Reason, structure or culture?

Since the emergence of the discipline in the 19th century, sociologists have debated the ontological underpinnings of human behaviour. One of the leitmotifs has been the debate over the primacy of structure over culture and vice versa. This debate is well represented through the opposition of a Marxist ontology with its focus on structural (economic) determinants of behaviour over cultural (ideological) epiphenomena, and a Weberian ontology stressing the fundamental role played by cultural aspects, including religion.

Another long-standing debate in the social sciences, which is partly linked to the previous one, is between structure and agency. While there have been laudable attempts to discuss and possibly overcome this (apparent) dichotomy (Archer Reference Archer1995; Bourdieu Reference Bourdieu1977; Giddens Reference Giddens1984), social theory has often stressed either the primacy of structure, leaving little margin for manoeuvre for human action or agency, while frequently neglecting the role of structural factors. Structural-functionalist thinking typically gives priority to structure, while rational choice theory – but also other approaches – tends to prioritize agency. This second debate is partly related to the first one as often culturalists have leaned towards agency, albeit with a different focus relative to rational choice theorists, while structuralists have obviously put structure first. Sometimes the same tension is present within the same author in different moments: for example, the early work of Talcott Parsons of the Structure of Social Action (Reference Parsons1937) is very different from the later one of the Social System (Reference Parsons1951).

If we combine these two fundamental debates in some way, we can outline three distinct ways of approaching the study of political participation, which has rarely engaged rationalist, structuralist and culturalist accounts in a constructive dialogue. To be sure, these three kinds of perspectives not only characterize theories of political participation. They have, for example, been discussed more broadly in relation to the underpinnings of comparative politics (Lichbach and Zuckerman Reference Lichbach, Zuckerman, Lichbach and Zuckerman1997). However, our point is that research on political participation has often followed one or the other of these perspectives, largely neglecting their possible combination.

Rationalist accounts emphasize the fact that actors’ decision to participate is self-interested and based on a cost/benefit calculus. In the most extreme version of the theory, individuals are understood as completely free from external constraints in taking decisions. In its more developed, moderate and plausible version, decision-making is influenced by external factors, and we can only speak of a ‘bounded rationality’ (Simon Reference Simon1957). This perspective is most common in studies of voting turnout and behaviour (Franklin Reference Franklin2004, Reference Franklin, Giugni and Grasso2022), although such approaches are also applied to other types of political participation (Whiteley Reference Whiteley, Giugni and Grasso2022) and inform work on social movements and protests (Oberschall Reference Oberschall1993; Oliver Reference Oliver1984; Opp Reference Opp1989). Rationalist accounts may also include work that is outside the rational choice tradition. Indeed, few today would maintain that actors are irrational. The difference is on the centrality given to this kind of motivation and the role played by self-interest and the maximization of individual utility in explanations of political participation.

Structuralist accounts are popular especially among students of social movements, less so in research on voting. In these accounts, political participation is strongly influenced, if not determined, by certain structural aspects of the broader environment, usually of the political-institutional context. When it comes to explaining participation in movements and protest activities, they most often refer to the constraining or channelling role played by political opportunity structures (Giugni Reference Giugni2009; Kriesi Reference Kriesi, now, Soule and Kriesi2004; Meyer and Minkoff Reference Meyer and Minkoff2004) and more generally by the political context of social movements (McAdam and Tarrow Reference McAdam, Tarrow, Snow, Soule, Kriesi and McCammon2019). Explanations stressing the role of the economic context might also be included among this kind of account (Della Porta Reference Della Porta2015; Grasso and Giugni Reference Grasso and Giugni2016; Kurer et al. Reference Kurer, Häusermann, Wüest and Enggist2019), with economic constraints sometimes conceptualized as a kind of opportunity structure (Kousis and Tilly Reference Kousis and Tilly2005).

If structuralist accounts stress structure, culturalist accounts emphasize the role of culture. Here we may include works in the political science tradition that look at the impact of values (Hitlin and Allyn Piliavin Reference Hitlin and Allyn Piliavin2004; van Deth et al. Reference van Deth, Jan and Scarbrough1998), but especially research in the field of social movement studies. Explanations of participation in social movements that stress cultural factors abound. This includes work by sociologists (Berstein Reference Berstein2005; Flesher Fominaya Reference Flesher Fominaya, Snow, Soule, Kriesi and McCammon2019; Goodwin et al. Reference Goodwin, Jasper and Polletta2001; Johnston and Noakes Reference Johnston and Noakes2005; Jasper and Polletta Reference Jasper, Polletta, Snow, Soule, Kriesi and McCammon2019; Polletta and Jasper Reference Polletta and Jasper2001) and social psychologists (Klandermans Reference Klandermans1997; van Zomeren and Iyer Reference van Zomeren and Iyer2009) that places the role of collective identity at centre stage. Others adopt framing theory with its emphasis on discourses and collective action frames (Benford and Snow Reference Benford and Snow2000; Snow Reference Snow, In Snow, Soule and Kriesi2004; Snow et al. Reference Snow, Vliegenthart, Ketelaars, In Snow, Soule, Kriesi and McCammon2019). New social movements theory may also be included in this group of works, for its emphasis on identity and insofar as it looks at how new movements – such as, for example, environmental, peace, squatters’ and women’s movements – are culturally distinct from older ones in many respects (Buechler Reference Buechler1995; Pichardo Reference Pichardo1997).

These three types of explanation have largely followed separate tracks, leading students of political participation to prioritize certain explanatory factors at the expense of others. However, there are some attempts to combine them, such as in the social movement and protest politics literature. For example, Bert Klandermans (Reference Klandermans1997) starts from a social-psychological perspective that places different kinds of grievances, collective identities and emotions – such as group anger – centre stage. However, he also takes into account the social-structural embeddedness of movement participants and the broader structural context of participation in social movements, while acknowledging that participation may be based on instrumental motivations, in addition to identity-based and ideological ones. Karl-Dieter Opp (Reference Opp2009) has developed a structural-cognitive model from a rational choice perspective that emphasizes the role of cognitive processes and selective incentives for participation, but which also considers structural factors such as political opportunity structures and cultural ones such as identity (conceptualized as a selective incentive) and framing processes.

More recently, Marc Hooghe (Reference Hooghe, Giugni and Grasso2022) has suggested two combinations of theoretical perspectives, which he sees as most promising for the study of political participation. The first combines rationalist and structuralist perspectives by acknowledging that the individual actors’ self-interest is largely shaped by their structural position within society. More generally, structural conditions – economic, social, political – set the boundaries for the actors’ assessment of self-interest and therefore help explain why they take the (rational) decision to participate. The second combination of theoretical perspectives suggested by Hooghe (Reference Hooghe, Giugni and Grasso2022) relates to structural and cultural accounts. He states that structural conditions are not sufficient to explain participation, but it is equally important to take into account the potentially liberating norms that are present in the political system itself. Although he warns us against the temptation to combine all kinds of theoretical approaches into a general framework, Hooghe (Reference Hooghe, Giugni and Grasso2022) has sketched an integrated framework that brings together various perspectives of political participation, starting from the assumption that the different dimensions of political approaches to participation are inherently linked to one another. This framework integrates the motivation to participate at the individual level, ideology and the alignment with social cultural frames, and political opportunity structures.

Based on the above, to bridge the rationalist, structuralist and culturalist accounts, we make the following propositions, echoing those of Hooghe (Reference Hooghe, Giugni and Grasso2022):

• • Proposition 1: Structural conditions set the boundaries for the actors’ assessment of self-interest and therefore for their decision to participate.

• • Proposition 2: The presence of structural conditions conducive to political participation must be accompanied by cultural norms that are present in the political system itself.

Cognition and affect

Common wisdom holds that heart and mind are two distinct and separate spheres of life. The first is the realm of reason and cognition, the latter that of emotions and affect. This oversimplistic but widespread view has informed much work on political participation and social science more generally. The result is a division of labour between those examining the cognitive side of individual predispositions – that is, political attitudes such as political interest, trust and efficacy – and those stressing accounts emphasizing their affective side – most notably, identity and emotions, but to some extent also personality traits and motivations. This division, however, is largely artificial, if not misleading. It can be seen, for example, in the concept of values, which has cognitive and affective dimensions (Marini Reference Marini, Borgatta and Montgomery2000), or in that of collective identity, which can be seen as resulting from people’s cognitive, moral and emotional connections with a broader community (Polletta and Jasper Reference Polletta and Jasper2001).

The distinction between cognition and affect partly overlaps that between rationality, structure and culture, or perhaps better, it is informed by it. Indeed, rationalist accounts stress the cognitive side of human behaviour, while culturalist accounts often – but not always – emphasize the role of emotions and affect. Nonetheless, the study of political participation – and perhaps even more so that of protest participation – has tended to follow two distinct and separate tracks: one stressing ‘rational’ factors such as resources and attitudinal predispositions as a basis for the decision to participate and another focusing on such aspects as identity and emotions, but also framings, memory and narratives.

We believe that this division of labour provides invaluable insights on the reasons why people participate or do not participate in politics, but in the long term might be harmful and prevent us from reaching a more comprehensive understanding of the wide range of factors leading to political participation. In particular, we suggest that cognitive and affective elements combine in various ways to lead people to become politically active. Once again, the literature already offers several insights as to how these two types of factors might combine.

While research on political participation has traditionally stressed (cognitive) political attitudes such as political interest, knowledge, sophistication, efficacy and the like, scholars have paid increasing attention to the role of emotions in politics. Emotions, affects, feelings, sentiments, moods, passions and so forth have made their breakthrough into the study of voting behaviour as well as social movements and protest politics. Political psychologists have investigated the role of emotions to explain political choice and how they may influence political attitudes and behaviours (Brader Reference Brader2005, Reference Brader2006; Marcus Reference Marcus2000; Marcus and MacKuen Reference Marcus and MacKuen1993; Marcus et al. Reference Marcus, Russell Neuman and MacKuen2000; Neuman et al. Reference Neuman, Marcus, Crigler and MacKuen2007; Valentino et al. Reference Valentino, Brader, Groenendyk, Gregorowicz and Hutchings2011, Reference Valentino, Hutchings, Banks and Davis2008; Weber Reference Weber2013). They have thus long acknowledged that the connections between cognition and affect have important implications for political evaluation, decision and action (Redlawsk Reference Redlawsk2006; Way and Masters Reference Way and Masters1996). For example, scholars have examined the relationship between emotions and political efficacy (Rudolph et al. Reference Rudolph, Gangl and Stevens2000; Valentino et al. Reference Valentino, Gregorowicz and Groenendyk2009), the role of anxiety as well as its distinct effect from anger (Best and Krueger Reference Best and Krueger2011; Huddy et al. Reference Huddy, Feldman, Casses, Neuman, Marcus, Crigler and MacKuen2007) or the impact of affect on political sophistication (Marcus and MacKuen Reference Marcus and MacKuen1993; Marcus et al. Reference Marcus, Russell Neuman and MacKuen2000; Miller Reference Miller2011).

But emotions have also entered the study of social movements and protest politics, after years of domination by (rationalist and structuralist) resource mobilization and political opportunity theories. Social psychologists were at the forefront in this regard, examining different kinds of emotions. A much-studied emotion in this respect, is anger, which is seen as particularly conducive to protest participation (Klandermans et al. Reference Klandermans, van der Toorn and van Stekelenburg2008; Leach et al. Reference Leach, Iyer and Pedersen2006, Reference Leach, Iyer and Pedersen2007; Mummendey et al. Reference Mummendey, Kessler, Klink and Mielke1999; van Stekelenburg and Klandermans Reference van Stekelenburg, Klandermans, Klandermans and Roggeband2007). Importantly, social psychologists have looked at how emotions combine with attitudes to explain protest participation or its withdrawal (van Stekelenburg and Klandermans Reference van Stekelenburg and Klandermans2013). In this vein, Martijn van Zomeren et al. (Reference van Zomeren, Spears, Fischer and Leach2004, Reference van Zomeren, Leach and Spears2012) have proposed a ‘dynamic dual pathway model’. This model understands collective action as the outcome of two distinct processes: an emotion-focused approach revolving around the experience of group-based anger, on the one hand, and a problem-focused approach revolving around beliefs in the group’s efficacy, on the other.

In addition to social-psychologists, sociologists and political scientists are also paying increasing attention to the role of emotions in social movements and protest politics (Goodwin et al. Reference Goodwin, Jasper and Polletta2000, Reference Goodwin, Jasper and Polletta2001; Goodwin and Jasper Reference Goodwin and Jasper2004; Jasper Reference Jasper2011; Van Ness and Summers‐Effler Reference van Ness, Summers‐Effler, In Snow, Soule, Kriesi and McCammon2019; see Jasper and Zhelnina Reference Jasper, Zhelnina, Giugni and Grasso2022 for a discussion in relation to political participation). Such attention is part of a ‘cultural turn’ in the study of social movements. Along with new social movements theory and framing theory, where the cognitive dimension of culture has largely overshadowed its affective dimension, these efforts have developed in directions different from the ‘structuralist bias’ of political opportunity theory (Goodwin and Jasper Reference Goodwin and Jasper2004). Early accounts of crowd behaviour in the late 19th and early 20th centuries stressed the role of panic. Later on, breakdown theories of collective behaviour of the 1950s and 1960s viewed frustration as central to explaining collective behaviour. Collective behaviour was seen as deviant if not irrational. On the other hand, this new work on emotions in social movements – emerging in part as a reaction to what was considered at the time as an excessive focus on resources and political opportunity structures – shared, among other aspects, the view that emotions are not to be understood as separate from rationality. This understanding has meant steps can be taken towards bridging cognition and affect. In other words, people can act rationally while showing strong emotions and, conversely, they can act emotionally while being fully rational. Moreover, while emotions were understood primarily in their negative declination in the past, they were now seen as both negative (e.g. anger, fear) and positive (e.g. enthusiasm, joy) emotions. A restrictive view of emotions as short-term outbursts of frustration and other negative feelings was replaced by a broader perspective comprising both short-term, reflexive emotions (e.g. disgust, surprise) and long-term moods and affective predispositions (e.g. frustration, love).

As we can see from this brief review of works by social-psychologists, sociologists and political scientists, while often following two separate tracks, they have also combined cognitive and affective explanations of political participation. In our own work on participation in demonstrations, we also found that both political attitudes and emotions concur to explain protest participation. This occurs through the link between political interest and anger, hence suggesting that cognition and affect both contribute to commitment to engage in protest activities (Giugni and Grasso Reference Giugni and Grasso2019). To take two examples, participants in protests staged by movements such as Black Lives Matter in the US or various anti-austerity movements across Europe were arguably driven by a desire for policy change, but also by feelings of righteous anger as well as indignation. These two aspects interact in complex ways. In a recent effort at reviewing works that bridge cognitive-based and affective-based explanations of political participation, Alessandro Nai (Reference Nai, Giugni and Grasso2022), echoing a broader model for the drivers of mobilization based on both psychological and communicational mechanisms, concludes that emotions and cognition interact in two fundamental ways. On the one hand, this occurs through a direct and joint effect, in particular the fact that emotions have a stronger role in participation at higher levels of cognitive attitudes. On the other hand, they further play a mediating role by fuelling the effects of political information on participation.

Based on the above, we make the following propositions with regard to bridging cognitive-based and affective-based explanations of political participation, echoing those made by Nai (Reference Nai, Giugni and Grasso2022):

• • Proposition 3: Beyond a simple additive effect, cognitive drivers (attitudes) and affective drivers (emotions) of political participation interact, so that cognitive drivers moderate the effect of affective drivers and vice versa.

• • Proposition 4: Beyond a simple additive effect, cognitive drivers (attitudes) mediate the effect of affective drivers (emotions) on political participation and vice versa.

The macro/micro gap

Moving from ontological underpinnings and theoretical perspectives to methodological approaches, explanations of political participation have traditionally followed two distinct paths when it comes to the level of observation or analysis. Analyses of political participation tend to focus on the individual (micro) level: voting, protesting or some other form of participation such as consumerism or digital forms (Giugni and Grasso Reference Giugni, Grasso, Giugni and Grasso2022). Scholars have also examined participation at the aggregate (macro) level, such as in studies of electoral volatility and electoral geography (Giugni and Grasso Reference Giugni, Grasso, Giugni and Grasso2022). This does not take into account the literature on social movements and collective action, which by definition are phenomena located at the meso or macro level and therefore require explanations at that level of analysis.

Especially in recent years, scholars have paid increasing attention to bridging the macro/micro link (Anduiza Reference Anduiza2002; Dalton et al. Reference Dalton, Van Sickle and Weldon2010; Grasso and Giugni Reference Grasso and Giugni2016; Kern et al. Reference Kern, Marien and Hooghe2015; Vráblíková Reference Vráblíková2013) in the study of political participation, including in protest activities, also helped by the development of (quantitative) multilevel analysis techniques. Multilevel models share two important methodological features which, in turn, offer remarkable advantages for theorizing about political participation. First, they combine a micro model with a macro model, hence allowing for modelling contextual heterogeneity – for example, variation across countries – in a more powerful and accurate way. Second, they allow for the inclusion of contextual characteristics in the models.

But multilevel models have another feature that is perhaps even more important for pushing the study of political participation further, namely the potential to model cross-level interactions. While research on political participation has mainly focused on the individual level, perhaps the most interesting and relevant hypotheses that we can think of often involve an interaction between individual and contextual factors. For example, we show (Grasso and Giugni Reference Grasso and Giugni2016) that the impact of relative deprivation on protest activism is conditional on context – specifically, unemployment – arguing this means individuals are more likely to understand their individualized experience in politicized terms.

The macro-level, contextual factors that may influence the relationship between individual characteristics and political participation are of different sorts. Students of social movements have long acknowledged the role of political opportunity structures for the emergence, mobilization and outcomes of social movements (Kriesi Reference Kriesi, now, Soule and Kriesi2004). They have stressed a variety of dimensions of opportunity, such as the multiplicity of independent centres of power within the regime, the regime’s openness to new actors, the instability of current political alignments, the availability of influential allies or supporters for challengers, the extent to which the regime represses or facilitates collective claim-making, as well as decisive changes in these aspects (McAdam Reference McAdam, In McAdam, McCarthy and Zald1996; Tilly and Tarrow [Reference Tilly and Tarrow2015 [2006]]). More recently, however, scholars have expanded the concept to include discursive or cultural opportunity structures (Koopmans and Olzak Reference Koopmans and Olzak2004) or economic opportunity structures (Kousis and Tilly Reference Kousis and Tilly2005). These different sorts of opportunity structures – or just opportunities – refer to the impact of the political-institutional, cultural-discursive and economic context on the options people have to mount social movements and collective action.

Although these contextual factors were developed by social movement scholars to account for collective action, they have been shown also to have a direct effect on individual-level political participation, especially participation in protest activities. A variety of contextual factors have been examined in the literature, from unemployment to social spending, GDP, Gini index, certain features of the political system, and so forth (Giugni and Grasso Reference Giugni and Grasso2019; Vráblíková Reference Vráblíková2013).

Based on the above, we make the following propositions with regard to bridging the macro/micro gap:

• • Proposition 5: The effect of micro-level, individual predispositions and attitudes on political participation is moderated by macro-level, contextual factors such as macroeconomic conditions, institutional arrangements and prevailing public discourses.

• • Proposition 6: Micro-level, individual predispositions and attitudes mediate the effect of macro-level, contextual factors on political participation.

The qualitative/quantitative divide

Among the most persistent debates in the social sciences, from a methodological point of view, is the one between advocates of qualitative and defenders of quantitative methods. Sometimes this has led to a clash between qualitative researchers and their quantitative counterparts, not always relying on concrete arguments in support of one approach over the other.

The adoption of one or the other of these methodological approaches has its roots in the different epistemological and ontological premises on which the methodologies rely (Howe Reference Howe1992). Quantitative methods are usually associated with a Durkheimian positivist ontology, while qualitative methods are seen as reflecting a Weberian interpretivist ontology. This has led students of political participation – but also scholars in other fields – to favour one or the other of these two methodological approaches.

In spite of attempts at reconciliation (see, for example, King et al. Reference KIng, Keohane and Verba1994; Mann Reference Mann1981) as well as the development of mixed methods approaches, the qualitative/quantitative divide sometimes continues to hamper research on political participation, or at least to erect barriers to a deeper and more comprehensive understanding of the reasons why people become politically active or refrain from doing so. To be sure, just as it is perfectly fine to study political participation only at the micro level, there is nothing wrong in adopting qualitative or quantitative methods. Our point is that by combining them the analysis would find itself enriched, able to answer different research questions, and reconcile internal and external validity.

As with the macro/micro link discussed earlier, here important steps have been made to bridge what Raymond Boudon (Reference Boudon1969) called a ‘false dispute’. While the different approaches are seen as belonging to different cultures with contrasting norms, practices and toolkits, there are calls to encourage exchange between these two cultures (Goertz and Mahoney Reference Goertz and Mahoney2013). In this vein, an increasing number of studies today adopt some kind of combination of qualitative and quantitative data and methods. Indeed, combining qualitative and quantitative methods today is usually referred to as the mixed methods approach and is becoming increasingly popular among social scientists.

Mixed methods research can serve different purposes (Nunes and Henn Reference Nunes, Henn, In Giugni and Grasso2022). Jennifer Greene et al. (Reference Greene, Caracelli and Graham1989) distinguish between five main purposes: triangulation, complementarity, development, initiation and expansion. John Creswell (Reference Creswell2013) suggests there are three main strategies of inquiry for mixed methods research, each implying a different way of bridging qualitative and quantitative methods of data collection and analysis: concurrent mixed methods, transformative mixed methods and sequential mixed methods. In the first, ‘the researcher converges or merges quantitative and qualitative data in order to provide comprehensive analysis of the research problem’ (Nunes and Henn Reference Nunes, Henn, In Giugni and Grasso2022: 353). The second ‘involve[s] the researcher using a theoretical lens as an overarching perspective within a design that contains both quantitative and qualitative data’ (Nunes and Henn Reference Nunes, Henn, In Giugni and Grasso2022: 354). In the third, ‘the researcher seeks to elaborate on, or expand beyond, the findings of one method with another’ (Nunes and Henn Reference Nunes, Henn, In Giugni and Grasso2022: 354). In this last approach, the sequence as well as the importance given to one or the other type of method can vary.

Based on the above, we make the following propositions with regard to bridging the qualitative/quantitative divide, which reflect the five purposes of mixed methods research outlined by Greene et al. (Reference Greene, Caracelli and Graham1989):

• • Proposition 7: Qualitative and quantitative methods can be used to address the same aspects or dimensions of political participation (in different sequences and with a different emphasis on one or other methodology) to corroborate with one method the findings obtained with another (triangulation), to increase validity of measurements and research findings (development) or to reach a more complete understanding of that aspect or dimension (complementarity).

• • Proposition 8: Qualitative and quantitative methods can be used to address different aspects or dimensions of political participation (in different sequences and with a different emphasis on one or other methodology) to generate new understandings of it (initiation) or to extend the breadth and range of inquiry by using different methods for different inquiry components (expansion).

Ballots, barricades and beyond

As we mentioned in the introduction to this article, the literature on political participation has traditionally followed two distinct paths with little dialogue between them. This has led scholars to focus on either institutional modes of participation – mostly voting – or non-institutional ones, often protest behaviour. Underlying this division of labour is the assumption that voting and protest – or ballots and barricades, to use Ronald Aminzade’s (Reference Aminzade1993) as well as Doug McAdam and Sidney Tarrow’s (Reference McAdam and Tarrow2010) apt metaphor – are part of a zero-sum game: those who vote do not participate in protests and vice versa. However, empirical studies have shown that protesters often do go to the polls, and that institutional and non-institutional participation are not mutually exclusive (Heaney and Rojas Reference Heaney and Rojas2015; Hutter et al. Reference Hutter, Snow, Soule, Kriesi and McCammon2019). This applies not only in democratic contexts but also in autocracies. Guillermo Trejo (Reference Trejo2014), for example, shows that the introduction of multiparty elections in a wide variety of autocracies around the world gave rise to major cycles of protest and that the relationship between the ballot and the street is a crucial factor for understanding the dynamics of stability and change in authoritarian regimes. More generally, it is important to acknowledge the iterative interactions between protest and policy (Meyer Reference Meyer, Meyer, Jenness and Ingram2005) and that activism and electoral politics are intertwined (Fisher Reference Fisher2012).

McAdam and Tarrow (Reference McAdam and Tarrow2010) have asked why two cognate literatures – social movements and electoral studies – have travelled along parallel paths with little conversation between them. They also propose several ways in which they can be linked, relying on the book Dynamics of Contention (McAdam et al. Reference McAdam, Tarrow and Tilly2001) and a focus on mechanisms and processes. More precisely, McAdam and Tarrow (Reference McAdam and Tarrow2010) specify six mechanisms and processes that link movement actors to routine political actors in electoral campaigns (although they only illustrate three of them): ‘[m]ovements introduce new forms of collective action that influence election campaigns; [m]ovements join electoral coalitions or, in extreme cases, turn into parties themselves; [m]ovements engage in proactive electoral mobilization; [m]ovements engage in reactive electoral mobilization; [m]ovements polarize political parties internally; [s]hifts in electoral regimes have a long-term impact on mobilization and demobilization’ (McAdam and Tarrow Reference McAdam and Tarrow2010: 533).

McAdam and Tarrow’s (Reference McAdam and Tarrow2010) call for a rapprochement of social movements and electoral studies has been answered. In fact, well before that call, proponents of the political process approach to the study of social movements had long acknowledged that electoral and non-electoral politics are intimately linked. The very concept of political opportunity structures and how they influence the rise and mobilization of social movements is all about the linkages between institutional and protest behaviour. Some of the most well-known and now ‘classic’ examples of this include Peter Eisinger’s (Reference Eisinger1973) study of the impact of local political opportunity structures on the incidence of protest activities in American cities in the late 1960s, Craig Jenkins and Charles Perrow’s (Reference Jenkins and Perrow1977) research on farmworkers’ movements in the United States, McAdam’s (1999 [Reference McAdam1999 [1982]]) work on the civil rights movements, Herbert Kitschelt’s (Reference Kitschelt1986) comparative analysis of the strategies and impacts of social movements, Tarrow’s (Reference Tarrow1989) study of the 1965–1975 protest cycle in Italy, and Hanspeter Kriesi et al.’s (Reference Kriesi, Koopmans, Duyvendak and Giugni1995) comparative study of new social movements in western Europe. The list is long and includes the vast body of literature looking at the institutional channelling of social movements or with one or other of the various components of political opportunities. In a similar fashion, historians and historical sociologists often point to the intimate relationship between elections and protest. The most obvious reference is the seminal work by Charles Tilly (Reference Tilly1986, Reference Tilly1995) on France and Britain, where the author shows how changes in the repertoires of contention relate to the emergence of new interests, opportunities and organizations due to the large-scale processes of the rise of capitalism and the formation of national states. While capitalism – and the related processes of urbanization and industrialization – refers above all to social and economic aspects, state formation clearly points to the fact that a transformation of political institutions may lead to new ways ordinary people have to make claims (Aminzade Reference Aminzade1993).

As hinted above with the work of Kitschelt (Reference Kitschelt1986), the literature on the political outcomes of social movements stresses the linkages between protest and institutions, albeit in the opposite direction (see Amenta et al. Reference Amenta, Andrews, Caren, Snow, Soule, Kriesi and McCammon2019 for a discussion). These works often stress the importance of certain features of the institutionalized political system and the role of political alliances as mediating the impact of protest activities on policy, as in the political mediation and joint-effect models of social movement outcomes (Amenta et al. Reference Amenta, Carruthers and Zylan1992; Giugni Reference Giugni2004).

Various recent works have examined the linkages between institutional and non-institutional politics, in particular between voting and protest. In this vein, research has, for example, shown the presence of a strong correlation between party membership and social movement activism (Giugni and Grasso Reference Giugni and Grasso2021). Scholars have made a strong case that electoral and non-electoral political behaviour need to be examined side by side (McAdam and Tarrow Reference McAdam and Tarrow2010). More recently, Endre Borbáth and Swen Hutter (Reference Borbáth, Hutter, Giugni and Grasso2022) have made an explicit attempt to assess the relationship between electoral and non-electoral political participation (more specifically, between voting and lawful public demonstrations) at the level of individual participants. They argue that individual-level predictors cannot sufficiently explain the variation in the extent to which citizens combine voting and demonstrating, and that to understand this, we need to look at structural conditions and opportunities for participation. Most importantly for our purpose, they illustrate the blurring boundaries between electoral and non-electoral mobilization and its repercussions on individual-level behaviour, showing the extent to which individuals combine voting and attending demonstrations in their action repertoire.

Based on the above, we make the following propositions with regard to bridging institutional and non-institutional forms of participation:

• • Proposition 9: Institutional participation is positively correlated with non-institutional participation.

• • Proposition 10: Partisanship is positively correlated with protest participation.