Introduction

The role of Catholic and Communist political traditions in forging the political and social sphere was a debated issue through the 90s, only to be almost forgotten in recent years. This debate, at the European level, reached its peak in the early 90s after the fall of the Berlin wall, but it focused on Eastern European countries (above all Poland and the Czech Republic) and, more rarely, on Western ones (Cento Bull 2000; Magner 2005; O’Mahony 2003). The debate was shaped differently according to the different roles played by Communism in the two blocs. This article foregrounds the Polish and the Italian examples in order to highlight the structure of the debate and to assess the current role of the political cleavage at the social and political level.

While Polish (like Hungarian) civil society was publicly perceived as one of the decisive forces contributing to the fall of the Communist regime—see for example the wide literature on Solidarność and its roleFootnote 1—the legacy of the regime in post-1989 civil society is more debated. Indeed, many authors stress the relevance of Polish pre-1952 civil society to account for the emergence of Solidarność (for example Frentzel-Zagorska 1990); in contrast to this observation, Magner (2005) argues that as the communists gained power, a large change was occurring within the sphere of associational life. This change was so radical that presently there is no link between contemporary organizations and pre-Communist ones, while the influence of post-Communist practices is still present. Nonetheless, this vision is often challenged by those who consider the influence of Communism limited if not irrelevant altogether (Gąsior-Niemiec and Gliński 2006) in modern Polish civil society.

As regards Italy, the genesis (but not the structure) of the debate is diverse. According to Cento Bull (2000), the (mainly) Italian debate on social identities and political culture was forged in Italy in the early nineties by three factors: the collapse of the old political system ‘which mirrored that of political regimes in Eastern Europe’ (ibid: 1), the Tangentopoli scandal (which forced a renewal in the party system) and the rise of Lega Nord (Northern League). The overall debate was about the normalizing electoral behaviour of Italian citizens. Scholars such as Sani (1992) and Pasquino and McCarthy (1993) stressed the importance of the 1992 election as a landmark of social change because of the deconstruction of the party system. Nonetheless, other authors are still debating the presence and the influence of political sub-cultures in political (Diamanti 2009) and social life (Biorcio 2007; Bassoli et al. 2011).

Overall, there are hints that the functioning of civil society, both in Italy and in Poland (Cento Bull 2000; Magner 2005) is still strongly influenced by cold-war divisions. Catholic or (post-)Communist affiliation of an institution still seems to be relevant in the social and political arena, and there has been growing tension between these two ideological and institutional traditions. In Italy in the post-WWII period, most civil society organizations were connected either to the Communist Party or to the Christian Democrats (Biorcio 2007), since they were the product of a deliberate political effort to influence civil society more broadly. In Poland, the majority of civil society organizations emerged either after being officially acknowledged as quangos by the former Communist regime or after being part of the oppositionary Christian Solidarity movement (Rose-Ackerman 2008, p. 54, Leś 2000, p. 193). Scholars have studied the impact of this duality in both countries both in the past and more recently. The ongoing effects of this division, however, are still debated, given also the increasing fuzziness between these sub-cultures (Tosi and Vitale 2009). For this reason, the article, using a social network analysis approach, dwells on this division in order to answer some key questions regarding both the societal and political spheres.

To what extent do these ‘dual origins’ of modern civil society in both countries matter? What is the role of Catholic and (post-)Communist affiliation of organizations in creating the structures of cooperation? Do Catholic and (post-) Communist affiliation influence organizations’ capacity to enter the policy making process? If so, does this mean that ideological affiliation creates exclusive groups and the structural divisions between civil society actors, which are then reproduced in the political arena? The focus on these two ideological affiliations stems from the attempt to assess whether the Catholic–Communist political cleavage, which developed as early as the first decades after 1917, still matters in local civil societies as well as in the broader political arena.

The first part of the article presents the importance of values and ideology in relation to network structure. The following section describes the theoretical importance of the research question, rooting it in the historical context of Polish and Italian civil societies, closely followed by the methodology deployed (third section) and the propositions (fourth). The fifth dwells on empirical investigation aimed at assessing to what extent organizations exchange information with those sharing similar norms within the social sphere, while the sixth section discusses the impact of ideological cleavage on the political sphere. Finally, the last section delineates general conclusions about the comparative role of ideological affiliation, both at the political and societal level.

Values, Networks and the Mediating Role of Civil Societies

Theoretical discussion on the role of values in organizations’ networks reveals a paradox. As Bassoli and Theiss (2014) showed, different theoretical approaches underline that values have a specific role vis-à-vis actors’ relations: network theory (Emirbayer and Goodwin 1994), Social Movement Theory (Snow 2004; Oliver and Johnston 2000) and Advocacy Coalition Framework (Sabatier ajnd Weible 2007; Zafonte and Sabatier 1998). Generally speaking, the differential importance of values relates also to the importance given to agents and their nature. While for individuals, agency may be rooted in values, the same is rarely suggested of organizations. In network theory, agents’ values are explicitly or implicitly (Adam and Kriesi 2007; Knoke and Yang 2008, p. 6; Borgatti and Halgin 2011, p. 1; Thatcher 1998, p. 409) considered as individually attributed. They are considered unproblematic features because their methodological order differs from that of network aspects (i.e. relations). Thus, one of the presumptions of the network approach is to focus on social structure (and not the values system). However, the use of individual attributes is well developed in the network field. Homophily, that is the grouping of cooperating actors along given features, is considered to be socially pervasive: it structures network ties of different types, including marriage, friendship, work, advice, support, information transfer, exchange and many other types of relationships. ‘People’s personal networks are homogeneous with regard to many socio-demographic, behavioural, and intrapersonal characteristics’ (McPherson et al. 2001).

For this reason, the deployment of network theory best suits the assessment of the role of Catholic and Communist traditions in post-Communist Europe. This latter subject is in itself quite a conundrum. While Catholicism is a religion, Communism is an ideology and a political frame. Catholicism played a major social and political role in the last twenty centuries of the European history against the scarce two centuries of Communism, even though the latter deeply influenced the twentieth century, favouring the creation of a bipolar system of international relation as well as a strong social divide within societies.

As a matter of fact, at the organizational level, the impact of both traditions is very strong across the different countries, because of the role they played in framing third sector in a competitive fashion, at least in western and Eastern Europe. Notably, the clash between the two traditions relies on the homogeneous attitude of both Catholic and Communist organizations towards the economic betterment, the material uplifting and the social inclusion of marginalized social strata, which creates room for competition. Given these similarities, the differences are many more.

The role of religion was never central during the Cold War (Kirby 2003), even though ‘that religion played a significant role in the Cold War might seem self-evident, given the atheistic nature of Communism and the powerful influence of Christianity on the lives of millions of people on both sides of the Iron Curtain’ (Merrilyn 2003). Only after the collapse of the Soviet Regime, the role of the Catholic Church has been assessed in the critical juncture (Anderson 2003; Donovan 2003). In more recent years, only rarely have scholars dwelt on the legacy of Communism in post-Communist countries (O’Mahony 2003; Saxonberg and Szelewa 2007), while an actual debate on Western Europe is almost non-existent anywhere but Italy (Bassoli et al. 2011; Tosi and Vitale 2009). The reason for this exception lies in the specific nature of the clash between Catholicism and Communism in Italy during the Cold War (Salvati 2003), which created an ongoing debate and continues to hold influence today (Cento Bull 2000).

Nonetheless, the legacy of both Catholicism and Communism in the structure of European civil societies and political systems has not until now been studied using network analysis. The only example limits itself to the division between Catholic and post-Communist organizations within civil societies (Bassoli and Theiss 2014), but does not tackle the impact this division has on access to the political arena. Indeed, while the impact of the collapse of the Soviet Union on national party systems has already been studied both in Eastern Europe (Lewis 2000) and Western Europe (Fabbrini 2009; Pennings and Lane 1998), its wider impact on civil societies has not been so central to the literature (Cento Bull 2000), or is limited to Eastern European idiosyncrasies (Elster et al. 1998).

Between Catholicism and Communism: Italy, Poland and Holy See

Ideological affiliation has always been of prime importance in Poland and Italy. There is, however, a significant difference in the role played by the Cold War in the two countries. While Poland suffered a Communist regime and a change in its territorial configuration after the end of WWII, the same does not apply to Italy. Nonetheless, in both countries, the Catholic Church acted as a catalyst for the formation of groups in the first stage of civil society formation (late nineteenth century for Italy and early twentieth century in Poland) and anchorage during the Cold War (Moro and Vannini 2006; Osa 2003). In addition, the Communist legacy is radically different in the two countries: while Polish post-Communist rhetoric still persists, the situation in Italy is more nuanced. While right-wing parties still use the political frame of anti-Communism for electoral success, left-wing parties rooted in the Communist party tradition are fully legitimized. Analogously, the two civil societies differ in the presence of post-Communist organizations, which have national presence and (a debated)Footnote 2 predominance in Poland, as opposed to the persistence of local political sub-cultures in modern Italy (Diamanti 2009). At the same time, both countries share the predominance of the Catholic Church among the active congregations (both Christians and non-Christians). A brief account of the development of the two civil societies is necessary in order to understand both the case selection and argumentation related to the direct/indirect impact the division may have on access to the policy making process.

Since the unification of Italy (1861), civil society organized itself around ideological affiliations. Italian civil society is rooted in the civil initiatives linked with ‘both the workers’ socialist movement and the development of the Catholic Church. These initiatives gave place to modern political parties, trade unions, cooperatives, mutual benefit societies, charitable organizations and networks of Community centres, poor people shelters and social canteens (Moro and Vannini 2006, p. 20). More importantly, the newborn organizations were not only separated from the state apparatus, but also formed in clear opposition to the liberal imprinting of the new country and opposing its (elite-driven) creation. On one side, the Socialist party (1892) could not access Parliament because of electoral rules, while on the other, Catholic politicians faced the threat of the Non Expedit issued by Pope Pius IX (1868) in order to avoid any acknowledgment of the Italian state.Footnote 3

For this reason, the two cultural traditions were already deeply rooted among the population, before entering the political sphere as key factors only with the first mass election (1919). After the fascist period, which suppressed both political parties and independent organizations, the country’s renewed political debate was forged along the lines of the clash between Communism and anti-Communism. The polarized party system was characterized by the presence of a post-fascist party (MSI), the biggest Western European Communist party (PCI), a centrist Christian Democratic party (DC) and many smaller parties (regionalist, socialist, liberal, etc.). Strong polarization at the political level was also mirrored in the structure of civil society, which was forged by external debates as well as constitutional setting.

The complex interplay of these factors (Salvati 2003), along with a strong fascist legacy on civil society, pushed the two major parties (DC and PCI) towards more direct social intervention. ‘The large numbers of Catholics and Leftists among the former Fascist masses […] had to be governed and channelled within and through a democratic party network. At the end of the war, the main mass parties were aware of the urgency of this pedagogic task and promptly devoted themselves to its realization’ (Salvati 2003, p. 557), by way of an increasing number of clearly identifiable flanking institutions (Ignazi and Ysmal 1998). The two parties were fundamentally aiming at promoting a unified framework of beliefs, behaviours and understandings, as well as recruiting people and seeking votes. ‘After the collapse of the state during the war as the symbol of national identity, the two main parties, Communist and Catholic, built up their respective influences by emphasizing the feeling of belonging to a large, integrated institution and to its protective network of relief organizations’ (Salvati 2003, p. 562).

While Italian civil society is deeply rooted in the 19th century, the Polish Communist era was a historical caesura of great importance, also given the role of WWII in fixing new national borders. Post-war Polish civil society is a mixed heritage developed during the Communist regime (Ekiert and Kubik 1999, 2014; Rose-Ackermann 2008; Leś 2000). Well before the collapse of the Communist regime (1989) came the creation of quasi-non-governmental organizations (strongly dependent on the governing Communist party), but also the existence of an opposition movement promoted by the Catholic Church, along with the rise of Solidarność in the early 80s (Kemp-Welch 1983).

However, it is important to note that the Communist regime could only reshape existing civil society, which was already well articulated and rooted in the Polish tradition. First of all, previously active organizations (such as the Polish Scouts’ Association, the Polish YMCA, Caritas) were abolished by the Communist government and made illegal until 1950. The standard practice was that of closing autonomous organizations followed either by the nationalization of their assets or their transferral to newly established organizations with a similar name, acting under full control of the ruling party. Groups of independent organizations active in similar policy fields were often forced to unite in order to achieve better control. In both cases, the charters and steering committees of newly established organizations were imposed by the Communist party. Moreover, the ruling Workers’ Party Steering Committee registered numerous organizations in order to transmit proper ideology to certain social groups (women, youths, the elderly, etc.). Thus, all organizations established in the 1940s and 1950s were fully dependent on the state: ideologically, economically and for wide membership (Magner 2005). In the meantime, the Catholic Church was able to create a space of associational freedom rooted in the Polish tradition. From 1956 onwards, there was a growth of new organizations under the auspices of the Church: Catholic Intelligentsia Clubs (KIK), political discussion clubs and parish charity clubs (Frentzel-Zagorska 1990).

Later on (during the seventies), Polish civil society witnessed the growth of an alternative society (distinct from the party-driven one) along with the emergence of a new ethos within civil society (Rymsza 2007, p. 26). As Osa (2003) points out, this period was a time when the ideological, political and above all social background for Solidarność was createdFootnote 4 thanks to the establishment of the first oppositional organizations (KOR-Workers’ Defence Committee, Human and Citizen Rights Defence Committee). The subsequent period was characterized by a more open clash around the Marshall Laws (1982–1989). Independent (and oppositional) civil society was directly sheltered by the Church, which could access financial support from the Western bloc.

On the emergence of the post-Communist regime under Lech Wałęsa (1990), financial support from the West increased significantly and took the form of economic as well as organizational and educational support to several civil society organizations. For this reason, according to some scholars (Gąsior-Niemiec and Gliński 2007, p. 31), ‘Modern Polish civil society has developed from next to nothing. Over 90 % of the Polish non-governmental organizations (NGOs) have been established after 1989. Their birth was triggered by the mechanism of self-education and self-development typical of civic movements as well as by foreign aid’. Despite the widespread change in Polish civil society, its dual origins are still relevant. In particular, the debate is not settled: while Magner (2005) supports the predominance of the Communist legacy on civil society organizations (considering the size of the coalitions and the relevance of organizations), others (Ost 1990, 2005; Linz and Stepan 1996) argue for a wider impact of the Catholic tradition because of its pervasiveness, although it is not always linked with the institutional Church (see also Magner 2005 on this point).

Overall, Catholic organizations maintain a privileged position, rooted in the role that they played during the Communist regime and reinforced by the ‘spiritual battle’. The link between the Catholic Church and civil society resulted in the emergence of a moral alliance for the good (Ost 1990, 2005; Linz and Stepan 1996): in other words, acting in the name of values rather than for the resolution of social problems is seen as a legitimization of a civil society organization. Many studies, including analyses of Solidarność’s structures, show the use of former opposition networks in creating contemporary civil society in Poland (Gliński 2006, p. 32).

Research Methodology

While addressing this issue at the national level would be the most appealing course of actions, it is also the most problematic (above all in Italy) for two distinctive reasons. On the one side at the national level, the tendency towards cross-cultural relationship is higher as compared to the local one, on the other, the degree of inclusiveness of the political arena is lower at the national level. Indeed, the national level and the local level are exposed to a different extent to what Tosi and Vitale (2009, p. 133) call cultural métissage (or contamination); that is the fact that different families of political actors have stronger cross-cultural relationship. At the national level, Italian civil society organizations (of any background) have a strong tendency towards mutual recognition and cooperation vis-à-vis the political level. “Forum Terzo Settore” is a clear example of cross-cultural relationship; it is the most representative Third Sector federation, as it is composed of organizations of both Catholic and Communist background. On the contrary, at the local level it is easier to find both concrete examples of hybridization, as shown by Tosi and Vitale (2009, p. 140), as well as overlapping membership. Therefore, it would be more important to assess the persistence of cultural division at the local level, rather than at the national level, because their resilience should be higher.

Moreover, given the low level of stakeholders’ involvement at the national level, the local level best suits this research thanks to the recent growth of local governance. Since the late nineties, the sub-national dimension has become increasingly salient since it allows to address complex issues such as active labour policies, social inclusion and economic development, thanks to the use of a wider set of inclusive tools (Bassoli 2010). This shift from the national to the local level has been driven primarily by a new collaborative attitude, developed among politicians and strongly supported by political institutions, with the committed support of all civil societies which were finally able to access local policy making directly. Therefore while it would be possible to assess concretely the role that private actors play at the national level, it is of greater importance to assess the place of these actors at the local level.

At the national level, the strong shift from corporatism to a more autonomous policy making (in Italy) and the persistence of a lack of multi-stakeholder involvement (in Poland) imply a low level of inclusion for all (kind of) actors, that does not allow the (eventual) political cleavage to frame the arena (Ranci et al. 2009; Rikmann and Keedus 2013). On the contrary, at the local level, the existing governance structures heavily require the direct involvement of the private actors. This involvement allows assessing the ongoing relevance of the political cleavage. For these various reasons, the impact of social identities and political culture can be better addressed at the local level.

Given this overall picture of Italian and Polish civil societies, the selection of two exemplary cases is important to test both whether cultural cleavage exists within civil society and what impact it has on access to policy making. Nonetheless, the choice of any exemplary locality would be difficult to be fully justified, given the lack of stringent selection criteria. Indeed, there are some conditions that must be filled (such as the presence of an active civil society and the absence of a hegemonic sub-culture for the Italian case) but an infinite number of localities (over 8000 Italian Comuni and 2400 Polish Gminy). For this reason, I opted to select from the existing database those which best suit the research needs. While for the Polish municipality it is important to have an active civil society, the situation in Italy is far more complex given the existence of different sub-cultures in different areas of the country.

I have therefore deployed the dataset developed by the YOUNEX project featuring different local civil society networks (Baglioni and Giugni 2014). The comparative design is based on the premise that Turin and Kielce share some common features in relation to the role played both by Catholic and (post-)Communist tradition in the social sphere, but are at the same time substantially different in terms, for instance, of the strength and length of civic tradition, as well as the impact of Communist heritage. On the choice of Turin, which is significantly more debatable than Kielce for the aforementioned reasons, we shall remember that (Bassoli and Theiss 2014, p. 184):

The city is an interesting case as regards the so-called Italian sub-culture (political sub-cultures) given its specific industrial tradition. First of all, the city is a former capital of Italy (1861–1871), where most intellectuals were liberals; secondly, in the city there was a strong tradition of charitable […] and educational [Catholic] institutions […] and finally, it also represents the industrial heart of Italy with a strong Trade Unions [sic] and Communist Party. Communist mayors ran the city from 1945 to 1951, and the Communist party of Turin often played an important role in forging national elites. Due to these intertwined variables, the city represents an interesting case of the role of Catholic and Leftist affiliation. Indeed, while in some regions, such as Emilia Romagna, the role of Communist tradition is mostly a matter of lack of other traditions, in Turin there is the co-presence of both traditions without a clear predominance (Diamanti 2009).

The datasetFootnote 5 is composed of 92 organizations dealing with labour market issues and combating youth social exclusion. The interview schedule was specifically designed to include not only qualitative in-depth questions, but also sets of standardized questions addressing inter-organizational networks within civil societies and across policy domains. The selection of actors started with the systematic examination of official lists of organizations active in Turin and Kielce. Thereafter, the selection was completed through examination of practitioners’ publications, access to online sources and through snowball techniques.

The project developed a set of three organizational networks: information, project and overlapping memberships. My analysis is confined to civic organizations, and political parties have not been included in the study as it is important to assess the influence of the ideological affiliation of organizations without the impact of party influence. As regards the network selection, the research deploys the information network because it allows for a more robust test of the hypotheses put forward. This is the widest network, and also the one expected to be less influenced by ideological cleavage because information exchange does not entail any form of cooperation among organizations with divergent values.

As regards the operationalization of ideological background and historical tradition of the organizations, the dataset coded an organization as Catholic if its charter explicitly mentions Catholic values. As for (post-)Communist organizations, the dataset considers those that were officially established during the Communist regime (in Poland) and those which were considered a flanking organization of the PCI (in Italy), or those more recent groups which are publiclyFootnote 6 acknowledged as being of far-leftist background. In both cities, I was eventually able to identify the existence of two ‘coalitions’—groups of organizations sharing common values and a shared idea of society.

In both cities, both the Catholic coalition and the post-Communist coalition were detected along with the ‘non-affiliated’ (third parties), within the field of unemployment and precariousness. Third parties are not a small minority as expected, but what is more surprising is the relative dimension of the two coalitions. Catholic organizations account for 34 % of existing organizations in Kielce (out of 32) and 28 % in Turin (out of 60), while the post-Communist organizations represent only 9 % in the Polish city as opposed to 42 % in Turin. Finally, the third parties are almost twice more numerous in Kielce than in Turin (56 vs. 30 %). The structure of information networks is depicted in Fig. 1, where squares are organizations that fulfil the criterion of being Catholic and black triangles are those fulfilling the (post-)Communist criterion. At first glance, the two coalitions can be clearly observed, as well as the presence of cross-cleavage ties.

Fig. 1
figure 1

The information network, Kielce on the left and Turin on the right (shape by ideological affiliation). Source Bassoli and Theiss, Inheriting Divisions? The Role of Catholic and Leftist Affiliation in Local Cooperation Networks, 2014, reproduced with permission of Palgrave Macmillan

Full size image

Propositions

I use two different networks to analyse the extent to which cultural traditions are still relevant in modern politics: an information network and a policy network. Networks connect pairs of actors within public space as well as civil society organizations and different tiers of government. Networks can be presented in a graphical format (used in the article) but are based on matrices. As regards the first network, the adopted matrices are square matrices (the so-called adjacency matrices, for which one finds the same sequence of actors across rows and columns) while for the policy network I deployed rectangular matrices. The latter are made up of organizations (node of one kind) that are involved at a given administrative level (node of another kind). The comparison of organizational involvement in Turin and Kielce is based on binary-directed relational data in the two networks so as to use different techniques.

I first give an account of the impact of ideological affiliation using the E–I (external–internal) indexFootnote 7 (Krackhardt and Stern 1988), rerunning the analysis of Bassoli and Theiss (2014). The index assesses the statistical difference between the actual homophily and the null hypothesis (the observed value is due to chance alone and not a systematic cause). The analysis was repeated for different partitions (both Catholic vs. non-Catholic, Communist vs. non-Communist, Catholic vs. Communist vs. others) to assess the relative importance of each ideological affiliation.

Proposition 1a

If there is a legacy of Cold-War era ideological conflicts, the division between Catholic and post-Communist organizations within civil society should produce a tendency towards homophily (above statistical average).

Proposition 1b

If there is a predominant coalition between the two as expected for the Catholic coalition in Poland, the group average centrality should be higher.

In the second part, I assess the impact of homophily on access to the political sphere. To do so, I control for the impact of affiliation either directly (considering the access granted to policy making), or by means of social centrality (centralities in the information network). Centrality can be measured using different variables (Freeman 1978), all based on the intuition that the more connection a node has, the more successful it is. The more a node is connected, the higher its ability to reach other nodes and gain access to information and resources.

Centrality can be measured in incoming ties or in outgoing ties. The two measures differ in their meaning. In the first case, the connection represents the prestige of the actors, in the second, it represents their outreach capacity (or desire for acknowledgement). In this article, only incoming ties are considered because they better reflect the relative importance of the organization in the field. Indeed, ‘prestige’ is a more accurate measure of power in contexts such as the one under analysis, in which the tendency of interviewees to project an image of connectedness is very strong (thus mentioning a large number of outgoing ties).

As regards the political relevance of ideological affiliation, the ANOVA routine (Borgatti et al. 2002) was used to correlate centrality in different networks. This procedure undertakes variance analysis comparing the distribution of centrality in two different networks (or between groups). The procedure was used to assess the correlation between centrality in the political network and three variables: ideological affiliation and societal importance (level of centrality in the local field of unemployment, as perceived by the organization or by others).

Proposition 2a

If the cultural tradition is a salient partition within civil society which also has a direct impact in the policy making network, centrality in the political network should be (statistically) different for Catholic, post-Communist and non-affiliated organizations.

Proposition 2b

If social capital plays a role as a mediating effect on the political sphere, the more central organizations will enjoy easier access to the public decision making process as compared to the more peripheral ones, independently by their cultural tradition. Therefore, the centralities of the social network should be correlated with those of the political network.

The Impact of Ideological Clustering at the Societal Level

As has already been shown (Bassoli and Theiss 2014), selective cooperation between Catholic and (post-)Communist organizations in both cities is not a coincidence (Proposition 1a) and the Catholic coalition plays a major role in both cities (Proposition 1b). The problem is thus to measure to what extent the Catholic and post-Communist organizations exchange information with those sharing their normative system, or to what extent the sharp cleavage is subsiding, given the presence of cross-cultural ties. Various conceptualisations can be used to measure the strength of the ideological cleavage. In this article, I have privileged the tendency towards homophily (the preference for organizations of the same kind) to other conceptualizations more focused on the strength of internal ties (such as Simmelian ties, cliques and so forth).

The network shows a strong tendency towards homophily (Proposition 1a). As is clear from Fig. 1, both in Turin and in Kielce organizations tend to prefer their own kind. To what extent can the presented situation be regarded as significant and worthy of further analysis, since the number of nodes (Catholic, post-Communist and non-affiliated) as well as the number of existing ties changes from city to city? The E–I index (Table 1) helps us solve this conundrum. The E–I measures the level of tendency towards homophily against a statistical benchmark. Its level is close to zero in Turin (−0.125) as in Kielce (−0.051). However, in both cities, the value is lower than expected and it is also statistically different (which indicates the presence of homophily).Footnote 8 Indeed, in Kielce, given the size of the three groups and the total number of ties, the expected value is 0.149 while in Turin it is 0.334. Moreover, the strength of homophily in Turin is much higher than in Kielce (considering the difference between the actual value and the expected one).

However, as it is clear from the description of the two different civil societies in the two countries, different behaviour is to be expected of Catholic organizations (which are more numerous in Kielce but not in Turin) and post-Communist ones (which are only a few in Kielce and half of the sample in Turin). It is also possible to assess the E–I index for each group separately (Table 1). In Turin, the post-Communist group shows a stronger tendency towards homophily than the Catholic one (the value of E–I index is lower: −0.447 vs. −0.122). More importantly, both groups score negative values, suggesting the same tendency. On the contrary, in Kielce the post-Communist actors are heterophilous (0.538), while the Catholic do not show any (statistical) predictable patterns.

Table 1 E–I index in different information networks in Turin and Kielce
Full size table

This indicator helps in assessing the phenomenon. First of all the division between the two worlds' works at the aggregate level, while it is much weaker (or non-existent) considering only one faction. The time that has elapsed since the collapse of the Soviet Union has encouraged the creation of many actors (Gasior-Niemiec and Glinski 2007, p. 31) which are most often non-affiliated to either coalitions. The situation as it is depicted by this index suggests ‘the presence of a hegemonic group in Kielce (the Catholic) and a bipolar situation in Turin’ (Bassoli and Theiss 2014, p. 190).

As regards the impact of cultural heritage on civil society, it is important to understand how central this division is (Proposition 1b). Indeed, the tendency to forge ties with other actors sharing a similar ideological background would be of little importance if the two coalitions were marginal. As to assess the relative importance of actors, social networks analysis provides a wide set of measures. A first measure, the so-called group centrality, is less used because it considers only the number of other actors (alters) that are connected to a given group, without taking into account the size of the group itself. Therefore, in the case of Kielce it would underestimate the relevance of the Catholic coalition.

I therefore deploy two individual-driven measures that help to assess the groups’ relative powers: average (normalized) in-degree centralityFootnote 9 and the average in-Bonacich power.Footnote 10 Both measures are calculated as averages of individual centralities, favouring those actors with more prestige. Both are measures of popularity, since they refer to the number of times that a given organization has been mentioned as a partner by other actors. In the first case, the (in-degree) alter’s centrality is not considered; in the second, the relative importance of the ‘sender’ is also taken into account. In both cities, both measures confirm the predominance of the Catholic coalition over the (post-)Communist one, while non-affiliated actors are definitively peripheral (Table 2). As in the previous case, the partition composed by three groups is the most significant with the best R-square, however similar results can be obtained using different partitions (see Online Supplementary Material).

Table 2 Group average centrality
Full size table

The Political Capacity of the Two Groups

Having observed that in both cities there is a high tendency towards homophily among civil society organizations, it is important to understand the political consequences of such a situation. In order to assess the political capacity of the coalition, I have opted for a simple correlation measure. I thus assess if centrality in the political network is influenced by ideological affiliation either directly (Proposition 2a), or by means of the social centrality (centralities in the information network) (Proposition 2b). In Fig. 2, a depiction of the actual participation of interviewed organizations in formal policy-making procedure is given by a tie between the association and the mentioned level of government. The affiliation is also depicted in the picture: dark squares are Catholic actors, dark triangles are (post-)Communist and grey squares represent the government tiers. Isolated actors (associations which are not involved in the political domain) are placed in the left-hand side of the pictures. The network has been graphed so as to have the actors grouped for homogeneous behaviour (number of ties): so for example, all organizations on the left of the political levels in Kielce and Turin are those with only one participation tie (usually the local level).

Fig. 2
figure 2

The political participation of interviewed organizations, Kielce on the left and Turin on the right (light grey squares are levels of participation, dark grey squares are Catholic organizations, circles are not affiliated and black triangles (post-)Communist). Source Network graphed with Netdraw® (Borgatti 2002) position given by principal components procedure adjusted for government level and repositioned across the 2-mode graph

Full size image

As in the previous test for correlation between centralities and ideological affiliation, it is possible to extract a vector containing centrality measures for each organization.Footnote 11 Therefore, I test political centrality both against affiliation (direct impact) and social centrality (indirect influence). While there is little to be discussed on the first approach, the idea informing the second one is that cultural affiliation is a strong driver among civil society organizations, but it may not directly translate into the political domain. Nonetheless, given the different size of the coalitions and the more central role of the Catholic organizations (in both cities) it is likely to have an indirect effect in the political sphere, although mediated by social structure.

As regards the first approach, ideological affiliation has a direct impact in the political sphere only in the city of Turin (Table 3): the distribution of political centralities among the three groups is not given by chance alone. Catholic organizations are more central as compared both to the organizations that are not affiliated and to the (post-)Communist ones. The main reason for such a clear result is linked to the poor access that new-leftist organizations have faced at the political level in Turin. Indeed, while in Turin all Catholic organizations are politically active, not all the (post)Communist ones are, and in Kielce the abundance of Catholic organizations makes them more evenly distributed in the participation network: some of them are very active (right-hand side of the figure), some are less active or even isolated. This result may be influenced by the local context. While Turin civil society constituency is of leftist background (as most Italian third sector) (Anzivino et al., forthcoming) and is permeated by the same attitude as the city is (Diamanti 2009); the same is not true for Kielce. Therefore, those Catholic organizations listed as such, have in Turin a “leftist” background which allows them to be recognized as relevant in the field by policy maker. There may be other peripheral Catholic organizations but they were completely overlooked in the sample. On the contrary, the pure (post-)Communist organizations are all equally acknowledgeable, but they are more heterogeneous influencing the whole group centrality.

Table 3 Ideological division and political relevance (direct influence)
Full size table

Given the evolution of the ideological cleavage in the two different contexts, this result is not a surprise. While the Polish-Polish war had a very strong impact on the social context, the Italian sub-cultures were primarily a political cleavage which later translated into social division. It is quite logical that the latter has a strong impact also on associations’ capacity to forge links with government associations. Of the two worlds, the Catholic is much more structured than the (post-)Communist, given the presence of new-leftist movements without connection to the political realms.

Moving to the indirect links, it is possible to test if the centrality of the political network is correlated with the one in the social sphere (which is correlated with ideological affiliation). The situation is completely different from what emerges from the direct links (Table 4). While social centrality (measured as in-Bonacich power) is significantly correlated with political activation in Kielce, the same does not apply in Turin. The same result holds considering out-Bonacich power. The main reason for this result has to be tracked back to the structure of the social networks. Moreover, the centralities help to neutralize the effect of Catholic hegemony. This would suggest that the organizations more active in creating social linkages are also those that are more active in forging political ties in the Kielce case, while the same cannot be stated for Turin.

Table 4 Social relevance and political relevance (indirect influence)—R-square
Full size table

Concluding Remarks

Assessing the relevance of the ideological legacy 25 years after the fall of the Berlin wall could be discounted as a futile theoretical exercise. However, the fact that the assessment of the relevance of Communist and Catholic legacies on the structure of civil society did not reach any conclusions (at least in the Polish case), Gąsior-Niemiec and Gliński (2006); Magner (2005) directly imply the importance of understanding both the reason for such a disagreement and the actual possibility of assessing the legacy. As I have suggested in this article, it is possible, both at the national and at the local level, to directly assess the ongoing legacy of ideological division using social network analysis techniques. The methodology deployed helps to measure the legacy not only within the public sphere but also in the political realm.

It is thus possible to conclude, at least for the selected (exemplary) cases, that there is a strong legacy of Communist ideology, of the Catholic tradition and of the cold-war division between the two worlds of values (Proposition 1a). In both cities, there is a tendency to prefer similar partners: Catholic organizations stick together, as do (post-)Communist ones. Both Catholic and (post-)Communist organizations tend to communicate, respectively, more often with those sharing similar values. However, the fact that in both localities there is a strong tendency towards homophily in the organizational field translates into different network properties and impacts at the political level in diverse ways.

It turned out that the ideological cleavage means something different in the two networks. If third parties are included in the analysis, the clustering level of Turin is higher than in Kielce because the non-affiliated are more prone to exchange information with the Catholic rather than post-Communist group members (see Table 1), while in Kielce the third parties are less relevant. Overall, there is a prevailing position of the Catholic group both in Kielce and in Turin, although the Italian case recalls a bipolar system, while the Polish recalls a hegemonic one. As is evident in Kielce, being of Catholic stock is an asset for an organization, which includes both social capital (social networks and contacts even from the times of Solidarność) (Osa 2003) and symbolic capital (Linz and Stepan 1996); indeed, the group average centrality is definitively higher (Preposition 1b). The same is true also in Turin but for a different reason. In the Italian city, the presence of many peripheral actors belonging to the new-left decreases the group average centrality. Notably, in both cities, the presence of a strong homophily tendency does not imply the absence of cross-cultural ties, simply it suggests that those ties are less important than internal ties. Thus, there is no real hint of the existence of a coalition grouping together (post)Communist and Catholic organizations.

The role of structural as well as symbolic contextual factors in explaining the role of organizations’ ideological affiliation in a policy network may be only tentatively answered. Indeed, the data deployed do not fully answer the question about the political relevance of the cultural clash, but only attest to the presence of structured network governance arrangements where cultural tradition plays a role. The data available only inform us about the direct involvement of these organizations during the policy making process, without clarifying the extent of their participation or its impact. As it is clear from Fig. 2, most Catholic organizations in Turin have direct access to the political arena, while the same does not apply in Kielce. Therefore, as stated in Proposition 2a, there is a direct impact of cultural division on the accessibility of the political sphere in Turin (but not in Kielce). Vice versa, most central actors in Kielce are politically active, while the same does not apply in Turin, confirming Proposition 2b in the Polish case only. Therefore, while belonging to the Catholic faction is a driver of political centrality in Turin, it is not in Kielce, where belonging itself needs to be supported by the central position of the organization. As already shown (Bassoli and Theiss 2014), the hegemonic role played by Catholic organizations in Poland features the presence of marginal organizations among their ranking; on the contrary (most), Catholic organizations in Turin are able to have direct access to policy making, without the mediating role of centrality.

These findings corroborate the idea that values are (still) central in Catholic-traditional countries. Indeed, the organizations that can be directly linked to this group have easy access to the political sphere both in Kielce and in Turin. In Kielce, thanks to the mediating role of the major Catholic organizations (such as Caritas) (as in Proposition 2b), in Turin, thanks to the ostracism faced by left-wing social movements (mainly movements in support of the unemployed and precarious youth in the analysed sample) (Proposition 2a). Following the methodology deployed here, I have opened up space for more research to be conducted in the scholarly literature about the impact of ideological legacy on civil society structuring as well as on policy making. Different links, summed up in a direct impact versus mediated role of cultural legacy, can be matched against specific findings that shed light on the function of local policy making; the study of networks can thus provide scholars with an important tool to assess the political relevance of societal division in general.