How can political parties integrate today?

Introduction

The integration of multiple, sometimes quite heterogenous, groups of citizens into the political process used to be one of the primary functions ascribed to political parties. In Europe, the parties that have historically achieved this feat were the ‘mass parties of integration’ that emerged in the early and mid-20^th century, in tandem with the advance of mass democracy, as well as the ‘people’s parties’ of the post-war era. Today, it is widely doubted that parties can still perform such a broadly integrative role. The decline of traditional class- and religious milieus, growing political polarisation and economic inequality, and an increasing tendency among citizens to prefer more individualised forms of participation – these are just some of the reasons why contemporary parties struggle to make citizens of different backgrounds feel like they are part of a common endeavour.

The purpose of this research note is to examine parties’ changing capacity to integrate citizens into the political process. Deliberately painting with a broad brush, I trace the main transformations from the age of mass parties until today, and discuss the new challenges facing parties that seek to appeal to wide and diverse constituencies – perhaps most of all social democratic parties. In closing, I make some suggestions as to how the contemporary integrative potential of parties could be studied as part of an alternative, more sociological research agenda on political parties.

Integration in the age of mass parties

Marxist thinkers were among the first to systematically theorise the integrative function of political parties. Antonio Gramsci (1971: 5-23, 30-31, 123, 168, 340) in particular argued that politically significant social blocs are constituted by parties, and not vice versa. Accordingly, parties can ‘dis- and rearticulate social groups partly by producing organic intellectuals who foster class alliances and cultivate the “good sense” of the masses’ (Mudge and Chen, 2014: 309). A few decades later, this theme resurfaced (in a less normatively-loaded language) in mainstream political science. Reflecting on the rise of the large labour and religious parties in the first half of the twentieth century, influential figures such as Maurice Duverger (1954), Otto Kirchheimer (1967) or Sigmund Neumann (1990) spoke of the emergence a new party type, the ‘mass party of integration’ or just ‘party of integration’. What makes this party form distinctive, wrote Neumann (1990: 47), is that it ‘demands not only permanent dues-paying membership … but, above all, an increasing influence over all spheres of the individual’s daily life’.

What Neumann meant was that the parties’ organisation reached, as it were, from the cradle to the grave. In the case of the continental European socialist parties, for example, it extended ‘from the workers’ infant-care association to the atheists’ cremation society’ (Neumann, 1990). Thus, the primary way in which the parties of integration made their constituents regard themselves as implicated in a common endeavour was by establishing and maintaining a wide-ranging network of party-related associations that foster and reinforce a sense of identity and community. These so-called milieu organisations were typically ‘more prominent than the local party branches’ (Scarrow, 2014: 162) and allowed the party’s followership to pursue many or indeed most of their everyday activities under the aegis of the party. In a sense, the party of integration really offered something for everyone: one could join one of the party’s football or gymnastics clubs, it’s women’s or youth organisation, and so on.

Of course, not all parties of integration needed to build a large network of milieu organisations from scratch. The early Catholic parties of integration, such as the German Zentrumspartei and later the CDU, could rely on the identity-building and mobilising capacities of pre-existing Church communities and religious associations like Catholic journeymen’s unions, fraternities, etc. (Bösch, 2002: 192) As one authoritative account of the Italian Democrazia Cristiana in the early post-war years stresses, ‘the party organisation remained weak, and the [Democrazia Cristiana] continued its heavy dependence on the “indirect party”, i.e. the Catholic organisations, to mobilise the vote during election campaigns’ (Leonardi and Wertmann, 1989: 126). In fact, when it came to shaping and sustaining a common sense of belonging among their constituents, and getting the latter to cast a vote for the party, these organisations proved just as effective as – if not more effective than – those specially established by the socialist parties.

The large Socialist and Catholic parties are the most prominent examples of parties of integration, but they were of course not the only parties that integrated different groups of citizens into the political process. For instance, agrarian parties – the most notorious being perhaps the American People’s Party of the 1890s (Kazin, 1995) – provided a way for the lower classes in rural areas to act together and make their voices felt. At any rate, what is important to note is that most parties, however great their integrative capacity, did not include those whom they integrated in intra-party decision making. As Neumann (1990: 47) observed about the parties of integration, while they take on ‘an ever increasing area of commitments and responsibilities assuring the individual’s share in society and incorporating him into the community,’ it is ‘only a small active core’ that decides on the party’s policies and more general direction (the classic study is Michels, 1989).

In addition to drawing on the support of milieu organisations, the parties of integration successfully availed themselves of at least three other integration strategies. The first and perhaps most common one was conjuring up shared enemies, that is, producing a collective ‘we’ by eliciting hostile reactions vis-à-vis an external other. This strategy was highly effective as a way of integrating diverse constituencies, especially because appealing to shared enemies rang plausible to many in an age where both the traditional class cleavage and the ideological struggle between East and West still were politically salient. For example, many commentators agree that the early CDU managed to unify the traditionally divided German Catholics and Protestants primarily because it continuously conjured up the common enemy of communism, stressing ‘the difference between Marxist “materialism” and Christian principles, and the need for all Christians to recognize their common interest in opposing communism’ (Granieri, 2004: 55).

Secondly, the parties of integration could make people feel like they are part of a greater collective endeavour by organising and orchestrating ritualistic practices, or profiting from the identity-building force of ritualistic practices performed in the wider milieu of the party. Think, paradigmatically, of the annual Labour Day celebrations; many socialist and social democratic parties saw (and still see) it as their responsibility to arrange these festivities, bringing together all their members and supporters in a joint affirmation of a shared political identity. Many of the Catholic parties of integration profited from identity-reinforcing practices enacted in local communities, ranging from the celebration of official Catholic holidays to particular rural festivities (Walter, 2009: 30). There was accordingly little need for those parties to ‘invent’ their own holidays or directly mobilise their own resources.

A third integration strategy to unify disparate segments of voters may be called being everything to everyone. This strategy tends to be associated with the 1960s and 70s ‘catch-all party,’ which is said to lack a distinct ideology and clearly identifiable constituency (Kirchheimer, 1967); but it is difficult to deny that many prominent parties of integration also integrated people in this way. The strategy mainly involves pandering to a range of different demands in order to attract different groups of voters; it is about gratifying as many people as possible. A French newspaper critically noted in 1946 that the German CDU is ‘socialist and radical in Berlin, clerical and conservative in Cologne, capitalist and reactionary in Hamburg, and counterrevolutionary and particularistic in Munich’ (cited in Granieri, 2004: 14). But this proved to satisfy very diverse constituencies, instilling in them a sense that the party really took their concerns seriously.

Integration in the age of ‘cognitive mobilisation’

Obviously, a lot has changed since the mid-twentieth century, when the parties of integration and ‘people’s parties’ had their heyday. The story of the fragmentation of parties’ social bases and the individualisation of mass publics has been told almost too often to bear further repetition. But it is still worth underlining that these twin tendencies are widely seen as amounting to a loss for democracy, precisely because they undermine parties’ capacity to integrate (e.g. Mair, 2013; Streeck, 2014). The thought is that milieu organisations, ritualistic practices and all sorts of group-based appeals lose their integrative force when citizens do not perceive themselves as belonging to a particular social group and sharing a particular way of life or core values with others.

Yet, any uniformly pessimistic interpretation of the decay of social segmentation and the increasing tendency of individualisation overlooks two things. First, it might well be that those who relate to politics in a more individualistic fashion and view their political allegiance in terms of a choice among alternative options could still be integrated using integration strategies that appeal to their individualised self-understanding and new participatory demands. Second, even if traditional milieus and political loyalties have largely disappeared, there are still large cohorts of voters who strongly identify with particular collectives, and who are also responsive to some of the aforementioned ‘classic’ integration strategies. While it is true that the social base of many parties ‘may no longer be amenable to the kind of collective action that parties traditionally inspired’ (Streeck, 2014: 127), new forms of social identity-based collective action can be traced.

The first thing to note in relation to possible new strategies of integration is that the gradual decline of parties’ social bases and the individualisation of mass publics have often been shown to correlate with a development that many sociologists and political scientists have described in terms of ‘cognitive mobilisation’. One influential scholar describes ‘cognitive mobilisation’ as involving the following developments:

First, the public’s ability to process political information has increased, as a function of higher levels of education and political sophistication among the electorate. Second, the cost of acquiring political information has decreased, such as through the expansion of the mass media and other information sources. Cognitive mobilization thus means that more citizens now possess the political resources and skills that better prepare them to deal with the complexities of politics and reach their own political decisions without reliance of affective, habitual party cues or other external cues. (Dalton, 2007: 276)

Typically, this is described as a general trend that unfolded over the second half of the twentieth century – and one that maps unequally onto citizenries. In particular, younger and more educated people tend to be more ‘cognitively mobilised’ in the just-described sense than older and less educated ones (Henn et al., 2018; Inglehart and Welzel, 2005; Li and Marsh, 2008; Milburn, 2019). Amongst those cognitively mobilised citizens who have an interest in political participation, moreover, many tend to demand ‘more individualised and direct forms of political participation’ that allow for greater self-actualisation (Gauja, 2015: 89). As Henn et al. (2018: 713) note, they exhibit a ‘tendency towards support for, and participation in, new styles of non-institutionalized political action that better fit their individualized life-styles and permit the actualization of their political aspirations’.

Some parties respond to these new participatory demands and capabilities by supplementing integration strategies that are aimed at the affective construction or affirmation of collective identities with strategies that give citizens room for voicing and shaping their views and connecting these views to collective decisions (Gauja, 2015; Invernizzi-Accetti and Wolkenstein, 2017; Wolkenstein, 2019). This typically involves diffusing power beyond the ‘small active core’ (Neumann, 1990: 47) of decision makers that tended to rule parties since the age of the party of integration, and doing so in ways that ‘cognitively mobilised’ citizens consider meaningful (Invernizzi-Accetti and Wolkenstein, 2017: 106).

For example, some parties make space for, and empower, ‘movements’ within parties that are driven by citizens who are committed to the party’s broader values and aims but want to transform the party in a bottom-up fashion. An instructive example is the Momentum movement within the British Labour party. This has contributed significantly to the re-politicisation and mobilisation of (especially young) people by establishing new, local fora of political discussion and debate that proved more dynamic and inclusive than traditional party branches. Animated by the notion that ‘politics as a spectator sport has lost traction with voters’ (Oltermann, 2018), Momentum also coordinated activities like phone canvassing, local campaigning, and even educational events where expert speakers could discuss current political affairs with lay audiences. All of this proved attractive for those favouring more or less non-institutionalised and highly self-actualising forms of political participation (Muldoon and Rye, 2020). It allowed them to take ownership of the party as a shared political project through directly engaging in discussion and debate.

As political theorists have recently suggested, parties seeking to integrate ‘cognitively mobilised’ citizens could also more systematically institutionalise non-conventional participatory channels that allow their members (and maybe also unaffiliated supporters) to voice their views and discuss them with others, either with the ‘positive’ aim of developing shared political agendas or the ‘negative’ aim of criticising those in power and holding them to account (for more discussion and examples, see Wolkenstein, 2016; Invernizzi-Accetti and Wolkenstein, 2017; Wolkenstein, 2019). Similar to the empowerment of intra-party movements, this might go some way in reconnecting certain citizen cohorts – notably young people without traditional party identities – to democratic political processes. A happy side-effect of this might be that even some of those who do not wish to participate themselves will evaluate parties more favourably. After all, evidence suggests that many citizens, especially politically disaffected ones, tend to think that open and democratic procedures of internal-decision making within parties are normatively desirable and evaluate internally democratic parties more positively than ones that are organised in a top-down fashion (Close et al., 2017).

Turning now to the second issue that I noted earlier, those who argue that social bonds and related political commitments are today exclusively ‘a matter of taste and choice rather than of obligation’ (Streeck, 2014: 126) also often overlook that there are still large numbers of citizens whose political self-understanding is tied to strong feelings of group loyalty. Indeed, despite the disappearance of traditional partisan milieus and identities, and despite the just-discussed trends of cognitive mobilisation, these group-based modes of political engagement remain pervasive in established democracies (compare the influential accounts by Achen and Bartels, 2016; Cramer, 2016; Gest, 2016; for an in-depth study of group loyalty among party members, see Wolkenstein, 2019). Cognitive mobilisation is very real, but it has far from crowded out identity-based and affective mobilisation.

Besides accounting for the voting behaviour of some of the older, still-loyal voters of long-standing class or religious parties, identity-based and affective mobilisation explains in large part the much-discussed rise of so-called ‘populist’ parties and leaders. Those parties and leaders tend to ‘emphasize a cultural cleavage, the national, ethnic, religious, or cultural identity of the “people” against outside groups who allegedly pose a threat to the popular will’ (Rodrik, 2017: 22). And here, they typically use the traditional integration strategy of conjuring up shared enemies that we have encountered earlier. As Rodrik observes,

[i]n the US, Donald Trump has demonized at various times the Mexicans, Chinese, and Muslims. In Europe, right-wing populists portray Muslim immigrants, minority groups (gypsies or Jews), and the faceless bureaucrats of Brussels as the “other”. (Rodrik, 2017: 22)

Just as with the shared enemies that politicians of the post-war era sought to construct, these ‘threats’ or ‘enemies of the people’ are evoked to produce divisive emotions that unify and mobilise specific constituencies (Richardson, 2019).

Can today’s parties integrate everyone?

All of this raises a difficult question: Could contemporary parties combine different strategies of integration, such that they make both those who are responsive to identity-based and affective appeals and the ‘cognitively mobilised’ feel like they are part of a shared endeavour? To be sure, not all parties might want to achieve such broad integration in the first place. For example, some minor parties in proportional representation electoral systems may limit their integration efforts to a small and homogenous constituency. But most parties will aim for more and try to reach out to a wider group of citizens. This is true not only for large ‘mainstream’ parties that have a history of representing diverse voter groups; research suggests that even niche parties that limit their platforms to very few or just a single issue often seek to appeal to heterogenous groups of voters that ‘cross-cut traditional partisan alignments’ (Meguid, 2005: 348).

Now, one obstacle for integrating very diverse voter groups arises from the fact that the different groups may not only relate in different ways to party politics, but also hold more or less irreconcilable views on salient political issues. This has been a major challenge for social democratic parties in particular, who often are incapable of reconciling the starkly diverging demands of younger, well-educated, highly skilled and mostly urban voters, on the one hand, and older voters with lower educational attainment and more specific skills, on the other hand (the standard account is Kriesi et al., 2008). What tends to divide the two groups are usually conflicting value-based commitments on such prominent issues as immigration, though one should be cautious with treating these as unconnected to economic grievances (Manow, 2018; Rodrik, 2017).

The example of the United Kingdom is instructive. Research suggests that value shifts that ‘shape the outlook of voters on a range of social and cultural issues, particularly on issues such as race and immigration, national identity, gender, rights for same-sex couples, Europe and ethnic diversity’ have created a divide between younger, more educated, urban voters – the group that is more likely to be ‘cognitively mobilised’ – and older, less educated and rural voters – the group that is more likely to be mobilised via appeals to collective identity (Ford and Goodwin, 2014: 278-279). Put baldly, what divides the two groups is that the former group regards the views of the latter group as ‘parochial and intolerant,’ while the latter group resents the former for its socially liberal views and supposed self-righteousness (Ford and Goodwin, 2014: 278-279).

Under such circumstances, implementing an integration strategy that empowers politically engaged members of the first group of citizens, and gives them power over the party’s political direction, is bound to further alienate the second group. For this would mean that the concerns of the second group remain unheard, while reinforcing that group’s sense of having no influence on the party anymore. This is exactly what happened in the British Labour Party when the political influence of the already-mentioned intra-party movement Momentum increased after the election of Jeremy Corbyn as party leader. As noted, Momentum mainly attracted younger and more educated citizens with socially liberal views; and the fact that the movement shaped the party line and Corbyn’s own views led older, more socially conservative Labour voters in rural areas to abandon the party. Reporting on Labour’s collapse in its former north-east heartlands in the 2019 UK general election, one journalist noted:

Talking to regulars the same allegations surface again and again. That Corbyn consorted with the IRA, that he is soft on terrorists. That he has remained silent on prosecuting veterans over the Bloody Sunday killings. The leader’s shifting agnosticism on Brexit [which was importantly influenced by the younger and more active party members], in this context, is portrayed as yet another failure of patriotism, just as symbolic as his unforgivable reluctance to sing God save the Queen at a Battle of Britain remembrance service. (Adams, 2019: np)

Given how much the two groups’ value commitments differ, it is difficult to see how Labour (or any other party that seeks to attract constituencies that are divided in this way) could successfully combine different integration strategies. There are for one thing no plausible candidates for shared enemies that could successfully be conjured up: while the younger, educated and urban voters might be inclined to assign to ‘big money’ and the (Conservative) politicians who act as its agents the role of a shared enemy, they would recoil at the thought of portraying immigrants or perhaps the EU in these terms – even if the less educated, older and rural voters that they tend to recognise as relevant constituents regard immigrants or ‘the faceless Brussels bureaucrats’ as major threat. For this reason, the party could also not try to be everything to everyone. It could only pander to the fears of the latter group of voters at the cost of limiting the influence of the former group.

That different, and divided, constituencies could be integrated by way of common ritualistic practices seems equally unlikely. First, even if older, formerly loyal constituents might still see the value of such practices and the collective identities they are meant to uphold, the younger constituents who have not been socialised to view themselves as part of a larger social group might see little value in traditions that seem like a relic from a time long past. The latter might develop new ritualistic practices that serve a similar purpose, or ‘re-purpose’ other collective practices in their milieu, such that they serve identity-formation (think of how in 2017 thousands of young festival-goers were chanting ‘Oh, Jeremy Corbyn!’ at Glastonbury festival) – but this might in turn exclude those who are not part of the same cultural milieu.

Second, research suggests that the different constituencies we are talking about are typically geographically divided, since younger, more educated and socially liberal people are more likely to live in cities while older, less educated and socially conservative people are more likely to live in the countryside (e.g. Maxwell, 2019). It seems plausible to assume that this geographical divide makes it harder to develop common ritualistic practices and traditions that parties could exploit in order to make very different kinds of people feel like they are part of a larger collective endeavour. There is not just little that connects the groups in question in terms of political commitments, they also tend to be spatially disconnected from one another.

Towards a renewed study of integration

As I have already noted, social democratic parties are probably most heavily affected by the trends I have discussed. The deep ideological gulf that often runs between their erstwhile constituencies and their new supporters makes it especially difficult for them to successfully employ multiple strategies of integration and integrate very diverse constituencies. But there is no reason to think that other parties remain unaffected. Conservative parties might on average find it easier to integrate diverse constituencies, not least because they have always drawn on the support of multiple groups of citizens and refrained from presenting themselves as the political arm of a particular class or other social group; however, to the degree that they are also losing voters to the various ‘anti-establishment’ parties that so effectively attract former social democratic loyalists, they will no doubt be confronted with the same challenges as their social democratic rivals.

It might be tempting to view these quite fundamental challenges purely as a matter of parties’ policy priorities, assuming that broad integration can be achieved by promoting those policies that appeal to the broadest group of voters (a recent book-length statement of this quite common view is Rosenbluth and Shapiro, 2018). But to reduce integration to a matter of individual policy preferences is to discount the affective and sociological bases of some forms of partisan mobilisation (Achen and Bartels, 2016; Richardson, 2019; Streeck, 2014: 127), as well as the fact that growing numbers of ‘cognitively mobilised’ citizens demand an altogether different way of organising politics (Henn et al., 2018; Milburn, 2019). Thinking of integration in such a reductive way might well be in conformity with the ‘relatively asociological paradigm’ (Mudge and Chen, 2014: 311) that dominates political science, but it overlooks important complexities that parties must navigate to achieve integration.

Most likely because of the ‘asociological’ way in which most research on parties is conducted, we currently know relatively little about these complexities. True, we know something about rural and working-class milieus who have increasingly come to resent ‘liberal elites’ and vote for anti-establishment parties and candidates. Pioneering this line of research were scholars like Eribon (2010), Cramer (2016), Gest (2016) and Hochschild (2016), all of whom have shown that place-based (e.g. rural) and class identities that tend to be only loosely connected to policy preferences profoundly influence how people understand politics and, by extension, relate to broader partisan political projects. We still know little, however, about emerging social milieus that give rise to new partisan identities, and their link to and interaction with those more traditional rural or class-based milieus.

To better understand the difficulties that contemporary political parties face in making different and heterogenous groups of citizens feel part of a shared political endeavour, more extensive research on different ‘socio-moral milieus’ (Lepsius, 1966) and their inter-linkages is needed. Of utmost relevance seems to be the question of where and how different milieus overlap (if they do). For instance, where do the ‘cognitively mobilised’ younger and urban voters with socially liberal views meet and engage with older, rural and socially conservative voters who resent all things urban and liberal? What practices and social spaces, if any, do they share? To give a satisfying answer to the broader question of how political parties can integrate today, we will have to answer those questions first.