suffering and well being” (2000, p.62). This statement, like consequentialism, is relatively flexible and allows for multiple interpretations. Therefore it is important to provide a detailed overview of the ethical constructs of this study in the following section.
The purpose of ethical guidelines Much like the experiment, the interview inherently deals with human subjects and relies on interaction with them to provide the researcher with data. In this process, it is impossible to operate in a moral vacuum. Therefore it is important to adhere to certain (scientific) standards and regulations so as not to treat the interviewees inappropriately. Even though European scholarship is more lenient and less standardised than Anglo Saxon scholarship when it comes to ethical guidelines, it is still a crucial part of interviewing as a data collection method that should never be neglected. Failure to account for certain ethical principles, especially in FRP research, can not only lead to biased information, but also to a failure to obtain any sort of relevant and objective data. The relationship between FRPs and the academic world is a complex one. On one hand, FRPs often accuse the academic world of being straw men of the more traditional parties, or more specifically, the traditional left wing parties. This accusation, in addition to their non traditional views, often comes across as harsh or judgemental. On the other hand, scholarship is often subjective, prejudiced and unfair in its analyses of FRPs, which contributes to the stigmatisation of the phenomenon. This, in its turn again, makes FRP functionaries wary and reluctant to cooperate with researchers. All together, this requires discretion, diligence and “principled sensitivity to the rights of others” (Bulmer, 2001, p.46), especially from the researcher’s part, to establish a successful interviewing relationship. Since all interviews have been deliberately done face to face, most of the ethical issues are most pertinent when the researcher is faced with his ‘data source’. It is the direct interaction with the interviewee that can influence the obtained information and the subsequent knowledge produced by the interview. Even though the interviewer needs to serve a ‘higher academic and scientific interest’, certain ethical considerations need to be made before the actual interview to guide and structure the interaction with the interviewee. This is even more the case when the interviewees are socially stigmatised, and the researcher represents a world that is too often unnecessarily and unfairly critical of FRPs.
112
The ethical guide: The most important ethical principles The ethical guidelines allow for specific ethical decisions the researcher makes throughout the interview process and the general data collection. In other words, as a part of the above described constructivist approach, this study follows a core set of ethical guidelines that allow the researcher to correctly interact with the interviewees, and to construct a study that follows the academic and scientific standards. The literature describes a wide variety of possible ethical principles and guidelines all sorts of studies should follow. For this study specifically, the researcher relied on the University of Berkeley’s Guidelines for the Protection of Human Subjects (1992) and APSA’s A Guide to Professional Ethics in Political Science (2 nd Edition, Revised 2008). The section below discusses the most relevant and important principles that comprise this study’s ethical guide. In an interview study, the importance of the interviewer should not be underestimated. The researcher serves as a crucial tool that makes decisions and can directly influence the obtained information. Therefore, the researcher (as a concept) also plays an important role in an interview’s ethical decisions. This goes much further than common ethical guidelines and principles (described below); the researcher needs to have moral integrity, commitment to a general set of principles and sensitivity. Since the researcher is the main instrument to obtain information, the awareness of value issues and ethical principles will help weigh ethical and scientific factors throughout the study. In short, a researcher’s integrity is an important explanatory factor when one evaluates a study (Brinkmann and Kvale, 2005). Due to the inherent interactive nature of an interview, the key potential problem is the tension between a professional and a personal relationship with the interviewee. A researcher might, accidently or purposely, start identifying with an interviewee to such an extent that he neglects to maintain a professional distance. Even though this does not necessarily constitute unethical behaviour from a ‘therapeutic relationship’, it does become an ethical problem when obtained information serves a scientific purpose, which is still the core intention of the study. Therefore, this researcher has always maintained an extremely professional relationship with each and every interviewee so as not to endanger the quality and the validity of the research and the possible contribution to the scientific literature. An essential ethical issue is that of confidentiality, which refers to an interviewee’s right and privilege to control the use of information obtained during an interview. Unlike for survey research, confidentiality is not assured via the anonymity of a computer, and the researcher must find a balance between the intrinsic contrast of ethical demands for confidentiality and the scientific demands for contributions. (Bulmer, 2001). In this study, the
113
interviewee is given absolute privacy and name, function, party and any other reference to his or her identity will remain confidential. No literal quotations are made, unless with explicit written consent of the interviewee. All of this is explained during the briefing so as to protect the interviewee, provide him or her with all the possible comfort and indicate what he or she can expect from the researcher. Generally, this enhances the interviewee’s security feeling, which can positively stimulate the information the interviewee reveals during the interview. Alternatively, some authors argue for confidentiality and subject anonymity in the research output in order to protect the researcher (Parker, 2005). Such reasoning does not allow the interviewees a voice in the study. Therefore, in order to respect the interviewee’s privacy but still produce academically significant research, when the researcher must use a reference it is done with a fictitious name and with the consent of the actual interviewee. When it is absolutely necessary to literally quote and refer to the interviewee by name, the researcher asks for the interviewee’s written permission before any sort of publication. Directly connected to this is the notion of informed consent, which indicates subjects should be informed they are subjects in an academic study and give consent for this (Bulmer, 2001). In other words, the researcher should be open about his intentions, the study’s purpose, and participation should not be coerced. Here, all interviews have been done with informed consent. Nonetheless, theoretically this issue can be problematic: “there are many situations where it is not possible to be completely open to all participants, and sometimes a full explanation of one’s purposes would overwhelm the listener” (Bulmer, 2001, p.52). Therefore, the researcher has resolved this potential issue by splitting this discussion in two. Before the interviews, during the briefing, the researcher gives a very concise overview of the purpose of the interview and only goes into details upon request (which never happened). The researcher emphasises that no one else will ever have access to the recordings and its transcripts. It is clearly stated the transcripts will never be reported or published as a whole. Rather, quotes and anonymous references (both approved by the interviewee) will be used to support claims and serve as the empirical basis of the study. After the interview, during the debriefing, the researcher can explain the design and purpose of the study in more detail. This is not done before the actual interview to maintain the spontaneity of the interviewee. More importantly, the researcher does not want to guide the interviewee towards any specific answers by revealing this beforehand. In other words, the opportunity for full information is given to the interviewee after the interview. Some of the other ethical principles are related to the relationship between the researcher and the interviewee, more specifically the trust between them. A researcher must
114
always seek to do good and cause no intentional harm. This is facilitated since the researcher is not part of a team. However, one must be careful not to justify every decision to the good of the study. Therefore, each step in the research project is held up to the scientific opinions of colleagues, researchers and advisors. Therefore, the researcher should avoid any and all forms of deception. It is such instances of covert research that have damaged FRPs’ trust in the academic world, but mostly the journalistic world. In practice, the researcher has always been open, honest and straightforward about his intentions, goals and objectives. Similarly, the researcher reports the results of his study in full and in a non selective manner. This is particularly important, as the researcher seeks to create a foundation for future research.
F. Conclusion This study seeks to shift the main focus from variables to processes, at least for the analysis of FRP development. Theoretically, it complements the existing frameworks with two underdeveloped notions, i.e. master frame and diffusion. Practically, it collects data, as there are no datasets that can serve as the empirical foundation of this study. In order to allow this study to gather the necessary components to support its claims, elite and expert interviews are the most appropriate tools to do so. The primary emphasis lies on FRP elites, as they have the hands on experiences and knowledge, whereas the experts are mostly used to cross validate the obtained information from the elite interviews (together with the other data sources). The primary advantage of this data collection method is its flexibility. This flexibility is also necessary to differentiate between two different stages within the data collection. The former stage refers to the early developmental stages of FRPs, whereas the latter stage refers to the more contemporary development. This differentiation is facilitated by the use of a semi structured form of interviewing. Following both these factors, the study risks to obtain a lot of information, without any scientific significance. Therefore, the researcher carefully structured the interview process in seven different stages. Before the actual interview, the researcher must meticulously prepare to maximise efficiency during and after the interview. At first it is important to be aware of the research field one is operating in. Following the important notion of triangulation, the researcher contextualises the research. Then, the researcher designs the researched process, taking into account the possible circularity of the interview process, the interviewee selection, and also the possible single interviewer bias. Once the actual interview process has been designed, the researcher must organise the actual interview by preparing an interview guide, which consists of a set of research questions and hypotheses coupled to a set of more concrete interview questions. The researcher also
115
discusses how he will maintain the quality of the interview process. Furthermore, the section describes the complexity of the transcription, the analytical purpose and how the researcher intends to report the interviews. Finally, the researcher provides an overview of the most important ethical guidelines he has followed not only during the interviews, but also throughout the entire research process.
5. Data analysis: The comparative method A majority of studies in today’s political science literature are either quantitative or, to a lesser extent, qualitative. As an alternative, comparative methodology has mostly been used in cross national studies to identify, analyse and explain patterns of diversity across different nations. The most distinctive feature of the comparative method is its study of patterns of diversity. It provides an analytical perspective for both differences and similarities, whereas most other methodologies only focus on one. Therefore, considering some of this study’s objectives and limitations, the researcher finds this an appropriate alternative. In general, the comparative method allows for the inclusion of a deeper understanding of socio cultural factors and how they can lead to diverse outcomes. Here, the researcher specifies why the comparative method was chosen for this study. The first section specifies the objectives of this methodological approach, together with some of its core strategies. The subsequent sections describe some of the more important strengths and weaknesses of the comparative approach. Case selection, sampling and the research design are the core components of the following section. The following two sections describe the importance of a proper conceptualisation and why it uses a case oriented approach, rather than a variable oriented approach. The final section answers the question how a smaller n study, such as a comparative study, can still propose valid and reliable causal inferences.
A. A study of diversity: Objectives and process The comparative method studies possible patterns of similarities and differences within a given set of cases (Ragin, 1987). It is often seen as a compromise or a bridge between quantitative and qualitative research, which is not entirely accurate. Qualitative research differs from comparative research because of their basic approach toward cases is very different. A qualitative study of similarities usually sees cases as multiple instances of the same phenomenon, whereas comparative research seeks to analyse the differences among these cases as well, i.e. their diversity. Quantitative research analyses differences in degrees and how different variables covary across a large number of cases, whereas comparative
116
research focuses on both similarities and differences, and focuses on becoming familiar with a limited number of cases. Therefore, the unique character of the study of diversity is its ambition to explain both the differences and the similarities within that given set of cases. The emphasis on patterns of diversity allows for the realisation of three important objectives: the examination of diversity, the interpretation of cultural or historical significance, and the advancement of theory (Ragin, 1994). Generally, the comparative method seeks to specify different causal conditions that are connected to different outcomes, thereby creating a possible classification within the cases. Particularly, the comparative method seeks to specify the different patterns of diffusion, connect them to FRP development, and identify which diffusion strategies are successful and which are not, thereby setting apart different FRP categories based on diffusion patterns. The main purpose here is to emphasise the diversity of diffusion patterns that can influence different FRP’s development. Contrary to qualitative research, comparative research has a good understanding of the larger category that includes this study’s (moderate number of) cases, as this has been determined well in advance and intentionally. As a rule, the larger category that establishes the boundaries and that comprises the cases is explicitly historically and geographically delimited. In this study, that would be the European FRP family. Such a primary classification (i.e. inclusion or exclusion in the party family, see chapter 2.1) provides a first analytical frame of interpretation. Following this, the discussion of historically or culturally significant events is often more a debate over the correct classification of events in general categories, which is a principal characteristic of comparative research. Among the different methods, comparative research is one of the best strategies to contribute to and advance existing theory because of its use of flexible analytical frames, its analysis of diversity patterns and its search for the causes of this diversity. Most of the literature uses the frame that FRPs have developed independently from one another, whereas this study analyses the validity of this frame, and investigates the legitimacy of a frame that specifically states that FRPs’ development is an interdependent process. Depending on the nature and the characteristics of the patterns of diversity, the study can change the original constructs of its analytical frames. Throughout the study, the initial analytical frames might need to be improved or revised in order to help make sense of some of the patterns of diversity. For this process, the comparative method suggests three unambiguous steps: First, the researcher must select his cases, then he must select his analytical frames so as to, finally, be able to analyse the patterns of diversity among the
117
different cases. Both the theoretical logic and the practical description of this study’s case selection are discussed extensively in one of the following sections. Following a case selection, the researcher chooses the analytical frames to help analyse them. Since the study is interested in diffusion patterns, the analytical frames constitute the different forms this can take, i.e. the diffusion mechanisms. The generally agreed upon frame in the literature is that of developmental independence between different FRPs. This study refutes such claims, and uses the independence frame as its null hypothesis. This study’s frames represent the different patterns, or mechanisms, of trans national diffusion that can influence the development of FRPs. In practice, the study describes four different diffusion mechanisms. It will show that both coercion and competition are not applicable to the case of FRPs, and that learning and emulation are the principal diffusion mechanisms that can affect FRP development. First, the study creates an econometric model that refutes the original null hypothesis and provides evidence for the importance of diffusion in FRP development. Then, the study argues that learning proves to be beneficial for FRP development, whereas emulation most likely has the opposite effect. As is the case in many studies, the researcher changes the composition of his analytical frames throughout the study, from four to two. The classification of the cases according to these diffusion mechanisms allows for the analysis of FRP development in each of these categories, and eventually for the identification of a possible causal relationship between certain diffusion mechanisms and successful FRP diffusion. The main objective is to examine similar and different patterns of diffusion and to identify a causal link with the success or the failure of the FRP development.
B. The comparative method: A valuable asset The methodological approach provides extensive opportunities for a detailed and comprehensive analysis of a complex phenomenon. Both quantitative and qualitative research would be limited in their analysis of trans national diffusion patterns between FRPs. A regular, variable oriented quantitative study would not be able to capture the dynamics between the FRPs, whereas a qualitative study would be able to capture the dynamics but would not be able to analyse the differences between FRPs and their diffusion patterns. A comparative approach allows the researcher to contextualise his information and gain in depth knowledge on the cases. This is a direct consequence of the limited number of cases a comparative study generally comprises. It is particularly interesting for this study to note that even if the focus lies on one specific (e.g. central) case, considering it in context of
118
other similar cases can also advance the knowledge of the original case. In other words with a heightened focus on the FN as the core of the diffusion patterns, the study also increases its information of other FRPs. Therefore, it is not necessary to include every West European FRPs in order to draw certain conclusions or make certain causal inferences. Generally, in the research process, classification precedes (comparative) analysis. Hence, a comparative study is only as good as the classification it is based on. In addition, a more comparative approach allows for the possible improvement of the original classification scheme that is proposed to capture the patterns of diversity. In this case, since the study claims diffusion patterns play an important role in FRP development, a classification based on diffusion mechanisms can help improve the original classification of successful/unsuccessful FRPs, or even the more general classification of which parties that belong to the party family. For example, if a party that is widely considered a member of the FRP family does not engage in trans national diffusion with any of the other party family members, one might ask serious questions as to whether this party belongs to the FRP family. A comparative approach also enables possibilities for the design and the actual testing of hypotheses. In this sort of study, hypothesis testing is not always the principal goal but it can certainly be achieved by isolating the effects of one variable. As a process, this can indicate how one might be able to refine or reformulate a hypothesis so as to improve its quality and design. Therefore, this methodological design allows for analytical circularity, which means one can constantly change and adjust its analyses based on previous examination, assumptions, outcomes, etc. Mostly following its particular case selection and methodological design, the comparative approach is also able to make certain limited predictions about the socio political environment. If a certain relationship between variables or a mechanism between cases is observed in a particular temporally and spatially limited setting, a researcher can possibly infer a similar relationship or mechanism would hold in an alternate temporal and spatial setting. For this particular study, this principle allows for the observation of the conditions and the attributes of certain diffusion mechanisms so as to infer their similar implementation, operation and outcome in another socio political setting. In general, these strengths and advantages make the comparative method a very valuable tool for this particular research project. It allows for the successful development of the arguments and claims this project formulates, and for the scientific contributions it ambitions. However, none of these benefits are necessarily included or inherent to this methodology. One has to carefully and properly construct the study so as to avoid that the
119
comparative approach becomes a “ (…) creative exercise of comparison that ultimately is meaningless” (Peters, 1998, p.85). In other words, this methodology also has certain limitations that are worth pointing out.
C. Putting in perspective: The possible drawbacks of comparative research The comparative method engages in the same sort of familiarity with its cases as qualitative research does. However, certain scholars argue that one cannot generalise from an in depth case study and social sciences are all about generalisation (Flyvbjerg, 2004). Others claim in depth studies can serve as pilot studies but are not suitable for actual research designs, or that in depth studies provide too much depth and scope for a researcher to interpret. Based on these arguments, certain scholars claim the validity of in depth studies is often subpar. However, such misconceptions neglect that, due to the exploratory nature of this study, it is considered the most appropriate methodology in this particular case (Flyvbjerg, 2004). A general misconception that leads to the underestimation of the value of in depth studies is that general and theoretically independent research is more valuable than some of the more concrete and practical research. However, since absolutes do not necessarily exist in the social sciences, the value of context specific (i.e. in depth) studies can be more valuable than the continuous search for generalisation and predictive theories. Similarly, certain scholars also claim the contribution of context specific studies to the literature is minimal because of their inability to generalise. Unfortunately, the scientific and intellectual value of generalisation is very often overestimated, whereas the force of example, especially for educational purposes, is often underestimated. At the same time, quantitative scholars often forget that in depth studies with a proper case selection can also lead to generalisation. Another misconception is that in depth studies are most valuable in the early stages of the research process, when hypotheses need to be designed. This, while other (more quantitative) methodologies are more suitable to test those hypotheses and construct theories or conclusions from them. Some researchers forget that in depth studies can play an important role in both the design and the testing of hypotheses. Directly following this dichotomous misconception, some researchers also claim that in depth studies tend to have sizeable confirmation biases, meaning they have a stronger likelihood to confirm the researcher’s original desires and thoughts than other, more quantitative designs. However, such claims are entirely unfounded and there is no scientific reason why confirmation bias should be greater for this method than for others. Nonetheless, in depth studies do have a greater bias toward falsification of original claims than toward verification.
120
Critical scholars often claim it is difficult to condense such research and develop general conclusions and propositions from it. Unfortunately, here researchers often mistake the difficulties to condense in depth research following as a research method, rather than as an actual consequence of the context and the reality of the reality studied. Often it is simply not possible, achievable or even desirable to summarise in depth studies and generalise from them. It might be better and more valuable to leave them in their entirety. Probably the most common critique is that of selection bias, which can be described as a non random selection of the cases included in the study. Based on the statistically unrepresentative nature of such a sampling scheme, the researcher cannot draw inferences for the larger population based on a non random case selection. This bias can result in the overrepresentation of certain cases that are positioned on one particular side of the distribution of an important variable. In addition, if a researcher would select on the dependent variable, this could also be particularly problematic because causal inference will be biased (Collier, 1995). The exact nature of the case selection is discussed on the following section; however, selection bias has not been an option in this study. Since the comparative method provides extensive in depth focus on the intrinsic value of each case, it is only possible to include a limited number of cases in the analysis. As the number of cases increases, the amount of time and resources necessary to gain familiarity with each case increases exponentially. Therefore, this style of research is incapacitated by a larger number of cases. When this relatively limited number of cases is accompanied by too many causal conditions, it becomes impossible to isolate the dynamics of the primary interest, i.e. diffusion patterns. Even when the number of cases is not much greater than the number of independent variables, i.e. when there are too few degrees of freedom, this can lead to a lack of robust results (Huber, Ragin and Stephens, 1993). Directly following from this so called small n dilemma is the possible problem of particularisation. More specifically, a researcher must be careful and adhere to the scope of the study so as not to resort to over generalisation of specific causal conditions based on a relatively small number of cases. In other words, the researcher must pay attention to the conclusions drawn from the study and make sure not to overestimate their scope. For this particular study, one must restrict the conclusions to West European FRPs, and not generalise to other FRPs, or even other party families. The study of FRP development deals with too many case and context specific factors, to fall into the trap of particularisation. When providing a complicated and extensive study with conclusions, the researcher faces additional dilemmas. On one hand, the researcher must incorporate two manifestations
121
of the so called travelling problem. Neither theoretical/analytical constructs, nor empirical operationalisation remain consistent across time and space. Therefore, it is increasingly difficult to provide generalisations from a comparative analysis. However, both the theoretical framework and the empirical operationalisations show considerable flexibility so as to incorporate comparison through time and space. The theoretical constructs of diffusion and its mechanisms do not change, however, their operationalisation can be different depending on the temporal or spatial setting. Some scholars would refer to this flexibility as inconsistency, however, this adjustment does not only follow empirical evolutions, it also allows for the practical comparison of concepts, dynamics and outcomes. On the other hand, as is often the case in academic research, one can question a researcher’s eventual interpretation of the research. In smaller n research such as this, the value free interpretation of results is often difficult, if not improbable. Even though this is impossible in absolute terms, the researcher does believe the systematic judgement and subjective analyses present in the some of the scholarship on FRPs needs to be balanced. Therefore, this study adheres to strict ethical norms and values, and submits itself to rigorous control from both academics and FRP functionaries. One of the more underestimated research dilemmas is Galton’s problem. It refers to the problem of drawing inferences from cross cultural data following the cross correlation of the data with itself (i.e. autocorrelation). The direct result would be a decreasing number of independent cases. In his original study, Galton attributes autocorrelation to (among others) a process of cultural diffusion among nations. Most of the existing FRP literature assumes FRPs have developed independently from each other. This study explicitly refutes such a claim. In other words, by the introduction of trans national diffusion dynamics, this study advocates an analytical approach that does not assume that FRP development happens without interdependence being present and significant (most specifically, see chapter 3.3).
D. Case selection and sampling scheme A case must always be specified as a case of something, otherwise it is merely a study. Its formulation for this study is not as evident as one might think. Due to the extensive emphasis of dynamics and processes, this study does not follow the typical steps in the design of a comparative study and is generally, more complicated. Regardless of this study’s dynamic disposition, it is possible to see FRPs and their diffusion characteristics as events of trans national diffusion. Following such a simplification, one could see each FRP’s diffusion
122
pattern as a totality of trans national diffusion characteristics and attributes that help describe the relationships with other FRPs. In general, the case selection is based on three principles that guide the researcher’s choice (Peters, 1998). First, a case must maximise the experimental variance. This means that by including cases that isolate the effect of the factors and variables that are being investigated, the logic of comparison is exploited as much as possible. Second, a case must minimise the error variance. In other words, in a comparative study, it is necessary to include representative cases, not atypical or deviant cases (Lijphart, 1971; Eckstein, 1975). Lastly, a case that is included must control for the extraneous variance, i.e. minimise the effect of other explanatory and spurious factors. Directly connected to this, a researcher should not select on the dependent variable so as to avoid influencing the variance artificially and to avoid extreme selection bias (Geddes, 1990). Due to several restrictions, a comparative study does not include an entire population, or use a random (i.e. representative) sample as the foundation of its case selection. Not every FRP politician, functionary, follower, voter, specialist or expert is interviewed, so this study uses an actual sampling scheme. As is often the case for studies that rely on elite interviews, sampling and case selection depends largely on access. The researcher is often faced with gatekeepers that either restrict access or are themselves ‘not available’ as interviewees. This can lead to severe selection bias and non random sampling error. Since such error do not cancel each other out, the case selection can become unrepresentative of the population. Normally, the number of cases one includes in a comparative research design depends on the number of suitable cases. Not only because of the availability discussed above, but also because qualitative data is generally too rich and complex to allow for the adoption of a large number of cases. Therefore, a typical comparative study’s case selection is not random but cases are selected precisely because they belong to a certain classification. In this case, the included cases belong to the FRP family and they all engage in some form of trans national diffusion. Practically, the explicit choice of cases has to be coherent and must have the potential to advance theory on the role of diffusion patterns in FRP development. This selection process must also indicate temporal (i.e. starting in the early 1980s) and spatial boundaries (i.e. Western Europe) to clearly delineate the cases. The case selection also experiences a minor instance of selection bias. In general, the upper echelons of FRPs have been extremely open and cooperating, and few politicians or functionaries have refused an interview, meaning that response bias has been minimal. The absence of the Austrian FRPs from the interview
123
sample does not lower the quality of this study because they are not necessarily a representative case, nor are they a crucial case. In addition, the instances of diffusion can be indicated based on other sources of information. Based on the earlier discussions, there are 13 possible cases that fit the criteria of a post 1980s West European FRP, hereby making them eligible for this research project’s case selection: the FN, the MNR, the VB, the FNb, the LN, the SVP, the FPÖ, the BZÖ, the BNP, the REP, the DF, the PS, and the SD. The most important case to include is the FN. On one hand, on a theoretical level, it serves as the Weberian ideal type FRP. On the other hand, the FN serves as an example and a re/source for many other FRPs. Its role and its position are of crucial importance to the claims this study makes. To such an extent, the FN can be called a theory confirming case (Lijphart, 1971) or a crucial case (Eckhart, 1975). Nonetheless, an important objective of this study is to obtain knowledge of different cases, events and incidents to maximise the contribution to the study of its different aspects, attributes, characteristics and instances of diffusion contributes to the analysis of patterns of diversity.
E. Durkheim versus Webber, or variables versus cases Emile Durkheim and Max Weber, both advocated for and assigned a central place to comparative methods. Certain similarities between the two, such as the substantive conceptions of order and change and some of the epistemological conceptions of social scientific knowledge, indicate the importance of this methodology (Ragin and Zaret, 1983). Nonetheless, both proposed a different basis. Weber’s case oriented approach results from a neo Kantian philosophy. According to him, the primary purpose of comparative research should be the causal constructs of historical diversity. Opposed to this stands Durkheim’s variable oriented approach, which parts from a positivist vision of society as a natural science. According to Durkheim, the principal purpose of the comparative method should be to obtain a substantive view of society as a system with historical diversity as a strain. Most classic comparative studies are rather qualitative in nature and case oriented in their approach to research (e.g. Moore, 1966; Lijphart, 1971; Skocpol, 1979; Lipset, 1979; Ragin, 1987; Peters, 1998). This indicates that most comparative studies focus on a rather comprehensive and in depth analysis of historically defined (as opposed to abstractly causal) cases. Their origin clearly follows from one of the central objectives of social science: The analysis and interpretation of the diversity of relationships and experiences of certain macro social units (Ragin, 1987). This is opposed to Durkheim’s more variable based research
124
tradition, which is more concerned with the analysis of similar relationships across macro social units and the generation of all explaining theories (Smelser, 1976). The most common objectives of case oriented research have a historically interpretative and causally analytic nature. The former seeks to provide explanations for certain historical events by creating a framework that incorporates socio historical factors and allows for certain context specific inferences. This appreciation for the intrinsic value of a case takes into account the complexity of a case and, at the same time, acknowledges that even the most general explanations are only partial explanations. Very often, the search for simultaneous simplicity and generality proves difficult. Specifically, this study contributes to the existing theoretical frameworks on FRP development by proposing the incorporation of case specific and context sensitive concepts such as the master frame and diffusion dynamics. The latter objective, i.e. the ambition to provide possible inferences from some of the analyses a study proposes, is not necessarily true for all case oriented comparative research. However, in effect, this researcher seeks to draw inferences from the trans national diffusion patterns between the FRPs that form the core of this study. The research project proposes that learning, as a diffusion mechanism, contributes to successful FRP development, whereas emulation is more likely to contribute to unsuccessful FRP development. One of the most distinctive features of case oriented research is its principal focus on the discovery of patterns of differences and similarities, which are most common and useful in studies with a relatively limited number of cases. Such an analysis between cases and within categories allows for a greater specificity in possible explanatory arguments, which can then contribute to the revision or development of classifications among the cases. Such an approach is relatively insensitive to the frequency distribution of different types of cases, what results in minimal attention to sampling distributions. More importantly, this also means that one single counter case can question the entire proposed causal dynamic, which puts the emphasis on the specificity and diversity of the cause and effect patterns. Another distinctive feature of case oriented research is that researchers consider the cases in their entirety, as a combination of case specific conditions that lead to a case specific outcome, and not merely as a collection of variables. This means the set of conditions and attributes that make up a case are seen in relationship to each other, not as a contextual side note of the analysis. For example, in more quantitative research, the relationships between sub parts of a case are seen as following from the sampling procedure, not in the context of the separate entities they compose.
125
6. Conclusion Both in its data collection methods and its data analysis methods, this study emphasises the importance of triangulation. The rejection of mono method approaches (Martin, 1989) lies at the foundation of how this study has been designed. Particularly in the case of an exploratory study, much like this one, a multi dimensional approach has been proven to be extremely beneficial. Not only does it allow for the analysis of both differences and similarities (rather than just one of those), it also allows the researcher to take historical and cultural factors into account, while simultaneously contributing to the existing literature. The first sections of this chapter have discussed the data collection methods. The triangulation approach has resulted in a wide variety of sources. Mostly, the primary, secondary and tertiary literature has been used to support and complement the principal data source: elite and expert interviews. Since there are no existing structural data sources, the former is mostly used to construct a pool of potential interviewees and to help develop the original theoretical foundation that serves as the basis of the interview guide. Subsequently, the interviews are used to both add to and improve the theoretical and analytical frameworks, and to gather empirical evidence that support or reject those frameworks. One of the primary purposes of this study is to complement the existing literature on FRP development with a theoretical framework that includes diffusion and to empirically illustrate this framework so as to serve as a foundation or reference for future research. The latter sections of this chapter focussed on the principal analytical approach this study takes, most specifically the comparative method. The rationale for such an approach is its emphasis on the analysis of patterns of diversity. It allows this study to examine both the differences and the similarities between FRPs when it comes to diffusion. Furthermore, such patterns of diversity can subsequently be analysed together with patterns of FRP development. Such an approach allows for the possibility to ascertain the exact role and the extent diffusion plays in the development of FRPs. At the same time, the comparative method is designed in such a way this latter observation might be concluded in causal terms.
126
3.3 Trans National Diffusion Dynamics between Far Right Parties: A Comprehensive Theoretical Framework
127
1. Introduction Preceded by vigorous debates, most of the literature now agrees FRP development can be largely contributed to a successful combination of three different factors: a fertile breeding ground, open political opportunities and favourable internal party dynamics (Eatwell, 2003; Mudde, 2007). Such a variable oriented approach has served as the basis of the vast majority of existing scholarship, however, it is important to not limit any explanatory models to this existing approach. More specifically, most of the existing literature ignores the important role the French FN and the diffusion of its master frame have played in the emergence of FRPs. In the late 1970s, the FN innovated itself by adopting a new master frame, thereby leaving the old and stigmatised master frame behind. The spread of this master frame has served as the basis of the current wave of FRPs throughout Western Europe. Only few authors have included this notion as part of their theoretical framework when explaining FRP development, thereby rendering them incomplete. The introduction of the concepts of master frame and diffusion allow for the analysis of FRP development as the European wide phenomenon it is, and not as the independent development of different FRPs. In addition, its inclusion helps avoid any possible ad hoc theorisations regarding FRP development. Before emphasising the importance of diffusion, the chapter provides an extensive conceptualisation of what a master frame refers to and puts this into relation with the more familiar concept of ideology. Subsequently the chapter describes the master frame designed by the FN, and illustrates how it spread throughout (Western) Europe. This will lead to a discussion on trans national diffusion, its importance for FRP development and the four core dimensions that compose diffusion: the transmitter, the adopter, the object and the channel. To conclude, the chapter briefly discusses its place in the current literature, and describes contributions, shortcomings, and future research avenues. Since the concept of diffusion has rarely been studied related to party development, and it is not considered a variable but rather a process, it is not evident to study this using raw empirical data. The absence of a full (quantitative) dataset restricts the possibilities to analyse this phenomenon (quantitatively). Additionally, with the limited (descriptive) data the study has, it is almost impossible to include the entire FRP family in its analysis. Therefore, the broader implications of this chapter are not necessarily to provide causal explanations, but rather to add to the current theoretical models of FRP development and to instigate further discussion and analysis of diffusion and its role in this process.
128
2. Far right parties and master frames The term frame was first used by the sociologist Goffman and defined as a “schemata of interpretations that enables individuals to locate, perceive, identify and label occurrences within their life space and the world at large” (1974, p.21). In other words, a frame was used to capture and structure an individual’s experiences and perception of society. These frames are then used as guidance through, and organisation of the world out there. Such a cognitive construction of frames had been used widely in psychology (Tversky and Kahneman, 1981; Klandermans, 1997) and various other disciplines (Tannen, 1993; Scheufele, 1999; Triandafyllidou and Fotiou, 1998). However, in that format, frames were termed too restrictive by social movement scholars (Klandermans, e.a., 1999). Collective action frames are not merely the aggregates of individual attitudes and perceptions (or individual frames). An addition of outcomes of negotiated shared meanings renders them more inclusive (Gamson 1992). Collective action frames were used as an interpretative function by simplifying and condensing aspects of the “world out there”, but in ways that are “intended to mobilise potential adherents and constituents, to garner bystander support, and to demobilise antagonists” (Snow and Benford, 1988, p.198). It is an “action oriented set of beliefs and meanings that inspire and legitimate the activities and campaigns of a social movement organisation” (Benford and Snow, 2000, p.614). Master frames play the same role as collective action frames do, only on a more aggregate scale. Or, as Snow and Benford state: “Master frames are to (…) collective action frames as paradigms are to finely tuned theories” (1992, p.138). They are not limited to the interests of a particular group or a set of related problems. Its scope and influence reach much further, and are designed to attract a broader audience or deal with a broader set of problems. A master frame influences and constrains the orientations and activities of other parties and can be considered a flexible and all inclusive framework that goes beyond a set of ideologies. In other words, a master frame can be considered rather generic, whereas a collective action frame is more exclusive and derivative (Snow and Benford, 1992). The strategic efforts to connect master frames to prospective constituents and resource providers, is referred to as frame alignment (Snow, e.a., 1986). There are four different processes: frame bridging, frame amplification, frame extension, and frame transformation (see below). Usually they complete each other, i.e. they are all present, be it in different degrees, and each process has a specific impact on the possible success of the newly implemented master frame. Most West European FRPs have used frame alignment processes to implement the FN master frame (e.g. the VB, the FPÖ, the REP, the LN, the DF, etc.).
129
Ideology and master frame are two highly interrelated concepts, and both refer to general constructs within which more specific ideas are understood. However, analytically, master frame must be clearly distinguished from ideology. Whereas master frame refers to a cluster of fairly broad and inclusive rhetorical strategies on which FRPs can draw (Swart 1995; Carroll and Ratner 1996), ideology refers to support for more specific articulations of theory and value nested within more general ones (Oliver and Johnston, 2000). In other words, a master frame does not necessarily include the socio political theory and normative value systems that characterise an ideology 31 .
3. The FN master frame Up until the 1980s, the FN was a stigmatised party, most notably due to its revisionist rhetoric. Based on pillars like biological racism (Proctor, 1988), anti Semitism (von Beyme, 1988), and anti democratic and anti system rhetoric were only able to attract marginal support throughout a Europe that had just overcome fascism and Nazism. The old master frame lacked electoral appeal and change was indispensable. At the end of the 1970s, the FN innovated a new master frame, which would become one of the most important foundations for future FRP success (Mayer and Perrineau, 1989). Based on the constructs proposed by Eatwell (2003) and Mudde (2007), the FN’s master frame is composed of four distinct elements: nationalism, xenophobia, authoritarianism, and populism 32 . Nationalism has deep historic roots in France and among French citizens and has always been exceptionally present in French political culture. It is something that is embedded in the French concepts of ‘Republicanism’ and ‘laïcité’. With the Italian MSI as an important example, the FN (heavily influenced by the Ordre Nouveau faction) adopted this framework in order to appeal to a broader right wing electorate. Contrary to the MSI, however, the FN abandoned all reference to a fascist or Nazi past and chose to focus on its innovative character. Due to increasing mass immigration, post industrialisation and globalisation (Kriesi, 1995), the salience of immigration increased. The FN perceived this as being in conflict with the traditional French model of individual integration. This led to the increasing feeling by French citizens that immigrants invaded their country, their religion, and their
31 For an excellent example of the distinction between ideology and master frame, see Aarelaid’s (1998), and Johnston and Aarelaid Tart’s (2000) studies on Estonian national movements. 32 Contrary to Mudde (2007), and for reasons that go beyond this paper, I do not combine xenophobia and nationalism into nativism. I believe nativism is too time and context specific to use as a simple contraction of xenophobia and nationalism.
130
neighbourhoods (Perrineau, 1997). Whereas traditional parties refused to include these challenges, the FN, led by the Nouvelle Droite 33 faction, did not hesitate to do so. Regardless of its initial use in the 1970s by the French PC and the British Torries 34 , the FN easily established issue ownership as the issue became more salient in the 1980s. Following the approach of the Frankfurt School (Adorno e.a., 1950), the authoritarian component is often connected to other attitudes and ideological features such as anti Semitism 35 and ethno centrism (exemplified by the welfare chauvinism advocated by the FN). The focus on socially conservative values was on one side, a reaction to the rise of post materialism (Ignazi, 1992), and on the other side a deliberate implementation of (successful) MSI strategies. The FN’s proposals included extensive law and order policies (e.g. “sécurité et liberté” voted in 1980 36 ) and scapegoating immigrants for unemployment and insecurity with for example the slogan “ Les Français d’abord ” (The French first). Populism was adopted by the FN due to its successful use in the past, and as a response to demand side processes. On one hand, Poujadism had successfully used populism in the 1950s to represent the middle class and those suffering because of socio economic change 37 . In the early 1980s, the FN used it to attract those who were considered the ‘losers’ of globalisation, industrialisation and modernisation. In addition, the FN rejected the heterogeneous nature of multiculturalism and the high degrees of elitism in French society. By adopting populism as an ideological feature, rather than just a style, the FN became the crusader for the ‘man in the street’ against the political, economic and intellectual elites. This can be considered the basis of the FN’s appeal to protest and issue voters (Bréchon, 2009). The combination of these four ideological frames, their subsequent rhetoric, and how they complement each other, provided the FN with a more moderate, less stigmatised and hence more attractive master frame. However, in order for this new master frame to be successful, socio political factors needed to be favourable as well in the early to mid 1980s. Following the “ Trente Glorieuses ”, France experienced a period of economic and social crisis: Inflation, unemployment, lack of integration by foreigners, increasing small
33 The Nouvelle Droite rejected biological racism but promoted a more moderate form of ethno pluralism, or so called cultural racism. Following this, it saw separate civilisations and cultures very much along the lines of Huntington’s ‘The Clash of Civilisations’ 34 More specifically, the Torries’ traditionalist conservative wing 35 Jean Marie Le Pen (the person) used this on occasion in his rhetoric; however, this was never part of the ideological features of the FN (the party). 36 This law, enacted on February 2 nd 1981, reinforces security and protects personnel freedom. It was implemented by Giscard d’Estaing, but Mitterrand abolished important sections of the law in 1983. 37 Poujade’s Union de Defense Commercants et Artisans (UDCA) was rather successful toward the end of the 4 th Republic, obtaining 53 representatives in 1956, among whom Jean Marie Le Pen.
131
delinquency, decreasing importance of culture, and a decreasing quality of public life (Bréchon, 2009). Until 1982 83, French traditional parties failed to adjust, and incorporate these societal concerns. As seen in many West European countries, France became a country of two speeds following post modernisation and post industrialisation (Wieviorka, 1992). Following this, the original cleavage structure (Lipset and Rokkan, 1967) lost some of its significance and gave room to new identification possibilities. The original economic bipolarisation disappeared and a new “ immense group central ” emerges (Giscard D’Estaing, 1976). In other words, the newly found social questions provided (new) parties, focussing on the socio cultural dimension, such as the FN, with political salience. Traditional parties’ failed to respond properly, and general societal frustration opened up political space for populism and parties that were critical of the (democratic) system. More specifically this was translated in a vast unhappiness with the PS (i.e. Mitterrand), and its inability to deal with the crisis, which eventually led to political dissatisfaction and disenchantment (Perrineau, 1997). Even compared to other European countries, France was characterised by a lower level of political confidence and a weaker partisan organisation (i.e. low party identification). The evolution from mass parties to cartel parties (Katz and Mair, 1995) has also been more prominent in France than in most European countries (Bréchon, 2005), thereby allowing the traditional left and right to crumble. Until 1986, the French electoral system has been a two round majoritarian system, with an effective threshold of about 37.5 percent (Lijphart, 1984), so the opportunities for smaller parties, like the FN, were limited. In 1986, the PS changed the electoral system to a d’Hondt proportional system to allow the FN to emerge and fragment the right. The effective threshold decreased to about 11 percent and allowed the FN to obtain 35 seats in the French Assemblée Nationale . The RPR UDF 38 obtained a majority after all, but the result gave the FN bargaining power and political legitimacy (DeClair, 1999; Shields, 2007). Before the change back to two round majority in 1988, there were no substantial subsidies available for the FN. Only if a candidate obtained more than five percent of the regular vote on the first ballot, which the FN did not, they would be reimbursed for the costs of the campaign (i.e. letters, flyers, posters, propaganda, etc.). Additionally, they would be reimbursed for their deposit of FFr. 1.000 (about USD 330 in 2012), which was lost for those
38 The coalition between the Rassemblement pour la République (Rally for the Republic) and the Union pour la Démocratie Française (Union for French Democracy), together comprising the traditional right in France at that time.
132
candidates who did not reach the five percent threshold. Generally, in France, public funding and state subventions to political parties were non existent until 1988. Regardless of these limitations, the FN did take advantage of the available political space. The change to a proportional system, and the coalition of the RPR and the UDF on the right opened up political space for a nationalist and populist right. The 1983 coalition with the RPR UDF list in the second round in Dreux 39 provided the FN with needed legitimisation and decreasing stigmatisation. A Figaro SOFRES barometer substantiates this in 1985, and shows that nine percent of French citizens declare they have a good opinion of Le Pen, and a same percentage would like him to play a bigger role in French politics 40 . Generally traditional parties had difficulties how to approach this new phenomenon. On one hand, the traditional left demonised the FN (e.g. via SOS Racisme ), and used anti racism campaigns to fill the void of its empty ideology during the early 1980s. At the same time it was perfectly content to let the FN compete with for the ‘rightist’ vote. Following Meguid (2008), we can see the traditional left took an ‘adversarial’ strategy toward the FN. The traditional right, on the other hand, took a more diversified approach. On the national level it opposed any kind of coalition with the FN, thereby (also) taking an ‘adversarial’ approach. On the subnational levels though, the traditional right did get into coalitions with the FN, thereby taking a more ‘accommodating’ approach. Following Meguid’s theoretical framework, this combination (on the national level) is beneficial for the FN, and eventually leads to an increasing vote share for the FN (which it did up until the early 2000s). In France, media access is generally very difficult and restricted for political parties. Even though paid political advertising used to be possible (until the early 1990s), media access for the FN in the early 1980s was difficult and minimal due to specific requirements and limited financial resources. In periods between elections, 20 minutes of free time was available every month for those parties represented in parliament (which the FN was not). During election campaigns, parties represented in parliament received additional airtime 41 . Up until 1982, French media actively refused the FN and its leader Jean Marie Le Pen any kind of access. In May and June of that year, Le Pen addressed several letters to president Mitterrand discussing the media boycott the FN suffered. Subsequently Mitterrand instructed French media to “devote equitable coverage to the FN” (Shields, 1997, p.196). This was followed by a number of media appearances by Le Pen, with as milestone his prime
39 This was followed by several other coalitions on the local and municipal level in 1984 and 1985 40 SOFRES, 1985, Opinion publique , Paris, Gallimard, pp. 182–83 41 Until 1986, this was about three hours of free airtime, which was equally distributed among the parties prior to the first ballot
133
time interview in L’heure de la vérité 42 where he presented himself as a politician rather than a social pariah, and described himself as a ‘Churchillian democrat’. In hindsight, both Le Pen and Mitterrand acknowledged the importance of this emission, and the more general presence in the media scene 43 , for the FN’s credibility and future success. The adjustment of the new master frame to the specific socio political characteristics of the French political system contributed to the FN’s development. In other words, frame alignment processes, and the proper adjustment to breeding ground factors and political opportunities are crucial factors for the successful development of the master frame, and the FRP. As a result of the FN’s innovative character, the initial success of the new master frame and the limited intellectual background and financial resources of most FRPs (Rydgren, 2005a), the FN has served as an example for other FRPs and it master frame has spread to other FRPs. Typically FRPs draw on an existing (and successful) master frame, strategically align this and adopt it. The goal is to maximise the compatibility of the master frame with its specific socio political environment.
4. The spread of the FN master frame Since the 1970s, we observe important transformations in West European political systems (Inglehart, 1990; Mair, 1997; Bauman, 20901; Karvonen, 2010; etc.). The most significant change was undoubtedly the development of a cultural axis in the political spectrum, mostly resulting from the increasing salience of issues related to this axis (Inglehart, 1990). Examples are immigration, law and order, globalisation, European integration, social rights, etc. This changing societal framework was an important part of the institutionalisation of FRP success in the 1980s and 1990s. In addition, the adoption of a more moderate master frame changed the image of FRPs as well. Only few West European political systems continued to perceive FRPs as political pariahs (de Lange, 2008). On one hand, the rise of new parties was (among others) a consequence of traditional parties’ failure to adjust to the ongoing socio political evolutions. On the other hand, the changing perception of FRPs and their growing popularity invoked a reaction from traditional parties. A general decrease in success, power and potential of traditional parties led to increasing political and electoral competition between traditional parties and FRPs
42 Broadcasted on France’s public channel France 2 (then Antenne 2 ) on February 13 th 1984. 43 In May 1984, SOFRES findings showed 18 percent of respondents declaring themselves ‘very or somewhat sympathetic to Le Pen’, while levels of support for Le Pen’s ideas on immigration, law and order, and anti communism reached 28 percent, 26 percent and 25 percent respectively (SOFRES, 1985, Opinion publique , pp. 178 180 and 182 183).
134
(Minkenberg, 2001). To such an extent that not only traditional parties increased the adoption of FRP master frame components, but they also increased the accommodative strategies vis à vis FRPs (Meguid, 2008). Examples of this are increased cooperation and collaboration between traditional parties and FRPs in France, Germany, Austria and Italy. This shift of traditional parties to the right, together with the increasing number of parties on the right (mostly FRPs), has been referred to as “ Verrechtsing ” (Heinisch, 2003; Bale, 2003). The changing West European political system contributed to the general development of the FRP phenomenon. Throughout the past three decades, each West European political system has experienced the presence of at least one FRP 44 . Several of those FRPs have successfully emerged (and gained electoral and/or political power), and some of them have consolidated. More specifically, in France, these societal evolutions contributed to the emergence of the FN. Based on the empirical evidence provided by Rydgren (2005b), and in line with the majority of the literature, it would be justified to consider the FN as a primus inter pares – the pater familias of FRPs (see also Backes, 1996; Ignazi, 1997). The FN adopted its master frame in the late 1970s and emerged in the early 1980s, with its first electoral success in the 1983 municipal elections. Formally, it can be seen as the principal innovator of the new master frame (for West European FRPs 45 ). Following this, the master frame spread to a small number of early adopters in the mid 1980s. Some of these FRPs turned out to be successful (Flemish VB and Austrian FPÖ), whereas others did not (German REP and Spanish FN). Following this initial adoption by a small number of FRPs, the new master frame spread to a multitude of West European FRPs in the 1990s (e.g. Walloon FNb, Italian LN, Finnish TF, Danish DF, Swedish SD, British BNP, and French MNR) 46 . In other words, during the 1980s and 1990s, a large number of political systems experienced the emergence of an FRP. Therefore, the past decade has seen fewer FRPs emerge, mostly due to saturation. However, there have been some important exceptions, usually referred to as laggards (e.g. Austrian BZÖ, Swiss SVP 47 , and Norwegian FrP).
44 There is one notable exception to this observation, and that is Ireland. In its entire history, Ireland has not experienced even an attempt of an FRP trying to emerge. 45 In Eastern Europe, the German REP originally functioned as the role model for aspiring FRPs. However, since the REP used the successful master frame provided by the FN to emergence, this project refers to the FN as the primary and principal innovator of the new master frame. 46 This included also a large number of FRPs in CEE (e.g. HSP, MIEP, PRM, etc.). 47 Due to a right mainstream history and issues of factionalisation, it is difficult to place the Swiss SVP in the political landscape. Some scholars claim the SVP is a traditional party (e.g. Hennecke, 2003), whereas others consider it an FRP (e.g. Husbands, 2000; Betz, 2004). From 2005 on, Christsian Blocher entered the Swiss government, thereby severely limiting the power of the more moderate Bern faction of his party and expanding the power of the Zurich faction (i.e. the FRP faction) (Skenderovic, 2005; Mudde, 2007).
135
It is clear from the above description that the number of FRPs similar to the FN, or at least using the new master frame the FN des igned, has been increasing steadily over the past three decades, both in Eastern and Western Europe. There is one innovator, some early adopters, a large group of adopters and a limited number of late adopters. If one would plot a frequency curve of this phenomenon, it would have the shape of a normal curve. If one would plot the cumulative adoption patterns, the curve would have an S shape. These curves are associated with the diffusion of innovations, or in this case, the diffusion of a new master fra me (see Figure 1). It begins with the reluctant adoption of the new master frame by only a few FRPs, followed by a rapid escalation of the adoption trend, and finally levelling off. This leads to a new equilibrium where all FRPs will have adopted this ne w master frame.
Figure 1: A bell-shaped frequency curve and an S -shaped cumulative curve for an adoption distribution
Source: Rogers, 1983, p.243
The main explanations for the above described distributions over time are based on the role of information and the levels of uncertainty in the spread of the new master frame . Following the normal distribution of the master frame adoption , the time of adoption is determined by a large number of socio political va riables and processes 48 that are not necessarily restricted to the political system in which the new master frame is adopted. The cumulative distribution has the same S shape as an individual’s learning curve (Gray, 1973). This indicates that the master f rame adoption process is similar to an individual’s learn ing process (Rogers, 1983, p.243 45). This indicates that FRPs do make mistakes in the early stages of adoption. However, they learn from the process and respond to its challenges and shortcomings. The curve indicates an interaction between FRPs because those who have already adopted the master frame influence those who have not (yet). In other words, those
48 This fits the require ment of a normal curve that “the value of each event is the result of the chance combination of a great many minutes and relatively equal factors” (Pemberton, 1936, p.549 550).
136
who have not (yet) see those who have already adopted as examples. The more FRPs adopt the new master frame, the greater the effect on non adopters will be (Rogers, 1983, p.243 45). Rogers (1983, p.245) argues that the first inflection point is the result of interpersonal networks spreading subjective evaluations of the new master frame between FRPs.
5. Defining (trans-national) diffusion In its most basic form, diffusion is but a function of connections between different organisations (Soule, 2004). However, the conceptualisation of diffusion has always been complex (e.g. Walker, 1969; Collier and Messick, 1975; Simmons, Dobbin and Garrett, 2006). Originally, Rogers claimed diffusion occurs when “an innovation is communicated through certain channels over time among members of a social system” (1983, p.14). Similarly, Strang and Meyer refer to diffusion as “the flow of social practices among actors within some larger system” (1993, p.488). The dilemma of these three descriptions is that they are relatively all inclusive and universal, which eventually leads to a tautological approach to diffusion. Therefore it is important to identify the different dimensions of diffusion. Michaelson defines diffusion as “the process by which an innovation (any new idea, activity or technology) spreads through population” (1993, p.217). Rogers later specifies this by referring to diffusion as “the spread of abstract ideas and concepts, technical information, and actual practices within a social system, where the spread denotes flow or movement from a source to an adopter, typically via communication and influence” (1995). Rogers emphasises the possible abstract nature of the ‘what’ and differentiates between communication and influence, which indicates an agent’s involvement in the process also needs to be taken into consideration. Katz (1968), however, is the most complete when describing diffusion as the acceptance of some specific item, over time, by adopting units – individuals, groups, communities – that are linked both to external channels of communication and to each other by means of both a structure of social relations and a system of values, or culture. This definition by Katz is narrow enough to specify the necessary dimensions of diffusion, and general enough to allow for multiple categorisations within it. At the same time it manages to be non tautological and non trivial, while being adequate and plausible enough to give the concept of diffusion explanatory power. To justify the use of diffusion in the discussion on FRP development, however, it is necessary to make one additional
137
specification. Similar to the claim made by Rydgren (2005a), and given the dynamics of party change, it is necessary (and sufficient) that the adopter is active in the process 49 . Much of the early diffusion literature focused on individuals and micro level diffusion 50 . In the past two decades, however, diffusion scholars are taking a more macro level approach, focussing more on organisations, social movements and protest. More specifically and more relevant to the field of FRP research, diffusion started playing an important role in the explanatory frameworks of collective action, symbolism and (technological) innovations. These applications differ from the classic diffusion literature because (i) it focuses on behavioural strategies and structures, (ii) it works with a larger historical and spatial canvas, and (iii) it sees diffusion as an explanation and not just some overarching theoretical framework (Strang and Soule, 1998). When conceptualising diffusion, it is important to make the following important distinction. Diffusion is either studied as an outcome (a dependent variable) or as a process. It is less interesting to study diffusion as an outcome since there are too many possible variables that can have an effect on it. In other words, it is almost impossible to parsimoniously model and explain diffusion as a dependent variable. In addition, diffusion has a dual character; it can be both a cause and a consequence (Elkins and Simmons, 2005). Therefore, this chapter treats diffusion as a process. On one hand, diffusion is seen as an environmental process that shapes both the agents and the object of diffusion. On the other hand, diffusion is seen as a decision consciously taken by the adopter as part of a rational choice process where benefits outweigh costs (Rogers, 1983). Diffusion of an innovation will only occur when its adoption appears beneficial compared to the current situation 51 .
6. FRPs and diffusion: Why is this essential? The applications of diffusion are widespread throughout the general academic literature (Rogers, 1983) and the political science literature more specifically (Walker, 1969; Gray, 1973; Berry and Berry, 1990). Even in comparative politics, its use has become more and more widespread (Collier and Messick; Bennnett, 1991; Simmons and Elkins, 2004).
49 Some scholars refer to the existence of ‘forced diffusion’ (Peters, 1997) or coercion (Dobbin, Simmons and Garrett, 2007), where the adopter is either passive or even resistant. Following a rational choice perspective, this does not stroke with the reality of FRP development, and the dynamics involved in the diffusion of a master frame. 50 This was especially the case in natural sciences like physics and molecular biology (Rogers, 1983). 51 A FRP will only adopt a new master frame if the aggregated benefits of this new master frame outweigh the value of the currently adopted master frame. This dilemma is extensive discussed and` modelled (through the combination of a change and a threshold model) in the following chapter, where it will contribute to the answer to ‘why’ master frames diffuse.
138
However, its application in the FRP literature is most often overlooked. The question remains why this is the case? Given the importance of diffusion in social movement literature (e.g. Tarrow, 1994), and the imperative similarities of social movements and FRPs (Goodwin and Jasper, 2009) 52 , disregarding diffusion can lead to a partial and incomplete theoretical framework. This chapter attempts to complement the existing framework by introducing diffusion as a significant and important process in FRP development. The existing literature points to a fertile breeding ground, or so called demand side factors (Eatwell, 2003; Mudde, 2007), as the catalysts for FRP emergence. These factors point to general economic, historical and social processes that take place on a broader level 53 . However, a fertile breeding ground alone cannot explain cross country differences in FRP development. Political opportunities 54 can explain such differences due to their cross national variation, and their facilitating or impeding nature 55 . Despite conceptual and operational difficulties, however, “the majority of studies agree that fixed or permanent institutional features combine with more short term, volatile or conjectural factors to produce an overall particular opportunity structure” (Arzheimer and Carter, 2006, p.422). Most of the literature looks at FRPs as developing independently from one another. In the diffusion literature, such a myopic situation is often referred to as the “null hypothesis” and serves as a benchmark against which we portray the situation of diffusion. This myopic situation assumes that any possible adoption patterns or similarities between FRPs are a direct consequence of FRPs facing similar challenges at about the same time. In other words, the null hypothesis assumes FRPs adopt the same new master frame irrespective of their interactions with one another (Berry, 1994; Volden, Ting and Carpenter, 2008). However, globalisation (Mudde, 2003) and internationalisation (Swank and Betz, 2003) make this highly unlikely. After all, FRPs are not inward looking and narrowly defined
52 For a more expanded discussion of this argument I refer to the previous chapter. 53 The shift to post modernism (Ingehart, 1977), the development of multiculturalism, economic crises (Stoss, 1991; Zimmerman, 2003), political crises (Daalder, 1992), demographic changes (Veugelers & Chiarini, 2002; Wendt, 2003), authoritarian background (Klandermans and Mayer, 2005), far right attitudes like resentment and xenophobia (Fennema and Tillie, 1998; Van der Brug e.a., 2000), insecurity (Dehousse, 2002; Christofferson, 2003) are just a few of the possible demand side factors mentioned in FRP literature. 54 For an overview of the different conceptual definitions of the political opportunity structure see the following: Eisinger, 1973; Kitschelt, 1986; Kriesi e.a., 1992; Gamson and Meyer, 1996; Tarrow, 1998; van der Heijden, 2010. 55 Throughout FRP literature, several dimensions of political opportunities help explain its emergence: political culture (Almond and Verba, 1965; Inglehart, 1990), fascist history (Coffé, 2005; Art, 2006), the rising levels of immigration (Husbands, 1992; Maddens and Hajnal, 2001), electoral and party system dynamics (Jackman and Vopert, 1996; Knigge, 1998; Golder, 2003), party positioning (Carter, 2005), the institutional framework (Evans, 2001; Abedi, 2002; Arzheimer and Carter, 2003), the degree of elitism (Jenkins and Schock, 1992; Decker, 2003), the media (Norris, 2000; Eatwell, 2003), and many more.
139
unitary agents 56 . Rather, they are a collection of organisations, networks, frames and individuals all striving toward party change 57 (McCarthy and Zald, 1977; Staggenborg, 1989; Buechler, 1990), but not limited to one political system. FRPs can affect other FRPs when tactics, ideas, styles, participants, organisations and more importantly master frames diffuse and spill over (Meyer and Whittier, 1994; Oliver and Myers, 1998). When analysing FRP development, cross national explanations are not accounted for in an extensive or systematic way 58 . However, it is important they are properly included because if they are ignored or neglected, which is often the case for in depth case studies (Gerring, 2007), ad hoc theorising can become a genuine problem 59 . In order to avoid this, many comparative studies focus on macro structural variables. Developmental clustering in time and the ambition to avoid ad hoc theorising (Hedström and Swedberg, 1998) can therefore result in a bias toward finding one universal cause (e.g. Betz, 1994; Kitschelt and McGann, 1995). Ultimately, one should find a middle ground, and look for causal mechanisms with a certain degree of generality. Such analytical pitfalls require an adjustment of the existing theoretical framework for FRP development. Most importantly, similarities between different FRPs are not only contributed to macro structures forming the demand side factors, but also to diffusion processes from successful FRPs to embryonic ones in other political systems. Following this, the trans national nature of diffusion, and the emphasis on the explanatory value of diffusion bring time and agency back into the analysis (Rydgren, 2005a). As Tilly (1984) points out, when things happen affects how things happen. Simultaneously, the inclusion of diffusion results in a renewed emphasis on FRPs, instead of macro structures, as the primary focus. Trans national diffusion is therefore an overlooked, yet important, process that can complement the existing analyses of FRP development. In the existing FRP literature, only three instances are found in which diffusion had been incorporated in the study of (West European) FRPs. Husbands (1996) claims the levels of support for ‘racist’ parties are
56 Choices made by developing FRPs should be seen as interactions with ex party, or even ex system, events and developments, not just as within system developments. 57 According to Harmel and Janda (1982), all political parties are conservative in nature when it comes to party change. However, since the socio political environment is constantly changing, political parties need to change if they want to survive politically and electorally. Therefore, all political parties constantly change, albeit not because of internal desire, but because of external pressure. 58 For the discussion of trans national diffusion processes in social movement literature see McAdam 1982; Jenkins and Eckert 1986; Olzak 1992; Rucht 1992; Olzak and Olivier 1994; Olzak, Shanahan, and West 1994; Kriesi, Koopmans, Duyvendak and Giugni 1995; Olzak, Shanahan, and McEneaney 1996; Rucht, Koopmans and Neidhardt, 1998. 59 Some examples of such in depth case studies are Ignazi (1996), Mudde (1996), Declair (1999), and many more.
140
partially caused by so called ‘contagion effects’. However, his results are rather mixed, and he does not account for cultural differences in his framework, which could also be a reason for differences in support. DeClair (1999) present a model connecting the FN (so called ‘mother party’) and other FRPs. However, his framework is rather narrow and cannot be applied more widely to study the extent and the scope of possible connections between FRPs. Rydgren (2005a) presents diffusion in a rather abstract manner and he does not present a tool to explain the differences in diffusion between FRPs. He does pose interesting questions, but he leaves them unanswered. In general, one can see “independent development” as the null hypothesis of this project. Most comparative FRP literature use a variant of this null hypothesis, and develop explanations based on specific, ad hoc factors. They often assume FRP development happens in isolation from other FRPs. This chapter advocates a more process oriented approach to this phenomenon (like the political process model suggested this for social movement developments). According to Katz’s definition (1968), following the theorisation of Rydgren (2005a) and applied to FRPs, diffusion involves four core dimensions: (i) a transmitter, (ii) an adopter, (iii) an innovation, and (iv) a channel. In the following sections, each of these dimensions will be carefully discussed and analysed.
7. The agents of diffusion: Who diffuses? This section is often theoretically underdeveloped because agents are usually considered of lesser importance than the object of diffusion. However, the theoretical emphasis on and the specification of both the transmitter and the adopter bring agency back into a discussion that has long been structure oriented. Therefore it is not only important to consider agents’ different roles in the different developmental phases (Coffé, 2005), but also to differentiate between those considering adoption and those looking to diffuse 60 . Each process of diffusion has these two sets of agents: a transmitter and an adopter. Based on Snow and Benford (1999), four different agent combinations can be observed. Theoretically, both agents can play either an active or a passive role in the diffusion process. However, as stated before, an adopter cannot be passive in this process. Snow and Benford (1999) state a passive adopter does not make sense in social sciences 61 .
60 Sometimes the literature indicates a third group of agents, more specifically go between agents (Stone, 1999), which provide a vertical dimension to the generally horizontal diffusion process (e.g. Shipan and Volden, 2006). In other words, go between agents provide elements of pressure to the diffusion process. 61 In the natural sciences, this is much more common (e.g. diffusion as contagion). Certain biological and physical processes such as the spread of an illness or genetic mutations can best exemplify this type of diffusion.
141
When both agents are actively involved in trans national diffusion, this process is referred to as “reciprocation”. When only the adopter is actively involved, and the transmitter remains passive, the trans national process is referred to as “adaption”.
A. Transmitting agents FRPs that have already adopted the new master frame are called transmitters. In several diffusion studies, such agents are neglected exactly because they have already adopted the innovation. This, however, is a misconception of the explanatory value of diffusion. Every master frame has roots somewhere and it is important to know where these lie, especially for those FRPs who adopted early, without much information and without outside pressures. For early adopters it is perhaps even more important to understand the origin of the master frame as this provides the opportunity to identify those adoption processes that exist regardless of diffusion. By making this analytical distinction, one can minimise the confusion between trans national and spurious diffusion (Volden, Ting and Carpenter, 2008). It is also possible a transmitter plays the role of innovator, as opposed to spin off (McAdam, 1995; Tarrow, 1998). Typically the former serves as an example for other FRPs (i.e. a primus inter pares). Such a role is usually for an FRP with sizeable expertise, leader capacities and a strong organisation. While one might be unaware of its innovator status, the creation of (successful) spin offs and active diffusion will give the innovator increasing legitimacy. In advanced stages of diffusion, spin offs can become innovators themselves because certain adopters might prefer the altered to the original version 62 .
B. Adopting agents FRPs willing to implement the new master frame are called adopters. Their characteristics, together with their preferences, goals, capabilities and the overall environment they operate in, are crucial for the diffusion process. Often, these factors, combined with the need and will to change a master frame, are referred to as “ prerequisites ” of diffusion (Collier and Messick, 1975). Without the appropriate prerequisites, diffusion, and the successful adoption of a new master frame, are impossible. The preferences of an FRP originate from several different angles, the most important one being from the individuals comprising an FRP. Their attitudes, judgements and experiences form the core of its preferences. In addition, other individuals (from traditional
62 This has been the case for the REP, which has aligned the FN master frame, but has also served as the primary example for most FRPs in CEE, most notably Jobbik (Hungary) and ATAKA (Bulgaria).
142
parties, other FRPs, etc.), institutions, or even the general electorate might also influence an FRP’s preferences. In general, an FRP’s preferences provide a solid indicator of the components of its master frame, and can therefore influence the likelihood of master frame diffusion (Walker, 1969; Collier and Messick, 1975). The goals of an FRP have a direct influence on its attitudes and preferences. Based on Strøm (1990) and Deschouwer (1992), an FRP can have four possible party goals: vote seeking, office seeking, policy seeking and/or inter party democracy 63 . Depending on which goals are being pursued, an FRP will adjust its preferences and subsequently adopt or not adopt a new master frame. This process can be influenced by several institutional and environmental factors, referred to as an FRP’s capabilities. Institutional constraints, such as cognitive shortcuts (Weyland, 2005), and the general political environment (e.g. political opportunities) can influence the adoption process in all its aspects (Volden, 2006).
C. Agent characteristics Additional to the level of initial involvement, six general FRP characteristics also influence diffusion (Wejnert, 2002). First, the societal character of an agent (i.e. its societal role) can influence the process of diffusion. More specifically, the mechanisms of diffusion 64 , the characteristics of the master frame, whether strong or weak ties are most prevalent, whether outcomes are mostly found on the micro or the macro level, etc. FRPs look to maximise electoral outcomes and influence policy as much as they can. Therefore, early in an FRP’s development, when structures have not yet been established and social ties between FRPs are still weak, indirect diffusion (e.g. through media processes) is more prevalent (see below). In general, sizeable collective agents, like FRPs, usually look to indirectly diffuse or adopt innovations with public consequences. Second, an FRP will more easily adopt a master frame as its perceived novelty decreases and its familiarity increases (Greve, 1998). Two crucial factors influence an FRP’s perception: the media and peer observation. The former refers to the influence media and expert opinions may have on the perceived risk of adoption (Newel and Swann, 1995). The latter indicates that familiarity with possible outcomes of diffusion can also influence the perceived risk of adoption (Rogers, 1995). Since FRPs are general conservative entities, their
63 Since all FRPs are authoritarian and hierarchic in nature (DeClair, 1999; Gomez Reino, 2001; Segert, 2005), internal party democracy is not often considered one of the primary party goals of an FRP. 64 For a more extensive and detailed discussion on the different mechanisms of diffusion, I refer to the following chapter.
143
risk taking behaviour is low. Therefore, perceived familiarity with the master frame should be high and uncertainty of outcome should be low in order for an FRP to adopt. A third important characteristic is an FRP’s social significance, which refers to its relative position within the FRP family 65 . When an FRP scores high on social significance, it is more likely it will be an innovator or adopt the master frame in an early stage, which increases the likelihood of becoming a transmitter. In other words, an FRP’s social position (determined by its status, wealth, legitimacy, success, etc.) significantly influences the diffusion decision, especially within culturally homogenous groups (Herbig and Palumbo, 1994). Since the 1960s, the media play an important role and they primarily affect those FRPs with a high degree of social significance (i.e. innovators and early adopters), which subsequently shape and diffuse the master frame among other FRPs (Weiman and Brosius, 1994). The new master frame is aligned with the existing norms and values because socially significant FRPs do not want to lose this status (Heyman and Mickolus, 1981). Previous research indicates that an agent’s socio economic conditions can also account for a variation in adoption rate (Hedström, 1994). In other words, they can influence how predisposed FRPs are to adopt a possible master frame. This indicates that macro , meso and micro level demand side variables influence the susceptibility of potential adopters to the new master frame 66 . Research has shown that for collective agents, like FRPs, economic variables are more prevalent than spatial or institutional factors (Hedström, 1994). Therefore, the absence of financial and intellectual means for most FRPs is a major contributor to the individual adoption and the general diffusion of a new master frame. One of an agent’s more important characteristics is its position in social networks. After all, the timing and success of diffusion is highly dependent on the interaction between different agents (Rogers, 1995). Four distinct interaction patterns between FRPs can be differentiated: interpersonal networks, organisational networks (with other FRPs similar in structure, content and ambition), structural equivalence (i.e. identification with the transmitter and limited social distance between agents), and social density (i.e. the concentration of adopters within a certain target arena) 67 .
65 Even though there is more cross country differentiation within the FRP family than there is within any of the traditional party families, the general literature still sees the FRPs as a party family (Mudde, 2007). Regardless of the continuous struggle to internationalise nationalist parties, and the academic disagreement on the parties constituting this family, they share variations of the same master frame and often depart from the same ideological background. 66 For an extensive discussion on these three levels of demand side factors, I refer to Mudde (2007, chapter 9). 67 In the emergence phase, it is impossible to already speak of a structural network between individuals. Based on numerous interviews, most contact in the early developmental phases was limited to sporadic interaction
144
For diffusion to occur, a certain form of identification is necessary. The greater the identification between agents is, the more likely diffusion can occur. Directly connected to this, and facilitating diffusion, is structural equivalence 68 , which indicates the tendency of organisations to identify with cross national counterparts as potential transmitters. Here, more homogeneous FRPs may see their trans national counterparts as salient transmitters (Strang and Meyer, 1993). This process becomes easier, more likely and facilitates diffusion when the distance between the adopter and the transmitter is small, i.e. there is homophily 69 between the two agents (DiMaggio and Powell, 1983; Burt, 1987) 70 . If the social distance between FRPs is limited, the diffused master frame is unlikely to be in conflict with the adopter’s predispositions and will increase the likelihood of diffusion. Lastly, the personal characteristics of the individuals who comprise FRPs also influence diffusion, especially those with authority. Compared to the characteristics discussed above, little research has been done on this and its overall value has been underestimated. Additionally, many of the personal traits are most important for diffusion between individual agents (e.g. self confidence, independence, strength, etc), and to a lesser extent for organisational networks. In the case of FRPs, this means that the personal characteristics are more important in the emergence phase when no (direct) structural relations between FRPs have been formed yet.
8. The object of diffusion: What diffuses? In order to discuss the diffusion of a new master frame, Kriesi e.a. (1995) provide the most appropriate and complete categorisation of possible objects of diffusion. Unlike other scholars, they emphasise that an agent cannot implement an innovation in a different socio political system without any adjustments, i.e. processes of alignment. Each socio political system is different; therefore each innovation requires careful modifications in order to be
between individuals, not FRPs as organisations. Such a network type is much more developed and prevalent once an FRP has emerged and is consolidated. 68 This is defined as “an actor’s perception of concordance with other members in a social and/or organisational network, particularly of comparable economic and social status” (Wejnert, 2002, p.308) 69 Rogers defines this as “the degree to which two or more individuals who interact are similar in certain attributes, such as beliefs, education, social status and the like” (1995, pp.18 19). 70 Here, distance needs to be interpreted in a sociological meaning. Most scholars see geographical distance as a proxy variable for social distance (Hedström, 1994; Myers, 2000; Andrews and Biggs, 2006), however this operationalisation can be questioned. Theoretically, geographical distance is not socially relevant and the only relevant distance is sociological in nature. Recent literature has shown that once political, socio economic and demographic measures of homophily are taken into account, geographical distance no longer matters, thereby suggesting geographical distance might not be so important (Braun and Koopmans, 2010).
145
successfully adopted. In other words, when a master frame is diffused, it needs to be salient, familiar and compelling in order for it to be successful. Kriesi e.a. (1995) indicate five sorts of possible innovations when theorising about trans national diffusion. The innovation can be the actual ‘content of mobilisation’, i.e. a particular ideology, goal, idea, etc 71 . Such an innovation allows for the introduction of new issues and repertoires. An innovation can be ‘the form of organisation’, i.e. the internal structure of an organisation. This includes an organisation’s hierarchical structure, its division of leadership and its extent of centralisation and institutionalisation. Besides tactical, an innovation can also be ‘a form of (collective) action’. This means an agent adopts a new form of action and adds it to its repertoire. An innovation may be ‘a model of action’, i.e. a transmitter’s mobilisation patterns can become an example for adopters in other political systems. Lastly, an innovation can be ‘the likely effect of collective action’. This refers to the general perceived chances of success the innovation has. Here, scholars have theorised a curvilinear relationship between chances of success and the level of actual mobilisation (Eisinger, 1973; Kitschelt, 1986). The former three categories of innovations refer to the adoption of particular features, and the entire diffusion process revolves around the adoption of these features. The latter two categories of innovations are more abstract and refer to a spread of collective action. Here, the transmitter of the innovation actually serves as an example for others. As discussed briefly before (and extensively in the next chapter), FRPs pursue goals similar to those of traditional parties and, therefore, mostly need to work via institutionalised channels to achieve such goals (and maximise their utility). Therefore, FRPs’ collective action dynamics are relatively similar across different political systems 72 and the characteristics of the master frame itself (i.e. the former three) are more important to describe diffusion. The literature focuses on different stages of the diffusion process (e.g. Kingdon, 1995), however most of the literature usually sets aside the early and latter stages in favour of the adoption stage and it looks at diffusion as a dichotomous process (e.g. Berry and Berry, 1990). Such assumptions limit the potential value of diffusion. The master frame plays a much more important role in the early stages of diffusion. However, including information on latter stages of adoption is not only necessary, but also provides a more complete and correct
71 This is in accordance with McAdam and Rucht (1993), who deal with the cross movement and trans national diffusion of some thoughts, ideas, goals, approaches, etc. between German and American new left movements. 72 It is important to differentiate here between the nature of collective action and the tactics used by FRPs. The former (which is discussed in this section) refers to the interaction and the communication with the establishment. The latter refers to the means of communication used by FRPs. These, of course, depend on various factors, like the socio political system, political opportunities, the media, etc.
146
view of how a master frame diffuses from one FRP to another (Boehmke and Witmer, 2004). Additionally, it is also important to analyse the scope of master frame change since a master frame itself might change through time and depending on the political system.
9. Channels of diffusion: How does it happen? After the ‘who’ and the ‘what’ of diffusion, the channel of diffusion describes how agents can be linked to each other: through direct and/or indirect channels (Soule, 1997). Direct channels suggest the rate and degree of diffusion is directly dependent on the levels of interaction: agents make their choices known to each other, develop shared understandings and explore the consequences of different innovations based on shared experiences (Strang and Meyer, 1993). The primary instruments that make this happen are (social) networks and the ties that facilitate inter node interaction (Diani and McAdam, 2003). These channels assume interpersonal ties are present, these ties are relevant enough to serve as diffusion channels, and they facilitate the diffusion of information concerning the master frame. To achieve this, direct diffusion can be either formal or informal. The former refers to official meetings or patterns of communication, usually on a higher level (e.g. participation at events, public speeches, international cooperation, etc.). The latter refers to non structural communication, usually on a lower level (e.g. friendships, readings of others’ publications, etc.). Typically, direct diffusion processes start out as informal and sporadic relations throughout the lower level tiers of organisations. Such non structural links between organisations are then possibly institutionalised, and made formal (Kriesi, e.a., 1995). Direct diffusion, however, cannot account for the apparent post diffusion similarities and resemblances between agents that do not have direct channels between them. If indirect channels are available to potential adopters, frameworks, ideologies, structures, themes, etc. can still diffuse between agents. An indirect channel requires the innovation to pass through a third party, from which the adopter will implement the master frame. Very often, such third parties are so called go between agents (see above), such as international institutions (e.g. EU), elected officials, think tanks, research institutes, etc. Since its growing importance at the end of the 1960s, however, most (horizontal) indirect diffusion passes through the mass media Dolowitz, 1997). In the case of indirect diffusion, however, it is important the involved agents show a significant degree of structural equivalence (Strang and Meyer, 1993) with a certain degree of homophily (McAdams and Rucht, 1993). In other words, indirect diffusion depends much more on the interaction between the master frame, the FRPs and the socio political
147
environment. The social distance between FRPs, and between an FRP and its new master frame should be limited. There should be no conflicts between the master frame and possible predispositions an FRP might have. Even though social distance and structural equivalence are important, agents are also selectively exposed due to the increasing importance of mass media. Koopmans and Statham (1999) describe mass media as a “discursive opportunity structure”, which Koopmans (2004, p.370) defines this as “a number of selective mechanisms (…) that affect the diffusion chances of messages in the public sphere”. Media can portray a certain issue or topic in various ways, regardless of its inherent characteristics. This refers to its societal functions such as agenda setting, priming, status conferral, stereotyping, gate keeping and framing (Koopmans, 2001; Koopmans and Olzak, 2004). Mass media’s role cannot be underestimated. Based on a theoretical framework provided by Gamson and Wolfsfeld (1993), FRPs need the media in their development for three reasons: mobilisation (collective action), validation and scope enlargement. First, since most people are part of the mass media community, using mass media is much more beneficial than using party oriented outlets, like party newsletters, to achieve maximum mobilisation. Second, media appearances provide recognition, credibility as a political party and increase electoral potential. Third, the media is needed to broaden the scope of discussion and the impact of its innovation. Conversely, mass media has a double function as an indirect channel of diffusion. Many forget (or leave out) that mass media is also responsible for the channelling of the innovation in the reverse direction (Koopmans, 2004). FRPs depend as much on the mass media as other parties do. They need mass media to provide them with information on other political parties, the functioning of the socio political system, new issues among the public, and most importantly, the public’s reaction to its master frame. Without any of this feedback, FRPs are not able to properly align their master frame with both the political needs and the electoral demands. Therefore, mass media is crucial as a source of strategic information and as an evaluation tool for the FRP’s proposed master frame (Koopmans, 2004).
148
3.4 Modeling Trans National Diffusion Mechanisms between Far Right Parties: Introducing a Three Step Model
149
1. Introduction Diffusion has been studied extensively throughout the literature. However, its presence in political science, and the FRP literature especially, has been limited. One of the main goals of this paper/chapter is to model the presence of diffusion in the process of master frame adoption. In this discussion, it is important to identify diffusion as a process, rather than an explanatory variable (Simmons and Elkins, 2004), and to see this process as interdependent. In other words, an FRP’s adoption decision depends on that of other FRPs, and vice versa. Explanatory variables have an influence on the diffusion process; they do not compete with diffusion as part of the explanation of a dependent variable. The diffusion literature is diverse and widespread. Therefore, just like for political opportunities, its conceptualisation and operationalisation can be problematic (see first paper/chapter). There does not exist one diffusion theory, and often, different interpretations compete with each other. This paper/chapter does not try to proclaim one single diffusion theory, but rather to introduce the notion of diffusion to the literature on FRPs and complement our comprehension on the adoption of the new master frame. As diffusion is a complex and multi layered process, further research and scholarly attention into this topic are needed to add to our general understanding. Diffusion literature recognises that diffusion like dynamics are not necessarily the consequence of a process of diffusion (Volden, Ting and Carpenter, 2008). It is possible that similar FRPs autonomously adopt a common master frame, without engaging in diffusion. This situation of independence is here referred to as the null hypothesis. However, normally, domestic socio political factors combine with factors of diffusion to reject the null hypothesis and provide evidence for its alternative (Simmons and Elkins, 2004). The literature has long recognised the probability that observations in different countries might not be as independent as one would like to assume (Simmons and Elkins, 2004). When using cross cultural data, the correlation between variables is most likely a function of time (difference). This comparative problem has been referred to as auto correlation or Galton’s problem. It becomes problematic when one wants to draw inferences from cross national data, which can lead to spatial dependence, spurious correlation, biased estimation and decreased statistical significance (Eff, 2004). Even though this often leads to unnecessary rejections of the null hypothesis (type I errors), this phenomenon is hardly ever taken into account. Therefore, instead of controlling for it, this paper/chapter attempts to model auto correlation when analysing master frame diffusion.
150
In order to do so, different diffusion mechanisms have to be identified and considered (Hedström and Swedberg, 1998, p.7). Each one of these mechanisms provides a systematic set of dynamics that describe a plausible account of how master frame diffusion between agents transpires (Schelling, 1978). The emphasis on such mechanisms, and the dynamics that compose them, can explain how and why FRPs influence each other. Which of these dynamics is most prevalent, and under what socio political circumstances they operate depends on the diffusion mechanism at play. Therefore it is important to provide a theoretical and an empirical differentiation between the four diffusion mechanisms considered: coercion, competition, learning and imitation (Simmons, Dobbin and Garrett, 2006). The main objective of the following paper/chapter is to complement the existing literature on FRPs, not to replace it. Its ambition is to provide a foundation for further research of the role (master frame) diffusion plays by bringing together different dynamics and mechanisms in a general framework. In the first section, I propose a master frame change model, based on expected utility functions, which analyses when and why an FRP changes its master frame. In the second section, I provide a brief theoretical discussion on diffusion and some of its more important aspects. In the third section, I provide an indication of how one FRP’s master frame decision can be influenced by other FRPs, i.e. how FRPs adoption decisions are interdependent. In the last section, I discuss the different adoption dynamics and how/if it is possible to aggregate these individual adoption decisions (see Hedström and Swedberg, 1998, p.23 on how the individual level can be connected to the aggregate level).
2. A master frame change model Ideology, and the change thereof, describe and define a party’s general behaviour, which can be considered an indicator of its motivations, wishes, ambitions, and aspirations. Departing from a rational choice perspective, a significant component of such an indicator is thus devoted to maximising a party’s objectives (Müller and Strøm, 1999). This means that when a party’s political behaviour changes, its objectives change as well. Building on rational choice theory and utility maximisation (Downs, 1957; Arrow, 1987), Strøm (1990) identifies three different types of party behaviour, each with a different primary party goal: Vote seeking, office seeking and policy seeking. Simultaneously, Harmel and Janda provide several models “to explain and predict which type of behaviour will predominate for a particular party on the basis of institutional and organisational factors” (1994, p.268). Depending on those factors, and since the above described party goals are not mutually exclusive, parties continuously adjust their behaviour, and hence their party goals
151
(Müller and Strøm, 1999). Additionally, Deschouwer (1992) discusses the need to consider more than just electoral and/or political performance as possible party goals. He describes ‘participatory power’ and ‘articulating members’ wishes’ as two additional motivators for party behaviour. These representation missions are embraced by the broader concern for ‘inter party democracy’ (Bruce, et al., 1991). A classic example of this is the Green movement in Germany and in Belgium 73 . Since all FRPs are authoritarian and hierarchic in nature (DeClair, 1999; Gomez Reino, 2001; Segert, 2005), internal party democracy is not often considered one of the primary party goals. At the same time Mudde (2007) describes how FRPs are not a normal pathology but rather a pathological normalcy. So, even though FRPs’ access to governmental and/or parliamentary functions is often either limited (e.g. electoral system) or refused (e.g. cordon sanitaire ), they can be perceived as ordinary vote and/or office maximisers (e.g. the FN’s 2012 bid for political and electoral power by using a clear vote maximising strategy). Often, FRPs are also important policy/ideology advocate. Generally, the exact combination of predominant party goals depends on both external (e.g. demand side factors, opportunity structures) and internal factors (e.g. party development) (Eatwell, 2003; Mudde, 2007). An FRP’s ability to achieve certain goals and objectives depends largely on the means it has at its disposition. One of the most important instruments and the core characteristics to do this is an FRP’s master frame. This plays an important role in the attractiveness, the ideological course and the possible success of an FRP. It is important a master frame adjusts, through various frame alignment processes 74 (Snow, et al., 1986), to its socio political surroundings, prospective constituents and resource providers in order to maximise a master frame’s payoff and effectiveness (Esser, 2001, pp.259 334; Braun, Gilardi, Füglister and Luyet, 2007; Volden, Ting and Carpenter, 2008). Depending on which party goals are considered imperative, an FRP will emphasise different aspects of its master frame. In other words, a master frame can be an important instrument of party change (incl. its goals and functions), which in its turn can influence a party’s ambitions. Seeing how FRP’s goals are multi dimensional, the utility they attribute to a certain master frame (in order to achieve these goals) is also multi dimensional. On one hand, a master frame’s utility is determined by the possible electoral outcome connected to it. On the
73 More recently, this is reflected in the implementation of (open) primaries within certain political parties such as the socio democratic parties in France ( Partie socialiste ) and Italy ( Partito democratico ). 74 There are four different alignment processes: frame bridging, frame amplification, frame extension, and frame transformation. Usually they complement each other, i.e. they are all present, be it in different degrees, and each process has a specific impact on the possible success of the newly implemented master frame (see Snow, et al., 1986; Snow and Benford, 1988; Benford and Snow, 2000; see paper one).
152
other hand, a master frame’s utility is influenced by its spatial closeness to the FRP’s ideal master frame. Maste r frames have known ideological locations, and FRPs have ideal positions alongside those ideological locations. It is this last dynamic that comprises an FRP’s policy seeking behaviour and its ideological advocacy. For an arbitrary master frame mf1 this means:
where U represents the master frame’s utility, V represents the payoff an FRP obtains as a result of an electoral outcome, P represents the payoff an FRP obtains as a result of its ideological location 75 , and a represents a probability coefficient weighting both factors. Both parameters are weighted because their importance can change throughout an FRP’s development, or between FRPs. In other words, for FRPs, certain master frame components’ payoffs can be different throughout time and space (S trøm, 1990), and the above mathematical equation reflects this. For example, since Marine Le Pen’s election as party president, the FN purposefully chose a vote maximising strategy, as opposed to its previously predominant policy seeking strategies. Ther efore, since then, the FN’s electoral coefficient a has increased in relative importance compared to its policy coefficient (1 – a) . Generally this is the case when elections get closer and campaigns start to become more important.
Umf1 refers to the mas ter frame’s anticipated payoff. Conversely, payoff is not the only parameter that influences its expected utility. In addition, an FRP is also influenced by the extent to which a master frame is effective (Volden, Ting and Carpenter, 2008). This can be defined as the relation between a master frame and an FRP’s goals, and has a continuous type distribution. A master frame can be considered effective when, regardless of its purpose, the result is the intended result. In other words, effectiveness is an indication of the degree to which the master frame achieves the proposed outcome. Combining the above parameters, the expected utility of a master frame is thus a function of electoral payoffs, ideological payoffs and master frame effectiveness. It is im portant to note the functioning of these parameters is not necessarily probabilistic, i.e. they are not necessarily independent. Master frame payoffs depend on and influence its effectiveness, and vice versa. If Umf1 corresponds to its payoffs and EF mf1 refers to its effectiveness, then master frame mf1 ’s expected utility is:
75 Therefore, as discussed before, this means that an FRP’s ideological position P is determined by its policy seeking behaviour and its ideological advocacy.
153
This equation indicates that a master frame with high electoral and/or ideological payoffs, but which is not very effective (i.e. has not properly aligned with the socio polit ical environment) will have a lower expected utility. For example, an FRP can uphold its ideological hard line stances, but the expected utility of such an approach might be rather limited since a more moderate approach has proven to be more effective in the electoral arena (Ignazi, 2003; Cole, 2005). Along the same lines, an FRP may emphasise its populist nature and oppose the current political class (generally, this is thought to be an effective approach), but the expected utility of this approach may b e rather limited if the existing political class is popular. This can result in an electoral backlash and thus a small or negative payoff for an FRP. In a more comparative context, one can look at the expected utilities of alternative master frames. Sin ce FRPs are rational choice actors/utility maximisers (Kahneman and Tversky, 2000), they adopt the master frame with the highest expected utility. In a framework of master frame diffusion, it would be adequate to assume an FRP has the possibility to adopt a master frame of its choosing, i.e. there are at least two alternative master frames available to FRPs. The question then is if/when FRPs would switch from their already adopted master frame to a new one. Or, more specifically, when will FRPs adopt the new FN tailored master frame and reject their old stigmatised master frame?
To go back to the algebraic model, Umf2 represents the payoffs associated with the alternative master frame mf2 , and EF mf2 represents its effectiveness. Similarly to mf1 , the expected utility function of the alternative master frame can then be described as:
An FRP will only change its master frame if the perceived expected utility of a new master frame mf2 outweighs that of the currently adopted master frame, which is t he expected utility of master frame mf1 . In other words:
The expected utility of change, i.e. the adoption of a new master frame, does not only depend on the characteristics of a master frame itself, but also on likely transaction costs C associated wi th the adoption of a new master frame. This can incorporate exploration and information costs (e.g. searching a new master frame), bargaining costs (cfr. game theory), administrative costs (when adopting a new master frame), and policing costs 76 (Dahlman, 1979).
76 Very often this happens through the legal system. A recent example is the lawsuit the FN (mostly Marine Le Pen) filed, and won, against the Walloon Fnb for the use of their symbol (the tri coloured flag). The FNb had been using the symbol since it unilaterally adopted the FN master frame in the early 1990s.
154
Additionally, the uncertainty that comes with a possible change, and the adoption of a new master frame, must also be taken into account. Not each modification of the master frame structure categorically leads to success. Along the same lines of the above expression, such uncertainty can be expressed in terms of probabilities: An FRP will adopt a new master frame with probability p, whereas the existing master frame will remain in place with probability (1 – p) . In other words, controlling for o ther independent variables, the more aligned a new master frame is, the higher its effectiveness will be, and the more likely its adoption will be successful. This lower degree of uncertainty therefore leads to a higher expected utility of the new master frame. In formula form this in indicated by:
;
Since FRPs look to achieve their goals and maximise their utility, they will only implement a new master frame if the expected utility of this new master frame outweighs the expected utility of the o ld master frame. In other words:
The last equation indicates that change occurs if C/p is smaller than a certain pre set threshold of success (different across time and space). More specifically applied to FRPs, an already adopted master frame ( mf1 ) is more likely to be discarded in favour of a new master frame ( mf2 ) when the original master frame’s effectiveness EF mf1 is perceived low, when the new master frame’s effectiveness EF mf2 is perceived high, and when the attractiveness of a new master frame increases compared to the older master frame ( Umf2 > U mf1 ). Here, it is important to point out and emphasise the interdependence of the two parameters comprising the expected utility of a master frame: payoff and effectiveness. The increased utility of a new master frame does not guarantee its adoption unless the new master frame is also effective in reaching its goals. It is possible a less than significant effectiveness is complement ed by a relatively high payoff. However, the model shows a new master frame will not be adopted based solely on its payoff. Conversely, and perhaps more plausible, a new master frame is rather unlikely to be adopted if it is perceived as extremely effect ive but fairly unpopular, i.e. its payoff is less than significant. In other words, an FRP might believe a new master frame is more effective than the currently
155
adopted master frame, however, it will be unlikely to go through the process of change if it faces opposition to said change (e.g. traditional parties, factions, the media, etc.).
3. The concept of diffusion Diffusion is not a new concept. Originally it was used in various natural sciences such as medicine (Coleman, Katz and Menzel, 1966) and biology (Bangham, Standish and Watkins, 1965). As early as the beginning of the 20 th century, diffusion also made its appearance in the non natural sciences like agriculture (Ryan and Gross, 1943), geography (Hägerstrand, 1952), anthropology (Wellin, 1955), communication (Deutschmann, 1963), and economics (Przeworski, et al., 2000). By the second half of the 20 th century, diffusion became a widespread concept in the literature on organisational behaviour (Katz and Lazarsfeld, 1955; Levitt and March, 1988; Rogers, 2003) and sociology (Tarrow, 1994; Strang and Soule, 1998; Della Porta, Kriesi and Rucht, 2009) 77 . The introduction of diffusion into political science has been limited (e.g. Gray, 1973; Stone, 1999; Karch, 2007), but is most significant in the field of policy diffusion (Berry and Berry, 1990; Levi Faur, 2005; Sabatier, 2007). More specifically, in the fields of party politics (Ladrech, 2002) and far right politics (Husbands, 1996; DeClair, 1999; Lloyd, 2003; Rydgren, 2005a), diffusion has known only limited attention and none of these studies provide one theoretical framework for diffusion or emphasise its structural importance and explanatory power. The literature on FRP development is extensive in its conceptualisations, operationalisations and theory formulations. The most comprehensive works, however, refer to a fertile breeding ground and favourable political opportunities as the principal explanatory factors for FRP emergence (Eatwell, 2003; Mudde, 2007). The former refers to general economic, historical and social processes that take place on a broader level. The latter refers to “fixed or permanent institutional features combine with more short term, volatile, and conjectural factors” (Arzheimer and Carter, 2006, p.422). Once FRPs have emerged and they are looking to sustain themselves, internal factors like a moderate ideology (de Lange, 2007), good leadership (Decker, 2004), a solid organisational structure (Norris, 2005) and the absence of internal struggle gain in importance (Marcus, 2000). While explanatory factors are converging more and more, most of the literature still perceives FRPs as developing independently from one another, each within their own political
77 The concept of diffusion is especially present in the literature on social movements (Strang, 1991; McAdam and Rucht, 1993; Strang and Tuma, 1993; Meyer and Whittier, 1994; Strang and Soule, 1998; Wejnert, 2002; Smith and Johnston, 2002; Snow, Soule and Kriesi, 2004)
156
system. Existing explanatory models to not include any possible indications of cross national influence. Such explanatory models imply each FRP acts independently and indifferently from other FRPs, and its influence does not reach beyond its own party in its own political system. In the diffusion literature, such claims are referred to as the null hypothesis: A situation of independence between agents. In this case, the null hypothesis refers to the independence of FRPs when adopting the new master frame (see below). Theories of diffusion explicitly reject such structural independence. Socio political evolutions in the past decades, such as globalisation and internationalisation (Mudde, 2000; Swank and Betz, 2003), make this unlikely. Theories of diffusion take into account that political parties, and more specifically FRPs, are not inward looking agents and they do not necessarily limit their influence to one political system. FRPs can influence other FRPs when tactics, ideas, styles, participants, organisations and more importantly master frames diffuse and spill over (Meyer and Whittier, 1994; Oliver and Myers, 1998). Trans national master frame diffusion can be described as the process of adoption of a new master frame that is systematically conditioned by prior master frame choices made by other FRPs in other political systems. It would be inaccurate to model trans national processes as exogenous factors of FRP development. The discussion of diffusion allows for the inclusion of interdependent and trans national relationships, and for master frame choices elsewhere to have an impact on the choices and behaviour of FRPs in one’s own political system. This requires a set of assumptions about the actors of diffusion, their motivation, the channels through which diffusion happens, and how the object of diffusion can change. While the importance of diffusion cannot be underestimated, other disciplines do not find systematic evidence to either accept of reject the diffusion hypothesis. It is therefore important to differentiate master frame diffusion from a process of independent adoption of a similar master frame as a response to breeding ground factors and political opportunities (Shipan and Volden, 2006). The latter process has two identifying components: a significant number of FRPs adopting a similar master frame, and independence of adoption (e.g. Kelemen and Sibbitt, 2004) 78 . Parting from a rational choice perspective, it is possible certain economic or socio political pressures warrant the adoption of a specific master frame. Hence, it will be important to carefully define and illustrate the different diffusion mechanisms that
78 The difference between interdependent adoption (i.e. diffusion) and independent adoption is important to substantiate the claim that diffusion is not only an important process, but also a very present one. Therefore, the differentiation between these two serves as the basis for the next paper/chapter. It discusses the causal mechanism of adoption patterns in more detail and illustrates more precisely how diffusion can be differentiated from the more myopic stage.
157
describe the inter FRP connection (Braun and Gilardi, 2006; Simmons, Dobbin and Garrett, 2006, 2007; Shipan and Volden, 2008) 79 .
4. Mechanisms of Diffusion Master frame diffusion can be defined as a process where master frame choices are not independent from one another. Regardless of whether the adoption process is independent or interdependent, it is based on the same criteria: payoff and effectiveness of alternative master frames, and the transaction costs and the uncertainty of the adoption process (see section one). The sole difference between the situations of independence and interdependence is that for the latter, the values of the parameters described in the above algebraic model depend on other FRPs’ behaviour. How exactly other FRPs influence an adoption decision depends on the diffusion mechanism(s) at work. This section discusses which are the most prevalent mechanisms, and under which circumstances they operate. All four of the possible mechanisms are described theoretically, and are amenable to thorough empirical testing.
A. Coercion Coercion refers to a power asymmetry between the transmitting and the adopting agent, which results in the imposition of the object of diffusion (i.e. a master frame) by the more powerful agent (i.e. transmitter). Therefore, this mechanism is often described as rather anti democratic. The more powerful FRP’s main intention would be to influence the weaker FRP’s decision to adopt a new master frame. Based on a rational choice approach, the more powerful agent’s intention would be to maximise its self interest, and therefore to manipulate the costs and benefits of the adoption process, even with (physical) threat if necessary. When there is a strict hierarchy between the two agents, and there is a definite will and capacity to manipulate, one can speak of hard coercion. On the contrary, when the weaker agent is given little choice but to adopt the new master frame, by the use of more indirect and implicit means of manipulation (e.g. framing, lack of support, no invitation to international colloquia), one can speak of soft coercion. Empirically such a mechanism can be observed by paying attention to three separate considerations. Firstly, it is important to identify the agents involved in the process: the transmitting and the adopting FRPs. This will indicate which FRPs are coercing and which
79 For a more detailed analysis of the different dimensions of diffusion (i.e. its agents, its channels and its object), I refer here to the previous dissertation chapter (chapter three) or paper (paper two), which comprises an elaborate explanation of the theoretical framework of diffusion (Rogers, 2003; Rydgren, 2005a)
158
are being coerced. Secondly, one must show that the coercing agent(s) support the master frame they intend to diffuse. Lastly, in order to make a case for coercion, the channel of diffusion must be identified, through which a master frame is transmitted trans nationally. In this case, however, coercion cannot be considered a proper diffusion mechanism. Its vertical relationship between agents, rather than a more horizontal interdependency (which is at the core of the conceptualisation of diffusion) makes it complicated to consider it as a possible diffusion mechanism (for a master frame). After all, the vast majority of FRPs operate independently from one another in their respective party systems. Hence, it would be complex for one FRP to trans nationally coerce another FRP into adopting a new master frame. Rationally, there would be no possible incentive for an FRP to coerce another FRP into adopting a similar master frame. If, by exception, the FRPs do operate in the same party system, they often target different demographics (e.g. the REP and the DVU in Germany) 80 . Even though it cannot necessarily be seen as a pure diffusion mechanism, its importance should not be underestimated. In order to be complete and provide a comprehensive overview of diffusion, it is only accurate to include coercion. For example, in the literatures on the diffusion of liberalism (Simmons, Dobbin and Garrett, 2006) and the diffusion of democracy (Simmons, Dobbin, Garrett, 2006), coercion is a relatively prevalent and discussed mechanism. Perhaps even more importantly, coercion illustrates that events (e.g. the adoption of a new master frame) in different political systems are not necessarily independent from one another. More generally, it indicates that auto correlation between political systems not only exists but also takes place; different cases are not always entirely independent from one another (Gerring, 2001, p.179).
B. Competition This section explains how and why competition between different FRPs drives the diffusion of a new master frame, given that the success of one FRP’s new master frame does not necessarily constitute or guarantee its success in another socio political environment. This mechanism makes several assumptions: (i) the adopting agent assumes a new master frame has the potential to advance its goals (e.g. increase its electoral payoff), (ii) the impact of a new master frame needs to be nearly instantaneous since elections follow each other quickly,
80 The only European example of two FRPs with an extremely similar master frame, operating in the same political system, is Austria. The FPÖ and the BZÖ are both FRPs, using a close variant of the original master frame designed by the FN. The difference between the two FRPs is limited to detail, most notably the FPÖ’s stronger emphasis on nationalism and Euro scepticism (both can be considered a heritage of the anti Haider faction that stayed in the FPÖ).
159
(iii) an environment with near perfect information, or the available cognitive shortcuts to achieve this (cfr. Downs, 1957), and (iv) horizontal, more decentralised relationships (as opposed to coercion). Certain studies infer that competition only happens when agents are located in proximity of one another, i.e. their geographical distance is limited (Strang and Soule, 1998; Karch, 2007). However, studies have shown that it is rather the social distance that facilitates competition (Braun and Koopmans, 2010). Simultaneously, competition is more likely when there is homophily between the agents, i.e. when social distance is limited, the master frame is unlikely to conflict with the adopter’s predispositions and increases the likelihood of diffusion (DiMaggio and Powell, 1983). Additionally, competition is more likely when transmitters are structurally equivalent, i.e. FRPs identify with each other and perceive the others as salient transmitters (Strang and Meyer, 1993). These two dynamics allow an FRP to observe alongside which dimensions other FRPs are sensitive to a new master frame. Typically, competition is most prevalent when agents belong to the same network. Examples of a competition mechanism can be found in the US’ between state diffusion of legislation (e.g. Berry and Berry, 1990; Berry and Baybeck, 2005). Mechanisms of competition indicate there is stimulation provided by outside agents to adopt the same (or rather a properly aligned) master frame. Following the algebraic model, such incentives are not the consequence of changes in payoff, but rather of changes, and possibly even tradeoffs, in effectiveness (Braun and Gillardi, 2006). For example, an FRP can prefer a more ideologically extreme master frame, however, when realignment and dealignment processes follow from its incompatibility, it does not only have a negative effect on the political state of affairs but it could also lead to a loss in electoral payoff. More moderate master frames (mode moderate far right or traditional parties) might become more appealing, and can therefore contribute to the effectiveness of an FRP master frame. Empirically, one must first identify this FRP network (and its agents) and only then can one analyse whether one FRP’s master frame adoption increases the probability of similar adoptions by other FRPs within the same (structurally equivalent and homophilic) network. The probability of such change is positively correlated with the degree of homophily, i.e. the social distance, between the different agents. In other words, the more structurally equivalent and the smaller the social distance between a transmitter and an adopter, the more likely diffusion and a change in master frame will occur. Unlike coercion, competition is a complete diffusion mechanism. Nevertheless, it is difficult to speak of master frame competition as a diffusion mechanism for the same reason it
160
was difficult to speak of coercion. An individual FRP adopts a new master frame, properly aligned with its socio political environment and within its own political system, thereby leaving little opportunity for trans national competition, influence or variance. The master frame differences are mostly the result of adjustments to the political opportunities rather than the result of conflicts with other FRPs.
C. Learning Learning refers to the change in beliefs or one’s perception of existing beliefs (Simmons, Dobbin and Garrett, 2006). Here, the emphasis lies on the FRP from which one learns, i.e. the innovator. The innovating FRP is important because its choices provide other FRPs (so called spin off FRPs) with additional relevant information about possible causal relationships and allow them to anticipate societal impact of a master frame. Its experiences will provide spin off FRPs with information on both the new master frame’s payoffs and its effectiveness (compared to the old master frame) 81 . Some research makes the difference between tactical learning, where new information results in differences in means but not necessarily in outcomes, and cognitive learning, where differences in means are accompanied by differences in outcomes (Levy, 1994; Elkins and Simmons, 2005). As discussed in an earlier paper/chapter, it is not useful to consider a passive adopter in social sciences; the adopter is always actively involved in the diffusion process (Snow and Benford, 1999). In other words, an adopter will not adopt a new master frame it does not wish to adopt. Conversely, transmitting FRPs can have both an active and a passive role in the diffusion process. Whereas the homophily and structural equivalence influence the actual diffusion decision, the willingness and strength of the transmitter determine whether it will be actively engaged in the diffusion process or not. When differentiating between the emergence and the consolidation phase, two distinct forms of learning mechanism can be distinguished. During the emergence phase, diffusion dynamics are mostly unilateral, predominantly serving the adopting FRP, and have an indirect disposition, often with the media as an intermediate. During its consolidation, FRPs have (possibly) developed more structural relationship, both on the personal and the partisan level,
81 Learning has an impact on both payoff and effectiveness because the algebraic model used in this paper assumes these two parameters interact. Meseguer (2003, p.14) even argues that effectiveness is the only parameter that can be influenced by the learning mechanism for the simple reason that she does not warrant “payoff” as a separate term in the algebraic model and includes it in the error term. But, since the function is additive, the effects of the error term (which includes the payoff effects) adds to those of effectiveness. This, however, does not give sufficient weight and importance to the theoretical effects of payoffs.
161
and the relational dynamic becomes more interdependent. In other words, once consolidated, FRPs (can) learn from each other. Most of the learning literature differentiates between rational and bounded learning (Meseguer, 2005, 2006; Weyland, 2007; Simmons, Dobbins and Garrett, 2008). Both forms of learning emphasise the decision making dynamic and the costs of the acquisition of information in this process. The difference between the two forms mainly lies in how and by which means decision are taken and information is acquired. Rational learning is based on the process of Bayesian knowledge (Roth and Erev, 1995). This refers to a dynamic process in which individuals add new experiences, or data, to prior knowledge and revise their behaviour accordingly. Before obtaining the new data, prior behaviour can be expressed as a set of probabilities over a network weight. Once the data has been observed, posterior behaviour can also be expressed as a set of probabilities over a network weight. A network weight that seemed reasonable before observation of the data but does not align with the data will see its probability decrease. Simultaneously, if the network weight fits the data well, its probability will increase. At the same time, no mechanism can guarantee an agent will actually properly interpret this new data and come to the “logical” (i.e. rational choice) conclusion. Concretely, a potential adopting FRP can analyse the experiences of a transmitting FRP and misinterpret its implications. How adopters interpret the experiences surrounding the master frame depends on factors like the data source and how the data is processed (before and after obtaining it). For example, when a potential adopting one FRP observes the behaviour of another FRP and how the latter’s master frame functions in its political system, it is important to take into consideration political opportunities like media exposure, the electoral system, campaign financing rules, etc. (e.g. Kriesi e.a., 1995; van der Heijden, 2010). Those factors can place similar master frames in completely different daylight. Bayesian learning is considered rational because it assumes optimal and efficient use of the information one obtains. However, adopting FRPs may draw on available cognitive shortcuts 82 or communication networks to inform themselves on the existing master frames rather than analyse all the available information, like the Bayesian approach would require. Regardless of the reasons 83 , this may result in “herd behaviour”, and eventually in a Pareto
82 FRPs can rely on cognitive shortcuts like representativeness, availability, and anchoring (Tversky and Kahneman, 1974; McDermott, 2001) 83 Such reasons can be high costs, a sampling problem, lack of time, lack of capacity, lack of perceived importance, etc. (March and Simon, 1993). However, the reasons themselves are not important in this context, only the use of intermediates is.
162
inefficient adoption of a master frame (Banerjee, 1992). This form of learning is referred to as bounded rationality (or bounded learning) and uses heuristics to obtain (less accurate) information on master frame payoffs and effectiveness (Tversky and Kahneman, 1986). Empirically, learning is fairly straightforward to observe. It requires the observation of an FRP’s successful master frame, followed by similar master frame adoptions made by other FRPs. In other words, measurable and favourable information on master frame (e.g. favourable electoral outcomes) will increase the probability of master frame adoption by other FRPs. The adopting FRPs can receive such information rationally and objectively or it can also be mediated through existing communication networks such as the media. When learning takes place, the process of master frame adoption can happen regardless of whether the transmitter is actively or passively involved 84 . As a mechanism, learning is a rather complex, but important. The theoretical explanation above shows that different forms of learning allow an FRP to update their knowledge on the effectiveness (and the payoffs) of a new master frame. When a change in its knowledge occurs, and this change favours a new master frame, this can, but does not necessarily, lead to the adoption of a new master frame. Such a process of information gathering and interpretation is more effective when done by rational learning than by bounded learning (Meseguer, 2006). This supports the claim that new information about the effectiveness of an adopted master frame is likely to lead to change when a new master frame proves to be more effective than the adopted master frame.
D. Imitation A last important diffusion mechanism is imitation 85 . Conventional arguments suggest that an FRP will adopt a new master frame when it is developmentally ready and when the master frame has proven to be favourable. However, in some instances, this is not the case and an FRP adopts a new master frame for mere symbolic rather than substantial reasons (Strang and Chang, 1993). This results from the imitation of an FRP with its self identified peer (or peers). Rather than structural equivalence and/or homophily, certain socio cultural linkages contribute to the “psychological” proximity between certain FRPs (Rose, 1993).
84 The next paper/chapter discusses extensively how an FRP can learn as a passive agent and as an active agent. It uses the examples of the FNb (passive) and the VB (active) as an illustration of such learning processes. 85 In the literature this is often referred to as emulation (e.g. Strang and Meyer, 1993; Strang and Soule, 1998; Simmons, Dobbin and Garrett, 2006). I prefer the term imitation because it emphasises a process, whereas the term emulation emphasises an outcome.
163
This proximity cannot be explained by previously discussed dynamics like learning and is often considered irrational (Simmons and Elkins, 2004). In this case, an FRP expects to instantly influence the outcome of the adoption process (e.g. increasing electoral outcome), without properly taking into consideration the effectiveness of the adoption (i.e. the alignment of the master frame). More often than not, due to the emphasis on immediate results, this mechanism takes place without proper master frame alignment, thereby making it difficult to successfully adopt a new master frame (Snow, et al., 1986; Benford and Snow, 2000). Empirically it is often difficult to differentiate imitation from other diffusion mechanisms. Bounded learning often overlaps with imitation because the former is attracted to the experiences of prestigious FRPs, much like the imitation process is, albeit for different reasons 86 . Here, it is important to analyse master frame alignment. When this is done properly and the master frame is adjusted to the country specific socio political circumstances, one can speak of learning. If this is not the case, the adopting FRP merely imitates the innovating FRP. At the same time, imitation can also overlap with coercion because both recognise that demand side factors might cause master frames to converge 87 . Even though imitation is considered to be a more horizontal process, there is still a dominant agent. The question then is, where does the influence of the dominant agent end? Such conceptual convergence makes it difficult to specify empirical tests for imitation mechanisms. There are three important empirical problems that one has to consider when attempting to empirically test diffusion models and their mechanisms at work 88 . Existing literature extensively discusses diffusion mechanisms, however, the role of politics has not been systematic analysed. It neglects the influence domestic politics has on the adoption of the innovation. Depending on the political system, a master frame might not diffuse, even if it has proven to be successful. Therefore it is important to take political opportunities into account. Furthermore, empirical studies often fail to take into account the heterogeneous situations an FRP operates in, both within its party and its party family. It is important to accurately portray the complex nature of the causal processes under study. Lastly, it is important to not only look at diffusion as an explosive event, but also as a gradual phenomenon. When only focussing on the former, this can lead to severe selection bias.
86 Bounded learners are interested in the prestigious FRPs because of their informative value, whereas imitating FRPs are interested because of their symbolic value (Finnemore, 1996). 87 In the case of imitation, such convergence is the consequence of persuasion or reproduction, whereas in the case of coercion this might be the consequence of force or obligation (this is often referred to as coercive isomorphism). 88 It is important to mention here that these shortcomings are of a rather normative nature and this discussion does not escape from them.
164
To conclude, diffusion can occur through different mechanisms, each affecting the parameters of diffusion in a distinctive way. More specifically, the latter two mechanisms play an important role in the diffusion of a new master frame between FRPs. Learning refers to the contribution of additional information, provided by other FRPS, which revises the effectiveness of the master frames (either through rational or bounded processes). Imitation refers to blind adjustment of the master frame to conformity, as exemplified by other FRPs, which alters the payoffs 89 provided by both the old and the new master frame. Therefore, diffusion can be very distinct, depending on which diffusion mechanism(s) is/are in play.
5. Diffusion Mechanisms at Work: A Threshold Model In the previous sections, this paper/chapter respectively dealt with how master frames changed, diffusion as a theoretical phenomenon, and how master frames are influenced by the different mechanisms of diffusion. The last section specified the interdependence of FRPs and how the nature of such a relationship influences the spread of a new master frame. One needs a model of how these individual level relationships interact and if/how they can be aggregated to a more general model of diffusion. Such an aggregate bandwagon model can be described by the use of a threshold model (Granovetter, 1978; Levi Faur, 2002). The key concept in such a model is that of the “threshold”, which refers to the proportion of other FRPs that must make a decision on whether to adopt before a given FRP does so (Schelling, 1971; Granovetter, 1978). Following different internal and external supply side factors (Mudde, 2007), it is fair to assume that FRPs are largely heterogeneous in nature and their preferences vary between them. Since such preferences are an important determinant of the threshold (Granovetter, 1978, p.1421), each FRP’s adoption threshold is different 90 . Therefore, a threshold model explicitly illustrates that individual level adoption decisions are interdependent. In other words, an FRP’s decision to adopt or reject a new master frame depends on its own characteristics, as portrayed by its preferences (i.e. the threshold), and other FRPs, as portrayed by the preferences of others (i.e. the proportion of adopters composing the threshold). In such a model, agents are rational and seek to adopt the master frame that maximises their utility. Therefore, a threshold model is extremely useful in situations where individual
89 Mostly the policy/ideology component ( P) of the payoff is altered, and not necessarily the electoral component (V) (Braun and Gilardi, 2006, p.314). 90 Directly following this, it should be noted a threshold model does not consider why an agent has the preferences it has, or how these preferences came about (Granovetter, 1978, p.1421). This is irrelevant for the actual adoption process.
165
level outcomes are not necessarily or intuitively consistent with the underlying choices and preferences. Such paradoxes might be more present that originally assumed since outcomes is too often considered the equivalent of preferences. A threshold model is most significant when agents rely on the behaviour and the preferences of other agents and there are few or no institutional structures or precedents for agents to rely on. This describes the situation of FRPs trying to emerge in the mid to late 1980s perfectly. Besides providing an indication that FRPs’ decisions on master frame adoption are interdependent, the combination of a change and a threshold model has several additional advantages. First and foremost, the threshold model provides a connection between the individual level decisions described in the earlier master frame change model, and the aggregate diffusion behaviour, which was something missing in the former model. At the same time, rather than merely providing a set of threshold parameters, the combined model is much more refined by also providing values for electoral payoffs, policy payoffs, effectiveness, transaction costs, and uncertainty. Altogether, the threshold model shows that the experiences and behaviour of other FRPs matter because it affects the perception of effectiveness and/or payoffs related to the master frame. In other words, it is not necessarily the absolute number of adopting FRPs that counts, but the implications, the experience and the effects of that number (Schelling, 1978, p.96). The combination of a change and a threshold model also allows for the differentiation between innovators, early adopters and spin off agents. Innovators are original and novel in the design of the object of diffusion. Most often, they do not require adoption by others since they create the innovation. In the case of FRPs, the FN is an example of such an innovator. Based on various outside sources and on in house intellectuals, they replaced the old stigmatised master frame by a new modernised master frame with significant electoral appeal. Early adopters are those FRPs who adopt a new master frame when no or few other FRPs have done so already. Therefore, their payoffs and effectiveness of the new master frame change independently of other FRP’s behaviour and preferences. Such strong preference for master frame change can be expressed algebraically by EF mf2 Umf2 – EF mf1 Umf1 > C/p. Typically, early adopters suffer from relatively high transaction costs and uncertainty (compared to the spin off FRPs) due to the innovative nature of the new master frame. Both of these ‘costs’ interact with each other, and can therefore reinforce one another 91 . Spin off FRPs also seek to adopt a new master frame, but will only do so when uncertainty and
91 Previous algebraic expressions show that when there is a significant level of uncertainty, the transaction costs will increase. Similarly, when the certainty of the adoption process increases, the transaction costs decreases.
166
transaction costs are low. In other words, spin off FRPs need a more significant pay off and effectiveness, and will therefore have a higher threshold than innovators and early adopters. This distinction illustrates once again that other FRPs influence the general adoption dynamic and the individual adoption behaviour of FRPs. However, the process is not necessarily as straightforward as described by the above aggregate model. Each FRP has different characteristics and decision making procedures. Additionally, the influence other FRPs have alters between FRPs and can be considered unique (e.g. threshold). Specifically, each FRP will require a different combination of conditions to be realised before it will (consider) to adopt a new master frame. As seen before, factors like payoff, effectiveness, transaction costs, uncertainty and a threshold can influence an FRP’s adoption decision. Therefore, the relevance and the impact of different mechanisms of diffusion will vary between FRPs. For example, some FRPs will choose to learn from another FRP (e.g. the VB learned extensively from the FN throughout the 1980s and 1990s), whereas other FRPs will limit themselves to mechanisms of mere imitation (e.g. the FNb imitated several characteristics of the FN in the early 1990s). Furthermore, the importance of different mechanism can also change throughout time. Imitation or indirect learning mechanism might be more prevalent in the early stages of FRP development, whereas direct (rational) learning mechanisms are more likely to become more established in the latter FRP developmental stages due to the implementation of institutional and organisational structures. The theoretical complexity of the dynamics at work at the aggregate level results in a difficulty to draw valid conclusions about the outcomes of diffusion. In other words, it is very difficult to identify and isolate the effects and the consequences of master frame diffusion for FRPs. One of several interesting observations we can make regarding master frame diffusion is that there appears to be a situation of convergence of master frame adoption. In the past three decades, most of the West European FRPs have adopted an aligned form of the master frame designed by the FN in the early 1980s. The West European political system is currently in a state where most of its FRPs are using a form of this master frame. Methodologically, the study of diffusion is slightly more complicated. Contrary to some of the literature, it is important to analyse diffusion as a process, and not as an independent variable separate from innovation (Lee and Strang, 2006). A possible strategy is to use a pooled cross sectional time series analysis that uses political, social, historical, and economic characteristics as independent variables. The dependent variable in such an analysis would be the time or the probability of adoption of the new master frame (Gray, 1973; Glick, 1981). Even though such an approach also has its drawbacks (e.g. Gray, 1976;
167
Berry and Berry, 1990), it can take into account both spatial and temporal influences in the same model, and thereby shield against mistaking a spurious relationship betweens FRPs’ years of adoption and those of others as evidence of trans national diffusion 92 . In this context, event history analysis can be very useful. It aims to explain a qualitative change in the behaviour of agents (i.e. the change in master frame, or the event) at a particular time. The dependent variable, or hazard rate, indicates how many FRPs at a certain time are still at ‘risk’ of experiencing the event. Specifically, event history analysis would use the master frame as a unit of analysis and analyse the transition from the old master frame to the new FN based master frame. In other words, it would determine the probability that the adoption of the new master frame will occur during a certain time period. A dataset would indicate when, where and if a certain FRP had gone through the transition of master frame during a certain period. For now, however, event history analysis remains an ideal methodological tool that not yet renders itself to empirical testing. Empirically, the available data and the data set collected by this researcher is not large enough yet to use for statistical analysis, and is better used serving as an illustration for the theoretical model. In other words, the number of observed events does not yet meet the requirements necessary to compose a valid model with a high enough statistical significance to serve as a basis for generalisations. These empirical limitations are inherent to the novelty of this phenomenon (or at least its study), and by no means diminish the quality of the above model and its theoretical implications.
6. Conclusion This paper/chapter started with the predicament of drawing inferences from cross cultural analysis due to auto correlation. The theoretical and methodological approach brought forward in this paper/chapter allows this dilemma to be modelled, rather than just discussed or controlled for. It is substantially important to consider that agents are almost never independent from one another, regardless of their functioning in different political systems (as opposed to the null hypothesis described in the diffusion literature). The master frame change model proposed in the first section illustrates that an agent’s decisions and behaviour are not only dependent on itself and its socio political situation (see FRP literature), but also on other agents (see diffusion literature).
92 If the described relationship would be spurious, the estimated effects of terms representing the behaviour of socially nearby FRPs would decline to almost zero because the spatial effects would be appropriately controlled for.
168
A limitation of the existing FRP literature is the assumption of independence between FRPs. Several strings of research provide generalisations without considering diffusion as a fundamental dynamic. This paper/chapter illustrates how FRPs are not independent in their development and how the original master frame (designed by the FN) plays a crucial role in this development. The proposed algebraic model has been one of the first steps toward a solid integrated theoretical framework, which takes into account the interdependence of FRPs. The model shows there are different diffusion mechanisms in play, which can influence the relative effectiveness and payoffs associated with an alternative master frame. Subsequently, other FRPs can influence the decision and/or adoption process of a new master frame. At the aggregate level there are complicated diffusion dynamics at play. The restricted dimensions of this paper/chapter do not allow for generalisations about a possible equilibrium at the aggregate level. In other words, at this point, it is not yet feasible or credible to conclude that diffusion mechanisms are resulting in a common output (i.e. the overall adoption of the same master frame). For now, it is not yet possible to identify the circumstances under which diffusion dynamics and its mechanisms lead to convergence of master frame adoption. In the future, continuing data collection will allow further investigation into this claim by using advanced statistical models like event history analysis. Due to its complexity, the continuous nature of a master frame, the relative difficulty to acquire observations on some of the internal, and more exclusive, characteristics of FRPs (Mudde, 2007), the empirical framework of this paper/chapter is restricted to the above estimations. By modelling ‘interdependence’, this paper/chapter attempts to specify and complement some of the existing literature, rather than suggest a crucial change. In the following paper/chapter, I will add to the empirical framework by providing an extensive illustration of trans national diffusion between FRPs (specification). This will highlight the continued efforts to understand, contribute to and test diffusion mechanisms empirically. This paper/chapter does not attempt to innovate or contribute to the general diffusion literature. As a theoretical concept, diffusion has been widely analysed and discussed, even in the social sciences. However, within the FRP literature, diffusion has been severely underemphasised and understudied. This paper/chapter identifies what diffusion is, how it plays a role in FRP development, how it can contribute and where the further opportunities lie. Continued research into this is absolutely imperative and will only contribute to existing explanations of FRP development.
169
4.1 Results and Analyses: Outline
170
Once the conceptualisation has been done properly, the comparative method described extensively, the most important concepts operationalised and the theoretical diffusion model designed, the research project allows for the discussion of the results, the empirical illustrations and a detailed analysis of different patterns of diffusion. For now, this research project’s final chapter(s) the researcher intends to systematically investigate the conditions under which trans national diffusion is most likely to occur, by which channels it most likely travels and which mechanisms are most likely in play. The principle ambition of these last chapter(s) is to deal with three empiric questions. Firstly, the researcher asks how one can empirically differentiate between a myopic state of diffusion, where all similarities are accidental and involuntary (i.e. the null hypothesis of this research project) and deliberate diffusion patterns, where similarities are an intentional consequence. Directly following this general question, the researcher is interested in whether trans national diffusion patterns are taking place in a specific temporal and spatial context. Finally, the researcher investigates whether a simulation model uphold and confirms the previous empirical claims. The previous chapters have created an elaborate theoretical and analytical framework that can help explain the development of FRPs. The introduction of diffusion as an explanatory dynamic complements existing theoretical frameworks, and indicates that diffusion is an important part of the explanatory framework for FRP development. Nonetheless, in order to be complete, one needs to make an additional specification. The multi dimensional similarities one can observe between FRPs are not necessarily a direct consequence of trans national diffusion patterns. It is very plausible that certain demand side factors have evoked similar evolutions or changes throughout different political systems, which is what most of the literature implicitly assumes. Therefore, if the researcher wishes to label certain similarities (or differences) as consequences of trans national diffusion, it is important to make a clear distinction between trans national diffusion patterns and a state of myopic diffusion, where similarities are also apparent, but they are accidental. The literature has the tendency to see FRPs as developing independently from one another. For this reason, researchers often design unique and case specific explanatory frameworks and forget the role trans national dynamics can play in FRP development. Regardless of this general assumption of myopic diffusion, one can observe that most FRPs’ development tends to cluster in time according to certain patterns, as shown earlier by Grey’s S shaped diffusion curve (see chapter 3.4). This observation suggests it is likely other, more interdependent and transcending factors also play a role in this process. While FRP development is certainly shaped and influenced by the unique socio political and cultural
171
factors that define its context, it might be very possible that an original incentive, motivation or idea has been imported (in some way) from somewhere else, regardless of the party’s developmental phase. This would mean that trans national diffusion is likely to play an important role throughout FRP development. Much of the evidence in favour of diffusion can also be the consequence of different FRPs responding independently to somewhat structural and universal demand side factors, without any form of interaction, communication, or learning from other FRPs’ experiences. However, as mentioned before, when responses or developmental patterns occur that closely follow or are almost identical to one another, it is highly unlikely such events or incidents are independent. Based on the existing diffusion literature and the empirical data obtained throughout this research project, the researcher will posit several causal mechanisms that reject the research project’s null hypothesis of myopic diffusion. In general, the researcher suggests that more academic efforts should be made to provide empirical evidence that allows us to differentiate between the states of myopic and deliberate diffusion. This differentiation clarifies what diffusion actually contains. In a following stage, therefore the researcher seeks to answer the question whether deliberate diffusion is taking place rather than myopic diffusion. The researcher designs a three step model that incorporates several necessary and sufficient conditions for trans national FRP diffusion. At first, it must be clear that temporal modelling is coherent with diffusion. This means that if one claims that one FRP diffuses trans nationally to another FRP, the latter’s reaction must be lagged to that of the former. Empirically, this step is the easiest one in the model to identify. The researcher shows that the timing and the pace of the FN’s emergence preceded the emergence of all the other proposed FRPs. Secondly, as already discussed in the previous chapters, when diffusion occurs, the transmitting and adopting agents must share specific common elements. This (perceived) similarity can be interpreted in two different ways, both with importance for the diffusion process. On one hand, the two agents must be similar in their being if diffusion is to occur between them (see chapter 3.3). For example, FRPs need to be structurally equivalent and have a limited social distance between them for diffusion to be successful. On the other hand, the two agents share an object of diffusion (e.g. master frame), which after adoption increases their similarity. Even though the object of diffusion can change forms in the respective political systems (e.g. frame alignment processes), it increases the agents’ similarity, their structural resemblance, and subsequently also their likelihood of successful development. This section identifies the constructs of the master frame, how certain FRPs resemble one
172
another (and why), and how both these factors contribute to the trans national diffusion process. Lastly, the most difficult step is to empirically identify how some of the common elements, most notably the object of diffusion, are diffused between FRPs. It means that it must be theoretically and empirically possible to identify the relational components that combine two or more agents and shape the diffusion process. Most importantly, this means one must be able to identify and describe the channels of diffusion. Here, the research project differentiates between direct and indirect diffusion. The theoretical framework emphasises that structural ties do not exist during the emergence phase, but that they are constructed. These contacts lead to an increasing familiarity and subsequently encourage the so called attribution of similarity between FRPs. Practically, the research project observes institutional equivalence, social characteristics, linguistic factors, the role of the media and the contradiction with a proposed strong national identity. At first, these three features are extensively conceptualised and operationalised so as to create a solid conceptual framework. Subsequently, this model is applied in a very specific comparative context, namely that of France and Belgium. By using the comparative method, and analysing patterns of diversity, the researcher seeks to provide an indication as to why and how (among other reasons) certain FRPs have successfully developed (and perhaps even consolidated), while others even failed to emerge. In general, the FN is seen as a mother party, i.e. an example for most West European FRPs. Both of the Belgian FRPs have their own unique history and relationship with the FN. These two unique relationships, which share the FN as a transmitting agent, allow the researcher to analyse diffusion as the comparative method intends to. In other words, the comparison between the FN VB and the FN FNb relationships provide diversity and allow the researcher to discuss their similarities and their differences. Through extensive illustrations, the researcher will analyse the time and the similarities between the two relationships, but the emphasis will be on the use of direct and indirect channels, and their in/direct consequences for the development of the Belgian FRPs. As extensively emphasised throughout the past chapters, there are no datasets or empirical studies this research project can fall back on and use as a foundation. The majority of the literature does not taken diffusion dynamics into account, and the exceptions that have taken this into account mostly do this theoretically and descriptively. Following some of the limitations mentioned in the methodological chapter, this research project mostly relies on empirical observations and illustrations to support its theoretical claims. It cannot rely on
173
large n dataset and draw statistical inferences from this. Nonetheless, the research project attempts to contribute to the field as an instigator and a standard for further research into trans national diffusion dynamics between FRPs, and perhaps more generally, political parties. Keeping this in mind, the researcher attempts to design a simulation model to substantiate the theoretical claims made throughout the research project. Such a simulation model allows the researcher to compare two (simulated) empirical situations, each describing a different reality. The first one describes the situation if diffusion would be myopic, i.e. non existent, whereas the second one describes a situation of genuine diffusion as indicated throughout the project. In other words, a simulation model allows this research project’s null hypothesis to be tested against a set of proposed alternative hypotheses.
174
E. REFERENCES
175
Aarelaid, A., 1998. Ikka kultuurile moeldes (Still Thinking about Culture). Tallinn: Virgela. Abedi, A., 2002. Challenges to established parties: The effects of party system features on the electoral fortunes of anti political establishment parties. European Journal of Political Research , 41(4), pp. 551 83. Abramowitz, A.I. and Saunders, K.L., 1998. Ideological Realignment in the U.S. Electorate. Journal of Politics , 60(3), pp.634–52. Adorno, T.W., Frenkel Brunswick, E., Levinson, D.J. and Sanford, R.N., 1950. The Authoritarian Personality. New York: Harper and Row. Abridged edition: 1982. New York and London: Norton and Company. Almond, G. and Verba, S., 1963. The civic culture . Princeton, NJ: Princeton University Press. Almond, G. and Verba, S. eds., 1989. The Civic Culture: Political Attitudes and Democracy in Five Nations. Newbury Park, CA: SAGE Publication Inc. Altemeyer, B., 1981. Right-Wing Authoritarianism . Winnipeg: The University of Manitoba Press. Aminzade, R., 1995. Between Movement and Party: The Transformation of Mid Nineteenth Century French Republicanism. In: J.C. Jenkins and B. Klandermans, eds. 1995 . The Politics of Social Protest: Comparative Perspectives on States and Social Movements . Minneapolis, MN: University of Minnesota Press, pp.39 62. Anckar, C., On the Applicability of the Most Similar Systems Design and the Most Different Systems Design in Comparative Research. International Journal of Social Research Methodology , 11(5), pp.389 401. Andersen, J.G., 2000. Radical right wing populism in Scandinavia: From tax revolt to neo liberalism and xenophobia. In: P. Hainsworth, ed. 2000. The Politics of the Extreme Right: From the margins to the mainstream . London: Pinter. Andersen, J.G., 2002. Denmark: a landslide to the right by trustful voters. In: Notre Europe, Sciences Po and EUI, 2002. Europe and the Crisis of Democracy . Elections in Europe: 1999- 2002. 6, Paris: Centre d'études européennes at Sciences Po, pp.13 15. Andeweg, R.B., 1995. The reshaping of national party systems. West European Politics , 18(3), pp.58 78. Andeweg, R.B., 2001. Lijphart versus Lijphart: the cons of consensus democracy in homogeneous societies. Acta Politica , 36, pp.117 28. Andrews, K.T. and Biggs, M., 2006. The dynamics of protest diffusion: Movement organizations, social networks, and new media in the 1960 sit ins. American Sociological Review , 71(5), pp.752 77. APSA, 2008. A Guide top Professional Ethics in Political Science . [pdf] Washington, DC: American Political Science Association. Available at:
176
Arzheimer, K. and Carter, E., 2003. Explaining variation in the extreme right vote: the individual and the political environment , Keele: KEPRU. Arzheimer, K. and Carter, E., 2006. Political Opportunity Structures and Right Wing Extremist Party Success. European Journal of Political Research , 45(3), pp.419 43. Back, H. and Teorell, J., 2005. Party Attachment and Political Participation. University of Karlstad: Karlstad Seminar in Studying Political Action . Karlstad, Germany 7 9 June 2005. Backes, U., 1991. Ideologie und Programmatik rechtsextremer Parteien – Unterschiede und Gemeinsamkeiten. In: J.W. Falter, H G. Jaschke and J.R. Winkler, eds. 1991. Rechtsextremismus: Ergebnisse und Perspectiven der Forschung . Opladen: Westdeutscher, pp.376 87. Backes, U., 2001. Gestalt und Bedeutung des intellektuellen Rechtsextremismus in Deutschland. In: Aus Politik und Zeitgeschichte (ApuZ) , Band 46. Backes, U. and Jesse, E., 1993. Politischer Extremismus in der Bundesrepublik Deutschland . Berlin: Propyläen. Baker, K., Dalton, R. and Hildebrandt, K., 1981. Germany Transformed: Political Culture and the New Politics. Cambridge, MA: Harvard University Press. Bale, T., 2003. Cinderella and her ugly sisters: the mainstream and extreme right in Europe’s bipolarizing party systems. West European Politics , 26(3), pp.67 90. Banerjee, A.V., 1992. A Simple Model of Herd Behavior. Quarterly Journal of Economics , 107(3), pp.797 817. Bangham, A.D., Standish, M.M. and Watkins, J.C., 1965. Diffusion of univalent ions across the lamellae of swollen phospholipids. Journal of Molecular Biology , 13(1), pp.238 52. Bardi, L., 1994. Transnational party federations, European Parliamentary Groups, and the building of Europarties. In: R.S. Katz and P. Mair, eds. 1994. How Parties Organise: Change and Adaptation in Party Organisations in Western Democracies . London: Sage, pp.357 72. Barker, M., 1981. The New Racism . London: Junction. Barnes, C., 1992. Qualitative Research: valuable or irrelevant?. Disability Handicap and Society , 7(2), pp.115 24. Bauman, Z., 2001. The Individualised Society . Cambridge, UK: Polity Press. Becker, H.S., 1996. The Epistemology of Qualitative Research. In: R. Jessor, A. Colby and R. Schweder, eds. 1996. Essays on Ethnography and Human Development . Chicago, IL: University of Chicago Press, pp.53 72. Bell, D., ed. 1963. The Radical Right . New York: Doubleday. Benford, R.D. and Snow, D.A., 2000. Framing processes and social movements: An overview and assessment. Annual Review of Sociology , 26, pp.611 39. Bennett, C.J., 1991. What is policy convergence and what causes it?. British Journal of Political Science , 21(2), pp.215 33. Berman, S., 1997. The life of a party. Comparative Politics , 30(1), pp.101 22. Bernstein, H., 1935. The Truth about the Protocols of Zion . New York, NY: Covici and Friede. Berry, B.F., 1994. Sizing Up State Policy Innovation Research. Policy Studies Journal , 22(3), pp.442 56. Berry, J.M., 2002. Validity and Reliability Issues in Elite Interviewing. PS: Political Science and Politics , 35(4), pp.679 82.
177
Berry, W.D. and Baybeck, D., 2005. Using Geographic Information Systems to Study Interstate Competition. American Political Science Review , 99(4), pp.505 19. Berry, F.S. and Berry, W.D., 1990. State Lottery Adoptions as Policy Innovations: An Event History Analysis. American Political Science Review , 84(2), pp.395 415. Betz, H G., 1993. The New Politics of Resentment: Radical Right Wing Populist Parties in Western Europe. Comparative Politics , 25(4), pp.413 27. Betz, H G., 1994. Radical Right-Wing Populism in Western Europe . New York, NY: St Martin’s Press. Betz, H G., 1996. The Radical Right in Western Europe: A Comparative Analysis (Review). Political Science Quarterly , 111(4), pp.716 17. Betz, H G., 1999. Contemporary Right Wing Radicalism in Europe. Contemporary European History , 8(2), pp.299 316. Betz, H G., 2001. Exclusionary Populism in Austria, Italy and Switzerland. International Journal , 56(3), pp.393 420. Betz, H G., 2002. Conditions favouring the success and failure of radical right wing populist parties in contemporary democracies. In: Y. Mény and Y. Surel, eds. 2002. Democracies and the Populist Challenge . Basingstoke: Palgrave, pp.197 213. Betz, H G., 2003. The growing threat of the radical right. In: P.H. Merkl and L. Weinberg, eds. 2003. Right-Wing Extremism in the Twenty-First Century . London: Frank Cass, pp.74 93. Betz, H G., 2004. La droite populiste en Europe: Extrème et démocrate?. Paris: CEVIPOF/Autrement. Betz, H G. and Immerfall, S. eds., 1998. The New Politics of the Right: Neo-Populist Parties and Movements in Established Democracies . New York, NY: St. Martin's Press. Billier, J. and De Witte, H., 1996. Attitudinal dispositions to vote for an extreme right wing party: the case of ‘Vlaams Blok’. European Journal of Political Research , 27(2), pp.181 202. Billig, M., 1995. Banal Nationalism . London: Sage. Blokker, P., 2005. Populist Nationalism, Anti Europeanism, Post Nationalism, and the East West Distinction. German Law Journal , 6(2), pp.371 89. Bobbio, N., 1994. Rechts und Links: Zum Sinn einer politischen Underscheiding. Blätter für deutssche und internationale Politik , 39(5), pp.543 9. Boehmke, F.J. and Witmer, R., 2004. Disentangling Diffusion: The Effects of Social Learning and Economic Competition on State Policy Innovations and Expansions. Political Research Quarterly , 57(1), pp.39 51. Boehnke, K.B. and Hadjar, A., 2004. Authoritarianism. Encyclopedia of applied psychology . Oxford, UK: Elsevier Academic Press. Bonjour, L., 2002. Epistemology: Classic Problems and Contemporary Responses . Lanham, MD: Rowman and Littlefield. Bonnell, V., 1983. Roots of Rebellion: Workers’ politics and organization in St Petersburg and Moscow. 1900-1914 . Berkeley, CA: University of California Press. Brandt, K W., 1986. New Social Movements as a Metapolitical Challenge: The Social and Political Impact of a New Historical Type of Protest. Thesis Eleven , 15(1), pp.60 68.
178
Braun, D. and Gillardi, F., 2006. Taking ‘Galton’s Problem’ Seriously – Towards a Theory of Policy Diffusion, Journal of Theoretical Politics , 18(3), pp.298 322. Braun, D., Gillardi, F., Füglister, K. and Luyet, S., 2007. Ex Pluribus Unum: Integrating the Different Strands of Policy Diffusion Theory. In: K. Holzinger, H. Jörgens and C. Knill, eds. 2007. Transfer, Diffusion und Konvergenz von Politiken . Wiesbaden: Verlag für Sozialwissenschaften, pp.39 55. Braun, R. and Koopmans, R., 2010. The Diffusion of Ethnic Violence in Germany: The Role of Social Similarity. European Sociological Review , 26 (1), pp.111 23. Bréchon, P. ed. 2005. Les Partis politiques français . Paris: La Documentation française. Bréchon, P., 2009. La France aux urnes . Paris: La Documentation française. Brinkman, S. and Kvale, S., 2005. Confronting the ethics of qualitative research. Journal of Constructivist Psychology , 18(2), pp.157 81. Bruce, J.M., Clark J.A. and Kessel J.H., 1991. Advocacy Politics in Presidential Parties. American Political Science Review , 85(4), pp.1089 105. Bryman, A., 2001. Social research methods . Oxford, UK: Oxford University Press. Budge, I., 1982. Electoral volatility: Issue effects and basic change in 23 post war democracies. Electoral Studies , 1(2), pp.147 68. Budge, I., Crewe, I. and Farlie, D., eds. 1976. Party Identification and Beyond . New York, NY: Wiley. Budge, I. and Farlie, D.J., 1983. Explaining and Predicting Elections: Issue Effects and Party Strategies in Twenty-Three Democracies . London: Allen and Unwin. Budge, I., Robertson, D. and Hearl, D., eds. 1987. Ideology, Strategy and Party Change: Spatial Analysis of Post-War Election Programmes in 19 Democracies . New York, NY: Cambridge University Press. Buechler, S., 1990. Women’s Movements in the United States . New Brunswick, NJ: Rutgers University Press. Bulmer, M., 2001. The Ethics of Social Research. In: N. Gilbert, ed. 2001. Researching Social Life . London: Sage, pp.45 57. Burgess, R., 1982. The Unstructured Interview as a Conversation. In: R. Burgess, ed. 1982. Field Research: A Sourcebook and Field Manual . London: George Allen and Unwin, pp.107 10. Campbell, A., Converse, P.E., Miller, W.E. and Stokes, D.E., 1960. The American Voter . New York, NY: John Wiley and Sons. Canovan, M., 1981. Populism. New York, NY: Harcourt Brace Javonovich. Canovan, M., 2002. Taking Politics to the People: Populism as the Ideology of Democracy. In: Y. Mény and Y. Surel, eds. 2002. Democracies and the Populist Challenge. New York, NY: Palgrave. Carroll, W.K. and Ratner, R.S., 1996. Master framing and cross movement networking in contemporary social movements. Sociology Quarterly , 37(4), pp.601 25. Carsey, T.M. and Layman, G.C., 2006. Changing Sides or Changing Minds? Party Identification and Policy Preferences in the American Electorate. American Journal of Political Science , 50(2), pp.464–77.
179
Carter, E., 2002. Proportional representation and the fortunes of right wing extremist parties. West European Politics , 25(3), pp.125 46. Carter, E., 2005. The extreme right in Western Europe: success or failure? . Manchester: Manchester University Press. Castells, M., 1997. The Power of Identity . Oxford, UK: Blackwell. Christie, R. and Jahoda, M. eds., 1954. Studies in the Scope and Method of “The Authoritarian Personality ”. Westport, CT: Greenwood. Christofferson, T., 2003. The French elections of 2002: the issue of insecurity and the Le Pen effect. Acta Politica , 38(1), pp.109 23. Clark, T.N. and Lipset, S.M. eds., 2001. The Breakdown of Class Politics. Washington, DC: Woodrow Wilson Center Press. Coffé, H., 2004. Groot in Vlaanderen, klein(er) in Wallonië: Een analyse van het electorale success van de extreme-rechtse partijen . Ph.D. Free Univeristy of Brussels (VUB) Coffé, H., 2005. Extreem-rechts in Vlaanderen en Wallonië: het verschil . Roeselare: Roulerta. Cohen, B.C., 1963. The Press and Foreign Policy. Princeton, NJ: Princeton University Press. Cohen, J.L., 1982. Between Crisis Management and Social Movements: The Place of Institutional Reform. Telos , 52, pp.21 40. Cohen, J.L., 1983. Rethinking Social Movements. Berkeley Journal of Sociology , 28, pp.97 113. Cohen, J.L., 1985. Strategy or identity: New theoretical paradigms and contemporary social movements. Social Research , 52(4), pp.663 716. Cole, A., 2005. Old right or new right? The ideological positioning of parties of the far right. European Journal of Political Research , 44(1), pp.203 30. Coleman, J.S., Katz, E. and Menzel, H., 1966. Medical Innovation: A Diffusion Study . Indianapolis, IN: Bobbs Merrill. Collier, D., 1995. Translating Quantitative Methods for Qualitative Researchers: The Case of Selection Bias. The American Political Science Review , 89(2), pp.461 66. Collier, D. and Mahon, J.E., 1993. Conceptual ‘stretching’ revisited: adapting categories in comparative analysis. The American Political Science Review , 87(4), pp.845 55. Collier, D., and Messick, R.E., 1975. Prerequisites versus Diffusion: Testing Alternative Explanations of Social Security Adoption. The American Political Science Review , 69(4), pp.1299 315. Committee for Protection of Human Subjects (CPHS), 1992. Guidelines for the Protection of Human Subjects . Berkeley, CA: University of California Press. Daalder, H., 1992. A crisis of party?. Scandinavian Political Studies , 15(4), pp.269–87. Dahl, R., 1989. Democracy and its Critics . New Haven, CT: Yale University Press. Dahlman, C.J., 1979. The Problem of Externality. Journal of Law and Economics , 22(1), pp.141 62. Dalton, R., McAllister, I. and Wattenberg, M., 2002. Political Parties and their Publics. In: R. Luther and F. Müller Rommel, eds. 2002. Political Parties in the New Europe . Oxford, UK: Oxford University Press, pp.19 42.
180
Dalton, R., Kuechler M. and Burklin, W., 1990. The challenge of new social movements. In: R. Dalton and M. Kuechler, eds. 1990. Challenging the Political Order . New York, NY: Oxford University Press, pp.3 20. DeAngelis, R.A., 2003. A Rising Tide for Jean Marie, Jörg and Pauline? Xenophobic Populism in Comparative Perspective. Australian Journal of Politics and History , 49(1), pp.75 92. de Benoist, A., 1977. Vu de droite. Anthologie critique des idées contemporaines . Paris: Copernic. de Benoist, A., 1994. Nationalisme: Phenomenology et critique. Paris: GRECE. Decker, F., 2000. Parteine Unter Druck: Der Neue Rechtspopulismus in Den Westlichen Demokratien . Opladen: Leske + Budrich. Decker, F., 2003. Rechspopulismus in der Bundesrepublik Deutschland: Die Schill Partei. In: N. Werz, ed. 2003. Populismus: populisten in Ubersee und Europe , Opladen: Leske + Budrich. Decker, F., 2004. Der neue Rechtspopulismus . Opladen: Leske + Budrich. DeClair, E.G., 1999. Politics on the fringe: The people, Policies, and Organization of the French National Front . Durham, N.C.: Duke University Press. Dehousse, R., ed. 2002. Europe and the Crisis of Democracy: Elections in Europe: 1999– 2002 . Paris: Notre Europe. de Lange, S.L., 2007. A New Winning Formula? The programmatic appeal of the radical right. Party Politics , 13(4), pp.411 35. de Lange, S., 2008. From Pariah to Power: The Government Participation of Radical Right- Wing Populist Parties in West European Democracies . Ph.D. University of Antwerp. Della Porta, D., Kriesi, H., Rucht, D. eds., 2009. Social Movements in a Globalizing World . 2nd ed. Basingstoke: Palgrave Macmillan. Delwit, P. and Poirier, P., 2007. Extrême Droite et Pouvoir en Europe . Brussels: Editions de L'Université de Bruxelles. Denzin, N.K. and Lincoln, Y.S., 1994. Introduction: Entering the field of qualitative research. In: N.K. Denzin and Y.S. Lincoln, eds. 1996. Handbook of qualitative research . Thousand Oaks, CA: Sage, pp.1 17. Deschouwer, K., 1992, The Survival of the Fittest: Measuring and Explaining Adaptation and Change of Political Parties , paper presented for presentation at the Workshop on ‘Democracies and the Organisation of Political Parties’, European Consortium for Political Research, Limerick, Ireland, 30 March to 4 April 1992. Deutschman, P.J., 1963. The Mass Media in an Underdeveloped Village. Journalism Quarterly , 40(1), pp.27 35. Devine, F., 1995. Qualitative Methods. In: D. Marsh and G. Stoker, eds. 1995. Theory and Methods in Political Science , London: Macmillan Press, pp.137 53. De Witte, H., 1998. Torenhoge verschillen in de lage landen: over het verschil in success tussen de Centrumdemocraten en het Vlaams Blok. In: J. Van Holsteyn and C. Mudde, eds. 1998. Extreem-rechts in Nederland . The Hague: Sdu, pp.157 73. Dexter, L.A., 1969. The Sociology and Politics of Congress. Chicago, IL: Rand McNally. Diani, M. and McAdam, D. eds., 2003. Social Movements and Networks . Oxford/New York: Oxford University Press. DiMaggio, P.J. and Powell, W.W., 1983. The iron cage revisited: institutional isomorphism
181
and collective rationality in organizational fields. American Sociological Review, 48(2), pp.147 60. Dobbin, F., Simmons, B. and Garrett, G., 2007. The Global Diffusion of Public Policies: Social Construction, Coercion, Competition, or Learning?. Annual Review of Sociology , 33, pp.449–72. Dolowitz, D.P., 1997. British Employment Policy in the 1980s: Learning from the American Experience. Governance , 10(1), pp.23 42. Downs, A., 1957. An Economic Theory of Political Action in a Democracy. The Journal of Political Economy, 65(2), pp.135 50. Dreyer, E.C., 1973. Change and stability of party identifications. Journal of Politics , 35(3), pp.712–22. Dülmer, H. and Klein, M., 2005. Extreme right wing voting in Germany in a multilevel perspective: a rejoinder to Lubbers and Scheepers. European Journal of Political Research , 44(2), pp.243 63. Dunleavy, P. and Ward, H., 1981. Exogenous Voter Preferences and Parties with State Power: Some Internal Problems of Economic Theories of Party Competition. British Journal of Political Science , 11(3), pp.351 80. Duyvendak, J.W. and Koopmans, R., 1992. Protest in een pacificatiedemocratie. Nieuwe sociale bewegingen en het Nederlandse politieke systeem. In: J.W. Duyvendak and R. Koopmans, eds. 1992. Tussen verbeelding en macht. 25 jaar nieuwe sociale bewegingen in Nederland. Amsterdam: SUA, pp.39 58. Dyer, R., 1993. The Matter of Images: Essays on Representation. New York, NY: Routledge. Eatwell, R., 1989. The Nature of the Right – The Right as a Variety of Styles of Thought. In: R. Eatwell and N. O’Sullivan, eds. 1989. The Nature of the Right. European and American Politics and Political Thought since 1789 . London: Pinter, pp.62 76. Eatwell, R., 2000. The Extreme Right and British Exceptionalism: The Primacy of Politics. In: P. Hainsworth, ed. 2000. The Politics of the Extreme Right: From the Margins to the Mainstream . London: Pinter, pp.172 92. Eatwell, R., 2003. Ten theories of the extreme right. In: P.H. Merkl and L. Weinberg, eds. 2003. Right-Wing Extremism in the Twenty-First Century . London: Frank Cass, pp.47 73. Eatwell, R., 2005. Charisma and the revival of the European extreme right. In: J. Rydgren, ed. 2005b. Movements of Exclusion: Radical Right-Wing Populism in the West . Hauppage: Nova Science, pp.101 20. Eatwell, R., 2006. The concept and theory of charismatic leadership. Totalitarian Movements and Political Religions , 7(2), pp.141 56. Eckstein, H., 1975. Case study and theory in political science. In: F.J. Greenstein and N.W. Polsby, eds. 1975. Handbook of political science (Volume 7) . Reading, MA: Addison Wesley, pp.79 137. Eder, K., 1993. The New Politics of Class: Social Movements and Cultural Dynamics in Advanced Societies . London: Sage. Eff, E.A., 2004. Does Mr. Galton Still Have a Problem? Autocorrelation in the Standard Cross Cultural Sample. World Cultures , 15(2), pp.153 70. Eisinger, P.K., 1973. The conditions of protest behavior in American cities. The American Political Science Review , 67(1), pp.11 28.
182
Eismann, W. ed., 2002. Rechtspopulismus: Österreichische Krankheit oder Europaïsche Normalität? . Vienna: Czernin Verlag. Elkins, Z. and Simmons, B., 2005. On waves, clusters, and diffusion: a conceptual framework. Annals of the American Academy of Political and Social Science , 598, pp.33–51. Ellinas, A.A., 2008. Parties after Success: Why some radical right parties persist and others collapse. In: APSA (American Political Science Association), 104 th annual meeting themed “Categories and the Politics of Global Inequalities” . Boston, United States of America, 28 31 August 2008, Boston, MA: APSA. Emirbayer, M. and Goodwin, J., 1994. Work Analysis, Culture and the Problem of Agency. The American Journal of Sociology , 99(6), pp.411 54. Epstein, L. and Mershon, C., 1996. Measuring Political Preferences. The American Journal of Political Science , 40(1), pp.261 94. Esser, H., 2001. Soziologie: Spezielle Grundlagen. Bd. 6: Sinn und Kultur . Frankfurt: Campus Verlag. Evans, G. and Need, A., 2002. Explaining ethnic polarization over attitudes towards minority rights in Eastern Europe: A multilevel analysis. Social Science Research , 31(4), pp.653 80. Evans, J., Arzheimer, K, Baldini, G., Bjørkland, T, e.a. 2001. Comparative mapping of extreme right electoral dynamics: an overview of EREPS (Extreme Right Electorates and Party Success). European Political Science , 1(1), pp.42 53. Evans, J. and Ivaldi, G., 2002. Les dynamiques électorales de l’extrême droite européenne. Revue politique et parlementaire , 104(1019), pp.67 83. Evola, J., 1969[1934]. Rivolta contro il mondo moderno. revised ed. Rome: Edizioni mediterranee. Evola, J., 1973[1961]. Cavalcare la Tigre. Orientamenti esistenziali per un epoca della dissoluzione . revised ed. Milan: Scheiwiller. Fantasia, R., 1988. Cultures of Solidarity . Berkeley, CA: University of California Press. Fennema, M., 1997. Some Conceptual Issues and Problems in the Comparison of Anti Immigrant Parties in Western Europe. Party Politics , 3(4), pp.473 92. Fennema, M. and Tillie, J., 1998. A Rational Choice for the Extreme Right. Acta Politica , 34, pp.223 49. Fetzer, J., 2000. Economic self interest or cultural marginality? Anti immigration sentiment and nativist political movements in France, Germany and the USA. Journal of Ethnic and Migration Studies , 26(1), pp.5 23. Fieschi, C., 2000. European institutions: The far right and illiberal politics in a liberal context. Parliamentary Affairs , 53(3), pp.517 31. Fieschi, C., Shields, J. and Woods, R., 1996. Extreme right wing parties and the European Union: France, Germany and Italy. In: J. Gaffney, ed. 1996. Political Parties and the European Union . London: Routledge, pp.235 53. Finkel, S.E. and Opp, K D., 1991. Party Identification and Participation in Collective Political Action. The Journal of Politics , 3(2), pp.339 71. Finnemore, M., 1996. National Interests in International Society. Ithaca, NY: Cornell University Press. Fiorina, M.P., 1981. Retrospective Voting in American national Elections. New Haven, CT: Yale University Press.
183
Fireman, B. and Gamson, W.A., 1979. Utilitarian logic in the resource mobilization perspective. In: M.N. Zald and J.D. McCarthy, eds. 1979. The Dynamics of Social Movements . Cambridge, MA: Winthrop, pp.8 45. Flacks, R., 2004. Knowledge for what? Thoughts on the State of Social Movement Studies. In: J. Goodwin and J. Jasper, eds. 2004. Rethinking social movements: Structure, meaning, and emotion , Lanham, MD: Rowman and Littlefield, pp.135 53. Flecker, J. ed., 2007. Changing Working Life and the Appeal of the Extreme Right . Aldershot: Ashgate. Flick, U., 2007a. Designing Qualitative Research (Book 1 of the SAGE Qualitative Research Kit). London: Sage. Flick, U., 2007b. Managing Quality in Qualitative Research (Book 8 of the SAGE Qualitative Research Kit). London: Sage. Flyvbjerg, B., 2004. Five Misunderstandings About Case Study Research. In: C. Seale, G. Gobo, J.F. Gubrium and D. Silverman, eds., 2004. Qualitative Research Practice . London and Thousand Oaks, CA: Sage, pp.420 34. Foucault, M., 1980. Power/Knowledge . New York, NY: Pantheon Books. Franklin, M.N., Mackie, T.T. and Valen, H. eds., 1992. Electoral change. Responses to evolving social and attitudinal structures in western countries. New York, NY: Cambridge University Press. Freeden, M., 1997. Editorial: ideologies and conceptual history. Journal of Political Ideologies , 2(1), pp.3 12. Freeden, M., 1999. The Ideology of New Labour. The Political Quarterly , 70(1), pp.42 51. French, J.R.P. and Raven, B., 2001. The Bases of Social Power. In: I.G. Asherman and S.V. Asherman, eds. 2001. The Negotiation Sourcebook. 2nd ed. Amherst, MA: HRD Press, pp.61 74. Gallagher, M., 1991. Proportionality, Disproportionality and Electoral Systems. Electoral Studies , 10(1), pp.33 51. Gallagher, M., Laver, M. and Mair, P. eds., 2005. Representative Government in Modern Europe” Institutions, Parties and Governments . 4 th ed. Boston, MA: McGraw Hill. Gamson, W.A., 1992. Talking Politics. New York, NY: Cambridge University Press. Gamson, W.A. and Meyer, D., 1996. Framing Political Opportunity. In: D. McAdam, J. McCarthy and M.N. Zald, eds. 1996. Comparative Perspectives on Social Movements . New York, NY: Cambridge University Press. Gamson W.A. and Wolsfeld, M.A., 1993. Movements and media as interacting systems. Annals of the American Academy of Political and Social Science, 528, pp.114–25. Geddes, B., 1990. How the Cases you Choose Affect the Answers you Get: Selection Bias in Comparative Politics. In: J. Stimson, ed. 1990. Political Analysis 2 . Ann Arbor, MI: University of Michigan Press. Gellner, E.J., 1983. Nations and Nationalism . Oxford, UK: Blackwell. Gellner, E.J., 1997. Nationalism . London: Weidenfeld and Nicolson. Gerhards, J. and Rucht, D., 1992. Mesomobilization: organizing and framing in two protest campaigns in West Germany. American Journal of Sociology , 98(3), pp.555–95. Gerring, J., 2001. Social Science Methodology: A Criterial Framework. Cambridge, UK: Cambridge University Press. Gerring, J., 2007. Case Study Research: Principles and Practices . Cambridge, UK:
184
Cambridge University Press. Gershman President, G., 2004. Democracy promotion: the relationship of political parties and civil society. Democratization , 11(3), pp.27 35. Giscard d’Estaing, V., 1976. Démocratie française . Fayard: Paris. Givens, T., 2002. The role of socioeconomic variables in the success of radical right parties. In: M. Schain, A. Zolberg and P. Hossay, eds. 2002. Shadows over Europe: The Development and Impact of the Extreme Right in Western Europe . New York, NY: Palgrave, pp.137 58. Givens, T., 2005. Voting Radical Right in Western Europe . New York, NY: Cambridge University Press. Gleditsch, K.S. and Ward, M., 2006. Diffusion and the International Context of Democratization. International Organization, 60(4), pp.911–33. Glick, H.R., 1981. Innovation in State Judicial Administration: Effects on Court Management and Organization, American Politics Quarterly , 9(1), pp.49 69. Goffman, E., 1974. Frame Analysis: An Essay on the Organization of the Experience. New York, NY: Harper Colophon. Golder, M., 2003. Explaining Variations in the Success of Extreme Right parties in Western Europe. Comparative Political Studies , 36(4), pp.432 66. Goldstein, K., 2002. Getting in the Door: Sampling and Completing Elite Interviews. PS: Political Science and Politics, 35(4), pp.669 72. Goldthorpe, J.H., 1997. Current Issues in Comparative Macrosociology: A Debate on Methodological Issues. Comparative Social research , 16, pp.1 26. Gomez Reino, M., 2001. Do new party organisations matters? Party dynamics and the new radical right wing family . Paper presented at the ECPR Joint Sessions of Workshops, Grenoble, April 6 11. Goodwin, J. and Jasper, J., 1999. Caught in a Winding, Snarling Vine: The Structural Bias of Political Process Theory. Sociological Forum, 14(1), pp.27 54. Goodwin, J. and Jasper, J. eds., 2009. The Social Movements Reader: Cases and Concepts . Oxford, UK: Wiley Blackwell. Gould, R.V., 1995. Insurgent Identities . Chicago, IL: University of Chicago Press. Gould, R.V., 1998. Political Networks and the local/national boundary in the Whiskey Rebellion. In: M. Hanagan, L.P. Moch and W. te Brake, eds. 1998. Challenging Authority . Minneapolis, MN: University of Minneapolis Press. Granovetter, M., 1978. Threshold models of collective behavior. American Journal of Sociology , 83(6), pp.1420 43. Gray, V., 1973. Innovation in the States: A Diffusion Study. American Political Science Review , 67(4), pp.1174 85. Gray, V., 1976. Models of Comparative State Politics: A Comparison of Cross Sectional and Time Series Analysis. American Journal of Political Science , 20(2), pp.235 56. Green, D.P. and Shapiro, I., 1994. Pathologies of Rational Choice Theory: A Critique of Applications in Political Science . New Haven, CT: Yale University Press. Greve, H., 1998. Managerial cognition and the mimetic adoption of market positions: What you see is what you do. Strategic Management Journal , 19, pp.967– 88. Greenfeld, L., 2001. Etymology, definitions, types. In: A.J. Motyl, ed. 2001. Encyclopedia of Nationalism. Volume I: Fundamental Themes . San Diego, CA: Academic Press, pp.251 65.
185
Gunning, J., 2008. Hamas in politics: democracy, religion, violence . London: C. Hurst & Co. Hägerstrand, T., 1952. The Propagation of Innovation Waves. Lund Studies in Human Geography , series B4, pp.3 19. Hague, R., Harrop, M. and Breslin, S., 1998. Comparative Government and Politics: An Introduction . 4th ed. London: Macmillan. Habermas, J., 1987. Lifeworld and System: A Critique of Functionalist Reason (translated by Thomas McCarthy). Cambridge, UK: Polity Press. Hainsworth, P. ed., 2000. The Politics of the Extreme Right: From the margins to the mainstream . London: Pinter. Hall, P.A., 2003. Aligning Ontology and Methodology in Comparative Research. In: J. Mahoney and D. Reuschemeyer, eds. 2003. Comparative Historical Analysis in the Social Sciences . New York, NY: Cambridge University Press, pp.373 404. Harmel, R. and Janda, K., 1982. Parties and Their Environments: Limits to Reform?. New York, NY: Longman. Harmel, R. and Janda, K., 1994. An Integrated Theory of Party Goals and Party Change. Journal of Theoretical Politics , 6(3), pp 259 87. Harmel, R. and Svåsand, L., 1993. Party leadership and party institutionalisation: three phases of development. West European Politics , 16(2), pp.67 88. Hedstrøm, P., 1994. Contagious Collectivities: On the Spatial Diffusion of Swedish Trade Unions, 1890–1940. American Journal of Sociology , 99(5), pp.1157–79. Hedstrøm, P. and Swedberg, R., 1998. Social Mechanisms: An Analytical Approach to Social Theory . Cambridge, UK: Cambridge University Press. Heinisch, R., 2003. Success in opposition – failure in government: explaining the performance of right wing populist parties in public office. West European Politics , 26(3), pp.91 130. Hennecke, H.J., 2003. Das Salz in der Wunden der Konkordanz: Christoph Blocher und die Schweizer Politik. In: N. Werz, ed. 2003. Populismus: Populiste in Übersee und Europe . Opladen: Leske + Budrich, pp.145 62. Herbig, P.A. and Palumbo, F., 1994. The Effect of Culture on the Adoption Process: A Comparison of Japanese and American Behavior. Technological Forecasting and Social Change , 46(1), pp.71 101. Hertz, R. and Imber, J.B. eds., 1995. Studying Elites Using Qualitative Methods . Thousand Oaks, CA: Sage. Heyman, E. and Mickolus, E., 1981. Imitation by Terrorists: Quantitative Approaches to the Study of Diffusion patterns in Transnational Terrorism. In: Y.G. Alexander and J.M. Gleason, eds. 1981. Behavioral and Quantitative Perspectives on Terrorism . New York, NY: Pergamon, pp.175 228. Himmelstein, J.L., 1990. To the Right. The Transformations of American Conservatism . Berkeley, CA: University of California Press. Hirsch, E.L., 1990. Sacrifice for the cause: group processes recruitment and commitment in a student social movement. American Sociological Review , 55(2), pp.243 54. Hobsbawm, E.J., 1990. Nations and Nationalism since 1780 . New York, NY: Cambridge University Press.
186
Hoffman, S., 1956. Le mouvement Poujade. Paris: Librairie Armand Colin. Holmes, D.R., 2000. Integral Europe: Fast-Capitalism, Multiculturalism, Neo-Fascism . Princeton, NJ: Princeton University Press. Huber, E., Ragin, C. and Stephens, J.D., 1993. Social democracy, Christian democracy, constitutional structure, and the welfare state. American Journal of Sociology , 99(3), pp.711 49. Hume, D., 1910 (1748). An Inquiry concerning Human Understanding . Berwyn, PA: P.F. Collier and Son. Hunt, S.A., Benford, R.D. and Snow, D.A., 1994. Identity Fields: Framing Processes and the Social Construction of Movement Identities. In: E. Laraña, H. Johnston and J.R. Gusfield, eds. 1994. New Social Movements: From Ideology to Identity . Philadelphia, PA: Temple University Press, pp.185–208. Hunter, M.L., 1998a. Nationalism unleashed: Jean Marie Le Pen, head of France’s National Front, moves east. Transitions , 5(7), pp.18 28. Hunter, M.L., 1998b., Oil, guns and money: The National Front goes to Chechnya. Transitions , 5(7), pp.29 32. Huntington, N. and Bale, T., 2002. New Labour: New Christian Democracy?. The Political Quarterly , 73(1), pp.44 50. Huntington, S.P., 1993. The Third Wave: Democratization in the Late Twentieth Century . Norman, OK: University of Oklahoma Press. Husbands, C.T., 1981. Contemporary Right Wing Extremism in Western European Democracies: A Review Article. European Journal of Political Research , 9(1), pp.75–99. Husbands, C.T., 1992. The Other Face of 1992: The Extreme Right Explosion in Western Europe. Parliamentary Affairs , 45(3), pp.267 84. Husbands, C.T., 1996. Racism, xenophobia and the extreme right: a five country assessment. In: S. Bekker and D. Carlton, eds. 1996. Racism, Xenophobia and Ethnic Conflict . Durban: Indicator, pp.97 118. Husbands, C.T., 1998. De Centrumstroming in perspectief: hoe verschillend is Nederland?, In: J. van Holsteyn and C. Mudde, eds. 1998. Extreem-Rechts in Nederland . The Hague: Sdu, pp.175 91. Husbands, C.T., 2000. Switzerland: right wing and xenophobic parties, from margin to mainstream?. Parliamentary Affairs , 53(3), pp.501 16. Ignazi, P., 1992. The Silent Counter Revolution: Hypotheses on the Emergence of the Extreme Right Wing Parties in Europe. European Journal of Political Research , 22(1), pp.3 34. Ignazi, P., 1996. The crisis of parties and the rise of new political parties. Party Politics , 2(4), pp.549 66. Ignazi, P., 1997. The extreme right in Europe: a survey. In: P.H. Merkl and L. Weinberg, eds. 1997. The Revival of Right-Wing Extremism in the Nineties. London: Frank Cass, pp.47 64. Ignazi, P., 2002. The Extreme Right: Defining the Object and Assessing the Causes. In: M. Schain, A. Zolberg and P. Hossay, eds. 2002. Shadows over Europe: The Development and Impact of the Extreme Right in Western Europe . London: Palgrave Macmillan, pp.21 38. Ignazi, P., 2003. Extreme Right Parties in Western Europe . Oxford, UK: Oxford University Press.
187
Immerfal, S., 1998. The neo populist agenda. In: H G. Betz and S. Immerfall, eds. 1998. The New Politics of the Right: Neo-Populist Parties and Movements in Established Democracies . New York, NY: St. Martin’s Press, pp.249 61. Inglehart, R., 1977. The Silent Revolution . Princeton, NJ: Princeton University Press. Inglehart, R., 1990. Culture shifts in advanced industrial society . Princeton, NJ: Princeton University Press. Ionescu, G., and Gellner, E. eds., 1969. Populism - Its Meaning And National Characteristics. London: Weidenfeld and Nicolson. Ivarsflaten, E., 2002. Re-considering the populist right’s challenge to contemporary democracies in Western Europe: A critical evaluation of Kitschelt’s account . M.Phil dissertation, University of Oxford. Jackman, R.W. and Volpert, K., 1996. Conditions favouring parties of the extreme right in Western Europe. British Journal of Political Science , 26(4), pp.501 21. Jagers, J., 2006. Stem van het volk! Populisme als concept getest by Vlaamse politieke partijen . Ph.D. University of Antwerp. Janda, K., 1993. Comparative political parties: research and theory. In: A.W. Finifter, ed. 1993. Political Science: The State of the Discipline . 2nd ed. Washington, DC: APSA, pp.163 92. Jaschke, H G., 1994. Die Republikaner, Profile einer Rechtaußen-Partei . 3 rd ed. Bonn: Dietz. Jasper, J.M., 1997. The Art of Moral Protest . Chicago, IL: University of Chicago Press. Jasper, J.M. and Poulsen, J.D., 1993. Fighting back: vulnerabilities, blunders and countermobilization by the targets of three animal rights campaigns. Sociological Forum , 8(4), pp.639 57. Jasper, J.M. and Poulsen, J.D., 1995. Recruiting Strangers and Friends: Moral Shocks and Social Networks in Animal Rights and anti Nuclear Protest. Social Problems , 42(4), pp.493 512. Jenkins, J.C. and Eckert, C.M., 1986. Channeling Black Insurgency: Elite Patronage and Professional Social Movement Organizations in the Development of the Black Movement. American Sociological Review , 51(6), pp.812 29. Jenkins, J.C. and Schock, K., 1992. Global structures and political processes in the study of domestic political conflict. Annual Review of Sociology , 18, pp.161 85. Johnston, H., 1994. New Social Movements and Old Regional Nationalisms. In: E. Laraña, H. Johnston and J.R. Gusfield, eds. 1994. New Social Movements. From Ideology to Identity , Philadelphia, PA: Temple University Press, pp.267 86. Johnston, H. and Aarelaid Tart, A., 2000. Generations, Microcohorts, and Long Term Mobilization: The Estonian National Movement, 1940 1991. Sociological Perspectives , 43(4), pp.671 98. Jost, J.T., Glaser, J., Kruglanski, A.W. and Sulloway, F.J., 2003. Political conservatism as motivated social cognition. Psychological Bulletin , 129(3), pp 339 75. Jungerstam Mulders, S., 2003. Uneven Odds: The Electoral Success of the Freiheitliche Partei Österreichs, the Vlaams Blok, the Republikaner and the Centrumdemocraten under the Conditions Provided by the Political System in Austria, Belgium, Germany and the Netherlands . Helsinki: Helsinki University Press. Kahneman, D. and Tversky, A., 2000. Choices, values, and frames . Cambridge, UK: Cambridge University Press.
188
Karch, A., 2007. Emerging Issues and Future Directions in State Policy Diffusion Research. State Politics and Policy Quarterly , 7(1), pp.54 80. Karvonen, L., 2010. The personalisation of politics. A study of parliamentary democracies. Colchester: ECPR Press. Katz, E., 1968. Diffusion (interpersonal influence). In: D.L. Shils, ed. 1968. International encyclopaedia of the social sciences . London: Macmillan. Katz, E. and Lazarsfeld, P.F., 1955. Personal Influence . Glencoe, IL: The Free Press. Katz, R.S. and Mair, P., 1995. Changing models of party organization and party democracy: the emergence of the cartel party. Party Politics , 1(1), pp.5–28. Keleman, R.D. and Sibbitt, E.C., 2004. The Globalization of American Law. International Organization , 58(1), pp.103 36. Kent, G., 2000. Informed Consent. In: D. Burton, ed. 2000. Research Training for Social Scientists: A Handbook for Post-Graduate Researchers . London: Sage, pp.61 67. Key, V., 1955. A Theory of Critical Elections. Journal of Politics , 17(1), pp.3 18. Key, V., 1959. Secular Realignment and the Party System. Journal of Politics , 21(2), pp.198 210. Kingdon, J.W., 1995. Agendas, Alternatives and Public Policies . 2 nd ed. New York, NY: Harper Collins. Kingdon, J.W., 2010. Agendas, Alternatives, and Public Policies. 2nd ed. Harlow: Longman. Kirchheimer, O., 1966. The Vanishing Opposition. In: R. Dahl, ed. 1966. Political Opposition in Western Democracies , New Haven, CT: Yale University Press, pp.237 59. Kirscht, J.P. and Dillehay, R.C., 1967. Dimensions of Authoritarianism: A Review of Research and Theory . Lexington, KY: University of Kentucky Press. Kitschelt, H., 1986. Political Opportunity Structures and Political Protest: Anti Nuclear Movements in Four Democracies. British Journal of Political Science , 16(1), pp.57 85. Kitschelt, H., 1991. Resource Mobilization Theory: A Critique. In: D. Gerdes and D. Rucht, eds. 1991. Research on Social Movements: The State of the Art . Boulder, CO: Westview, pp.235 71. Kitschelt, H. and McGann, A., 1995. The Radical Right in Western Europe: A Comparative Analysis . Ann Arbor, MI: University of Michigan Press. Klandermans, B., 1984. Mobilization and participation: social psychological expansions of resource mobilization theory. American Sociological Review , 49(5), pp.583 600. Klandermans, B., 1991, New social movements and resource mobilization: The European and the American approach revisited. Politics & the Individual , 1(2), pp.89 111. Klandermans B. 1992. The social construction of protest and multiorganizational fields. In: A.D. Morris and C.M. Mueller, eds. 1992. Frontiers in Social Movement Theory. New Haven, CT: Yale University Press, pp.77 103. Klandermans, B., 1997. The Social Psychology of Protest. Oxford, UK: Blackwell. Klandermans, B., de Weerd, M., Sabucedo, J.M. and Costa, M., 1999. Injustice and adversarial frames in a supranational political context: farmers’ protest in the Netherlands and Spain. In: D. della Porta, H. Kriesi and D. Rucht, eds. 1999. Social Movements in a Globalizing World. London: Macmillan, pp.134 47. Klandermans, B. and Mayer, N. eds., 2006. Extreme Right Activists in Europe: Through the Magnifying Glass . London: Routledge.
189
Klandermans, B. and Tarrow, S., 1988. Mobilization into Social Movements Synthesizing American and European Approaches. International Social Movement Research , 1(1), pp.1 42. Knigge, P., 1998. The ecological correlates of right wing extremism in Western Europe. European Journal of Political Research , 34(2), pp.249 79. Koch, K., 1991. Back to Sarajevo or beyond Trianon? Some thoughts on the problem of nationalism in Eastern Europe. Netherlands Journal of Social Sciences , 27(1), pp.29 42. Koopmans, R., 1996. Explaining the rise of racist and extreme right violence in Western Europe: grievances or opportunities?. European Journal of Political Research , 30(2), pp.185 216. Koopmans, R., 1999. Political. Opportunity. Structure. Some Splitting to Balance the Lumping. Sociological Forum , 14(1), pp.93 105. Koopmans, R., 2004. Movements and Media: Selection Processes and Evolutionary Dynamics in the Public Sphere. Theory and Society, 33(3), pp.367 93. Koopmans, R. and Olzak. S., 2004. Discursive Opportunities and the Evolution of Right Wing Violence in Germany. American Journal of Sociology , 110(1), pp.198–230. Koopmans, R., Statham, P., 1999. Political Claims Analysis: Integrating Protest Event and Political Discourse Approaches. Mobilization, 4(2) Koopmans, R. and Statham, P., 2000. Migration and ethnic relations as a field of political contention: an opportunity structure approach. In: R. Koopmans, and P. Statham, eds. 2000. Challenging Immigration and Ethnic Relations Politics: Comparative European Perspectives. Oxford, UK: Oxford University Press, pp.21 82. Kreml, W.P., 1977. The anti-authoritarian personality . Oxford, UK: Pergamon Press Kriesi, H., 1989. New Social Movements and the Social Class in the Netherlands. American Journal of Sociology , 94(5), pp.1078 116. Kriesi, H., 1995. Bewegungen auf der Linken, Bewegungen auf der Rechten: Die Mobilisierung von zwei neuen Typen von sozialen Bewegungen in ihrem politischen Kontext. Swiss Political Science Review , 1(1), pp.9 52. Kriesi, H., 1999. Movements of the left, movements of the right: putting the mobilisation of two new types of social movements into political context. In: H. Kitschelt, P. Lange, G. Marks and J.D. Stephens, eds. 1999. Continuity and Change in Contemporary Capitalism . New York, NY: Cambridge University Press, pp.398 423. Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S., Frey, T., 2005. Globalisation and the transformation of the national political space: six European countries compared . unpublished paper. Kriesi, H., Koopmans, R., Duyvendak, J.W., and Giugni, M.G., 1992. New Social Movements and Political Opportunities in Western Europe. European Journal of Political Research , 22(2), pp.219 44. Kriesi, H., Koopmans, R., Duyvendak, J.W., and Giugni, M.G., 1995. The Politics of New Social Movements in Western Europe: A Comparative Analysis. Minneapolis, MN: University of Minnesota Press. Krosnick, J.A. and Kinder, D.A., 1990. Altering the Foundations of Support for the President Through Priming, American Political Science Review , 84(2), pp.963 75. Kvale, S., 1994. Ten Standard Objections to Qualitative Research Interviews. Journal of Phenomenological Psychology , 25(2), pp.147–73.
190
Kvale, S., 1996. Interviews – An Introduction to Qualitative Research Interviewing . Thousand Oaks, CA: Sage. Kvale, S., 2007. Doing Interviews (Book 2 of the SAGE Qualitative Research Kit). London: Sage. Ladrech, R., 2002. Europeanization and Political Parties towards a Framework for Analysis. Party Politics , 8(4), pp.389 403. Lave, J. and Kvale, S., 1995. What is anthropological research? An interview with Jean lave by Steinar Kvale. Qualitative Studies in Education , 8(3), pp.219 28. Lazarsfeld, P.F. and Merton, R.K., 1948. Mass communication, popular taste and organized social action. In: L. Bryson, ed. 1948. Communication of Ideas . New York, NY: Harper and Row. Lee, C.K. and Strang, D., 2006. The international diffusion of public sector downsizing: network emulation and theory driven learning. International Organization , 60(4), pp.883– 909. Levi Faur, D., 2002. Herding towards a new convention: On herds, shepherds and lost sheep in the liberalisation of the telecommunications and electricity industries. [working paper in politics] No 6. Oxford, UK: Nuffield College. Levi Faur, D., 2005. The global diffusion of regulatory capitalism. Annals of the American Academy of Political and Social Sciences , 598(1), pp.52–66. Levitt, B. and March, J.G., 1988. Organizational Learning. Annual Review of Sociology , 14, pp.319 40. Levy, J.S., 1994. Learning and foreign policy: sweeping a conceptual minefield. International Organization , 48(2), pp.279–312. Lichtman, A.J. and French, V., 1978. Historians and the Living Past . Arlington Heights, IL: Harlan Davidson, Inc. Lijphart, A., 1971. Comparative Politics and the Comparative Method. American Political Science Review , 65(3), pp.682 93. Lijphart, A., 1984. Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries . New Haven, CT: Yale University Press. Lipset, S.M., 1960. Political Man , London: Heinemann. Lipset, S.M., 1979. The first New National: The United States in Historical and Comparative Perspective . New York, NY: Norton. Lipset, S.M. and Raab, E., 1978. The Politics of Unreason. Right Wing Extremism in America 1790-1977 , 2 nd ed. Chicago, IL: University of Chicago Press. Lipset, S.M. and Rokkan, S., 1967. Party Systems and Voter Alignments: Cross-National Perspectives . New York, NY: The Free Press. Lloyd, J., 2003. The Closing of the European Gates? The New Populist Parties of Europe. The Political Quarterly , 74 (issue supplement s1), pp.88 99. Lubbers, M., 2001. Exclusionistic Electorates: Extreme Right-Wing Voting in Western Europe . Ph.D. Radbout University (Nijmegen). Lubbers, M., Gijsberts, M. and Scheepers, P., 2002. Extreme right wing voting in Western Europe. European Journal of Political Research , 41(3), pp.345 78.
191
Lucardie, P., 1998. The Netherlands: The Extremist Center Parties. In: H G. Betz and S. Immerfall, eds. 1998. The new politics of the Right: neo-Populist parties and movements in established democracies . London: Palgrave Macmillan, pp.111–42. Luther, K.L., 2008. Electoral Strategies and Performance of Austrian Right Wing Populism, 1986 2006. In: G. Bischof and F. Plasser, eds. 2008. The Changing Austrian Voter (Contemporary Austrian Studies XVI). New Brunswick, NJ: Transaction Publishers. Luther, K.R., 2009. Governing with right wing populists and managing the consequences: Schüssel and the FPÖ. In: G. Bischof and F. Plasser, eds. 2009. The Schüssel Era in Austria . New Orleans, LA: University of New Orleans Press. Luther, K.R. and Deschouwer, K. eds., 1999. Party elites in divided societies. Political parties in consociational societies . London: Routledge. Maddens, B. and Hajnal, I., 2001. Alternative models of issue voting: The case of the 1991 and 1995 elections in Belgium. European Journal of Political Research , 39(3), pp.319 46. Maguire, D., 1995. Opposition Movements and Opposition Parties: Equal Partners or Dependent Relations in the Struggle for Power and Reform. In: J.C. Jenkins and B. Klandermans, eds. 1995. The Politics of Social Protest: Comparative Perspectives on States and Social Movements. Minneapolis, MN: University of Minnesota Press, pp.199–228. Mair, P., 1993. Myths of electoral change and the survival of traditional parties: The 1992 Stein Rokkan Lecture. European Journal of Political Research , 24(2), pp.121 33. Mair, P., 1997. Party system change: Approaches and interpretations. Oxford, UK: Oxford University Press. Manin, B., 1995. Principes du gouvernement représentatif, Paris: Calman Lévy. March, J.G. and Simon, H.A., 1993. Organizations , Oxford, UK: Blackwell. Marcus, J., 2000. Exorcising Europe's Demons: A Far Right Resurgence?. The Washington Quarterly , 23(4), pp.31 40. Martin, P., 1996. Le vote Le Pen: L’éléctorat du Front National . Paris: Saint Simon. Mayer, N., 2002. Ces français qui votent Le Pen , Paris: Flammarion. Mayer, N., 2005. Radical right populism in France: How much of the 2002 Le Pen votes does populism explain?. In: The Centre for the Study of European Politics and Society, Symposium on globalization and radical right populism . Ben Gurion University of the Neguev. Beer Sheva, Israel 11 12 April 2005. Beer Sheva: Centre for the Study of European Politics and Society. Mayer, N. and Perrineau, P. eds., 1989. Le Front national à découvert . Paris: Presses de la Fondation Nationale des Sciences Politiques. Mayer, N. and Perrineau, P., 1992. Why do they vote Le Pen?. European Journal of Political Research , 22(1), pp.3 41. Mays, N. and Pope, C., 1995. Rigour and qualitative research. British Medical Journal , 311(6997), pp.109 12. McAdam, D., 1982. Political Process and the Development of Black Insurgency. 1930-1970 . Chicago, IL: University of Chicago Press. McAdam, D., 1983. Tactical Innovations and the Pace of Insurgency. American Sociological Review , 48(6), pp.735 54. McAdam, D., 1994. Culture and Social Movements. In: E. Laraña, H. Johnston and J.R. Gusfield, eds. 1994. New Social Movements: From Ideology to Identity. Philadelphia, PA: Temple University Press, pp.36 57.
192
McAdam, D., 1995. Initiator” and “Spinoff ” Movements: Diffusion Processes in Protest Cycles. In: M. Traugott, ed. 1995. Repertoires and Cycles of Collective Action . Durham, NC: Duke University Press, pp.217 40. McAdam, D., 1996. The Framing Function of Movement Tactics: Strategic Dramaturgy in the American Civil Rights Movement. In: D. McAdam, J.D. McCarthy and M.N. Zald, eds. 1996. Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings . New York, NY: Cambridge University Press, pp.338 55. McAdam, D., 1999. Political Process and the Development of Black Insurgency, 1930-1970. 2nd ed. Chicago, IL: University of Chicago Press. McAdam, D., McCarthy, J.D. and Zald, M.N., eds. 1996 . Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings (Cambridge Studies in Comparative Politics) . New York, NY: Cambridge University Press. McAdam, D. and Paulsen, R., 1993. Specifying the Relationship Between Social Ties and Activism. The American Journal of Sociology , 99(3), pp.640 67. McAdam, D. and Rucht, D., 1993. The Cross national Diffusion of Movement Ideas. The Annals of the American Academy of Political and Social Science, 528(1), pp.56–74. McCarthy, J.D., 1996. Constraints and opportunities in adopting, adapting and inventing. In: D. McAdam, J.D. McCarthy and M.N. Zald, eds. 1996. Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings . New York, NY: Cambridge University Press, pp.141 51. McCarthy, J.D. and Zald, M.N., 1977. Resource Mobilization and Social Movements: A Partial Theory. American Journal of Sociology , 82(6), pp.1212–241. McCombs, M.E. and Shaw, D.L., 1972. The Agenda Setting Function of the Mass Media. Public Opinion Quarterly , 36(2), pp.176–85. McDermott, R., 2001. The Psychological Ideas of Amos Tversky and Their Relevance for Political Science. Journal of Theoretical Politics , 13(1), pp.5 33. Meffert, M.F., Norpoth, H. and Ruhil, A.V.S., 2001. Realignment and Macropartisanship. American Political Science Review , 95(4), pp.953 62. Meguid, B.M., 2008. Party competition between unequals: Strategies and electoral fortunes in Western Europe . New York, NY: Cambridge University Press. Melucci, A., 1985. The Symbolic Challenge of Contemporary Movements. Social Research , 52(4), pp.789 816. Melucci, A., 1989. Nomads of the Present . London: Hutchinson Radius. Melucci, A., 1994. A Strange Kind of Newness: What’s “New” in New Social Movements?. In: E. Laraña, H. Johnston and J.R. Gusfield, eds. 1994. New Social Movements: From Ideology to Identity , Philadelphia, PA: Temple University Press, pp.101–30. Melucci, A., 1996. Challenging Codes: Collective Action in the Information Age . Cambridge, UK: Cambridge University Press. Mény, Y. and Surel, Y., eds. 2002. Democracies and the Populist Challenge. New York, NY: Palgrave Macmillan. Merkl, P.H. and Weinberg, L. eds., 1993. Encounters with the Contemporary Radical Right . Boulder, CO: Westview. Meseguer, Y.C., 2003. Learning and economic policy choices: a Bayesian approach . [EUI working paper], RSC No. 2003/5, San Domenico: European University Institute.
193
Meseguer, C., 2005. Policy learning, policy diffusion, and the making of a new order. Annals of the American Academy of Political and Social Science, 598(1), pp.67–82. Meseguer, C., 2006. Learning and Economic Policy Choices. European Journal of Political Economy , 22(1), pp.156–78. Meyer, D.S., 2004. Protest and Political Opportunities. Annual Review of Sociology , 30, pp.125 45. Meyer, D.S. and Minkoff, D.C., 2004. Conceptualising Political Opportunity. Social Forces, 82(4), pp.1457 92. Meyer D.S. and Staggenborg, S., 1996. Movements, counter movements, and the structure of political opportunity. American Journal of Sociology. 101(6), pp.1628– 60. Meyer, D.S. and Whittier, N., 1994. Social Movements Spillover. Social Problems , 41(2), pp.277–98. Meyer, J.W. and Rowan, B., 1977. Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology , 83(2), pp.340 63. Michaelson, M.G., 1993, The development of a scientific speciality as diffusion through social relations: The case of role analysis. Social Networks , 15(3), pp.217 36. Mikhaylov, S., Laver, M. and Benoit, K.R., 2012. Coder Reliability and Misclassification in the Human Coding of Party Manifestos. Political Analysis , 20(1), pp.78 91. Mill, J.S., 1910. Utilitarism, Liberty and Representative Government . London: J.M. Dent. Minkenberg, M., 1998. Context and Consequence: The Impact of the New Radical Right on the Political Process in France and Germany. German Politics and Society , 16, pp.1 23. Minkenberg, M., 2000. The Renewal of the Radical Right: Between Modernity and Anti Modernity. Government and Opposition , 35(2), pp.170 88. Minkenberg, M., 2001. The radical right in public office: Agenda setting and policy effects. West European Politics , 24(4), pp.1 21. Mische, A., 1996. Projecting democracy: the construction of citizenship across youth networks in Brazil. In: C. Tilly, ed. 1996. Citizenship, Identity, and Social History . Cambridge, UK: Cambridge University Press, pp.131 158. Mill, J.S., 2002. A System of Logic . Honolulu, HI: University Press of the Pacific. Miller, J.M. and Krosnick, J.A., 1996. News media impact on the ingredients of presidential evaluations: a program of research on the priming hypothesis. In: D.C. Mutz, P.M. Sniderman and R.A. Brody, eds. 1996. Political Persuasion and Attitude Change, Ann Arbor, MI: University of Michigan Press, pp.79–99. Mishler, E.G., 1986. Research Interviewing: Context and Narrative . Cambridge, MA: Harvard University Press. Moore, B., 1966. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World . Boston, MA: Beacon Press. Mudde, C., 1996. The Paradox of the Anti Party Party. Party Politics , 2(2), pp.265 76. Mudde, C., 1998. The Party Family and Its Study. Annual Review of Political Science , 1, pp.211 29. Mudde, C., 2003. The Ideology of the Extreme Right . Manchester: Manchester University Press. Mudde, C., 2004. The Populist Zeitggeist . Government and Opposition , 39(3), pp.541 63.
194
Mudde, C., 2007. Populist Radical Right Parties in Europe . New York, NY: Cambridge University Press. Mudde, C., 2011. Who’s Afraid of the European Radical Right?. Dissent , Fall 2011, pp.7 11. Mudde, C. and Kaltwasser, C.R. eds., 2012. Populism in Europe and the Americas: Threat or Corrective for Democracy? , New York, NY: Cambridge University Press. Müller, W.C. and Strøm, K., eds. 1999. Policy, office, or votes? Cambridge, UK: Cambridge University Press. Myers, D., 2000. The Diffusion of Collective Violence: Infectiousness, Susceptibility, and Mass Media Networks. American Journal of Sociology , 106(1), pp.173–208. Newell, S., Robertson, M. and Swan, J., 1997. Professional associations as ‘brokers’, facilitating networking and the diffusion of new ideas: Advantages and disadvantages. In: J.L. Alvarez, ed. 1997. The Diffusion and Consumption of Business Knowledge. Basingstoke: MacMillan Press. Nimmo, D.D. and Savage, R.L., 1976. Candidates and their images Concepts, methods and findings . Pacific Palisades, CA: Goodyear Pub. Co. Norris, P., 2000. A virtuous circle: Political communications in postindustrial societies . New York, NY: Cambridge University Press. Norris, P., 2002. Democratic Phoenix: Reinventing Political Activism . New York, NY: Cambridge University Press. Norris, P., 2005. Radical Right: Voters and Parties in the Electoral Market. New York, NY: Cambridge University Press. Office the United Nations High Commissioner for Human Rights, 2007. International Convention on the Elimination of all Forms of Racial Discrimination . [online] Available at:
195
Papadopoulos, Y., 2005. Populism as the other side of consociational multi level democracies. In: D. Caramini and Y. Mény, eds. 2005. Challenges to Consensual Politics: Democracy, Identity and Populist Protest in the Alpine Region . Brussels: PIE, pp. 71 81. Parker, I., 2005. Qualitative Psychology – Introducing Radical Research . Buckingham: Open University Press. Pedersen, M.N., 1979. The dynamics of European party systems: Changing patterns of electoral volatility. European Journal of Political Research , 7(1), pp.1–26. Pedersen, M.N., 1982. Towards a new typology of party lifespans and minor parties. Party Politics , 10(4), pp.367 83. Pemberton, H.E., 1936. The Curve of Culture Diffusion Rate. American Sociological Review , 1(4), pp.547 56. Perrineau, P., 1997. Le symptôme Le Pen. Radiographie des électeurs du Front National. Paris: Fayard. Perrineau, P. ed., 2001. Les Croisés De La Société Fermée: L'Europe Des Extrêmes Droites . Paris: L'Aube Essai. Perrineau, P., 2002. Le vote d’êxtrème droite en France: adhésion ou protestation?. Futuribles , 276, pp.5 20. Peters, B.G. 1997. Policy Transfers Between Governments: The Case of Administrative Reforms, West European Politics , 20(4), pp.71 88. Peters, G., 1998. Comparative Politics: Theory and Methods . London: Macmillan. Plasser, F. and Ulram, P.A., 1995. Wandel der politischen Konflikdynamik: Radikaler Rechtspopulismus in Österreich. In: W. Müller, F. Plasser and P.A. Ulram, eds. 1995. Wählerverhalten und Parteienwettbewerb: Analysen zur Nationalratswahl 1994 . Vienna: Signum, pp.471 503. Platt, J., 1981. Evidence and proof in documentary research. Sociological Review , 29(1), pp.31 52. Plotke, D., 1990. What’s so new about the new social movements?. Socialist Review , 20, pp.81 102. Plummer, K., 1983. Documents of life . London: Allen and Unwin. Poguntke, T. and Webb, P. eds., 2005. The presidentialization of politics. A comparative study of modern democracies. Oxford, UK: Oxford University Press. Pedazhur, A. and Brichta, A., 2002. The institutionalisation of extreme right wing charismatic parties: a paradox?. Party Politics , 8(1), pp.31 49. Polletta, F., 1990. It was like a fever: narrative and identity in social protest. Social Problems , 45(2), pp.137 59. Polletta, F. and Jasper, J.M., 2001. Collective Identity and Social Movements. Annual Review of Sociology , 27, pp.283 305. Prezworski, A., Alvarez, M.E., Cheibub, J.A. and Limongi, F., 2000. Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990 . New York, NY: Cambridge University Press. Probst, L., 2003. Jörg Haider und die FPÖ: Anmerkungen zum Rechtspopulismus in Österreich, In: N. Werz, ed. 2003. Populismus: Populisten in Übersee und Europe . Opladen: Leske + Budricht, pp.113 25. Proctor, R., 1988. Racial Hygiene: Medicine Under the Nazis . Cambridge, MA: Harvard University Press.
196
Putnam, R.D., 2000. Bowling Alone: The Collapse and Revival of American Community. New York, NY: Simon and Schuster. Radaelli, C.M., 2003. The Europeanization of public policy. In: K. Featherstone and C.M. Radaelli, eds. 2003. The Politics of Europeanization . New York, NY: Oxford University Press. Ragin, C., 1987. The Comparative method: Moving Beyond Qualitative and Quantitative Strategies . Berkeley, CA: University of California Press. Ragin, C., 1994. Constructing Social Research: The Unity and Diversity of Method . Thousand Oaks, CA: Pine Forge Press. Ragin, C., 1997. Turning the Tables: How Case Oriented Research Challenges Variable Oriented Research. Comparative Social Research , 16, pp.27 42. Ragin, C., 2000. Fuzzy-Set Social Science . Chicago, IL: University of Chicago Press. Ragin, C. and Zaret, D., 1983. Theory and Method in Comparative Research. Social Forces , 61(3), pp.731 54. Raschke, J., 1985. Soziale Bewegungen: Ein historisch-systematischer Grundriß . New York, NY: Campus. Ray, J.J., 1984. Alternatives to the F scale in the Measurement of Authoritarianism: A Catalog. The Journal of Social Psychology , 122(1), pp.105 19. Richardson, B., 1991. European Party Loyalty Revisited. American Political Science Review, 85, pp.751–75. Robinson, J.P., Shaver, R.P. and Wrightsman, L.S. eds., 1991. Measures of Personality and Social Psychological Attitudes (Volume 1 in Measures of Social Psychological Attitudes Series), Waltham, MA: Academic Press. Rooduijn, M. and Pauwels, T., 2012. Measuring the Thin ideology of Populism in Comparative Research: Two Content Analysis Methods Compared in a European Context . unpublished paper, University of Amsterdam. Rogers, E.M., 1983. Diffusion of innovations. 3rd ed. New York, NY: Free Press. Rogers, E.M., 1995. Diffusion of innovations . 4th ed. New York, NY: Free Press. Rogers, E.M., 2003. Diffusion of innovations . 5th ed. New York, NY: Free Press. Rose, R., 1991. Comparing Forms of Comparative Analysis. Political Studies , 39(3), pp.446 62. Rose, R., 1993. Lesson-Drawing in Public Policy: A Guide to Learning Across Time and Space . Chatham, NJ: Chatham House. Rosenof, T., 2003. Realignment: The Theory that Changed the Way We Think about American Politics. New York, NY: Rowman and Littlefield. Roth, A.F. and Erev, I., 1995. Learning in Extensive From Games: Experimental Data and Simple Dynamic Models in the Intermediate Term. Games and Economic Behavior , 8(1), pp.164 212. Rucht, D., 1988. Themes, Logics, and Arenas of Social Movements: A Structural Approach. In: B. Klandermans, H. Kriesi and S. Tarrow, eds. 1988. From Structure to Action: Social Movement Participation across Cultures . Greenwich, CT: JAI, pp.305–28. Rucht, D., 1992. Studying the Effects of Social Movements: Conceptualization and Problems . Paper presented at the Joint Sessions of the European Consortium for Political Research, Limerick, March 30–April 4. Rucht, D., Koopmans, R. and Neidhardt, F. eds., 1998. Acts of Dissent. New Developments in
197
the Study of Protest . Berlin: Sigma. Ryan, B. and Gross, C., 1943. The Diffusion of Hybrid Seed Corn in Two Iowa Communities. Rural Sociology , 8(1), pp.15 24. Rydgren, J., 2004a. Explaining the emergence of extreme right wing populist parties: the case of Denmark. West European Politics, 27(2), pp. 474 502. Rydgren, J., 2004b. The Populist Challenge: Political Protest and Ethno-Nationalist Mobilization in France . New York, NY: Berghahn. Rydgren, J., 2005a. Is extreme right wing populism contagious? Explaining the emergence of a new party family. European Journal of Political Research , 44(3), pp.413 37. Rydgren, J. ed., 2005b. Movements of Exclusion: Radical Right-Wing Populism in the Western World . New York, NY: Nova Science Publishers. Sabatier, P.A. ed., 2007. Theories of the Policy Process . Boulder, CO: Westview Press. Sainsbury, D., 1980. Swedish Social Democratic Ideology and Electoral Politics 1944-1948. A study of the Functions of Party Ideology , Stockholm: Almqvist and Wicksell. Sartori, G., 1976. Parties and Party Systems , New York, NY: Cambridge University Press. Sartori, G. eds., 1984. Social Science Concepts: A Systematic Analysis . London: Sage Publications. Saunders, K.L. and Abramowitz, A.I., 2004. Ideological Realignment and Active Partisans in the American Electorate. American Politics Research , 32(3), pp.285 309. Schain, M.A., 1997. The Radical Right in Western Europe (Review). Comparative Political Studies , 30(3), pp.375 79. Schelling, T., 1971. Dynamic Models of Segregation. Journal of Mathematical Sociology , 1(2), pp.143 86. Schelling, T.C., 1978. Micromotives and macrobehavior. New York, NY: Norton. Scheuch, E.K. and Klingeman, H.D., 1967. Theorie des Rechtsradikalismus in westlichen Industriegesellschaften. Hamburger Jahrbuch fur Wirtschafts- und Gesellschafspolitik . 12, pp.11 29. Scheufele, D.A., 1999. Framing as a theory of media effects. Journal of Communication , 49(1), pp.103–22. Schmidt, J., 2003. Der Front National und Jean Marie Le Pen. In: N. Werz, ed. 2003. Populismus: Populisten in Übersee und Europa . Opladen: Leske + Budrich, pp.89 111. Schmitter, P., 2001. Participatory governance in a multi level context. In: J.R. Grote and B. Gbikpi, eds. 2001. Participatory Governance: Political and Societal Implications. Opladen: Leske and Budrick. Schwartz, J.M., 1993. Left. In: J. Krieger, ed. 1993. The Oxford Companion to Politics of the World . Oxford, UK: Oxford University Press, pp.531 32. Scott, J., 1990. A Matter of Record: Documentary Sources in Social Research . Cambridge, UK: Polity Press. Segert, D., 2005. Der tschechische Allparteienpopulismus: post sozialistische Instabilitat als Grundlage fur eine populistische Versuchung in Parlament und Regierung. In: S. Frolich Steffen and L. Rensmann, eds. Populisten an der Macht: Populistische Regierungsparteien in West- und Osteuropa . Vienna: Braumuller, pp.191–208. Seiler, D., 1980. Partis et Familles Politiques . Paris: PUF.
198
Semetko, H.A., 1996. The Media. In: L. LeDuc, R.G. Niemi and P. Norris, eds. 1996. Comparing Democracies: Elections and Voting in Global Perspective . Thousand Oaks, CA: Sage. Shields, J., 2007. The Extreme Right in France . London: Routledge. Shipan, C.R. and Volden, C., 2006. Bottom Up Federalism: The Diffusion of Antismoking Policies from U.S. Cities to States. American Journal of Political Science , 50(4), pp.825–43. Shipan, C.R. and Volden, C., 2008. The Mechanisms of Policy Diffusion. American Journal of Political Science , 52(4), pp.840–57. Siegel, S., 1956. Nonparametric Statistics for the Behavioral Sciences . New York, NY: McGraw Hill. Simmons, B., Dobbin, F. and Garrett, G., 2006. Introduction: The International Diffusion of Liberalism. International Organization , 60(4), pp.781 810. Simmons, B., Dobbin, F. and Garrett, G., 2007. The Global Diffusion of Public Policies: Social Construction, Coercion, Competition, or Learning?. Annual Review of Sociology , 33, pp.449–72. Simmons, B., Dobbin, F. and Garrett, G. eds., 2008. The Global Diffusion of Markets and Democracy. New York, NY: Cambridge University Press. Simmons, B. and Elkins, Z., 2004. The Globalization of Liberalization: Policy Diffusion in the International Political Economy. American Political Science Review , 98(1), pp.171 89. Simonson, P., 1999. Mediated Sources of Public Confidence: Lazarsfeld and Merton Revisited. Journal of Communication , 49(2), pp.109–22. Skenderovich, D., 2005. The Radical Right in Switzerland: Postwar Continuity and Recent Transformations: A study of Politics, Ideology and Organisations . Freiburg: Cric Print. Skocpol, T., 1979. States and Social Revolution: A Comparative Analysis of France, Russia and China . New York, NY: Cambridge University Press. Smelser, J.N., 1976. Comparative Methods in the Social Sciences . New Jersey, NJ: Prentice Hall. Snow, D.A. and Benford, R.D., 1988. Ideology, Frame Resonance, and Participant Mobilization. International Social Movement Research , 1, pp.197–217. Snow, D.A. and Benford, R.D., 1992. Master frames and cycles of protest. In: A.D. Morris and C.M. Mueller, eds. 1992. Frontier in Social Movement Theory . New Haven, CT: Yale University Press. Snow, D.A. and Benford, R.D., 1999. Interactive types of cross national diffusion in the social movement arena. In: D. Della Porta, H. Kriesi and D. Rucht, eds. 1999. Social Movements in a Globalizing World . London: Macmillan, pp.23 39. Snow, D.A., Rochford, B.E., Worden, S. and Benford, R., 1986. Frame Alignment Processes, Micromobilization, and Movement Participation. American Sociological Review , 51(4), pp.464–81. Snow, D.A., Soule, S.A. and Kriesi, H., eds. 2004. The Blackwell Companion to Social Movements . Malden, MA: Blackwell Publishing. Snow, D.A., Zurcher, L.A. and Ekland Olson, S., 1980. Social Networks and Social Movements: A Microstructural Approach to Differential Recruitment. American Sociological Review , 45(5), pp.787 801. SOFRES, 1985, Opinion publique , Paris, Gallimard, pp. 178 180 and pp. 182–83.
199
Soule, S.A., 1997. The Student Divestment Movement in the United States and the Shantytown: Diffusion of a Protest Tactic. Social Forces , 75(3), pp.855 83. Soule, S.A., 2004. Diffusion Process Within and Across Movements. In: D.A. Snow, S.H. Soule and H. Kriesi, eds. 2004. The Blackwell Companion to Social Movements . Oxford, UK: Blackwell, pp.294 311. Spencer, P. and Wollman, H., 1998. Good and bad nationalism: a critique of dualism. Journal of Political Ideologies , 3(3), pp.255 74. Spicker, P., 2000. A Third Way?. The European Legacy , 5(2), pp.229 39. Staggenborg, S., 1989. Stability and innovation in the women's movement: A comparison of two movement organizations. Social Problems, 36, pp.75 92. Stedward, G., 1997. On the record: an introduction to interviewing. In: P. Burnham, ed. 1997. Surviving the research process in politics. London: Pinter. Stone, D., 1999. Learning Lessons and Transferring Policy Across Time, Space and Disciplines, Politics, 19(1), pp.51 59. Stöss, R., 1991. Politics against Democracy: Right-Wing Extremism in West-Germany . Oxford, UK: Berg. Strang, D., 1991. Global Patterns of Decolonisation, 1500 1987. International Studies Quarterly , 35(4), pp.429 54. Strang, D. and Chang, P., 1993. The international labor organization and the welfare state: institutional effects on national welfare spending, 1960–1980. International Organization , 47(2), pp.235–62. Strang, D. and Meyer, J.W., 1993. Institutional conditions for diffusion. Theory and Society , 22(4), pp.487–511. Strang, D. and Soule S.A., 1998. Diffusion in organizations and social movements: from hybrid corn to poison pills. Annual Review of Sociology, 24, pp.265 90. Strang, D. and Tuma, N.B., 1993. Spatial and Temporal Heterogeneity Diffusion. American Journal of Sociology , 99(3), pp.614 39. Strange, S., 1996. The Retreat of the State . Cambridge, UK: Cambridge University Press. Strøm, K., 1990. A Behavioral Theory of Competitive Political Parties. American Journal of Political Science , 34(2), pp.565 98. Sully, M.A., 2002. Austria: Jörg Haider uses his power base in Carinthia as a springboard for broader ambitions. Federations , 2(3), pp.1 6. Swank, D. and Betz, H G., 2003. Globalisation, the welfare state and right wing populism in Western Europe. Socio-Economic Review , 1(2), pp.215 45. Swart, W.J., 1995. The League of Nations and the Irish question: master frames, cycles of protest, and 'master frame alignment'. Sociology Quarterly , 36(3), pp.465 81. Swyngedouw, M., 2000. Explaining the relationship between the Vlaams Blok and the city of Antwerp. In: P. Hainsworth, ed. 2000. The Politics of the Extreme Right. From the Margins to the Mainstream . London and New York: Pinter, pp.121 43. Swyngedouw, M., 2001. The subjective cognitive and affective map of extreme right voters: using open ended questions in exit polls. Electoral Studies , 20(2), pp.217 41. Taggart, P., 1995. New Populist Parties in Western Europe. West European Politics , 18(1), pp.34 51.
200
Taggart, P., 2000. Populism . Buckingham: Open Press. Taguieff, P A., 1984. La rhétorique du national populisme. Les règles élémentaires de la propagande xénophobe. Mots , 9(9), pp.113 39. Taguieff, P A., 1990. The new cultural racism in France. Telos , 83, pp.109–22. Tannen, D., 1993. What’s in a Frame? Surface Evidence for Underlying Expectations. In: D. Tannen, ed. 1993. Framing in Discourse . New York, NY: Oxford University Press. Tarrow, S., 1983. Struggling to Reform: Social Movements and Policy Change During Cycles of Protest. Ithaca, NY: Cornell University Press. Tarrow, S., 1991. Struggle, Politics, and Reform: Collective Action, Social Movements, and Cycles of Protest (Western Societies Program, Occasional Paper No.21, 2 nd ed.). Center for International Studies, Cornell University, Ithaca NY. Tarrow, S., 1994. Power in Movements: Social Movements, Collective Action and Politics . New York, NY: Cambridge University Press. Tarrow, S., 1998. Power in movement: Social movements and contentious politics . New York, NY: Cambridge University Press. ten Have, P., 1999. Doing conversation analysis: A practical guide. London: Sage. Thomassen, J., 1976. Party Identification as a Cross National Concept: Its Meaning in the Netherlands. In: I. Budge, I. Crewe and D. Farlie, eds. 1976. Party Identification and Beyond . New York, NY: Wiley, pp.63 80. Tilly, C., 1984. Big Structures, Large Processes, Huge Comparisons . New York, NY: Russell Sage Foundation. Touraine, A., 1981. The Voice and the Eye . Cambridge, UK: Cambridge University Press. Traverso, E. ed., 2001. Le Totalitarisme, Le XXe siècle en débat . Paris: Edition du Seuil. Triandafyllidou, A. and Fotiou, A., 1998. Sustainability and modernity in the European Union: a frame theory approach to policy making. Sociological Research Online , 3(1),
201
Van der Brug, W., Fennema, M. and Tillie, J., 2005. Why some anti immigrant parties fail and others succeed: a two step model of aggregate electoral support. Comparative Political Studies , 38(5), pp.573 73. Van der Brug, W. and Mughan, A., 2007. Charisma, Leader Effects and Support for Right Wing Populist Parties. Party Politics , 13(1), pp.29 51. Van der Heijden, H A., 1997. Political opportunity structure and the institutionalisation of the environmental movement. Environmental Politics , 6(4), pp.25 50. van der Heijden, H A., 2010. Social movements, public spheres and the European politics ofthe environment . Green Power Europe?. Houndmills: Palgrave Macmillan. Van Hauwaert, S., 2012. The psychology of conspiracy theory: The role of cognitive shortcuts, [online] Available at:
202
Weinberg, L., 1993. The American Radical Right: Exit, Voice and Violence. In: P.H. Merkl and L. Weinberg, eds. 1993. Encounters with the Contemporary Radical Right . Boulder, CO: Westview Press. Wejnert, B., 2002. Integrating Models of Diffusion of Innovations: A Conceptual Framework. Annual Review of Sociology , 28, pp.297 326. Wellin, E., 1955. Boiling water in a Peruvian town. In: B.D. Paul, ed. 1955. Health, culture and community . New York, NY: Russell Sage Foundation, pp.78 103. Wendt, C., 2003. Toward a Majoritarian Mobilization Model for Western Europe: Explaining Changes in Far Right Vote Support . Paper presented at the APSA Annual Meeting, August 28 31 2003. Weyland, K., 2005. Theories of Policy Diffusion Lessons from Latin American Pension Reforms. World Politics , 57(2), pp.262 95. Weyland, K., 2007. Bounded rationality and policy diffusion. Princeton, NJ: Princeton University Press. Wievorka, M., 1992. La France raciste . Paris: Seuil. Wieviorka, M., 1998. Le racisme, une introduction. Paris: La Découverte. Wiles, P., 1969. A Syndrome, Not A Doctrine. In: G. Ionescu and E.J. Gellner, eds. 1969. Populism - Its Meanings and National Characteristics, London: Weidenfeld and Nicolson. Wilson, J., 1973. Introduction to Social Movements. New York, NY: Basic Books. Wright, E.O., 1985. Classes . London: Verso. Wright, E.O., 1989. The Debate on Classes . London: Verso. Yow, V.R., 1994. Recording Oral History . Thousand Oaks, CA: Sage Publications. Zald, M.N. and McCarthy, D., 1979. Social Movement Industries: Competition and Cooperation among Movement Organisations . [working paper]. 201. Michigan, MI: University of Michigan. Zaslove, A., 2004. Closing the Door? The Ideology and Impact of Radical Right Populism on Immigration Policy in Austria and Italy. Journal of Political Ideologies , 9(1), pp.99 118. Zimmermann, E., 2003. Right wing extremism and xenophobia in Germany: escalation, exaggeration, or what?. In P.H. Merkl and L. Weinberg, eds. 2003. Right-Wing Extremism in the Twenty-First Century . London: Frank Cass, pp.220 50.
203