Abstract
Participation in organizational democracy is viewed in widely different ways: both positively, as schools of democracy, and more negatively, as elite pools of democracy. However, most of the research on this topic has focused on participation in voluntary associations. This paper looks beyond the classic understanding of association democracy to explore the democratic governance in a church, a consumer cooperative, and in student organizations, and examines their links to Helsinki region city councils and parliamentary politics. Using public access data, we created a database of 4141 people to map how these democratic organizations connect with institutional democracy through a social network of democratically elected representatives. Drawing on social network analysis and the literature on voluntary associations, our study revealed a tightly interlocked interorganizational network and an interpersonal network of democratic elites in Finnish society.
Introduction
In Nordic societies, democratic organizations form a substantial part of our daily lives. We live in housing cooperatives (Kohl and Sørvoll, 2021), buy our groceries from consumer cooperatives (Ekberg, 2017), and entrust our finances to credit unions (Spicer, 2022)—all of which are governed by representative councils elected by members. Even religion, in the Lutheran view a very private matter, is infused with the democratic spirit: National churches, although somewhat separate from the state, are run by a large body of elected representatives who convene to decide on both theological and secular church matters (Nelson, 2017).
What does the democratic nature of these organizations mean for democratic participation and political culture? This question is not limited to Nordic societies, although it might have a pronounced resonance here (see e.g. Alapuro and Stenius, 2010). The current body of literature on extra-parliamentary democracy has mostly drawn from studies of voluntary associations.
The neo-Tocquevillian view emphasizes how voluntary associations build and maintain a space in which local democracy can prosper and in which more inclusive participation and enlightened self-interest, in the form of civic action, not only make life better but teach citizens the skills for being active members of a democratic society (Dodge and Ospina, 2016; Eliasoph, 2011; King and Griffin, 2019; Skocpol, 2003; Warren, 2001). A more critical perspective, informed by recent European research, argues that the civic-minded individuals who gravitate toward associational life are already active in political life, thus voluntary associations are more like ‘pools of democracy’ (van der Meer and van Ingen, 2009). However, many are skeptical of the assumed link between associations and democratic participation (Burrmann et al., 2019; Ibsen et al., 2019; Jo, 2020; van der Meer, 2016).
Recently, scholars have called for both conceptual and operational revitalization of civil society and the public sphere in political sociology, to expand the idea beyond universal, a priori categories toward a more grounded, historically located and relational understanding (Egholm and Kaspersen, 2021a, 2021b; Stewart and Hartmann, 2020). Inspired by this, we looked beyond the classical understanding of voluntary associations to explore how three democratic organizations connect with institutional democracy through a social network of elected representatives in the greater Helsinki region. Drawing methodologically from social network analysis and the literature on voluntary associations, as well as institutionalist perspectives on local governance and elite networks, our study shows extensive interlocks between democratic organizations and institutional politics.
Background
Our case consisted of representatives of three democratic organizations with elaborate democratic structures, extensive membership, a strong ethos, and considerable influence within politics in the Helsinki metropolitan region in Finland: the consumer cooperative HOK(Helsingin Osuuskauppa)–Elanto, the Lutheran Church, and university student unions. We examined the social network of thousands of elected representatives in these organizations and combined it with the elected representatives in the Helsinki, Espoo, and Vantaa city councils, the Finnish parliament, and the European parliament. Despite operating in vastly different fields, these organizations share a set of features, on the basis of which we selected them for the case. They all have massive membership bases (tens to hundreds of thousands of members), they are all governed through electoral democracy modeled on similar political cultures and practices, and they are all committed to open governance, which allows public access to data. The similarities meant that we were not looking at a haphazard collection of relations but a network of relations across organizations. Empirically, we focused on the interlinkages between these three organizations and their official politics in our data set, which comprised all the members (
Finnish society and political life is characterized by highly organized relations between official politics, the state, and non-governmental organizations. Contrary to more contentious political cultures, in Finland, civil society organizations are recognized and well represented in decision-making and governance, and the state and associations have a longstanding tradition of mutual support. This tradition has also been extensively researched (e.g. Alapuro, 2005, Luhtakallio, 2012) and provides a clear picture of the role of voluntary associations in Finnish national politics. Membership of voluntary associations in Finland is high, but most members are not involved in the everyday running of the associations, creating a culture of passive membership throughout civil society (Alapuro, 2005). Elections tend to follow the same pattern: Everything is highly organized, voter turnout is relatively high, but most citizens play a passive role in campaigning. This aspect of Finnish political culture is reflected in the three non-governmental organizations we look at in this paper.
The results show extensive interlinkages between these organizations and official political institutions at the local, state, and in our case, EU level. This connectedness shows that civic-minded, democratically governed voluntary associations can network with institutional politics in many ways: In our empirical setting, their connective patterns suggest ladders to political careers as well as grassroot activism in official politics, but also notable consolidation of power among local elites.
The paper proceeds as follows. First, we look at the current literature on social network studies, more precisely on the topics of urban politics, voluntary associations, and elites. Then, we present our empirical case and findings from a social network analysis of the dataset. Finally, we conclude by discussing the implications of our exploration for future research in social networks and voluntary associations.
Voluntary associations and civil society
A central argument to the Tocqueville-inspired project of democratic theory has been that voluntary associations are significantly linked on a macro level to the well-being of democracy. According to the theory, vibrant civic life in associations produces “more democracy in more domains of life,” building, spreading and enabling democratic citizenship capacities (Warren, 2001: 14). Associational life both inspires and teaches citizens about the public good, representative democracy, and the articulation of heterogenous voices in society. Being a member of a large-scale association is also credited with making meaning in the civic lives of individuals; not only honing civic skills but also creating the feeling of being a part of something bigger and working together toward a civic good (Skocpol, 2003). Membership in voluntary associations has been a core feature of Nordic societies (Alapuro and Stenius, 2010; Wollebæk et al., 2010), and its decline in the US has famously been credited as a harbinger of fundamental changes in society, in line with Putnam (2000).
A longstanding critique of the neo-Tocquevillean theory of associations and democracy is that it lacks any clear mechanisms beyond macro-level correlations. Closer empirical examinations claim that, rather than associations making active citizens, it is the already active small minority of citizens that gravitate toward associational life—they are “pools of democracy,” not schools of democracy (van der Meer and van Ingen, 2009). The importance of the macro-level link to the general argument however has resulted in an imbalance in the research literature, as the internal workings of association democracy have been studied much less from the perspective of the sociology of democracy (Ibsen et al., 2019).
The research on civic organizations and networks argues that they have a tendency to link together in ways that are conducive to collective action (Baldassarri and Diani, 2007). These linkages are formed on the basis of issues, informal collaboration and transactions, social tie bonds, and shared individual members, either on the rank-and-file level or on the directorial level (Baldassarri and Diani, 2007; Cornwell and Harrison, 2004). The approach shows these links in a positive light as the
Networks and democracy
Modeling the connections of persons in groups, in his work, Ronald Breiger emphasized the inherent duality of social networks in which interorganizational and interpersonal networks are mutually constitutive (Breiger, 1974; Breiger and Ennis, 1997). This is an important notion, as the social network of democratic organizations in our case is very much dualist in nature. The interpersonal network of elected representatives and the interorganizational network of elected organs of democratic governance are two sides that can only be separated from each other analytically.
Prior research has also studied the effects of formal and informal networks on local governance in metropolitan areas. Rooted in the theoretical framework of institutionalist approaches, particularly the collective institutional action approach (Feiock, 2009), these studies primarily investigate the connections between policy networks, elected officials, and governance institutions. The shape that these networks take vary, their topology depending on the type of transactional “goods” or “services” that flow within them (Andrew, 2009) and the building of trust and sharing of information to reduce uncertainty (Feiock et al., 2010). Studies using the institutional action approach share the view that the effectiveness of policies and democratic decision-making is hindered by the fragmented nature of metropolitan governance, which assumes that self-organized networks will emerge to tackle these problems of interorganizational coordination and cooperation.
Another strand of research on governance gives social networks a different role. It sees them as the “bottom-up, self-organizing” side of local democracy that stands against the “top-down policy frameworks” created by local governance (Bramwell, 2012). This approach is important, because it highlights the role that the actors and institutions outside official governing structures play in the totality of governance. A networked governance model continuously negotiates the relations between different institutions and can thus create a space in which the decision-making includes heterogeneous voices and better sensibility than top-down policy frameworks (Rosenthal and Newman, 2019). At the local level, the mapping out of the network connections of neighborhood-based civic organizations has demonstrated their importance for urban political participation (Lelieveldt et al., 2009).
The network of elected officials in democratic organizations also constitutes a type of elite network. The relatively open and loosely interconnected structure of the Finnish power elite has changed very little in recent decades (Ruostetsaari, 2013, 2015). Whereas holding a dual mandate in parliament and a municipal council is common in European countries (Hajek, 2017; Navarro, 2009, 2013; Van de Voorde and de Vet, 2020), it is strikingly ubiquitous in Finland (Arter and Söderlund, 2023). Although these findings shed light on the features of institutional politics and the electoral system, they do not address the networks between the grassroots civil society and institutional democracy, suggesting that these relationships need to be further examined.
Our thinking in this paper stretches the classical notion of civil society organizations. In the next section, we look in more detail at three organizations that are democratic and civic-minded but fail to tick all the boxes of the classic definition of voluntary associations.
First, the cooperative society is an economic enterprise that produces (mostly) economic services for its members. De jure, this does not have to be for-profit, but de facto, it has to be profitable to survive. This deviates from the classic definition of a voluntary organization—that it must be non-profit. Second, the church has devolved from the state church and still has several state-like features, most notably the right to collect taxes from its members. This deviates from the classic understanding of voluntary associations that emphasize their separateness from the state. In Finland, however, this separation has not been as strict as elsewhere (e.g. Alapuro, 2005). Third, membership of student associations is not voluntary; it is mandatory for all university students.
However, despite stretching the idea of voluntary associations, we argue that it is fruitful to compare these organizations to the voluntary associations in the literature, for two reasons. First, the organizations and their ethos present them as such. The cooperative self-identifies as a movement, the church as a civic organization, and the student associations as organizing and articulating the interests of students. Second, the voluntary association theory works for its purpose, as we attempt to show in this paper, offering a new point of view to the way in which these democratic organizations interact with politics and challenging the way in which we think about associations and democracy. Additionally, because we are interested in the interlinkages of the mandates and the elite network this produces, the structure, record-keeping, electoral system, and size of these organizations make them uniquely suited for this kind of study.
Thus, informed by the prior research literature on civil society networks, networked urban governance, and elite networks, we are primed to ask the following questions in our empirical setting.
First, if associations truly are the cement of civil society, what exactly do they glue together? Does the network data support both sides of the dualistic processes of “schooling and pooling” democracy in democratic organizations?
Second, how viable in our case is the idea that civic networks work as a counterweight to top-down policies? If we test the argument against our empirical data, how conducive are civic networks to grassroots political issues obtaining a voice rather than acting as merely another avenue for those already in power.
Last is the issue of elite consolidation in democratic organizations. In our Finnish case, in a culture strongly permeated by political parties’ organization, what does the topology of the network tell us about the role of elites in voluntary associations?
Case and dataset
Helsinki, the city of a thousand democratic institutions
This study was carried out in the Helsinki metropolitan area, which includes the nearby cities of Espoo and Vantaa. With around 1.5 million inhabitants, it is by far the most populous urban region in Finland, as well as the center for business, politics, and culture. We chose the following three civic-minded organizations with substantial membership in the region: the HOK–Elanto consumer cooperative, the Finnish Evangelical Lutheran Church, and university student organizations.
The consumer cooperative HOK–Elanto is wholly owned by its 600,000 members (covering 80% of households in the greater Helsinki region). Membership is open to anyone for a small fee and provides members with small discounts and a share of the yearly surplus that the cooperative distributes back to its members. As already mentioned, HOK–Elanto is the largest retailer in the Helsinki region, and operates hundreds of grocery stores, ranging from corner shops to hypermarkets, as well as restaurants, hotels, and gas stations.
For a long time, the consumer cooperatives have had tight connections to city councils and urban planners in the Helsinki region. Their influence on urban development and zoning is epitomized by Ylermi Runko, the longstanding CEO of Elanto, who chaired the Helsinki city planning committee from 1968 to 1992 (Komulainen, 2015). High-profile social democrats have been visible in leading cooperative positions, most notably president Tarja Halonen, who was a member of the Elanto council from 1975 to 2000. The Elanto and HOK cooperatives were instrumental in legitimizing the new consumer culture in post-war Finland; for example, they pioneered the first car-friendly supermarkets in the 1970s (Seppälä, 2018).
HOK–Elanto members elect a 60-member representative body every four years. The representative body elects the CEO and the supervisory board, who are responsible for the strategic and business management of the cooperative. In the latest elections in 2020, over 120,000 members cast their vote (HOK–Elanto, 2020), roughly a 20% voter turnout. This makes the HOK–Elanto elections one of the largest local elections in Finland in terms of votes cast (second only to the Helsinki municipal elections). The HOK–Elanto representative body has both direct and indirect power over what the stores, restaurants, and hotels offer, as well as over the regional expansion, zoning, and use of land that the cooperative seeks to develop commercially.
The second organization whose elected representatives were included in the dataset is the Evangelical Lutheran Church of Finland. The membership of the national church has waned steadily due to secularization and immigration, but still today, 69% of all Finns and more than 50% of Helsinki region residents are members. In the Helsinki region, local parishes are active in the civic sphere through youth work, social work, and providing support and facilities for smaller civic organizations.
The church has a complicated four-tiered democratic system and elections are held every four years to elect the lowest level, the parish council, which then proceeds to elect the higher representational levels. The highest national level is the synod, which convenes twice a year and handles many theological to practical issues (e.g. Bible translations, women in priesthood, same-sex marriages, or church legislation). Since parishes are a major provider of social services, such as children's afternoon clubs and food provisions, the lower levels of the democracy work close to the grassroots.
The third part of our dataset was student unions’ elected representatives. Student union membership is mandatory for all undergraduate students and voluntary for graduate students, creating a membership base of tens of thousands of students in the region. Representative councils, elected every two years, govern the unions and elect boards that oversee the strategy and everyday management. Research on student unions has revealed the student movement as a ladder in the career trajectories of Finnish politics, as well as an important driver of historical changes in political culture (Kolbe, 2008). The two largest student unions in the Helsinki region, from the University of Helsinki (HYY) and Aalto University (AYY) have, by historical chance, found themselves to be important real-estate owners and developers, with properties valued north of 500 and 100 million euros. In addition to their presence on the commercial real-estate market, student unions also support student associations, run major restaurant chains and provide healthcare for the student population. Student unions coordinate on the national level through national unions.
The last section of our data is official politics. We complemented our civic network by including elected representatives of political institutions at the city, national and European levels. This included representatives of the city councils of the region (Helsinki, Espoo, and Vantaa), the national parliament of Finland, and the European parliament. Throughout Finnish political history, the political system, including party organizations, has been in close contact with civil society and non-governmental organizations (Alapuro, 2005, 2006). Following the Northern European model of local governance, Finnish municipalities have autonomy from the state, the power to collect taxes, and can carry out wide-ranging tasks to provide public services (Haila and Le Gales, 2005: 118–119). As the political, economic and cultural hub of Finland, the Helsinki metropolitan area city councils are closely related to national politics and feature high-profile politicians.
Party affiliation manifests differently in our three case organizations. In the church institutions and student unions, most members campaign without explicit party affiliation, whereas in the city councils, practically every member has a party affiliation. HOK–Elanto elections feature party-organized lists and nominees are actively recruited through the parties’ organizations. Finnish general elections use a proportional representation system with open lists, meaning that voters vote for individual candidates from the parties and the party organizations cannot choose the ranking of these candidates. The institutional elections in our data used the same principle. Even though the parties in our case were ubiquitous, the local political culture severely limited their power and coordination.
Data types and collection of data
For our database, we collected network data on the democratically elected representatives in our case organizations (in technical terms, this is a positional strategy to network boundary specification; Knoke and Yang 2008: 16–45). Our dataset included the central bodies that were democratically elected in each of the organizations, resulting in a database of 21 different bodies. These are listed in Appendix 1. Deputy members were included in the organizations in which they played a meaningful role. In total, the database consisted of 8184 representative seats held by 4141 different people between 1999 and 2020.
The number of seats varied between the democratic organizations due to the different structures of their democratic governance (see Table 1). The most member-elected bodies were in the church democracy, whereas the HOK–Elanto cooperative had only one body and the fewest number of elected seats in the database.
Number of elected representative seats and elected persons in each section of the database.
The data were manually collected and checked for possible duplicate names or name changes. In such rare cases, we ensured their correct identification by confirming the institutional background of the representative(s) in question and cross-checking which organization they were part of.
All the non-governmental organizations in our data were publicly committed to Nordic-style transparency and openness in their democratic governance. In practice, however, with many of the selected bodies, data collection was not straightforward. Some data, ranging from Excel sheets to grainy scans of archived documents, had to be excavated from neglected organizational archives. Sometimes, our data requests were declined, or we never received responses.
The dataset we collected was almost complete. It contained every member of the included organizations, except for a few that were missing because of the poor archival availability in the Helsinki and Vantaa parishes in the early 2000s. The missing data were marginal and did not affect the overall analysis.
The dataset included several election cycles in all the organizations. This was important, as some network effects happen longitudinally. Regarding the interorganizational network, the time dimension gave us a solid robustness check, as the overall pattern of connectivity between the organizations remained relatively stable at the different timepoints, despite individual members changing. However, for the interpersonal network, the timeframe was not long enough for a closer analysis of career trajectories.
Constructing the network
Our data consisted of two kinds of nodes, elected representatives and the institutions in which they had been elected. This created a bipartite network of which the edges were the different seats any given representative held in this network at some point in the timeframe. This type of network structure is often called an
The reduction of a bipartite network is based on the idea of co-affiliation (Borgatti and Halgin, 2014). We reasonably assumed that there was a connection between two people if they had both been members of the same body within the timeframe of our dataset. Similarly, institutions are connected when they share members. In our analysis, we interpreted an edge in our bipartite network to represent a
When assessing the representativeness of our dataset it is worth noting that our database was neither a sample nor an estimated network created on the basis of a survey (see e.g. Diani, 2015); it was a complete set of all the individual representatives in these organizations from 2004 to 2020. Due to the relational nature of network data, statistical models that assume full data independence are not useful for testing the significance of most networks, as they would compare the observed results to outcomes that are not possible in real life (Croft et al., 2011; Snijders, 2011). Therefore, we opted to test the significance of our results using a Monte Carlo simulation that compares the likelihood of our observed sector-specific interorganizational links with simulated alternative outcomes. In the simulated alternatives, the sources of the edges were the same but the target nodes were randomized. The probability distribution in this test is relative to the size of the sectors, meaning that in the simulations, a link to church entities is more likely than a link to HOK–Elanto. For a full breakdown of the simulations, see Appendix 2.
Based on this simulation test, it appears unlikely that the connections we found are random. In fact, most of the connections between the 21 boards had
Results
This section continues to follow the dualist nature of the social network. First, we look at the overall structure of the interorganizational network to explore the connections of the three democratic organizations with institutional politics in the Helsinki region. Then, we look at the interpersonal network—the connections between individuals mediated through these institutions—to explore the personal network of the elected representatives through their positions of power in democratic organizations.
The interorganizational network
Figure 1 illustrates the interorganizational network of all the boards in our dataset. It shows all the connections as shared elected members at some point (but not necessarily simultaneously). The thickness of the lines between the organizations represents the number of shared members. The greater the number of connections, the thicker will be the lines.
Interorganizational network of democratic organizations and institutional politics in the greater Helsinki region 2000–2020. Church organizations, including local parish councils, city-level parish union councils (JCPU), regional church councils (the Diocesan chapter), and national councils (Synod and Church Council), are squares. Student associations are circles, city councils are triangles, and parliamentary organs are pentagons.
Here, we see the general topology of the network. The core of the network consists of key political institutions, the city councils and the Finnish parliament, whereas the more grass-root bodies in the church democracy and student unions are positioned at the periphery. The church organization was well connected to established politics, but the connections were highly local in their respective cities and reached national politics mostly through city council members. In comparison to the literature on civil society and democracy, our results show that the network analysis supported the notion of the church as grassroot democracy (Pessi and Grönlund, 2012), which engaged many people and connected with politics through the local or city level.
In contrast, the consumer cooperative HOK–Elanto held a much more central role in the network. This is a notable finding, as it was also the organization with the fewest people in our dataset. The cooperative's representatives were extremely well connected to city councils and parliamentary politics but had weaker connections to church democracy or student politics. The literature on the HOK–Elanto cooperative in Finnish politics notes some top-level connections (Komulainen and Siltala, 2018) but our results show that the cooperative democracy is connected to political institutions in a way that goes beyond anecdotal evidence, and this calls for further research on the cooperative as a democratic organization.
Our data showed a link between student democracy and official political institutions but it was not as strong as that claimed in the literature on the Finnish student movement (Alapuro and Stenius, 2010; Kolbe, 2008). The student organization connections with other institutions persisted over time but were limited in number, illustrating the weak ties one would expect from typical “feeder” organizations in career politics. This also brings us to the limitations of our dataset. Due to our limited timeframe, we were unable to provide a full analysis of the career trajectories across different sectors.
The interpersonal network
Next, we look at the way in which the individuals in our data were connected in the network. In the literature, we found varying notions of what participation in democratic organizations produces for civil society. On the one hand, more democracy creates more democracy, bringing people into grassroot civic life and schooling them in ways of collective decision-making (Putnam, 2000; Skocpol, 2003; Warren, 2001). On the other, more democracy often translates into more democracy for the same set of active participants, creating a pooling effect of elite consolidation rather than schooling a wider range of people (van der Meer and van Ingen, 2009).
In our case, we saw that the distribution of the seats between the people in the network was uneven. On average, each person in this dataset was a representative or board member for two terms (either two terms in the same seat or one term in two different seats), about 60% of the people in our database only had one term in one body (see Figure 2). The top 10% had three or more seats, and 276 people also sat five or more terms. Here again, the church and student organizations were at the grassroot fringes of the network, and only a few people were connected to the more elite organizations such as the Finnish parliament.
Distribution of seats per individual in the database.
Of the 10 most connected individuals, nine were Members of the Parliament of Finland and one was a former chairperson of the Helsinki City Council. This was not party-specific, the top 10 included people from all except one of the major political parties (the populist–nationalist Finns Party only grew to be a major player in the last years of our timeframe). Of the 88 people with three institutional affiliations, over 70% (63) were either current or former MPs or MEPs. This, at least in our case, demonstrates that the top stratum of democratic organizations is deeply integrated into hard political power and does not form any sort of counterweight to it.
The memberships in our database were also highly concentrated within one board and one section. About 80% of the people were on only one board (for one or more terms, see Figure 3), whereas less than 5% were on three or more different boards (not necessarily at the same time). When the 21 boards were grouped into sections according to type of organization, about 90% of the people were members of boards within only one section (see Figure 4). According to the “schooling” argument, the sections with the largest number of representatives would be the most likely to teach the participants something about democracy in organizations, but this did not appear to spur them into continuing to participate in it.
Distribution of representative bodies held by the people (for one or more terms) in our database.
Distribution of sections in which a person held a seat in our database. The sections are all the representative boards grouped as belonging to the church, student organizations, the consumer cooperative, city councils, or parliaments.
Instead of a bottom-up network of organization democracy activists as a counterweight to institutional politics, the data showed signs of a small group of super networkers holding positions in all sectors. This elite formation tendency appears robust if we look at the turnover of people at five-year intervals, as only a small group of people were part of the network for an extended time, staying with these institutions for years and likely accumulating power along the way. This suggests that in Finland, a type of democratic elite exists, for whom more democracy produces more positions of power.
Student unions, on the other hand, showed a connective pattern in which they fed active participants to larger, more central organizations. The link to democratic institutions was constant but weak and illustrated, in a network form, the idea of voluntary associations as schools of democracy (Putnam, 2000; Skocpol, 2003; Warren, 2001) by identifying important feeder organizations. In our case, the ladder effect would likely have been more prominent if we had had a longer time to cover more of the careers of the representatives in our dataset.
Conclusion
In this paper, we explored the social network of representatives in democratic organizations. The network was inherently dualistic, consisting simultaneously of two networks—persons and organizations—both of which were mutually constitutive.
First, our results show emphatically that the democratic organizations on the border of or beyond the classical definitions of voluntary associations have extensive interlocks with the political establishment in the Helsinki region. The church democracy and student organizations showed connective patterns typical of grassroot civil society organizations, engaging many different individuals and connecting mostly with city-level politics. In contrast, the HOK–Elanto consumer cooperative democracy consisted of representatives who were very well connected to political organizations at the city and national levels. This offers new insights into the literature on democratic organizations, as the relation between HOK–Elanto and Helsinki politics has previously only been discussed on an anecdotal basis.
Second, our analysis of the personal network in our case showed a tendency toward elite consolidation of positions of power, even on the somewhat obscure representative boards of these democratic organizations. Although the church democracy and student organizations engaged many people, only a few of them made the jump to local or national politics in the 20-year timeframe of our database. This contradicts the common view in the Finnish civil society literature (e.g. Kolbe, 2008) of student organizations as a central ladder for political careers.
Civil society discussions often refer to democratic decision-making in voluntary associations as schools of democracy. Our network analysis presented some critical issues regarding this notion. First, a spectrum of democratic organizations wider than classical voluntary associations needs to be studied in more detail. In our data, the consumer cooperative was very well connected to core politics—yet we found practically no contemporary research on what goes on within the cooperative democracy and what civic or political implications it has. Second, what exactly did the participants learn in these schools? As others have pointed out, active volunteering does not necessarily make active politics (Burrmann et al., 2019; Jo, 2020), and our dataset of thousands of people with a one-and-done term in democratic organizations suggests a similar trend. Further, more qualitative research is needed on whether the lessons that participation in democratic organizations teaches are conducive or corrosive to democracy at large. Echoing the calls for a post-sectoral concept of civil society (Egholm and Kaspersen, 2021a, 2021b), we call for a review of the established ideas of how the associations between democratic organizations and politics interact in Finland, and more broadly in the Nordics.
Footnotes
Acknowledgements
Juulia Heikkinen would like to thank Juha Heikkinen, Research Professor, applied statistical methods, Natural Resources Institute Finland (Luke) for his hugely helpful advice on the methods and analysis in this paper.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Koneen Säätiö.
Author biographies
Appendix 1.
Institutions, representative bodies, and public access data sources of database.
Appendix 2.
Strength of ties between organizations, estimated by comparison to 999 Monte Carlo simulations of the distribution of the same number of connections over the same number of nodes. P values indicate the results are statistically significant.
Results of Monte Carlo simulation on sector-specific interorganizational edges. Columns b and c show the maximum and minimum values of the simulation results (i.e. sampling distribution) with 95% confidence intervals (2.5% off both tails). If the observed values (column A) fall outside the confidence interval, the observed results are statistically significant. This is also expressed by the
Appendix 3.
Number of different bodies, individuals and seats in the database betweenness centrality (degree to which nodes stand between each other) and closeness centrality (degree of how close the node is to other nodes) for the different boards calculated.