Sage Journals: Discover world-class research

Abstract

People-centrism is a key defining characteristic of the dominant ideational approach to populism. Yet, who do populist actors consider their ingroup? Research on the empirical characteristics of who is deemed “the people” is scarce. Bringing together previous work on populism, group references, and social identity theory, this article develops a typology of populist in- and outgroup references. Applying word embeddings, a novel text-as-data approach, the empirical characteristics of these groups are identified and contextualized in election manifestos from Austria, Germany, and Switzerland between 1983 and 2021. In line with the theoretical expectations, the results show that the ingroups’ identity remains imprecise and is primarily constructed indirectly, namely through opposition to and in distinction from the outgroup. Right-wing populist parties can be distinguished from other parties through their negative references to outgroups such as foreigners or Muslims. These references indirectly create the ingroup ex negativo. These findings have implications for the representation of minority groups and political polarization in liberal democracies.

Keywords

group references people-centrism populism word embeddings

Introduction

The demonization of social groups, and particularly the antipathy towards the elite, provides populists with an enemy, but it is also a crucial component of the attempt to construct an identity (Taggart, 2000: 94).

While reference to “the people” is “at the very core, the minimal defining element” of populist political communication (De Vreese et al., 2018: 427), who is considered to be “the people” is not a priori fixed (De Cleen, 2019: 21). Rather, it is an empty vessel (Mudde and Kaltwasser, 2013), which is filled by political actors. Previous research on the empirical characteristics of the populist ingroup is scarce and we still know very little about the (constructed) identity of “the people” (Reinemann et al., 2017). However, it has been shown that even between right-wing populist parties, conceptualizations of the ingroup differ (Heinisch and Werner, 2019). Moreover, Meijers and Van Der Velden (2022) have found that different translations of “the people” into German (“Volk” versus “Bürger”) influence respondents’ agreement to populist attitudes. Theoretically, the populist ingroup is first defined by its homogeneity and unity, which is the precondition for the formulation of a common will (volonté générale). Second, the ingroup is defined through its opposition to other (out-)groups (Stanyer et al., 2017). The consideration of the outgroups is of particular importance when studying the ingroup (Mudde and Kaltwasser, 2013; Taggart, 2000). Building on the three populist core concepts (people-centrism, anti-elitism, and anti-pluralism) as well as social identity theory (Tajfel, 1982), this study distinguishes three types of populist group references: ingroup, vertical outgroup, and horizontal outgroup. These groups differ in their specificity, how they contribute to ingroup identity construction, and the sentiment of the relationship toward them.

To analyze the (constructed) identity of populist in- and outgroups, this study uses a novel text-as-data approach and makes use of word embeddings (Kozlowski et al., 2019), more specifically “à la carte on text (conText) embeddings” (Rodriguez et al., 2023). Word embeddings rest on the logic that the meaning of words can be deduced from their surroundings (or context). Through this semantic contextualization, the method allows for the comparison of the identity of and sentiment toward these groups between parties. As all parties use group references in their political communication, appeals to groups are not unique to populist communication. Studies looking at group references in political communication more generally (Dolinsky, 2022; Horn et al., 2021; Huber, 2022; Thau, 2019) have so far relied on manual coding. This approach is not only time- and resource-intensive, but the coding schemes were open to extension and additional group references had to be added during coding. This need for post-analysis extension of target terms also creates an obstacle for automatized approaches such as dictionaries: only group references present in the dictionaries can be found and possible synonyms would be overlooked. Word embeddings, on the other hand, offer the opportunity to extract information about populist in- and outgroups from the so-called nearest neighbors of group references. This method is applied to election manifestos (n = 153) from 33 political parties in Austria, Germany, and Switzerland between 1983 and 2021.

In line with the theoretical expectations, the results show that the ingroups’ empirical identity remains imprecise and is primarily constructed indirectly, namely through opposition to and in distinction from the outgroups. Right-wing populist parties can be distinguished from other parties through their negative references to horizontal outgroups such as foreigners or Muslims. These references indirectly create the ingroup ex negativo. This is consistent with the theoretical expectation that right-wing populist actors focus on outgroup rather than ingroup definition (Hawkins, 2009), which is why they are also called “exclusionary populist” (Mudde and Kaltwasser, 2013).

Right-wing populist parties can exploit this ingroup imprecision strategically. By not being explicit about who belongs to the ingroup, populist parties can attract a diverse set of voters with potentially divergent interests (Reinemann et al., 2017; Schmuck et al., 2017). However, this also poses a challenge to the norms of modern liberal democracy and fosters polarization. When ingroup identity construction by right-wing populist parties primarily works through the exclusion of certain groups, this does not only concern symbolic exclusion (Mudde and Kaltwasser, 2013) but in fact undermines the “political inclusion” dimension of democracy (Dahl, 1971) and endangers minority rights. It can lead to “fragmentation within the national community and promote inter-group antipathies” (Huber, 2022: 14), which in turn has ramifications for political polarization.

The contribution of this study is threefold. Theoretically, it brings together previous work on populism (Heinisch and Mazzoleni, 2017) as well as group references (Huber, 2022) and social identity theory (Tajfel, 1982) and develops a typology of three types of populist group references. Substantially, it adds to the ever-growing research on populism in political texts with a focus on people-centrism, which has been so far neglected in empirical research (Reinemann et al., 2017). It shows that the right-wing populist ingroup is not only theoretically, but also empirically imprecise and ambiguous. Instead, ingroup identity construction primarily works through distinction from outgroups. Methodologically, it introduces the use of word embeddings for research on populism and group references and presents a workflow that offers opportunities for future research beyond populist in- and outgroups.

Populist In- and Outgroups in Political Communication

The ideational approach to populism, which has become dominant in the European literature on populism (Aslanidis, 2016; Hawkins and Rovira Kaltwasser, 2017; Rooduijn, 2019), defines populism

as an ideology that considers society to be ultimately separated into two homogeneous and antagonistic camps, “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: 543).

The first group mentioned in this definition, “the people,” constitutes the populist ingroup with positive qualities. On the other hand, the second group, “the elite,” has negative traits and represents the outgroup. The two groups not only stand in opposition to each other, but their relationship is antagonistic and morally founded: the outgroup is seen as corrupt insofar as they betray the “general will” of the ingroup. This common sense should be the ultimate basis for all politics and therefore be expressed as directly as possible.

“The people” does not simply refer to all citizens of a country, but rather to “a virtuous and unified population” of an imagined and constructed “heartland” (Taggart, 2000: 95), representing a status quo ante. The ingroup is depicted as homogeneous (Rooduijn, 2018), but as such ultimate homogeneity does not exist in reality, it is a constructed, simplified entity. The identity of the ingroup is not a priori fixed (De Cleen, 2019: 21), but rather an empty vessel (Mudde and Kaltwasser, 2013), which is filled by political actors through reference to groups. The identity of the ingroup is hence initially constructed through articulation (Wilde, 2013; Zicman de Barros, 2021).¹ Consequently, political actors have leeway to use various terms to refer to this group.² For instance, Reinemann et al. (2017: 17–18) distinguish between political, economic, nationalist and cultural conceptions of the ingroup.

The theoretical homogeneity of the ingroup makes it difficult to refer to specific groups that might have diverging and possibly contradictory interests. Instead, broad and rather imprecise terms are expected to be used for reference to the ingroup (Canovan, 1999; Jagers and Walgrave, 2007). Moreover, the “presumed unity of the people also implies that populism cultivates antagonistic relationships towards those who do not fit in and therefore threaten the homogeneity” (Abts and Rummens, 2007: 409). The construction of the ingroup’s identity can hence not only be direct but also indirect “as when political actors define ‘the people’ by virtue of not being member of an out-group” (Stanyer et al., 2017: 360). This allows populist actors to not explicitly state who they stand for (direct identity construction), but to rather stress who they claim not to represent (indirect identity construction) (Hawkins, 2009; Hogg et al., 2004; Mudde and Kaltwasser, 2013; Oliver and Rahn, 2016). However, we still lack an empirical examination of this argument.

To uncover and understand the construction of the ingroup’s identity, it is therefore necessary to not only look for “a positive relationship between those that represent and those being represented” (Heinisch and Werner, 2019: 1), but to also consider these (indirect) references ex negativo, that is, exclusionism of groups that do not belong to the ingroup. It might even be the case that the identity of the ingroup is never revealed explicitly, but can only be deduced from (and in contrast to) the outgroups’ identity. So, to identify the identity of “the people,” one must also consider the identity of its opponents, the “constitutive outside” (De Cleen, 2019: 32).

When studying groups in (populist) political communication, one can build on two concepts of social identity theory (Hogg et al., 2004; Tajfel, 1982; Turner et al., 1979). First, ingroup favoritism describes the tendency to ascribe positive traits to the ingroup and evaluate it more positively than outgroups. Second, and connected, outgroup discrimination concerns the negative evaluation and description of outgroups. Transferring this to (populist) political communication by parties, references to the ingroup are expected to appear in a positive context. References to the outgroup, on the other hand, are expected to appear in a negative context.

Two types of outgroups can be distinguished by the direction (vertical or horizontal) of contrast to the ingroup. Anti-elitism denotes the vertical dimension of populism and describes an antipathy toward all kinds of elites. The elites can be defined abstractly or explicitly, it can be political, economic, cultural or media elites (Heinisch and Mazzoleni, 2017; Jagers and Walgrave, 2007; Reinemann et al., 2017). As such, references to vertical outgroups exist on a spectrum from low to high specificity. Populism can also have a horizontal dimension, which is characterized by anti-pluralism, that is, the rejection of the idea that a society benefits from diversity. As the ingroup is depicted as a homogeneous group with one general will, special minority rights are rejected and “specific population segments” (Jagers and Walgrave, 2007: 324), who are seen as the “dangerous other” (Albertazzi and McDonnell, 2008; Hameleers and De Vreese, 2018), are excluded. The identity of these outgroups is specific and can vary between for instance immigrants, religious, ethnic, or sexual minorities.

While some do not see anti-pluralism as a “core characteristic of populism, but instead specific of right-wing populism in Western Europe” (Blassnig et al., 2019: 110), others disregard the idea to treat it as just host ideology, but rather see it as a “part of populisms’s core” (Hameleers and De Vreese, 2018: 17). Following Mudde (2004, 2013), who identifies two direct opposites of populism, namely elitism and pluralism, I argue that when populism is both people-centrist and anti-elitist, and considers the two groups to be homogeneous, it cannot be pluralist, but is anti-pluralist (Huber et al., 2023; Kriesi and Pappas, 2016). Moreover, from a perspective of social identity theory, both anti-elitism and anti-pluralism underlie the mechanisms of ingroup favoritism as well as outgroup discrimination (Turner et al., 1979) and are therefore implicit integrals of people-centrism (Reinemann et al., 2017; Stanyer et al., 2017). The three core concepts have an antagonistic and reciprocal relationship: the definition of the ingroup (“the people”) depends on the construction of the outgroups (“the elites” and “the others”) and vice versa (Reinemann et al., 2017). Anti-elitism and anti-pluralism are used as markers to delineate the borders of the populist ingroup. Table 1 presents a typology of the three types of populist group references and summarizes their characteristics.

Table 1.

Typology of Populist Group References.

Group	Populist core concept	Ingroup identity construction	Sentiment of context	Specificity
ingroup	people-centrism	direct	positive	low
vertical outgroup	anti-elitism	indirect	negative	low to high
horizontal outgroup	anti-pluralism	indirect	negative	high

Data and Methodology

Case Selection

This study analyzes political communication in 153 election manifestos from 33 parties from Austria, Germany, and Switzerland between 1983 and 2021. Existing studies applying computational text analysis are primarily focused on texts in English (Baden et al., 2021) and multilingual textual analysis poses additional challenges (Licht and Lind, 2023). Therefore, this study focuses on texts in one language, namely German. As Austria, Germany, and Switzerland share the language, this allows for a mono-lingual text analysis.³ The three neighboring countries are established democracies in Western Europe with a multi-party system. However, the history and success of right-wing populist parties differ.

The Austrian FPÖ is the oldest member of the party family and was founded in 1956. The party has been in parliament ever since and until now has been part of five government coalitions. The BZÖ, a splinter party formed by previous FPÖ members, was founded in 2005 and immediately replaced the FPÖ in the government at that time. The BZÖ was represented in the Nationalrat until 2013. In Germany, on the other hand, it was only in 2017 that a right-wing populist party, the AfD (founded in 2013), entered the national parliament and has so far been a marginalized opposition party in the Bundestag. Finally, the Swiss SVP was founded in 1971 and has been the largest party in Switzerland since 1999. With a very short exception of a few months in opposition in 2008, the SVP has always been represented in government due to the particular Swiss political system. All four parties are classified as populist, far-right and Eurosceptic by the Populist (Rooduijn et al., 2023).⁴

Previous research has compared Austria and Germany and found that the AfD and the FPÖ, even though they are both right-wing populist parties, “differ in their understanding of ‘the people’” (Heinisch and Werner, 2019: 14). However, in- and outgroup references can not only differ between right-wing populist parties but between parties in general. Therefore, this study does not only look at political communication by the mentioned right-wing populist parties but also analyzes other (non-populist) parties, which function as the comparison group, to examine the right-wing populist parties in the context of their respective party system. Table A1 in the Online Appendix gives an overview of the parties included in this study and their respective party family (based on Lehmann et al. (2022)).

Analyzing election manifestos has several advantages. First, since parties address potential voters in their election manifestos, one can expect that group references appear in this type of document. Therefore, previous research on group-based appeals (Dolinsky, 2022; Thau, 2019) has also looked at manifestos. Second, manifestos are official documents, which often build the basis for potential coalition negotiations. As such, they “constitute the most authoritative statement of parties’ policy positions prior to elections” (Proksch and Slapin, 2009: 329) and represent the party as a whole. Finally, looking at manifestos allows for a longitudinal perspective. In this study manifestos from 1983 until 2021 are analyzed, since the early 1980s are seen as the starting point for the “third wave” of the populist radical right in Western Europe (Mudde, 2013). As the AfD was only founded in 2013, the manifestos from the latest three federal elections (2013, 2017, and 2021) are studied in the German case.

The election manifestos are retrieved from the Manifesto Corpus (Lehmann et al., 2023). Separate corpora for Austria, Germany, and Switzerland are constructed and the text is tokenized on the word level. Table 2 summarizes the number of documents and tokens (after preprocessing) per country. Details on the selection of documents and preprocessing can be found in the extensive workflow in the Online Appendix B.

Table 2.

Overview of the Three Manifesto Corpora.

Country	N manifestos	N tokens	Time frame
Austria	57	712748	1983 - 2019
Germany	20	780941	2013 - 2021
Switzerland	76	456948	1983 - 2019

Measuring (Populist) Group References in Political Communication

This article uses a novel approach to study populist group references in political texts. It introduces the application of word embeddings, a method which is “able to interpret the contextual meanings of the selected words” (Rooduijn and Pauwels, 2011: 1279), to study populist group references in political communication. Through semantic contextualization, word embeddings offer the opportunity to extract information about populist in- and outgroups from the so-called nearest neighbors of group references. This methodological approach departs from previous research on populist communication as well as group appeals and overcomes their shortcomings.

Previous studies on populist communication have either relied on manual coding or automatized dictionary approaches.⁵ Building on the seminal study by Jagers and Walgrave (2007), numerous studies have developed codebooks and relied on manual coding to study populist political communication (Cranmer, 2011; Manucci and Weber, 2017; March, 2017; Rooduijn et al., 2014; Wirth et al., 2019). However, in none of these studies, the concrete identity of the populist ingroup and their horizontal and vertical counterparts has been studied. Moreover, while classical content analysis produces reliable results, it is both labor- and time-intensive. Following studies tried to automatize the measurement process to be able to study large amounts of text data with dictionaries (Gründl, 2022; Rooduijn and Pauwels, 2011). But these dictionaries do not capture the identity of the populist in- and outgroups either. To summarize: “Most empirical studies . . . do not look for who might actually be meant by “the people” in the context of the specific message” (Reinemann et al., 2017: 17).

Previous studies analyzing group appeals in election manifestos more generally have also relied on manual coding (Dolinsky, 2022; Horn et al., 2021; Huber, 2022; Thau, 2019). Due to the exploratory nature of these studies and to ensure comprehensiveness, the coding schemes were broad and open for extension. Only after the coding was completed, group references were systematized, reclassified, and aggregated into higher-level group categories. This need for post-analysis extension of coding schemes cannot be overcome by dictionaries: only group references present in the dictionaries can be found and possible synonyms would be overlooked, which results in false negatives. To address and overcome these shortcomings, this article introduces word embeddings to the study of populist group references.⁶

Over the last years, word embeddings have become popular in the social sciences (Rheault and Cochrane, 2020; Rodman, 2020). Word embeddings (Mikolov et al., 2013; Pennington et al., 2014) rest on the logic of the “distributional hypothesis” (Rodman, 2020: 91), which states that the meaning of words can be deduced from their surroundings. Word embeddings translate semantic relations between words into relationships (or distances) between vectors in a high-dimensional space, where each word is represented as a vector with unique coordinates. The vectors closest to the target term vector are its context terms and are either words that often appear in similar contexts (e.g. appear in the same sentence) or which are used in a similar meaning. This is because “the optimal distance between two vectors is a function of shared context rather than strict co-occurrence” (Kozlowski et al., 2019: 910). For example, if we imagine “fries” as the target term, both “ketchup” and “chips” would be context terms since fries are often served with ketchup (the two words often appear in the same sentence) and “chips” is the British English equivalent to the American English “fries” (the two words can be used interchangeably in the same sentence). The distance between word vectors can be measured through cosine similarity. The higher this similarity, the closer the terms. Accordingly, the context terms are also called “nearest neighbors”:

A word’s nearest neighbors are often either its synonyms or syntactic variants. A word’s broader neighborhood in the embedding space is typically populated by a host of terms with related meanings. Therefore, a great deal of semantic and cultural information is available simply by examining the word vectors that surround a word of interest (Kozlowski et al., 2019: 911, emphasis added).

This article is interested in the nearest neighbors of populist in- and outgroups references since these nearest neighbors can reveal more about these groups’ identity. Moreover, it is possible to compare nearest neighbors between different “subcorpora” (Rodriguez et al., 2023). Such subcorpora cannot only be manifestos from different countries but also manifestos from different parties. To compare subcorpora, the nearest neighbors cosine similarity ratio is calculated. This ratio captures how discriminant a context term is for a subcorpus. A value larger than 1 means that a context term is more discriminant for the numerator (e.g. the right-wing populist party), while a value smaller than 1 means that the context term is more discriminant for the denominator (e.g. all other parties).

The selection of target terms, that is, words that refer to groups, is based on linguistic and theoretical considerations.⁷ In a first step, I systematically extracted all group references that have been included in codebooks and dictionaries (Gründl, 2022; Rooduijn, 2010; Rooduijn and Pauwels, 2011; Rooduijn et al., 2014). Second, I considered group references that were analyzed in previous qualitative studies (Linares, 2019; Truan, 2019; van Leeuwen, 2019). Afterwards, I went through this list of group references and checked whether these terms first fulfill the theoretical assumptions to be considered as a populist in- or outgroup and second appear in the manifestos analyzed here. This resulted in a list of six ingroup target terms and four outgroup target terms.⁸

The literal German translation of “the people” is Volk, which has a historically negative connotation (Hoffmann, 1991). It is an “ideologically loaded term” (Linares, 2019: 286), still closely connected to right-wing extremism and defined as the “nativist notion” (Wodak, 2022: 29) or “ethnic conception” (Meijers and Van Der Velden, 2022) of the populist ingroup. Therefore, previous research has included German terms that are (allegedly) synonymous to “the people” in order to capture references to the ingroup. These are “humans” (Menschen),⁹ “population” (Bevölkerung) as well as “citizens” (Bürger). Finally, also references to “Austrians” (Österreicher) and “Swiss” (Schweizer) respectively are selected as ingroup target terms.¹⁰

Populist outgroups exist in two directions of opposition to the ingroup. Anti-elitism creates vertical outgroups, while anti-pluralism produces horizontal outgroups (Hameleers and De Vreese, 2018). For the vertical dimension, references to “elites” (Eliten) and “major corporations” (Großkonzerne) are selected as target terms.¹¹ For the horizontal dimension references to “foreigners” (Ausländer) as well as “Muslims” (Muslime) are included. Both groups are seen as “dangerous others,” who threaten the ingroup (Albertazzi and McDonnell, 2008; Benveniste et al., 2016; Wodak, 2015).

The analysis in this article builds on both quantitative and qualitative steps (Hawkins and Rovira Kaltwasser, 2019; March, 2017). After the qualitative selection of target terms, the nearest neighbors are calculated with word embeddings. These nearest neighbors are then interpreted qualitatively. As the next section shows, some of the nearest neighbors are intuitive and understandable at first glance, but for the interpretation of others the “domain knowledge of the researcher” is crucial (Rheault and Cochrane, 2020: 122). As such, the method applied here constitutes an advancement of an “integrated man-machine approach” (Rooduijn and Pauwels, 2011: 1279).

Results

In the first step the nearest neighbors can be compared on the country level, which describes the context in which in- and outgroup references appear in the three countries.¹² The first remarkable observation is that references to people are the least frequent ingroup target term in all three countries, but especially rare in German and Austrian manifestos. References to humans, on the other hand, are most frequent (Supplemental Table B2). Supplemental Table C1 shows that the nearest neighbor with the highest cosine similarity for people is “democracy” in Austria and Switzerland and placed second in Germany. Similarly, “participation” has the highest cosine similarity with humans in Austria and Switzerland, while it is “promote” in Germany. Generally, verbs such as “strengthen,” “enable,” “improve” and “contribute” often appear among the nearest neighbors of ingroup target terms. These verbs describe the actions of the parties for the ingroup and their situation. References to both “Austrians” and “Swiss” have a high cosine similarity with “citizens,” indicating that these terms are used synonymously.

References to outgroups, on the other hand, are on average even less frequent in manifestos from Austria, Germany, and Switzerland (Supplemental Table B2). Moreover, Supplemental Table C2 reveals that the nearest neighbors of elites in Swiss manifestos are adjectives that describe the identity of this outgroup more clearly: it is about “political” and “bourgeois” elites. Concerning major corporations, the nearest neighbors in Austrian manifestos reveal that these outgroup references are used in the context of “taxes,” “tax havens,” “tax avoidance” and “loopholes,” denoting a negative connotation of this vertical outgroup. In German manifestos, references to Muslims have a high cosine similarity with terms such as “Islam,” “minorities,” “religious” and “mosques.” In Austria, on the other hand, verbs such as “fight” and “oppose” point to a more negative context. The nearest neighbors of foreigners display the synonym “immigrant” (“Zuwanderer” in Austria, “Einwanderer” in Germany and Switzerland). In Germany, references to foreigners moreover appear in the context of “criminality” and “crimes.” Terms such as “asylum procedure,” “deportation,” “right of residence,” or “security authorities” also have a high cosine similarity with this horizontal outgroup, pointing toward a negative context.

While the results on the country level give an overall impression of the general context of target terms, turning to the party level allows for an even better insight into the identity of and the sentiment toward these groups. First, comparing nearest neighbors within countries shows differences and similarities between the right-wing populist party and other parties in the respective country. Second, the comparison of the right-wing populist parties across countries reveals differences and similarities within the right-wing populist party family. In the next sections, I discuss the nearest neighbors on the party level and present exemplary quotes.¹³

Ingroups

The results for ingroup target terms first reveal that right-wing populists refer to people in the context of terms describing the political system of democracy (Supplemental Table C3). For instance, “democracy” appears among the nearest neighbors of people in the manifestos of AfD, BZÖ, FPÖ, and SVP. The latter two also mention “freedom” in the context of this group. The FPÖ speaks of “democracy” and “freedom” in the context of citizens (Supplemental Table C5), thereby using references to this ingroup in a similar way as references to the people. The right-wing populist parties discuss the people in the context of constitution and jurisdiction, connecting it with its sovereign role in the democratic political system. As such, when right-wing populists refer to people, they conceptualize the populist ingroup in the sense of “the people as sovereign” (Reinemann et al., 2017: 17). This “political” conception (Reinemann et al., 2017: 17) of the populist ingroup is exemplified by the following quotes:

Consequently, the BZÖ is committed to assuming responsibility for Austria in a government and enabling more co-determination for the people by expanding direct democracy along the lines of the Swiss model (BZÖ, 2008).

Direct democracy without ifs and buts. According to the Federal Constitution, the Austrian people are the sovereign (FPÖ, 2013).

The people as sovereign determine the fate and future of Switzerland in freedom and independence (SVP, 2015).

The constitution may not be amended and no significant international treaty may be concluded without the consent of the people (AfD, 2017).

Second, the ingroup is the target of protection by right-wing populist parties. For example, not only BZÖ and FPÖ, but also the ÖVP connect the population to terms concerning safety, such as “guarantee,” “protection,” “security,” “feeling of security” (Supplemental Table C4). Similarly, the FPÖ “save” and “promote” Austrians (Supplemental Table C7). Among the BZÖ’s nearest neighbors of Austrians, two terms stand out: “reintroduction” and “border controls,” again referring to the security of this ingroup. Interestingly, the BZÖ’s top five nearest neighbors of citizens are connected to Europe (Supplemental Table C5). When looking at the sentences in the BZÖ manifesto of 2006 in which this ingroup is addressed, the Euroskeptic profile of the party becomes manifest:

In addition to the sovereignty of the member states, the citizens of the EU must also be more involved in the opinion-forming process and decision-making through instruments of direct democracy (BZÖ, 2006).

No EU constitution without involving the citizens of Europe (BZÖ, 2006).

Third, for all parties but AfD, BZÖ, FPÖ, and SVP the female plural version of citizens (“Bürgerinnen”) is the term with the highest cosine similarity (Supplemental Table C5). The right-wing populist parties do not use gender-inclusive forms when referring to this ingroup. Fourth, references to humans are not only most frequent, but also the broadest and most inclusive. The semantic context of references to humans differs very little between parties in general (Supplemental Table C6). Referring to humans seems to be the lowest common denominator between right-wing populist parties and other parties. Finally, terms indicating affluence, such as “prosperity” and “wealth,” appear among the nearest neighbors of Austrians, irrespective of party ideology (Supplemental Table C7). Also in Switzerland Swiss are mentioned in the context of financial terms such as “taxes,” “income,” “wealth” and “taxation” by all party families. In general, and in line with social identity theory, references to ingroups mostly appear in a neutral or positive context. Overall, the empirical identity of the ingroup remains imprecise.

Outgroups

References to outgroups are overall numerically less frequent than ingroup references (Supplemental Table B2). Elites are an outgroup that is very rarely addressed in manifestos in Austria and Germany. Supplemental Table C8 shows that the SVP refers to elites in the context of “democracy,” “self-determination” and “sovereignty.” When looking at sentences with such references, the SVP’s anti-elitist opposition to this vertical outgroup becomes apparent:

The sell-out of Swiss sovereignty and self-determination by the political elites must be stopped (SVP, 2015).

If Switzerland is not a member of the EU today, it is not thanks to complacent elites in politics, business, society, and the media, but solely to the special case of direct democracy with the right of the people to have their say (SVP, 2019).

However, the nearest neighbors in the manifestos of SPS and FDP reveal that this group is also discussed in the context of education, shown for instance through terms such as “qualified,” “educate,” “colleges” or “promotion”:

In the form of colleges, where we want to train the elite of tomorrow (FDP, 1995).

The admission requirements are only aimed at an elite and force the less qualified into illegality (SPS, 2003).

The other vertical outgroup, major corporations, is dominantly named in a financial context, indicated by nearest neighbors such as “banks,” “taxation,” “investments,” “earnings” or “profit” (Supplemental Table C9). Moreover, the nearest neighbors reveal adjectives that describe this outgroup more explicitly: “global,” “agricultural” or “international.” The SVP does not refer to major corporations at all. Only the ÖVP, a non-populist party, refers to major corporations in a negative context, indicated by terms such as “tax havens,” “letterbox companies” or “loopholes,” which describe practices of tax fraud by major corporations:

Introduce digital permanent establishment to counteract tax avoidance by major corporations. Prevent profit transfers to letterbox companies in tax havens. More transparency and stronger audits of major international corporations (ÖVP, 2017).

What differentiates the right-wing populist parties in Germany, Austria, and Switzerland from other parties, is how they construct the identity of the horizontal outgroups. References to Muslims are rare in Swiss and Austrian manifestos (Supplemental Table B2). While the FPÖ and SVP do not refer to Muslims at all, the religion “Islam” and the radical ideology “Islamism” appear in their manifestos. While the top five nearest neighbors in the AfD manifestos are different grammatical cases of the word (both as a noun and adjective) as well as to the religion “Islam,” terms such as “radicalization” and even “terror” appear among the top 20:

Anti-constitutional associations [. . .] must be prohibited from building and operating mosques due to the concrete danger that imams will spread teachings there that contribute to the political-religious radicalization of Muslims directed against our legal system (AfD, 2017).

We want to prevent Muslims from becoming radicalized to the point of violent Salafism and terror (AfD, 2017).

Integration does not mean that Germany is adapting to Muslims. Integration means that Muslims adapt to Germany (AfD, 2017).

Whereas the AfD mentions Muslims in a clearly negative context, German and Austrian Green parties as well as the German Left party discuss this group in a more positive context, indicated by terms such as “societal cohesion,” “prevention of violence,” “freedom of religion,” “minorities” and “discrimination” (Supplemental Table C10). This shows that Muslims do not constitute an outgroup for Green parties:

Pushing back the influence of political Islam also means strengthening social cohesion, for example with democratic Muslims (Austrian Greens, 2017).

At the same time, Muslims are particularly affected by structural discrimination and violent attacks (German Greens, 2021).

That is why we defend Jewish people, Muslims and all other religious minorities when they are discriminated against because of their religion (The Left, 2021).

When it comes to foreigners, the divide between right-wing populist parties and other parties is most pronounced (Supplemental Table C11). The Green parties display an explicitly positive connection to this group by talking about integration and naturalization indicated by nearest neighbors such as “integration,” “suffrage,” “naturalization” and even “immigration society”:

Facilitation of the naturalization process, political rights at municipal level, including voting and electoral rights for foreigners, residence status for women regardless of marital status (GPS, 1999).

More democracy for the immigration society. We want foreigners to become nationals with equal rights and obligations as soon as possible. We are committed to a liberal citizenship law that not only enables rapid naturalization, but also birthright citizenship and multiple nationality (German Greens, 2021).

The right-wing populist parties, on the other hand, create a very negative image of foreigners by connecting this group to criminality, repatriation, and deportation. The AfD refers to foreigners in a negative context of “criminality” and “crimes.” Moreover, “unaccompanied” and “minors” are terms that refer to the so-called “refugee crisis” in 2015, when unaccompanied adolescent refugees arrived in Germany. Foreigners are put in the context of “delinquent” and “deported” by the BZÖ, and “repatriation” and “deportation” by the FPÖ:

An aliens police force based on the Swiss model should be set up to deal exclusively with the concerns of foreigners, including the multifaceted crime committed by foreigners (FPÖ, 2006).

Nevertheless, an ever-growing parallel society of poorly educated foreigners and immigrants who are unwilling to integrate is beginning to establish itself (BZÖ, 2006).

Unaccompanied, allegedly underage foreigners (UMA) abuse the law on foreigners and asylum (AfD, 2017).

If the juvenile offenders are foreigners, the expulsion of the entire family should also be considered in the event of a repeat offense (SVP, 2003).

Clear restrictions on family reunification and social benefits should ensure that fewer foreigners who cannot support themselves and only want to live off the state immigrate to Switzerland from the outset (SVP, 2015).

As such, foreigners are the outgroup that clearly separates right-wing populist parties from other parties. This is corroborated by the calculation of the nearest neighbors cosine similarity ratios for references to foreigners, which are plotted in Figures 1 to 3. Here, the ratio is calculated between right-wing populist parties (FPÖ and BZÖ in Austria, AfD in Germany, and SVP in Switzerland) and all other parties in the respective country. A value larger than 1 means that the context term is more discriminant for the right-wing populist party, and a value smaller than 1 means that the context term is more discriminant for all other parties. In Austria (Figure 1), FPÖ/BZÖ address foreigners in the negative context of crime and repatriation. While other parties in Germany (Figure 2) refer to this group in the context of “Germany,” “German,” “possibility,” “obtain” or “eased,” the AfD speaks about foreigners in the same negative way as FPÖ and BZÖ, indicated by terms describing delinquency and deportation. By naming this outgroup in the context of expulsion from their country FPÖ, BZÖ and AfD clearly differentiate themselves from other, non-populist, parties in Austria and Germany. In Switzerland (Figure 3), other parties refer to foreigners in a positive context of “support,” “foster,” “possibilities” and “integration.” The SVP, on the other hand, is not as negative as the other right-wing populist parties, but also puts this outgroup in the context of crime.

Figure 1.

Nearest Neighbors Cosine Similarity Ratio for “Foreigners” in Austrian manifestos. A cosine similarity ratio value larger than 1 means that the context term is more discriminant for the numerator (FPÖ, BZÖ), a value smaller than 1 means that the context term is more discriminant for the denominator (all other parties).

Figure 2.

Nearest Neighbors Cosine Similarity Ratio for “Foreigners” in German manifestos. A cosine similarity ratio value larger than 1 means that the context term is more discriminant for the numerator (AfD), a value smaller than 1 means that the context term is more discriminant for the denominator (all other parties).

Figure 3.

Nearest Neighbors Cosine Similarity Ratio for “Foreigners” in Swiss manifestos. A cosine similarity ratio value larger than 1 means that the context term is more discriminant for the numerator (SVP), a value smaller than 1 means that the context term is more discriminant for the denominator (all other parties).

Discussion and Conclusion

Even though people-centrism is a core feature of populism, research on the empirical characteristics of this populist ingroup is scarce. This article develops a typology of populist in- and outgroup references that brings together previous work on populism, group references, and social identity theory. To answer the question about the constructed identity of these groups, this study makes use of word embeddings. This automatized approach offers the opportunity to extract information about populist in- and outgroups from the so-called nearest neighbors of group references in election manifestos by political parties from Austria, Germany, and Switzerland between 1983 and 2021.

The results show that literal references to people are very scarce, especially in manifestos from Austria and Germany. This is in line with previous research, which has also found a “significant underuse of the lexeme Volk” (Truan, 2019: 208) in parliamentary speeches (see also Dilling and Krawatzek, 2024: 1310). In Austria and Germany, the term “Volk” is still closely connected to the Nazi ideology, which can explain the parties’ avoidance to use this term. In manifestos of AfD, BZÖ, FPÖ, and SVP, the people appear in the context of terms describing the political system of democracy. This aligns with the definition of the populist ideology, in which “politics should be an expression of the volonté générale (general will) of the people” (Mudde, 2004: 543). Moreover, the findings indicate that right-wing populist parties in Austria see the people as a group that needs to be protected, which is in line with earlier findings by Heinisch and Werner (2019). On the other hand, humans is the most used ingroup reference and also the most inclusive and least specific one, allowing for homogeneity. The semantic context of references to humans differs very little between parties. This ingroup seems to be the lowest common denominator between right-wing populist parties and other parties. References to ingroups mostly appear in a neutral or positive context, which is in line with social identity theory (Hogg et al., 2004; Tajfel, 1982; Turner et al., 1979) and the concept of ingroup favoritism. Overall, the empirical identity of the ingroup remains imprecise and is left open to imagination. Right-wing populist actors can exploit this imprecision strategically. By not being explicit about who belongs to the ingroup, they can attract a diverse set of voters with potentially divergent interests (Reinemann et al., 2017; Schmuck et al., 2017).

What differentiates the right-wing populist parties in Germany, Austria, and Switzerland from other parties, is how they construct the identity of horizontal outgroups. While FPÖ and SVP do not refer to Muslims, Islam and Islamism are mentioned by these right-wing populist parties. This can be seen as a strategy of collectivization, as the parties do not refer to the group of Muslim people, but only to the religion and the extremist political ideology. The AfD even mentions Muslims in the context of radicalization and terror. The German and Austrian Green parties, on the other hand, have a more positive image of this group, discussing it in the context of freedom of religion. When it comes to foreigners, the divide between right-wing populist parties and all other parties is most pronounced. While again the Green parties display an explicitly positive connection to this group by talking about integration and naturalization, AfD, BZÖ, FPÖ, and SVP create a very negative image of foreigners by connecting this group to criminality, repatriation, and deportation.

The theoretical expectation was that the identity of the ingroup is primarily constructed in distinction from the outgroup’s identity. Indeed, right-wing populist parties use references ex negativo to create their ingroup: they do not explicitly state who belongs to their ingroup (e.g. humans), but they are clear about who does not (foreigners and the Islam). The constructed identity of the ingroup and even the vertical outgroups, which build on the two populist core concepts people-centrism and anti-elitism, do not differ that much between parties. What differentiates right-wing populist parties from other parties, is their anti-pluralism, which creates (and excludes) horizontal outgroups. As such, the ingroup’s identity is primarily constructed indirectly. This is in line with the theoretical expectation that right-wing populist actors focus on outgroup rather than ingroup definition (Mudde and Kaltwasser, 2013).

The definition of the ingroup does not only concern symbolic inclusion (Mudde and Kaltwasser, 2013), but also the right to political participation (Huber and Schimpf, 2017). The fact that ingroup identity construction by right-wing populist parties primarily works through the exclusion of certain groups undermines the “political inclusion” dimension of democracy (Dahl, 1971). It endangers minority rights and can lead to “fragmentation within the national community and promote inter-group antipathies” (Huber, 2022). This, in turn, has ramifications for political polarization and poses a challenge to liberal democracy.

Certainly, this study has limitations and leaves potential for future research. Here, the focus was on one communication channel, namely election manifestos. It has been shown the articulation of populist ideology through populist political communication is context-dependent and there should be differences in references to populist in- and outgroups between different communication channels, as these diverge in their audiences (Pauwels, 2011) and degree of publicity (Cranmer, 2011). Hence, the next step would be to test if the findings about group references in manifestos travel to other channels of political communication, such as press releases, parliamentary speeches, or social media posts.

Moreover, the populist parties analyzed here are all right-wing populists. Whether the German Left party is left-wing populist is a contested issue and the PopuList for instance classifies it as “borderline populist, borderline far left” (Rooduijn et al., 2023). Therefore, further research on countries with left-wing populist parties is necessary to see whether these parties also primarily construct their ingroup through demarcation and, if so, which groups they depict as outgroups.

As Baden et al. (2021) have noted, computational text analysis is still heavily dominated by and focused on the English language, and one has to be careful when applying certain techniques to corpora in other languages. Studies that want to compare multilingual political texts face additional challenges. Therefore, this study focused on texts in one language, namely German. However, the methodological workflow presented in this article can be adapted and extended to other languages, when corpora in different languages are used for the training of word embeddings.

The analysis has shown that some of the nearest neighbors are intuitive and understandable at first glance, while for others “the interpretation ultimately involves the domain knowledge of the researcher to detect an overarching pattern” (Rheault and Cochrane, 2020: 122). Hence, while computer-based approaches to text analysis offer the opportunity to efficiently study large corpora, human knowledge remains crucial. It has been shown that word embeddings can be used to expand dictionaries and improve their recall (Liang et al., 2023). In a similar vein, they could also be used to extend instructions for manual coding, so that post-analysis expansion is not necessary anymore. This is not restricted to group references, but word embeddings can identify, contextualize, and systematize other concepts of interest, such as ideology or sentiment.

Supplemental Material

sj-docx-1-psx-10.1177_00323217251341784 – Supplemental material for Identity Through Distinction: Contextualizing Populist In- and Outgroup References

Supplemental material, sj-docx-1-psx-10.1177_00323217251341784 for Identity Through Distinction: Contextualizing Populist In- and Outgroup References by Lisa Zehnter in Political Studies

Footnotes

Acknowledgements

For helpful comments on previous versions of this article, I wish to thank Heiko Giebler, Robert Huber, Sophia Hunger, Christoph Ivanusch, Werner Krause, Thomas Meyer, Daniel Saldivia Gonzatti, Teresa Völker as well as the UNDA members, the two anonymous reviewers and participants of the EPSA 2022 Conference and ECPR GC 2022.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author thankfully acknowledges financial support received from the German Research Foundation (DFG) under grant number WE1974/7-4 in the context of the long-term funding of the Manifesto Research Project on Political Representation (MARPOR). The open access publication was funded by the WZB Berlin Social Science Center.

ORCID iD

Lisa Zehnter

Supplemental Material

Additional Supplementary Information may be found with the online version of this article.

Contents

A. Parties

B. Methodological Workflow

Table B1. German Terms for Target Words.

Table B2. Target Terms in Manifestos (Share of All Tokens in Percentages in Brackets).

C. Nearest Neighbors

Table C1. Top 10 Nearest Neighbors of Ingroup Target Terms in Manifestos.

Table C2. Top 10 Nearest Neighbors of Outgroup Target Terms in Manifestos.

Table C3. Top 5 Nearest Neighbors of References to People.

Table C4. Top 5 Nearest Neighbors of References to Population.

Table C5. Top 5 Nearest Neighbors of References to Citizens.

Table C6. Top 5 Nearest Neighbors of References to Humans.

Table C7. Top 5 Nearest Neighbors of References to Austrians/ Swiss.

Table C8. Top 5 Nearest Neighbors of References to Elites.

Table C9. Top 5 Nearest Neighbors of References to Major Corporations.

Table C10. Top 5 Nearest Neighbors of References to Muslims.

Table C11. Top 5 Nearest Neighbors of References to Foreigners.

Notes

Author Biography

Lisa Zehnter is a postdoctoral research fellow in the Manifesto Research Project on Political Representation (MARPOR) at the WZB Berlin Social Science Center. Her research focuses on populism, political communication, and text-as-data.

References

Abts

Rummens

(2007) Populism Versus Democracy. Political Studies 55 (2): 405–424.

AfD (2017) Programm für die Wahl zum Deutschen Bundestag am 24. September 2017. Election manifesto for the 2017 elections.

Albertazzi

McDonnell

(2008) Introduction: The Sceptre and the Spectre. In: Albertazzi

McDonnell

(eds) Twenty-first Century Populism: The Spectre of Western European Democracy. Houndmills; Basingstoke; Hampshire; New York: Palgrave Macmillan, pp.1–11.

Aslanidis

(2016) Is Populism an Ideology? A Refutation and a New Perspective. Political Studies 64: 88–104.

Aslanidis

(2018) Measuring Populist Discourse with Semantic Text Analysis: An Application on Grassroots Populist Mobilization. Quality & Quantity 52 (3): 1241–1263.

Austrian Greens (2017) Das ist Grün. Wahlprogramm der Grünen Nationalratswahl 2017. Election manifesto for the 2017 elections.

Baden

Pipal

Schoonvelde

, et al. (2021) Three Gaps in Computational Text Analysis Methods for Social Sciences: A Research Agenda. Communication Methods and Measures 16 (1): 1–18.

Benveniste

Lazaridis

Puurunen

(2016) Populist Othering and Islamophobia. In: Lazaridis

Campani

(eds) Understanding the Populist Shift: Othering in a Europe in Crisis. London: Routledge, pp.50–69.

Blassnig

Büchel

Ernst

, et al. (2019) Populism and Informal Fallacies: An Analysis of Right-Wing Populist Rhetoric in Election Campaigns. Argumentation 33 (1): 107–136.

10.

BZÖ (2006) Wahlprogramm. Election manifesto for the 2006 elections.

11.

BZÖ (2008) Deinetwegen. Österreich. Das Wahlprogramm des BZÖ. Election manifesto for the 2008 elections.

12.

Canovan

(1999) Trust the People! Populism and the Two Faces of Democracy. Political Studies 47 (1): 2–16.

13.

Cranmer

(2011) Populist Communication and Publicity: An Empirical Study of Contextual Differences in Switzerland. Swiss Political Science Review 17 (3): 286–307.

14.

Dahl

(1971) Polyarchy: Participation and Opposition. Political Science. New Haven, CT: Yale university press.

15.

Dai

Kustov

(2022) When Do Politicians Use Populist Rhetoric? Populism as a Campaign Gamble. Political Communication 39 (3): 383–404.

16.

De Cleen

(2019) The Populist Political Logic and the Analysis of the Discursive Construction of “the People” and “the Elite.” In: Zienkowski

Breeze

(eds) Imagining the Peoples of Europe: Populist Discourses across the Political Spectrum. Amsterdam: John Benjamins Publishing Company, pp.19–42.

17.

De Vreese

Esser

Aalberg

, et al. (2018) Populism as an Expression of Political Communication Content and Style: A New Perspective. The International Journal of Press/Politics 23 (4): 423–438.

18.

Dilling

Krawatzek

(2024) The Populist Radical Right as Memory Entrepreneur? The Prominence, Sentiment, and Interpretations of History in German Parliament. British Journal of Political Science 54 (4): 1296–1317.

19.

Dolinsky

(2022) Parties’ Group Appeals across Time, Countries, and Communication Channels—examining Appeals to Social Groups Via the Parties’ Group Appeals Dataset. Party Politics 29: 1130–1146.

20.

FDP (1995) En avant la Suisse. FDP – damit es aufwärts geht. Election manifesto for the 1995 elections.

21.

FPÖ (2013) Liebe deine Nächsten. Für mich sind das unsere Österreicher. Election manifesto for the 2013 elections.

22.

German Greens (2021) Deutschland. Alles ist drin. Bundestagswahlprogramm 2021. Election manifesto for the 2021 elections.

23.

Gründl

(2022) Populist Ideas on Social Media: A Dictionary-Based Measurement of Populist Communication. New Media & Society 24 (6): 1481–1499.

24.

GPS (1999) Für eine zukunftsfähige Schweiz. Election manifesto for the 1999 elections.

25.

Hameleers

De Vreese

(2018) To Whom Are “the People” Opposed? Conceptualizing and Measuring Citizens’ Populist Attitudes as a Multidimensional Construct. Journal of Elections, Public Opinion and Parties 30 (2): 255–274.

26.

Hawkins

(2009) Is Chávez Populist? Measuring Populist Discourse in Comparative Perspective. Comparative Political Studies 42 (8): 1040–1067.

27.

Hawkins

Rovira Kaltwasser

(2017) What the (Ideational) Study of Populism Can Teach Us, and What It Can’t. Swiss Political Science Review 23 (4): 526–542.

28.

Hawkins

Rovira Kaltwasser

(2019) Introduction. The Ideational Approach. In: Hawkins

Carlin

Littvay

, et al. (eds) The Ideational Approach to Populism: Concept, Theory, and Analysis. Routledge Studies in Extremism and Democracy. London; New York: Routledge, pp.1–24.

29.

Hawkins

Carlin

Littvay

, et al. (eds) (2019) The Ideational Approach to Populism: Concept, Theory, and Analysis. Routledge Studies in Extremism and Democracy. London; New York: Routledge.

30.

Heinisch

Mazzoleni

(2017) Analysing and Explaining Populism: Bringing Frame, Actor and Context Back in. In: Heinisch

Holtz-Bacha

Mazzoleni

(eds) Political Populism. Glashütte: Nomos, pp.105–122.

31.

Heinisch

Werner

(2019) Who Do Populist Radical Right Parties Stand for? Representative Claims, Claim Acceptance and Descriptive Representation in the Austrian FPÖ and German AfD. Representation 55 (4): 475–492.

32.

Hoffmann

(1991) Das, Volk‘: Zur ideologischen Struktur eines unvermeidbaren Begriffs. Zeitschrift für Soziologie 20 (3): 191–208.

33.

Hogg

Abrams

Otten

, et al. (2004) The Social Identity Perspective: Intergroup Relations, Self-Conception, and Small Groups. Small Group Research 35 (3): 246–276.

34.

Horn

Kevins

Jensen

, et al. (2021) Political Parties and Social Groups: New Perspectives and Data on Group and Policy Appeals. Party Politics 27 (5): 983–995.

35.

Huber

(2022) Beyond Policy: The Use of Social Group Appeals in Party Communication. Political Communication 39 (3): 293–310.

36.

Huber

Schimpf

(2017) On the Distinct Effects of Left-Wing and Right-Wing Populism on Democratic Quality. Politics and Governance 5 (4): 146–165.

37.

Huber

Jankowski

Juen

(2023) Populist Parties and the Two-Dimensional Policy Space. European Journal of Political Research 62: 989–1004.

38.

Jagers

Walgrave

(2007) Populism as Political Communication Style: An Empirical Study of Political Parties’ Discourse in Belgium. European Journal of Political Research 46 (3): 319–345.

39.

Kozlowski

Taddy

Evans

(2019) The Geometry of Culture: Analyzing the Meanings of Class through Word Embeddings. American Sociological Review 84 (5): 905–949.

40.

Kriesi

Pappas

(2016) Populism in Europe during Crisis: An Introduction. In: Kriesi

Pappas

Aslanidis

, et al. (eds) European Populism in the Shadow of the Great Recession. ECPR—Studies in European Political Science. Colchester: ECPR Press, pp.1–29.

41.

Laclau

(2005) On Populist Reason. London; New York: Verso.

42.

Lehmann

Franzmann

Burst

, et al. (2023) Manifesto Project Dataset. DOI:10.25522/MANIFESTO.MPDS.2023A.

43.

Lehmann

Jovicic

van Rinsum

(2022) Manifesto Party Family Handbook. https://manifesto-project.wzb.eu/information/documents/further_handbooks

44.

Licht

Lind

(2023) Going Cross-Lingual: A Guide to Multilingual Text Analysis. Computational Communication Research 5: 1–31.

45.

Linares

(2019) The Volk (“People”) and Its Modes of Representation by Alternative Für Deutschland-AfD (“Alternative for Germany”). In: Zienkowski

Breeze

(eds) Imagining the Peoples of Europe: Populist Discourses across the Political Spectrum. Amsterdam: John Benjamins Publishing Company, pp.285–314.

46.

Manucci

Weber

(2017) Why the Big Picture Matters: Political and Media Populism in Western Europe since the 1970s. Swiss Political Science Review 23 (4): 313–334.

47.

March

(2017) Left and Right Populism Compared: The British Case. The British Journal of Politics and International Relations 19 (2): 282–303.

48.

Meijers

Van Der Velden

(2022) The People as “Volk” or “Bürger”? The Implications of Ethnic and Civic Conceptions of the People for the Measurement of Populist Attitudes. European Journal of Political Research 62: 1403–1413.

49.

Mikolov

Sutskever

Chen

, et al. (2013) Distributed Representations of Words and Phrases and Their Compositionality. DOI:10.48550/arXiv.1310.4546.

50.

Mudde

(2004) The Populist Zeitgeist. Government and Opposition 39 (4): 541–563.

51.

Mudde

(2013) Three Decades of Populist Radical Right Parties in Western Europe: So What? European Journal of Political Research 52 (1): 1–19.

52.

Mudde

Kaltwasser

(2013) Exclusionary Vs. Inclusionary Populism: Comparing Contemporary Europe and Latin America. Government and Opposition 48 (2): 147–174.

53.

Oliver

Rahn

(2016) Rise of the Trumpenvolk: Populism in the 2016 Election. The ANNALS of the American Academy of Political and Social Science 667: 189–206.

54.

ÖVP (2017) Der neue Weg. Neue Gerechtigkeit & Verantwortung. Election manifesto for the 2017 election.

55.

Pauwels

(2011) Measuring Populism: A Quantitative Text Analysis of Party Literature in Belgium. Journal of Elections, Public Opinion and Parties 21 (1): 97–119.

56.

Pauwels

(2017) Measuring Populism: A Review of Current Approaches. In: Heinisch

Holtz-Bacha

Mazzoleni

(eds) Political Populism. Baden-Baden: Nomos Verlagsgesellschaft mbH & Co. KG, pp.123–136.

57.

Pennington

Socher

Manning

(2014) GloVe: Global Vectors for Word Representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar, pp.1532–1543. Stroudsburg, PA: Association for Computational Linguistics

58.

Proksch

Slapin

(2009) How to Avoid Pitfalls in Statistical Analysis of Political Texts: The Case of Germany. German Politics 18 (3): 323–344.

59.

Reinemann

Aalberg

Esser

, et al. (2017) Populist Political Communication: Toward a Model of Its Causes, Forms, and Effects. In: Reinemann

Aalberg

Esser

, et al. (eds) Populist Political Communication in Europe. London: Routledge, pp.12–25.

60.

Rheault

Cochrane

(2020) Word Embeddings for the Analysis of Ideological Placement in Parliamentary Corpora. Political Analysis 28 (1): 112–133.

61.

Rodman

(2020) A Timely Intervention: Tracking the Changing Meanings of Political Concepts with Word Vectors. Political Analysis 28 (1): 87–111.

62.

Rodriguez

Spirling

Stewart

(2023) Embedding Regression: Models for Context-Specific Description and Inference. American Political Science Review 117: 1255–1274.

63.

Rooduijn

(2010) Coding Scheme. Populism in Election Programs.

64.

Rooduijn

(2018) What Unites the Voter Bases of Populist Parties? Comparing the Electorates of 15 Populist Parties. European Political Science Review 10 (3): 351–368.

65.

Rooduijn

(2019) State of the Field: How to Study Populism and Adjacent Topics? A Plea for Both More and Less Focus. European Journal of Political Research 58 (1): 362–372.

66.

Rooduijn

Pauwels

(2011) Measuring Populism: Comparing Two Methods of Content Analysis. West European Politics 34 (6): 1272–1283.

67.

Rooduijn

de Lange

van der Brug

(2014) A Populist Zeitgeist? Programmatic Contagion by Populist Parties in Western Europe. Party Politics 20 (4): 563–575.

68.

Rooduijn

Pirro

Halikiopoulou

, et al. (2023) The Populist 3.0: An Overview of Populist, Far-Left and Far-Right Parties in Europe. DOI 10.17605/OSF.IO/2EWKQ.

69.

Schmuck

Matthes

Boomgaarden

(2017) Austria: Candidate-Centered and Anti-immigrant Right-Wing Populism. In: Aaalberg

Esser

Reinemann

, et al. (eds) Populist Political Communication in Europe. London: Routledge, pp.85–98.

70.

SPS (2003) Gleichheit, soziale Sicherheit und Lebensqualität! Election manifesto for the 2003 elections.

71.

Stanyer

Salgado

Strömbäck

(2017) Populist Actors as Communicators or Political Actors as Populist Communicators: Cross-National Findings and Perspectives. In: Reinemann

Aaalberg

Esser

, et al. (eds) Populist Political Communication in Europe. London: Routledge, pp.353–364.

72.

SVP (2003) Wahlplattform 2003 bis 2007. Election manifesto for the 2003 elections.

73.

SVP (2015) SVP – die Partei für die Schweiz. Parteiprogramm 2015 bis 2019. Election manifesto for the 2015 elections.

74.

SVP (2019) Frei und sicher. Ich will’s. Ich wähl’s. Parteiprogramm 2019 bis 2023. Election manifesto for the 2019 elections.

75.

Taggart

(2000) Populism. Concepts in the Social Sciences. Buckingham; Philadelphia, PA: Open University Press.

76.

Tajfel

(1982) Social Psychology of Intergroup Relations. Annual Review of Psychology 33 (1): 1–39.

77.

Thau

(2019) How Political Parties Use Group-Based Appeals: Evidence from Britain 1964–2015. Political Studies 67 (1): 63–82.

78.

The Left (2021) Zeit zu handeln! Für soziale Sicherheit, Frieden und Klimagerechtigkeit. Wahlprogramm zur Bundestagswahl 2021. Election manifesto for the 2021 elections.

79.

Truan

(2019) The Discursive Construction of the People in European Political Discourse. Semantics and Pragmatics of a Contested Concept in German, French, and British Parliamentary Debates. In: Zienkowski

Breeze

(eds) Imagining the Peoples of Europe: Populist Discourses across the Political Spectrum. Amsterdam: John Benjamins Publishing Company, pp.201–228.

80.

Turner

Brown

Tajfel

(1979) Social Comparison and Group Interest in Ingroup Favouritism. European Journal of Social Psychology 9 (2): 187–204.

81.

van Leeuwen

(2019) Measuring People-Centrism in Populist Political Discourse. A Linguistic Approach. In: Zienkowski

Breeze

(eds) Imagining the Peoples of Europe: Populist Discourses across the Political Spectrum. Amsterdam; Philadelphia, PA: John Benjamins Publishing Company, pp.315–339.

82.

Watanabe

(2021) Latent Semantic Scaling: A Semisupervised Text Analysis Technique for New Domains and Languages. Communication Methods and Measures 15 (2): 81–102.

83.

Watanabe

Zhou

(2022) Theory-Driven Analysis of Large Corpora: Semisupervised Topic Classification of the UN Speeches. Social Science Computer Review 40 (2): 346–366.

84.

Wilde

(2013) Representative Claims Analysis: Theory Meets Method. Journal of European Public Policy 20 (2): 278–294.

85.

Wirth

Wettstein

Wirz

, et al. (2019) Codebook. NCCR Democracy Module II: The Appeal of Populist Ideas and Messages. DOI: 10.17605/OSF.IO/RYX42.

86.

Wodak

(2015) The Politics of Fear: What Right-Wing Populist Discourses Mean. London: Sage.

87.

Wodak

(2022) Far-Right Populist Discourses. In: Manucci

(ed.) The Populism Interviews. London: Routledge, pp.29–33.

88.

Zicman

Barros

(2021) ‘Not All Claims Are Representative Claims’: Constructing “The People” in Post-representative Movements. Representation 57 (4): 515–530.

89.

Zollinger

(2024) Cleavage Identities in Voters’ Own Words: Harnessing Open-Ended Survey Responses. American Journal of Political Science 68 (1): 139–159.

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.07 MB