Overcoming a conundrum in the realignment debate
The political sociology of Western European cleavage structures today faces a conundrum, with research showing both the continued groundedness of 21st-century electoral politics in social structure and a dissolution of the thick organizational and associational understructure that characterized older periods of cleavage politics. Mounting evidence indicates that despite a decline in classic patterns of religious and class voting, political allegiances continue to be structured by socio-structural characteristics, notably those of occupational class, education, or region (Ford and Jennings Reference Ford and Jennings2020; Lindh and McCall Reference Lindh and McCall2020). For a crucial new divide that has been reorganizing party systems in the last decades – that between the nativist far right and a liberal or Green new left – very similar patterns of structural alignment have occurred in most Western countries (Kriesi et al. Reference Kriesi, Grande, Lachat, Dolezal, Bornschier and Frey2008; Oesch and Rennwald Reference Oesch and Rennwald2018; Häusermann and Kriesi Reference Häusermann, Kriesi, Beramendi, Häusermann, Kitschelt and Kriesi2015): Far-right parties have been disproportionately successful in mobilizing lower-educated groups, workers, older males, and rural residents, who see themselves as disadvantaged by the transformation from industrial to globalized “knowledge societies” (Iversen and Soskice Reference Iversen and Soskice2019; Hall Reference Hall2022). Meanwhile, the new left has developed a core base among those who tend to benefit from these changes, such as female, more highly educated urban professionals (Gingrich and Häusermann Reference Gingrich and Häusermann2015; Marks et al. Reference Marks, Attewell, Hooghe, Rovny and Steenbergen2022; Abou-Chadi and Hix Reference Abou-Chadi and Hix2021). At the same time, amidst increased volatility, declining turnout rates in many countries, and an unmooring of ‘machine politics’ from civil society (Dalton and Flanagan Reference Dalton and Flanagan2017, Schäfer and Schwander 2019, Dassonneville Reference Dassonneville2022), a top-down logic of political articulation appears to prevail over the mere channelling of pre-existing structural antagonisms (De Vries and Hobolt Reference De Vries and Hobolt2020; De Leon, Desai, and Tuğal Reference De Leon, Desai and Tuğal2020). Evidence for a loosening of socio-political allegiances, as emphasized by research on party strategy, thus stands side by side with evidence for a continued rootedness of political oppositions in social structure, as highlighted by structuralist approaches.
At the frontier of addressing this puzzle is work concerned with groups and group identities, on which both structuralist accounts and research focused on party strategy seem to be (re)converging, and of which mainstream political science had long lost sight (Dodd, Lamont, and Savage Reference Dodd, Lamont and Savage2017). On the more structuralist side, visions of “dealigned” and individualized political behavior have been countered by a series of studies that update classical cleavage theory for the 21st century, tracing not only contemporary links between socio-structural groups and parties but also showing that these emerging electorates are developing distinct group identities. In other words, these studies indicate that a modern “cleavage” articulated by new left and far-right parties encompasses a social, a political, and a group identity element. While many voters may certainly disengage and lack strong political loyalties, important segments of electorates today are evidently politically anchored in such a new cleavage (Stubager Reference Stubager2009; Bornschier et al. Reference Bornschier, Haffert, Steenbergen and Zollinger2021; Zollinger Reference Zollinger2022; Steiner, Mader, and Schoen Reference Steiner, Mader and Schoen2023; Sczepanski Reference Sczepanski2023). In another strand of research, even work that insists on the importance of political actors’ short-term strategies in a fragmented electoral landscape extends the study of issue competition and entrepreneurship to include research on “group appeals,” thus explicitly incorporating group identity politics (Huber Reference Huber2022; Dolinsky Reference Dolinsky2022; Robinson et al. Reference Robinson, Stubager, Mads and Tilley2017; Thau Reference Thau2019; Elder and O’Brian Reference Elder and O’Brian2022, Zuber et al. Reference Zuber, Howe and Szöcsik2023).
In this article, we argue that newer takes on cleavage theory which center on groups and group identities have the potential to reconcile these seemingly contradictory perspectives: of a politics (re)structured along group lines, on the one hand, and of dissolving formal associational ties, on the other. However, this requires a more sociological understanding of groups and group identities than that which characterizes much current work on political behavior and party strategy. We propose that political scientists who study group-based politics can more effectively speak to each other’s work and build a cumulative research agenda if we (1) study groups in the “thicker” sociological sense of groups as collectives with shared forms of identification or consciousness and if we (2) explicitly make the mechanisms and processes through which groups are formed an object of empirical inquiry. Recent political science too often adopts a minimalist understanding of groups, for instance as objectively and socio-demographically defined. It tends to treat group identities as static or given – there for parties to appeal to – rather than as continually contested, constructed, and articulated into being. In the following, we argue for a more sociological perspective that allows us to ask: Which processes drive the formation of new cleavage-related group identities in an environment without dense organizational linkages between civil society and party politics? What mediates between parties articulating new lines of conflict and their socio-structural bases in a historical moment of fragmentation and class demobilization? Which aspects of parties’ strategic use of group appeals reflect a realignment of core constituencies in 21st-century politics (and which do not)?
To conceptualize and study group identities in this way, we propose to draw on a prominent sociological tradition that contemporary cleavage theory has only just started to incorporate, namely Bourdieusian research on distinction and group-making (Swartz Reference Swartz2006; Bonikowski Reference Bonikowski2015; but see Damhuis Reference Damhuis2020). This perspective emphasizes that groups do not simply exist but are rather made and continually maintained through (bottom-up) social closure and (top-down) classification struggles. This makes it compatible with cleavage theory, which proposes that socioeconomic change creates tensions, grievances, new interests and value patterns, and with them potentials for new forms of group consciousness (bottom-up) which political actors must then mobilize (top-down) (Bornschier Reference Bornschier2010).
As inherent in this understanding, cleavage research investigates the triad of social structure, group identities, and politics (see Figure 1 below and Bartolini and Mair Reference Bartolini and Mair1990). The perspective we develop spells out more concretely how–in this triad–group identities are linked with structure and politics and how we can study the links between these elements. We identify two key processes: closure and classification struggles. Closure refers to the link between social structure and group identities: here we look at patterns of everyday interaction, association, and demarcation through which new collective identities emerge ‘bottom-up’ from structural inequalities. Classification struggles denote the contentious processes through which political actors shape, articulate, and cohere identity potentials ‘top-down’ (e.g., by appealing to specific class segments, excluding sexual or ethnic minorities from the category of “ordinary people” or defining the interests of a particular group as the “national interest”). While the first term captures how cleavage identities are elaborated in the everyday practices of ordinary citizens, the second looks at the way cleavage identities are “made” by political actors.
Figure 1. Model of Group Identity Processes in Cleavage Formation (extended based on Bornschier et al. Reference Bornschier, Haffert, Steenbergen and Zollinger2021).
The contribution of this article is threefold: We review the state of existing research on contemporary cleavage and identity formation, provide a new overarching theoretical framework that incorporates sociological insights, and discuss avenues for empirical investigation that align with this framework. We also demonstrate how our approach can bring cleavage research into conversation with newer work on party strategy and political entrepreneurship. We thus add to a number of recent studies that update older theories of group politics for an age of fundamentally transformed electoral politics (Dodd, Lamont, and Savage Reference Dodd, Lamont and Savage2017; Gidron and Hall Reference Gidron and Hall2017; Elder and O’Brian Reference Elder and O’Brian2022).
The current state of cleavage research on group identities
Group identity formation has been theorized as a central component of the cleavage concept since its inceptionFootnote 1 (Bartolini and Mair Reference Bartolini and Mair1990). Identities give cleavages their long-term structuring power by linking structural and political divides, anchoring individuals socially and politically (Bornschier Reference Bornschier2009). The formation of working-class identities in the industrial era, to cite a paradigmatic example, arose from common structural dislocations and shared social experiences of workers across capitalist countries. But their linkage to the political left depended on the capacity of parties to articulate and organize identities that centered on the class conflict (Bartolini Reference Bartolini2000, Reference Bartolini2005; Eidlin Reference Eidlin2016). The question is whether and how–in group politics of the current stage of advanced capitalism–similar processes of articulation are at play.
The antagonism of far-right and new left parties, described as an emerging new cleavage above, is the central testing ground for these questions: Realignment research has established that green/left-libertarian parties today are increasingly anchored in the educated middle classes who tend to have relatively more progressive attitudes on migration, societal liberalization, and welfare deservingness–an attitudinal pattern described as universalism (Häusermann and Kriesi Reference Häusermann, Kriesi, Beramendi, Häusermann, Kitschelt and Kriesi2015; Oesch and Rennwald Reference Oesch and Rennwald2018). Far-right parties draw disproportionately on the support of workers and small business owners whose worldviews lean relatively more to the particularist opposite pole, marked by more skeptical views of immigration, minority recognition, and the deservingness of welfare recipients. Yet, whether “universalists” and “particularists” are forming as social groups with an internal sense of identification and antagonistic group belonging is a question that has only very recently started to receive attention, despite being so central to the diagnosis of a new cleavage.
Often, the question of group identities is elided in the form of two reductionisms. On the one hand, a fallacy that may be called structural reductionism too easily takes socio-structural correlates of political behavior for group identities proper. The statistical overrepresentation of production workers and sociocultural professionals in parties’ electorates, for instance, does not automatically warrant ascribing a universalist group identity to one group, or a particularist one to the other, as long as this identity cannot be shown to be salient for members of these classes themselves, in their everyday identification processes. Political reductionism, on the other hand, infers the existence of collective identities from political partisanship, for instance by deducing the presence of cleavage identities from lowered electoral volatility across voter blocks (Bartolini and Mair Reference Bartolini and Mair1990). Such inter-block stability or differences in relative rates of party identification alone cannot be interpreted as evidence of cleavage identities, but at most as a starting point for a deeper investigation of group formation. Similarly, research on cleavage identities needs to go further than studies of partisan identities, issue identities, and affective polarization (Achen and Bartels Reference Achen and Bartels2016; Mason Reference Mason2018; Gidron, Adams, and Horne Reference Gidron, Adams and Horne2020; Hobolt, Leeper, and Tilley Reference Hobolt, Leeper and Tilley2021; Helbling and Jungkunz Reference Helbling and Jungkunz2020; Wagner Reference Wagner2021; Reiljan Reference Reiljan2020). Interesting in their own right, these studies attest to cleavage identities at best partially, capturing identitarian semantics of political conflict, but not the penetration of political principles into the wider formation of social groups.
Newer cleavage research has started to move past these narrower understandings of groups and identities, initially by studying whether well-known structural correlates of far-right and green/left-libertarian voting translate directly into self-perceptions (see pioneering work by Stubager Reference Stubager2009 on educational identities), and more recently by investigating a broader range of culturally connoted group categories through which structural divides might become indirectly articulated (Bornschier et al. Reference Bornschier, Haffert, Steenbergen and Zollinger2021, Damhuis and Westheuser Reference Damhuis and Westheuser2024, Zollinger Reference Zollinger2022, Reference Zollinger2024). This work (so far focused on a small number of countries) indicates that core far right and new left electorates indeed tend to antagonistically demarcate themselves from each other in rather culturalist terms, e.g., with a “down-to-earth” nativist national or rural identity contrasting with the “open-mindedness” of self-proclaimed cosmopolitans and feminists.
While this work indicates that modern cleavage identities exist, it provides scarcely any evidence on how these identities, which evidently consistently correlate with socio-demographic characteristics, emerge from the transformed socio-structural conditions of post-industrial societies. What is the contemporary equivalent, for instance, of social interactions on the shop floor, at social democratic sports clubs, or in church choirs? And why do culturally connoted group boundaries resonate particularly well with voters’ everyday self-understandings, rather than, say, the educational or occupational categories that “objectively” characterize groups? Where do these notions of group belonging come from and how are they stabilized?
Part of the answer is likely that specific group boundaries are supplied by political actors, but here too, existing work in the cleavage tradition has provided very little direct evidence on how political actors succeed in activating and maintaining a specific sense of group belonging among their core constituencies. If this no longer works through the partisan press, unions’ communication with members, or weekly preaching from the pulpit, how then? While there is a promising body of work on the strategic use of group appeals from a perspective of political entrepreneurship (e.g., Huber Reference Huber2022; Dolinsky Reference Dolinsky2022), this has so far remained largely disconnected from cleavage studies. Integrating these bodies of research is hindered partly by the diverging temporal perspectives adopted (ranging from electoral cycles in the party strategies literature to epochal structural shifts in the cleavage literature). Further, the expanding group appeals literature so far tends to conceptualize groups in what we have here described as a structurally reductionist perspective, staying close to socio-demographic groups and ascriptive characteristics.Footnote 2 It also often treats groups as given, rather than constructed by political actors through various channels. This narrower treatment of groups may originate partly from methodological obstacles (e.g., the need to delimit the concept of group appeals for quantitative text analysis of manifestos or campaign material), but these are not unsurmountable, given ongoing advances in the measurement of group appeals (see below). What is needed then, is a more encompassing understanding of group identities and consciousness.
The Bourdieusian contribution
To theorize how group identities and group formation are positioned between structural change and political mobilization, political studies can draw on a long tradition of sociological research. A classical source of this tradition was Marx’s writings about the way in which shared economic experiences but also political struggles, representation, and ideological self-education help classes “in themselves” gain self-awareness and collective agency, thus turning them into classes “for themselves” (see e.g. Marx Reference Marx1968).Footnote 3 This conceptual metaphor inspired a long thread of research on class formation, class consciousness, and top-down political articulation (Katznelson and Zolberg Reference Katznelson and Zolberg1986; Gramsci Reference Gramsci1971; Chibber Reference Chibber2022; Thompson Reference Thompson1968). On the other hand, sociological studies of group formation have drawn on Max Weber’s theorizing about “status groups” defined by common “lifestyles”;Footnote 4 as well as his observations on the emergence of ethnic groups from processes of “social closure” (Weber Reference Weber1978, 302ff.; 385ff.; Murphy Reference Murphy1986; Parkin Reference Parkin1979). This gave birth to a fertile tradition of research on symbolic boundary-making and the formation of collectives like ethnicities and nations as “imagined communities” (Anderson Reference Anderson2006). As this research points out, feelings of “groupness” and belonging together are the outcome both of everyday practices of ethnic sociability and of cultural homogenization processes manufactured top-down by nation-building elites (Wimmer Reference Wimmer2013; Brubaker Reference Brubaker2004; Barth Reference Barth1969). Both the Marxian and the Weberian traditions, in other words, concur in looking at groups as making themselves and being made from the potentials provided by objective material conditions.
In our own contribution, we mainly draw on the synthesis of these older lines of inquiry developed in the work of Pierre Bourdieu. While many aspects of this perspective have also been articulated in other contexts, the Bourdieusian theoretical paradigm offers a particularly fruitful entry point for a dialog between political science and sociology, not only due to its immense influence in contemporary sociologyFootnote 5 but also because it offers a systematic conceptual apparatus capturing both bottom-up and top-down processes of group-making.Footnote 6 Picking up from Marx, the Bourdieusian perspective shifts the focus away from “groups on paper,” such as classes constructed by researchers, and towards the social and political processes by which groups are made and unmade.Footnote 7 “Groups are not found ready-made in reality,” waiting to be appealed to and mobilized; instead “they are always the product of a complex historical work of construction” (Bourdieu Reference Bourdieu1987, 9). Two processes are at the core of this work of construction:
First, groups form through everyday processes of boundary-making in which ingroups are demarcated against specific others (Lamont and Molnár Reference Lamont and Molnár2002; Abbott Reference Abbott1995; Jenkins Reference Jenkins1996). Expanding on Weber’s notion of closure, the Bourdieusian perspective centers on the ways in which lifestyle orientations, cultural habits, and a sense of social proximity and distance function as markers of position in social hierarchies (Bourdieu Reference Bourdieu1991). Categories like those of being “respectable,” “hard-working” or “ordinary” are not only logically defined by their “dishonest”, “lazy,” or “aloof” opposites, but also derive their meaning from a social demarcation against others associated with these categories (Skeggs Reference Skeggs1997; Lamont Reference Lamont2000; Rehbein et al. Reference Rehbein2015). Boundary-making describes an important way in which latent structural inequalities become salient in everyday life through an idiom of demarcation, identification, and categorization that defines groups both by what they are and by what they are not. This process rests on a bedrock of objective material relations, but the set of categories by which practices of boundary-making structure networks – the symbolic code of closure – is empirically heterogeneous and cannot simply be read off structural positions.Footnote 8
Second, groups are made in what Bourdieu (Reference Bourdieu2018) calls “classification struggles.” These are struggles over the power to name, define, and represent groups – e.g., in the form of collective names, symbols, metaphors, and cultural signs. “The social world,” Bourdieu (Reference Bourdieu2000, 187) writes, is the product of
political symbolic struggles over […] recognition, in which each pursues not only the imposition of an advantageous representation of himself or herself, […] but also the power to impose as legitimate the principles of construction of social reality most favourable to his or her social being.
Group formation involves struggles over the legitimacy of “principles of vision and division,” that is, the question of which categorical distinctions can claim public legitimacy and become part of the common sense view of the world (Bourdieu Reference Bourdieu1989; Swartz Reference Swartz2013). Returning to the example of the class cleavage, clerics and Christian Democrats sought to impose a group distinction of believers and nonbelievers in interaction with and against the distinction between workers and owners pushed by trade unions and Socialists. Classification struggles are shaped by claims to group identity which are made by political or cultural entrepreneurs, such as politicians and interest group representatives, journalists, pundits, intellectuals, or artists, when speaking about a group or in its name. Such agents
struggle to impose representations […] which create the very things represented, which make them exist publicly, officially. Their goal is to turn their own vision of the social world, and the principles of division upon which it is based, into the official vision. […] A ‘class,’ be it social, sexual, ethnic, or otherwise, exists when there are agents capable of imposing themselves, as authorized to speak and to act officially in its place and in its name, upon those who, by recognizing themselves in these plenipotentiaries […] recognize themselves as members of the class, and in doing so, confer upon it the only form of existence a group can possess. (Bourdieu Reference Bourdieu1987, 13ff.)
In sum, we can identify two complementary and interlocking processes by which group identities are made: the bottom-up closure of social circles in everyday life, and top-down public classification struggles over the boundaries of groups. For research on cleavage identity formation, this clarifies a number of central sites, units, and mechanisms we ought to look at: Wherever antagonistic groups of “cosmopolitans” and “communitarians,” or “universalists” and “particularists,” are forming, we should be able to trace the salience of new cleavage categories a) in the codes of everyday processes of boundary-making; and b) in public interventions of political entrepreneurs struggling over the identities and boundaries of groups. As illustrated in Figure 1, these two processes link the group identity dimension to the two other elements of the cleavage triad (Bartolini and Mair Reference Bartolini and Mair1990): Closure describes how cleavage identities emerge from changes in the structure of social relations, while classification struggles stand for the way cleavage identities are molded by political contestation.
This directly picks up the ball from previous research on cleavage formation, itself often described as a form of social closure (Bartolini and Mair Reference Bartolini and Mair1990). Cleavage theory has long emphasized how the homogeneity of social networks and structural obstacles to social mobility foster the emergence of ‘identity potentials’ (Lipset and Rokkan Reference Lipset and Rokkan1967; Bartolini Reference Bartolini2000; Bornschier Reference Bornschier2010). Our perspective adds to this a focus on boundary-making as a way to grasp the ‘symbolic code’ (c.f. Bartolini Reference Bartolini2005) active in bottom-up processes of group identity formation. The notion of classification struggles, on the other hand, conceptualizes the ‘entrepreneurial’ or ‘top-down’ element of identity conflicts and cleavage formation (De Vries and Hobolt Reference De Vries and Hobolt2020, Evans Reference Evans2010) as an inherently dynamic and contested process.
In the following, we demonstrate the usefulness of our framework for empirical research on cleavage formation. We mainly show how political science research can benefit from an interdisciplinary dialog with sociological scholarship on group-making; but the reverse is also true: Bourdieu’s theorizing about politics, for instance, mostly remained on the level of sketches, and he never truly translated his theoretical observations about classification struggles into a systematic empirical framework for studying political entrepreneurs and their interactions. Nonetheless, as we review below, a host of studies have made important empirical findings shedding light on the formation of new cleavage identities in processes of closure (section 4) and classification struggles (section 5). For both processes, we theorize relevant actors, resources, and symbolic codes. We also outline research agendas that could help fill in what is currently missing.
Closure
Our conceptual focus on closure resonates with research on electoral behavior showing that the key micro-level mechanism behind long-term partisan attachments is repeated interaction with a politically homogeneous environment (Berelson, Lazarsfeld, and McPhee Reference Berelson, Lazarsfeld and McPhee1954). Historical cleavages – such as the class or religious cleavages – were stabilized not just by political parties, but also by the identities conferred by the networks and memberships of churches, unions, social clubs, children’s camps, conventions, newspapers, and so on, which brought groups to life and delimited social (e.g. business, personal, and romantic) interactions “from cradle to grave” (Bartolini Reference Bartolini2000). While there is ample research on how these older forms of social closure have lost their structuring power (Dalton et al. Reference Dalton, Flanagan and Allen Beck1984; De Vries and Hobolt Reference De Vries and Hobolt2020; Dassonneville Reference Dassonneville2022), there is much less research on whether, where, and how new forms of social closure are forming group identities relevant for contemporary cleavage politics. In the following, we propose a research agenda that could answer these questions by locating processes of closure beyond the level of formal organizations, across sites, and in the context of unequal social positions.
The code of closure
As noted, the decline of mass organizations poses a challenge to the study of cleavage identities because it makes it much more difficult to discern linkages between politics and the pre-political sphere. In fact, as we outlined at the outset of this article, robust evidence of structured political behavior today (Häusermann and Kriesi Reference Häusermann, Kriesi, Beramendi, Häusermann, Kitschelt and Kriesi2015, Oesch and Rennwald Reference Oesch and Rennwald2018, Marks et al. Reference Marks, Attewell, Hooghe, Rovny and Steenbergen2022, etc.) in the absence of large-scale formal social organizations is puzzling from the perspective of cleavage theory. The puzzle concerns the mechanisms and processes through which durable links between voters and parties form today, and it has implications for whether we can expect contemporary electoral alignments to be as stable as those of the late 20th century.
The focus on closure as boundary-making is helpful for addressing this puzzle because it directs the attention to practices of distinction situated on the level of lifestyles, habits, and everyday culture, that is, beyond the level of formal organizations. A field in which this kind of research is far advanced but has remained largely isolated from cleavage research is the sociology of cosmopolitan identities and cultural consumption practices. Prieur and Savage (Reference Prieur and Savage2013; Reference Prieur, Savage, Coulangeon and Duval2014), Savage et al. (Reference Savage, Hanquinet, Cunningham and Hjellbrekke2018), and others have noted a reconfiguration of divisions by cultural capital which relativizes the traditional differentiation of high- and low-brow tastes and instead increases the salience of an opposition between forms of cultural capital that do and do not valorize cultural openness (Jarness Reference Jarness2015; Bennett et al. Reference Bennett, Savage, Silva, Warde, Gayo-Cal and Wright2009; Lizardo and Skiles Reference Lizardo and Skiles2016). Transnational engagements confer a form of “cosmopolitan cultural capital” that serves as a status marker in the distinction practices of educated professional class fractions in urban metropoleis (Coulangeon Reference Coulangeon2017; Igarashi and Saito Reference Igarashi and Saito2014; Meuleman and Savage Reference Meuleman and Savage2013; Flemmen, Jarness, and Rosenlund Reference Flemmen, Jarness and Rosenlund2018; Helbling and Teney Reference Helbling and Teney2015).
Studies have found some evidence for a link between identities and lifestyles of this kind and new cleavage politics: Debus (Reference Debus, Weßels and Schoen2021), for instance, finds “cosmopolitan” or “parochial” leisure activities correlated with Green Party or AfD voting, respectively (see also Jackman and Vavreck Reference Jackman, Vavreck, Sniderman and Highton2012), while Ollroge and Sawert (Reference Ollroge and Sawert2022) show that cosmopolitan lifestyles of fictitious others are read as signals of their liberal political attitudes, and affect the willingness of respondents with cosmopolitan attitudes to interact with them (see also Iyengar, Sood, and Lelkes Reference Iyengar, Sood and Lelkes2012; Hobolt, Leeper, and Tilley Reference Hobolt, Leeper and Tilley2021; Sczepanski Reference Sczepanski2023). Linking this to another strand of political research, we can think of cosmopolitan closure as a mechanism of social sorting, with homophily based on lifestyle signals deepening the correspondence between social identities and political preferences along the new cleavage (Rawlings Reference Rawlings2022; DellaPosta, Shi, and Macy Reference DellaPosta, Shi and Macy2015; Hetherington and Weiler Reference Hetherington and Weiler2018).
On the other side of the new divide, a number of studies have captured similarly lifestyle-based forms of what could be called communitarian or particularist closure, based on the valorization of “down-to-earth” or locally rooted identities (Gest Reference Gest2016; Hochschild Reference Hochschild2016; Cramer Reference Cramer2016; Zollinger Reference Zollinger2022). In an evocative, albeit largely speculative contribution, Reckwitz (Reference Reckwitz2020) relates both processes, theorizing that a cultural code of “singularization,” the curation and production of a uniqueness of the self and its experiences, has become a central way for progressive ‘new middle class’ knowledge workers to draw boundaries against the more conservative lower and “old middle class.” By devaluing the sense of “ordinariness” central to the identities of these latter classes (Savage, Bagnall, and Longhurst Reference Savage, Bagnall and Longhurst2001; Sachweh and Lenz Reference Sachweh and Lenz2018) as “run-of-the-mill” or “boring,” distinction strategies around “creative” and “interesting” identities are said to provoke a counter-distinction of the devalued, and thus feed into the crystallization of the new cleavage. Although doubts remain about the pervasiveness and real-life importance of such processes (for a skeptical perspective see e.g. Beck and Westheuser Reference Beck and Westheuser2022), it is clear that further research on the formation of new cleavage identities through lifestyle-based closure has the potential for extensive synergies between political science and sociology.
The sites of closure
Another aspect of closure concerns the alignment of group formation across different social sites, including families, workplaces, schools, neighborhoods, places of leisure, clubs, and associations. Again, previous research had shown the increasing disassociation of these life spheres from one another and from politics, with individuals said to be assembling idiosyncratic patchwork identities (Bauman Reference Bauman2000). Yet, some recent findings point to a reconfiguration of linkages between them in the context of new cleavage formation. An important example is the increased centrality of educational institutions for forming identities, social ties, and political attitudes (Stubager Reference Stubager2009; Marks et al. Reference Marks, Attewell, Hooghe, Rovny and Steenbergen2022). Higher education has long been shown to correspond to more liberal positionings on second-dimension issues (Kriesi et al. Reference Kriesi, Grande, Lachat, Dolezal, Bornschier and Frey2008; van de Werfhorst and Dirk de Graaf Reference van de Werfhorst and Dirk de Graaf2004); it acts as a site of social closure e.g. via the trend towards educational homogamy or “assortative mating” rooted in a function of education systems as marriage markets (Iversen and Soskice Reference Iversen and Soskice2019), as well as the gendered composition of specific fields of study (Hooghe et al. Reference Hooghe, Marks and Kamphorst2024). Similarly, the knowledge economy occupations and workplaces that many higher-education careers feed into might function as sites of social closure (Kitschelt Reference Kitschelt1994), and their clustering in urban spaces may foster the formation of place-based identities (Savage et al. Reference Savage, Hanquinet, Cunningham and Hjellbrekke2018, Zollinger Reference Zollinger2024). Studies of localities “left behind” by socioeconomic transformations show signs of a similar overlap of educational, place-based, as well as, at times, age-based, and gendered identities aligning with particularist political leanings (Damhuis Reference Damhuis2020; Cramer Reference Cramer2016). In this vein, some recent work in the cleavage tradition uses survey data to indicate the importance of homogeneous (but not formalized!) educational or class networks in reinforcing and stabilizing attitudes and identities associated with the new cleavage (De Jong and Kamphorst Reference De Jong and Kamphorst2024, Zollinger and Attewell Reference Zollinger and Attewell2023). These social networks are theorized as emerging initially from socio-structural conditions (parental background, stratified educational paths, occupational trajectories, etc.) but also as consolidating and stabilizing through social closure behavior.
Future research could more systematically assess the degree to which closure across different sites leads to the reinforcement of cleavage-related group identities, as well as typical forms in which cross-cutting group formation diffuses the potential for cleavage formation. While the majority of existing case studies center on already mobilized groups, especially of radical right voters, or on “left behind” localities in which factors conducive to cleavage formation compound, cross-cutting factors could be studied by looking at more ambivalent e.g. periurban or thriving rural places, or by reconstructing how identities made salient in one site (e.g. connected to being a worker) pull in different directions than those formed in others (e.g. connected to being a homeowner). Open questions also remain concerning the role of civil society or religious associations, which were central to older cleavages but can today be seen to both act as sites of closure (say, in the case of urban refugee support groups or rural gun clubs), or as spaces for sociability across cleavage lines (e.g., in churches or neighborhood initiatives) (see also Hutter and Weisskircher Reference Hutter and Weisskircher2022).
The resources of closure
As mentioned, the making of group boundaries ties in with the demarcation of status positions. In this vein, empirical studies show how members of various social milieus draw boundaries to maintain a sense of dignity and worth, and to depict their ingroup in favorable terms (Lamont Reference Lamont2000, Damhuis Reference Damhuis2020). This helps address a puzzle that political scientists have grappled with, namely that identities on the two sides of the new divide tend to follow an asymmetrical pattern instead of that of a Manichaean opposition. Rather than self-declared “particularist” “losers” demarcating themselves from “universalist” “winners” and vice versa, we see how both sides define themselves positively and through distinct and idiosyncratic vocabularies, e.g., as “open,” “informed,” and “active” (universalists) and “hard-working,” “morally upright,” and “patriotic” (particularists) (Zollinger Reference Zollinger2022). While particularist identities are partly defined by hostility towards urbanites and the highly educated, a reverse demarcation from rural dwellers or the less educated is not salient for universalists (Bornschier et al. Reference Bornschier, Haffert, Steenbergen and Zollinger2021).
When situated in unequal social relations, these differences make sense, as those in dominant positions can remain blind to their own privilege (while often nonetheless drawing subtle boundaries to consolidate it; see Jarness and Friedman Reference Jarness and Friedman2017). Going beyond the familiar point of social identity theory – that individuals seek to define their ingroup in positive terms (Tajfel and Turner Reference Tajfel, Turner, Austin and Worchel1979; Shayo Reference Shayo2009) – sociologically informed studies can investigate the idiosyncratic ways in which distinct electorates and class fractions develop cleavage identities and how this relates to their respective social positions (for a first empirical study looking at both sides of the universalism-particularism cleavage see Damhuis and Westheuser Reference Damhuis and Westheuser2024). The new cleavage may look very different from different social positions, and there may be multiple and competing ideas of what – and who – this divide is “really about.” This point takes us directly to the more top-down process of cleavage identity formation: classification struggles.
Classification struggles
While closure concerns the bottom-up emergence of groups, classification struggles stand for processes in which groups are made “top-down,” via political and cultural entrepreneurs’ performative claims about group identities and group boundaries. Sociological contributions here help us rethink political identity claims and group appeals: While the political science literature often tends to talk about “mobilizing” or “activating” identities – i.e., treating them as latent and given – the perspective we seek to revive foregrounds the making of groups as a contentious process of higher-level boundary work or “symbolic labor” (Boltanski Reference Boltanski1979) by which groups are transformed from “an ‘analytical construct’ into a ‘folk category’” (Bourdieu Reference Bourdieu1987, 8). This notion of a laborious forging of groupness ties in nicely with the concept of “cross-pressured” voters already well-known from cleavage theory (Bornschier Reference Bornschier2010, Dassonneville Reference Dassonneville2022): the multidimensionality of voters’ identities–which in turn can be viewed as rooted in socio-structural conditions–arguably only creates pressures for the individual once classification struggles between social and political actors force voters to take a stand on “who they are and which side they are on.”Footnote 9
By focalizing classification struggles, the construction of universalist and particularist identities is treated as analogous to the historical “making” of classes and national groups (Wimmer Reference Wimmer2013; Thompson Reference Thompson1968). As introduced above, appeals to groups only resonate where the groundwork has been laid through a process of the repeated “inculcation” of group boundaries (Bourdieu Reference Bourdieu2018, 82) in which political and cultural entrepreneurs establish the salience and “groupness” of identity categories (Brubaker Reference Brubaker2004).Footnote 10 To give an example, appeals to “communitarians” and “universalists” would – at least currently – not have a political effect, while appeals to, e.g., “this country’s hard-working families” mobilize an entire set of implicit, morally charged classifications defined by previous classification struggles (e.g. by gesturing towards but avoiding the term “working class,” opposing working people to non-workers, invoking national belonging, the deservingness of hard work, the respectability of family life, etc.).
Cleavage theory makes room for political elites’ role in actively shaping electoral potentials (see e.g. Evans Reference Evans2010); and a number of recent studies trace parties’ references to specific group categories over a long time (Thau Reference Thau2019; Mierke-Zatwarnicki Reference Mierke-Zatwarnicki2023; De Leon, Desai, and Tuğal Reference De Leon, Desai and Tuğal2020). But there has been much less systematic investigation into exactly how parties and other actors have contributed to the emergence of new group identities in the current conjuncture, e.g., by providing voters with the repertoires to express social identities and categorize those of others; and how competing actors struggle over the meaning of identity categories, each seeking to define group boundaries in a way that serves their own political purpose.
The code of classification struggles
There is an important strand of research that studies classification struggles around the new cleavage without using this term: This is research on the disarticulation of working-class identities and their rearticulation along nationalist, conservative, and producerist lines by the radical right. As Evans and Tilley (Reference Evans and Tilley2017) and others show, established parties have progressively ceased to center working-class identities in their appeals, leaving a void that radical right actors fill by defining the category “worker” through an opposition between productive, majority-ethnic “makers” and ostensibly lazy welfare-dependent “takers” of minority-ethnic origin (Attewell Reference Attewell2021; Rathgeb Reference Rathgeb2021; Somers Reference Somers, Mackert and Turner2017), through a distinction between categories of workers (Hürtgen Reference Hürtgen, Book, Huke, Klauke and Tietje2019), including a culturalization and moral devaluation of migrant workers (Yilmaz Reference Yilmaz2016); as well as by associating working-class norms of respectability, speech, and gender with political conservatism (Prasad et al. Reference Prasad, Perrin, Kieran Bezila, Hoffman, Manturuk, Smith Powers and Payton2009; Peck Reference Peck2019). Similarly, other studies show how actors of the right connect localist, place-based identities of rural populations with ethnic nationalism, effectively shifting what it means to be “from here” (Kalb and Halmai Reference Kalb and Halmai2011; Stacul Reference Stacul, Kalb and Halmai2011). By comparison, the making of universalist identities, e.g., in the middle class or urban milieus, has received much less scholarly attention; and there is a dearth of studies showing why the rearticulation of group identities along the new cleavage succeeds in some cases but not in others. The Bourdieusian concept of classification struggles can here provide a framework for a holistic research agenda looking at competing efforts of group-making across the cleavage, and the historical longue durée of political articulation (e.g., of remaking the working class as “communitarian” and “white” or the middle class as “progressive”).
Within political science, such an approach presents an extension of existing literature on group appeals in party strategy (Huber Reference Huber2022; Dolinsky Reference Dolinsky2022, also see a review article by Zuber et al. Reference Zuber, Howe and Szöcsik2023). Beyond its current method of registering the way parties “mention” already defined socio-demographic groups as part of their shorter-term strategy, this extension would look to historical and qualitative work to capture culturally connoted, morally charged “universalist” and “particularist” group boundaries emerging among electorates on opposite sides of the new cleavage, and to trace how identity boundaries now salient among voters have been established in political discourse over time.Footnote 11 Such a broadening of the perspective on the dynamic construction of group boundaries is greatly facilitated by current methodological progress. Advances in quantitative text analysis based on large language models, for instance, make the identification of group references in political text much less reliant on pre-defined key terms (Licht and Sczepanski Reference Licht and Sczepanski2023). Relatedly, approaches to harnessing images as data, for instance in social media communication, are making great strides (Webb Williams, Casas, and Wilkerson Reference Webb Williams, Casas and Wilkerson2020). All of this could make it more feasible to capture symbolic appeals that political actors make in the attempt to create and reorder group boundaries and impose their specific vision of the social order.
The field of classification struggles
A second extension concerns the field of actors involved in classification struggles. While the discussion so far has revolved largely around parties, the Bourdieusian perspective points towards a mapping of actors developing and disseminating identitarian vocabularies, including not just political actors like parties or movements but also cultural and intellectual producers. We can think of media actors contributing to “manufacturing conflict,” be it consciously, as in pundits intervening in social and political debates (see Green Reference Greenforthcoming), or more subtly, through the framing of social and political divides and the production of stereotyped images and symbols for group identities, as in the imagery of “chai latte” drinking liberals or an ethnically defined and inherently right-wing “white working class” (Mondon and Winter Reference Mondon and Winter2019; Bartholomé, Lecheler, and de Vreese Reference Bartholomé, Lecheler and de Vreese2015; Bergfeld Reference Bergfeld2019). Similarly, intellectuals and artistic producers can contribute to the elaboration of group identities (see e.g. Göpffarth Reference Göpffarth2020).
For cleavage scholarship, this broadening of perspective is not only valuable because it opens studies of cleavage formation to evidence, data, and methods from intellectual history, cultural sociology, or media studies (cf. Martin Reference Martin2023). Understanding struggles over classifications as part of cleavage formation also aids scientific reflexivity: In what Bourdieu (Reference Bourdieu2000) calls a “theory effect,” categories created by the social sciences may feed back into the processes of group-making they were designed to describe. This has arguably been the case with diagnoses of a new cleavage, or terminology like that of “cosmopolitans” and “communitarians,” or “winners” and “losers of globalization,” which have transformed from analytical shorthand into talking points of political and lay discourse. Scientific artifacts like group labels, analytical dimensions, or survey items, too, can also form part of the field of classification struggles. The advantage of this concept is that it allows us to reconstruct connections and strategic interactions across different spheres, each with their own mode of developing and propagating competing group identity frames. This may be particularly important in a historical moment in which parties have ceased to hold the monopoly of symbolic production in the realm of politics and in which voters have constant and easy access to alternative sources of information.
The resources of classification struggles
Not all actors in the field of classification struggles have the same chance of having their interpretations and identity claims heard and recognized (Wacquant Reference Wacquant2013), that is, of successfully imposing their “principles of vision and division” on society. Indeed, many groups do not have access to these struggles at all. What this points to is that – just like processes of closure – classification struggles are situated in unequal social relations. Classification is not a merely symbolic operation. On a material level, it builds on socio-structural realities, such as shared experiences and grievances, e.g., of lower-educated workers, or younger urban women, which make claims about their common identities as a group plausible and resonant. Successfully making these claims is further underpinned by resources for campaigns, events, and local organizational presence, as well as access to or outright ownership of media institutions (De Jonge Reference De Jonge2019). A less obvious but important resource for universalist and particularist group-making is the power of the state to create and inculcate legitimate social categories (Starr Reference Starr1992; Fourcade Reference Fourcade, Morgan and Orloff2017). State categorization is central to contested group boundaries at the heart of universalist and particularist identities, such as that between native citizens and foreigners; between the behaviorally, sexually, or criminally “normal” and “deviant”; or between more or less “deserving” categories of citizens and welfare recipients. The control over this capacity of the state forms another material basis for classification struggles. Lastly, the group-making power of actors depends on unequally distributed forms of discursive power, conferred e.g. by positions of authority, as well as cultural influence, recognition, and legitimacy. Stigmatized or fringe actors like neo-Nazi groups here stand in contrast to established actors, like parties in government. Besides codes and actors, the resources of group-making form a third dimension for analyzing the field of classification struggles around new cleavage identities.
Conclusion
Scholars of political sociology, political behavior and party strategy are reconverging on studying group-based politics. After a period in which the decline of the classic 20th-century cleavages appeared to give way to a more individualized and dealigned politics, evidence is mounting that in post-industrial, secularized, advanced capitalist democracies, important segments of the electorate continue to make sense of politics through a lens of group belonging. Similarly, political elites still appeal to social groups, and not just enlightened issue voters. Yet, two sets of observations about contemporary politics sit uneasily side by side: on the one hand, there is strong evidence that the success of far-right or new left “challenger” parties is rooted firmly in the social structure of the post-industrial knowledge society. On the other hand, the ongoing fragmentation of political landscapes and the gaping vacuum left by once-dominant mass social and political organizations raises questions about where and how that structure comes about – and how it sticks.
We have argued here that recent attempts to update cleavage theory for 21st-century group politics indicate potential answers to this paradox, specifically if they are complemented with conceptual tools and empirical insights from Bourdieusian sociology. A “thicker” sociological understanding of group formation – in the sense of the forging and attainment of collective consciousness – can also allow cleavage scholars to build more effectively on recent advances in the study of strategic group appeals. Concretely, we have suggested that, while contemporary cleavage research already provides evidence on the existence of three constitutive elements of a new cleavage – structure, identity, and politics – Bourdieusian scholarship helps us to identify, conceptualize, and empirically study two central sets of processes connecting these elements: social closure, the process through which identity potentials arising from socio-structural conditions become realized in everyday practices of distinction; and classification struggles through which political actors impose their vision of relevant group and identity boundaries in society. The agenda we propose examines the way new cleavage categories become salient in everyday processes of distinction and in public symbolic struggles over the categorization of groups. Especially if combined, both component processes form a holistic agenda of research on the emergence and making of new cleavage identities, which to date, is lacking. To demonstrate the insights that such an agenda can bring to light, we have highlighted empirical work that already sheds light on aspects of new cleavage identity formation in these processes, and we have outlined avenues for future research at the intersection between cleavage theory and sociology.
The macro-sociological perspective of cleavage theory is uniquely positioned to capture the epochal transformations in electoral politics that come with a shift to post-industrial, globalized, knowledge-based economies (Iversen and Soskice Reference Iversen and Soskice2019, Hall Reference Hall2022). The concept of cleavage identities, meanwhile, provides an obvious opening for integrating this strand of work with the rich synthesis of sociological thought achieved in the Bourdieusian research tradition. The latter generates valuable theoretical and empirical insights into contemporary social divisions, which should be integrated into a long-durée perspective on the transformation of political cleavage structures.
Acknowledgments
For important comments on earlier drafts, we would like to thank Jane Gingrich, Simon Bornschier, Koen Damhuis, Sigrid Roßteutscher, Dave Attewell, the three anonymous reviewers and the editors at EPSR. We are grateful for feedback we received at the IPZ Publication Seminar of the University of Zurich, the German Political Science Association (DVPW) congress, and the European Graduate Network conference at SciencesPo.
Funding
Delia Zollinger thankfully acknowledges funding from the Swiss National Science Foundation (SNSF Doc.CH Grant No. 188365).
Competing interests
None.
Open access
