The limits to bridging social capital: Power,social context and the theory of Robert Putnam

Abstract

Although there is general agreement on the definition of social capital as the benefits to be derived from social connections, the type of advantages (more effective transacting across all fields versus contacts for personal advantage) and the beneficiaries (community versus individual) identified by social capital scholars differ. This variety can be addressed with a distinction between the so-called schools of cooperation and competition. This article focuses on the former, particularly the work of Robert Putnam. The author uses Nicos Mouzelis’s critique of rational choice theory, and his distinction between micro and macro actors, as a diagnostic tool to highlight the shortcomings of Putnam’s work and the cooperation school more generally. The author argues that Putnam’s notion of bridging social capital as a solution to problems of intolerance and more general social ills is overblown, given that both diverse social networks and increasing tolerance are the result of deeper social processes not analysed by Putnam. In support of this, the article lists a number of ways in which macro actors influence the ability of social networks to form, and once formed, constrain and enable their agency in either a cooperative or competitive direction. It also criticises the tendency of cooperation theorists to generalise the solution of the collective action problem on the micro level to the macro level. Finally, the article emphasises the importance of analysing the interaction between different collective action problems, as well as the connection between cooperation and competition.

Keywords

cooperation Mouzelis Putnam social capital social theory

Introduction

In the first of these schools, as exemplified by Coleman and Putnam, the focus is on explaining how cooperation for the common good is possible (the so-called collective action problem), given the conception derived from rational choice theory of people as selfish utility maximisers. The cooperation school solves this problem by incorporating the impact of social norms in orienting the choices of people in a more cooperative direction. If agents perceive the actions of others to be governed by norms, they are less likely to be distrustful of their motives.

The second approach, of which Bourdieu’s work is the most important exponent (but see also Lin, 1999, 2008), regards social capital as a resource that limits some and advantages others. For Bourdieu (1997), social capital is one of four types of capitals valid in all fields, the other three being cultural, economic and symbolic capital, with which it can be exchanged to varying degrees. People possess different amounts of these four types of capital. As a result of the differential distribution of capital, the objective possibilities that are open to individuals are not the same. This is why capital can be seen as ‘a force inscribed in the objectivity of things so that everything is not equally possible or impossible’ (Bourdieu, 1997, p. 46). Those with much capital find their agency enhanced; those with little capital are stymied in their projects. People compete to enhance their portfolios of capital. The emphasis on competition and stratification in Bourdieu’s work is therefore clear, although there are also elements of the cooperation school in his theory. For reasons of space, I cannot pursue those here.

Within each of these traditions, there are further subdivisions. I discuss some of the complexity of the cooperation tradition below. With regard to the competition school, I can mention the work of Burt (2004). Unlike Bourdieu, his focus is purely on the structural aspects of social networks. He explored the opportunities provided to ‘information brokers’ who straddle the ‘structural holes’ between groups. Given that information emerges from social connections, Burt believes that individuals who belong to more than one group have an advantage. This is because they can creatively combine information coming from more than one source and are consequently more successful. On the other hand, those who belong to only one group are more likely to access redundant information.

A typical response to the existence of different schools of social capital theory is to choose one over the other. For example, Denord, Hjelbrekke, Korsnes, Lebaron, and Le Roux (2011, p. 90) make a clear choice for Bourdieu’s theory. His theory is superior, they say, because he emphasises the historical and institutional dimensions of social capital, analyses the relationship between the different kinds of capital and has produced a theory of group formation and class. I do not believe that it is necessary to make such a choice. Rather than choosing between theories of cooperation and competition, it is better to construct a theory of social capital that fleshes out the relationship between the cooperative and competitive elements of social capital. Both are essential elements of an adequate theory of social capital – groups benefit from their ability to work together as much as some individuals (and groups) are able to enhance their position relative to others. Cooperation within a small group may make competition between this group and another more intractable, for example. The search for the conditions for cooperation in society, as pursued by social capital theorists, is therefore a worthwhile exercise, provided that it is not isolated from conflict and competition.

My ultimate aim is to construct a synthetic theory of social capital in which the various dimensions of the concept are fleshed out. In this article my aim is more modest, however. I focus specifically on the cooperation school of social capital, particularly the work of Robert Putnam, who is the most influential exponent of the concept. I therefore hope to reach a halfway house towards the ultimate aim, so to speak. Consequently I do not pursue the work of Bourdieu (or Burt, for that matter) any further in this article.

Putnam’s work has been subjected to severe criticism, as will be apparent from the references cited below. He has, inter alia, been attacked for ignoring power relations as well as the effects of broader social structures. I generally agree with these criticisms. I argue however that it is possible to improve on the diagnosis of what went wrong with his theory. An improved diagnosis will also provide a better road map for escaping these problems. Despite this, I also believe that Putnam’s aim of discovering the conditions for cooperation is worthwhile. It is essential, however, to analyse the relationship between cooperation and competition, and I go some way in addressing this issue in what follows.

My principal diagnostic tool is Nicos Mouzelis’s (1974, 1995, 2008) theoretical framework. His framework directs our attention to relationships that go beyond the mesolevel factors that form the basis of cooperation theories (see for example Woolcock, 2004). By focusing on how social capital is embedded within macrolevel processes, specifically through his conception of macro actors, it allows us to explore the role of inequality and power differences between actors. It also helps to avoid the unrealistic and inequitable policy prescriptions of the cooperation school (see for example Fine, 2010). Through his emphasis on the importance of considering the actual relations between groups, and thus the role of individual and collective agency, it helps us to move away from the formalistic and abstract formulations of theorists such as Putnam, embedded as they are within a rational choice framework. In this respect Mouzelis’s theory invites a consideration of the interrelationship between social capital situations, thus transforming our understanding of the collective action problematic. His framework also allows us to flesh out the historical and institutional dimension of social capital.

The article proceeds as follows. After a brief overview of the cooperation school of social capital, I use Mouzelis’s theory, particularly his critique of rational choice theory and his emphasis on the importance of considering both the vertical and horizontal relationships between social groups, to diagnose what is wrong with Putnam’s theory. I give particular attention to Putnam’s more recent work on religion in America, and his argument that religious bridging social capital explains the increasing tolerance of Americans for religious diversity. I argue that diverse religious networks cannot be an original explanation for such tolerance, and that both such networks and religious tolerance are in fact the result of deeper social change that Putnam’s theory is not capable of explaining.

Robert Putnam and the cooperation school of social capital

According to Coleman and Putnam, high trust based upon informal norms and personal relationships leads to cooperation and thus greater efficiency in all endeavours. This is captured by Putnam’s (1993, p. 31) definition: ‘Social capital here refers to features of social organisation, such as trust, norms and networks, that can improve the efficiency of society by facilitating coordinated actions.’ It facilitates economic exchange because it reduces the need for expensive lawyers and complicated contracts and thus transaction costs (Fukuyama, 1995, pp. 26–27, 336; Williamson, 1985). An example of this is the diamond merchants of New York City who circulate expensive parcels of diamonds without contracts or insurance because they trust one another (Coleman, 1988). Social capital also improves the functioning of state institutions, with knock-on effects on the success rate of development initiatives.

While they all share these concerns, the cooperation literature is divided when it comes to the size of the collective that is their analytical concern. Some, such as James Coleman (1988), view social capital as a property of small groups. For them, the solution to the collective action problem is provided by trust sustained by norms in social networks. In the case of Coleman the structure of the network, specifically whether it is open or closed, helps to determine whether social capital can be generated. Others, typically political scientists and economists, regard social capital as a property of communities and nations. They tend to emphasise the connection between values of generalised trust and positive macrosocial outcomes such as economic development and low levels of corruption. Societal characteristics are explained by the attributes of individuals; in this case the types of values that they hold. The collective is reduced to the typical individual, and social networks on the meso level do not carry any explanatory weight (cf. Fukuyama, 1995, pp. 26–27; Uslaner, 2009). An exception is Putnam (1993, 2000), who is also interested in social capital as a property of large collectivities, but sees voluntary organisations and informal groups on the meso level as the main locations where people are socialised into habits of cooperation.

In the rest of the article I focus exclusively on the work of Robert Putnam. According to Putnam (1993), only horizontal relationships, such as civic relationships in voluntary organisations, can produce social capital. Vertical relationships, for example patron–client relationships, destroy social capital because they imply dependency that breeds mistrusts among the clients as they vie for the favour of the patron. In addition, the patron can defect from the relationship without sanctions from the clients because of the power advantage possessed by the patron. Consequently, Putnam was, especially initially, particularly concerned about the health of voluntary organisations that organise people into horizontal relationships of mutual help or in support of a common cause. These organisations are important sources of generalised trust in society, according to him. It is where people learn the habits of cooperation that are an essential lubricant for democratic political action and an economically prosperous society.

In his first book on social capital, Putnam (1993) explains the greater effectiveness of regional government in the North of Italy relative to the South as being due to the type of social relations typical of each region. He contrasts the vertical social connections of the South of Italy with the horizontal relations typical of the North (embodied in thousands of voluntary organisations). He traces the development of these relationships historically over a period stretching from the Middle Ages to the present. While his grasp of history is impressive, it is peculiar that the vertical versus horizontal forms that he espies are themselves completely unaffected by this history. It is as if these social relationships are unchanging essences that survived intact all these centuries despite the complete transformation of the Italian social landscape over the centuries.

This contrasts with his most famous book, Bowling Alone, as Ponthieux (2004) points out. Instead of seeing social networks as relatively unchanging, in this book he tries to show how the health of voluntary organisations in America had been declining since the late 1960s, after reaching a high point in the postwar period, and secondly to working out what the causes of this change are. The first cause he identifies is the passing of the highly civic generation that came of age during the Second World War. Other causes are long commuting times, the increasing movement of women into the labour market, and an increase in television viewership that took up time previously devoted to participation in voluntary organisations.

Putnam (2000) claimed that the decline of voluntary organisations in the US bode ill for the future health of American democracy. These claims were highly contested, and generated a large literature that cannot be summarised here. I concentrate here on his argument about the importance of voluntary organisations for the health of a democracy. It was questioned, for example, whether participation in voluntary organisations helped to socialise people into prosocial values, as Putnam had claimed. Perhaps these organisations attracted people with such values, so that the causal connection runs in the opposite direction (from prosocial values to participation in voluntary associations) (Hooghe, 2008). Outside of North America and Western Europe there is also no empirical indication that membership of voluntary organisations is associated with generalised trust (Rossteutscher, 2008).

Putnam responded to these criticisms by arguing that more informal types of social networks such as friendship groups and neighbourhood get-togethers can also fulfil this socialising function (Hooghe, 2008). Over time, he became more sceptical about the positive effects of secular voluntary organisations. By 2010 he and his co-author (Putnam & Campbell, 2010, p. 249) quoted Theda Skocpol to the effect that ‘fraternal and civic organizations no longer bring together people from different social and economic backgrounds as once they did’. Instead, they (Putnam & Campbell, 2010, p. 475) now argued the case for informal religious associations such as Bible study and prayer groups. These networks produce civic attitudes and behaviours such as generalised trust, volunteering and voting. According to Putnam and Campbell (2010, p. 473), the crucial variable is ‘religious belonging’, not ‘religious believing’, thus emphasising the role of networks in producing good neighbours and citizens.

Another criticism of the social capital literature was that it ignored the negative aspects of social networks (e.g. Portes & Landolt, 2000, p. 532). Putnam and Goss (2002, p. 8) responded to this by acknowledging that social networks may carry negative as well as positive externalities for outsiders. He used the distinction between bonding and bridging social capital to elaborate on this phenomenon (Putnam, 2000, p. 22; Putnam & Goss, 2002, pp. 11–12). According to him, there is a greater likelihood that bonding social capital (connections between similar people) will produce negative externalities such as exclusion than bridging social capital (connections between diverse people). Consequently, he increasingly emphasised the potential of social connections between diverse people for social solidarity in plural societies (e.g. Putnam, 2000, pp. 22–23; Putnam & Campbell, 2010, p. 527). For example, in states in India where civic associations bridge religious divides between Hindus and Muslims, he claims there is less intercommunal violence (Putnam & Goss, 2002, p. 12).

As a result, Putnam and Campbell (2010), in their book on religion in America, are particularly interested in the evidence for increasing amounts of bridging social capital across religious divides. This is seen in the growth in religiously diverse marriages, neighbourhoods and friendship circles (Putnam & Campbell, 2010). They argue that this leads to increasing tolerance of people across religious divides. According to Putnam and Campbell (2010), the direction of causality is from diverse social networks to increasing acceptance and not the other way round. They also reject the alternative explanation that the decline of religious animosities is due to a decline in religiosity (Putnam & Campbell, 2010, p. 538). The changing nature of religious social networks also had another positive spinoff: religious connections have become increasingly diverse in race and class terms (Putnam, 2007, p. 161), thus helping to reduce the intensity of racial and class divisions. This is another example of bridging social capital producing greater tolerance.

In response, one can argue that while it may well be true that, once they are established, diverse social networks can help to reduce communal tensions in diverse societies. Being friends with people from a different religion, race or class can reduce misunderstandings across lines of division, provided that people have the ability to generalise from the individuals they know to the groups that they belong to. Beyond this, however, it is hard to escape the conclusion that bridging social capital represents less the solution to problems of communal mistrust than an indication that this problem has already been solved, at least to some extent. Under normal circumstances, diverse social networks face major obstacles to their formation because of the extent of mistrust and animosity that needs to be overcome. Because of these obstacles, diverse networks cannot be an original solution to problems of communal conflict because they are unlikely to form in the first place. Diverse social networks can therefore only be a secondary, derived, solution to problems of communal strife and not an original solution. Putnam (2007) himself has indicated that neighbourhood ethnic and class diversity produce (at least in the United States) social isolation and distrust. This throws doubt on his claim that bridging social capital can serve as an independent variable producing increasing tolerance.

Putnam and Campbell (2010, p. 149) are aware of how exceptional the current degree of interreligious socialisation is in the United States. Over the last hundred years, religious intermarriage has moved from being exceedingly uncommon to being fairly common, for example. Neighbourhoods that had previously been segregated on the basis of ethnicity and religion have become desegregated. These changes have facilitated the current degree of religious bridging social capital. However, they so strongly want to emphasise the causal role of bridging social capital as a solution to religious intolerance that they underplay the extent to which bridging social capital is itself the effect of deeper social changes.

With regard to the latter, one has to be satisfied by vague statements such as the following: ‘Americans have more or less deconstructed religion as a salient line of social division over the last half century’ (Putnam, 2007, p. 160). This leaves many questions unanswered: Were some Americans perhaps more influential than others in this process? What role did the institutional structure in which religion is embedded in the United States play in this? As will become clear from the next section, there are many additional issues to do with the changing role of religion that are relevant in explaining both the increasing amount of religious diversity in social networks and increasing amounts of religious tolerance.

This observation is a variant of the criticism of the social capital literature to the effect that it ignores problems of endogeneity and spurious relationships (Fine, 2010; Ponthieux, 2004; Portes & Landolt, 2000). Diverse religious networks and religious tolerance are both the effect of deeper social variables, and are likely to vary together. This does not exclude the possibility that religious networks, once they have formed, can also independently affect religious tolerance. However, it does suggest that the exclusive focus on social networks, or norms and trust, as causal agents is very limited. In the next section I apply Nicos Mouzelis’s theory to the social capital literature in order to show how one can overcome this limitation.

Applying Mouzelis’s theory to the social capital literature

According to Mouzelis (1995, 2008), we need to understand society both as a structure of actors and as a structure of institutions and roles. Social hierarchies that consist of individuals and groups/organisations with unequal amounts of power are central to society as a structure of actors. This leads Mouzelis to suggest that there are micro, meso and macro actors. Micro actors are those with the least power in social hierarchies. The consequences of their actions do not transcend the local level. Macro actors have the power to make a difference on a society-wide (or even global) scale. They thus have a level of agency that is unavailable to less powerful actors. Macro actors are typically powerful politicians and business people. They can also be collective actors, such as political parties and transnational corporations. Finally, like agency, institutional structures can also manifest on micro, meso and macro levels.

The power differentials between micro and macro actors imply that the agency of the former is often subsumed under that of the latter. Most directly, this means that less powerful individuals and groups often have to take orders from the more powerful. In general, there are myriads of ways in which the ability of less powerful groups to form, and once formed, to act, is limited (as well as facilitated) by the projects and preferences of more powerful actors. This provides a new perspective on Putnam’s contention that social capital can only form among people who are in horizontal relationships, as it means that these relationships often have less autonomy from vertical relationships than Putnam imagines.

The subsumption of a group of micro actors under a macro actor also means that the welfare of any single one of them is connected (either positively or negatively) to the welfare of others. For example, if one gets a contract or a raise, this may exclude the possibility of the other receiving one. Moving up one level, the same applies to the relationships between groups of micro actors. This means that their actions are interdependent, rather than independent of one another. Mouzelis (1995, 2008) uses this interdependence to criticise rational choice theorists. They tend to see the relation between small-group interactions (micro) on the one hand and society (macro) simply as one of aggregation. However, microlevel interactions can only be added together to form a whole if they are independent of one another. The reality of subsumption means that this is often not the case. Even if they are considered separately from their relationship to macro actors, micro interactions form chains of interdependencies on the horizontal level (such as division of labour, conflict and competition) that render the strategy of aggregation unworkable. More on this below.

The influence of macro actors on micro and meso situations

In applying Mouzelis’s critique of rational action theory to the social capital literature, I first provide a list of some of the ways in which macro actors can influence both the capacity of micro and meso actors to cooperate and the likelihood of competition between them. In some cases, this influence is via their impact on institutional structures, in the sense that they have used institutional rules and procedures to exert the degree of control that is apparent here. As pointed out above, Mouzelis emphasised the importance of both actors and institutions in explaining social change. Here I will concentrate only on the former to simplify matters. In addition, the distinction between horizontal and vertical relations is an analytical one, and it is sometimes difficult in practice to separate the two. Popular mobilisation of micro and meso actors may have played a role in some of the trends mentioned here. None of this however affects the essence of the argument I am making.

Macro actors regulate access to resources

By increasing conditions of scarcity, for example through deficit cutting in the case of governments or downsizing by firms, macro actors can bring on conflict. This can occur either directly, as a result of increasing competition prompted by diminishing resources, or indirectly, as in the case of cost-cutting in voluntary organisations that increases the load on volunteers and reduces the incentive to be part of the organisation (Sennett, 2013). Macro actors can alternatively increase the potential for cooperation by actively mediating in the conflicts over distribution of resources among subordinate groups.

They can also reduce conflict by increasing resources among the poor. In the United States the New Deal, expansion of higher education, and programmes such as the G.I. Bill for military veterans, which subsidised education opportunities and offered assistance towards home ownership, played a major role in reducing the class disadvantages of (especially white) poor people. During the younger Bush presidency, rules were changed so that federal funds could be used by religious organisations for social programmes. Increasing the funds available facilitated the activities of grassroots religious organisations.

Macro actors develop/enforce norms

In contrast to the cooperation school, which sees social norms arising from the bottom up as a solution to collective action problems, norms are in many cases imposed from the top down. By formulating clear norms and enforcing them equitably, macro actors can guarantee contracts and increase trust among subordinates (see Rothstein & Stolle [2008] for a detailed empirical and theoretical argument along these lines). They can also encourage competition among them, as in the case of transnational corporations that expect their divisions to compete for contracts to manufacture particular products.

Macro actors impinge on the settings of interaction

Settings of interaction are understood here as the space–time contexts within which interaction takes place. Micro actors interact within spaces that are strongly influenced by the actions of macro actors. For trust to develop it is essential that there is a degree of stability in the relationships between actors (Sennett, 2013). If this is not forthcoming, social capital will be weakened, or generally struggle to develop. In the community interaction setting this involves residential stability, as demonstrated by Sampson (1991), who has made a connection between what he calls community efficacy and average length of residence in a neighbourhood. Desmond (2016) also points to a connection between residential stability and what he calls psychological stability, as well as community stability. Because the powerless have little capacity to resist, poor communities are vulnerable to relocation initiatives that flow from housing and infrastructure projects, mining, proclamation of nature reserves, and so on, and their consequent erosion of community solidarity (see for example Chris De Wet’s [1995] study of the corrosive impact of apartheid era relocation initiatives in the Eastern Cape province of South Africa). Enduring social relations are also an important factor in group solidarity in the workplace. As is well known, the trend over the last 30 years in the Western world has been towards the erosion of job security. Due to this increase in short-term work, it has become more difficult to build long-term relationships in the workplace, with a negative impact on workplace solidarity (Sennett, 2013). This increases the likelihood of competitive rather than cooperative relationships.

Desmond (2016) describes how in American cities a combination of declining incomes and increasing rents made housing unaffordable for poor people. Incomes declined because well-paid manufacturing jobs were replaced by less lucrative service sector jobs, unemployment increased in inner cities, and welfare payments were cut. Rents increased inter alia because the state withdrew from public housing, thus reducing the supply of housing. Poor people were squeezed in this vice, and the result was skyrocketing numbers of evictions in American cities, which in itself was a cause of increasing poverty. The crack and later the opioid epidemics, increasing incarceration of especially black men, and the outmigration of middle-class black families weakened the extended family connections of especially African-Americans (Desmond, 2016). This, together with the instability produced by increasing evictions, produced what Desmond (2016) calls ‘disposable ties’. These are short-term ties of mutual help, especially between neighbours, characterised by low trust.

Macro actors also shape the social media sphere of interaction. While it is true that social media such as Twitter and Facebook have created new spaces for citizens to express themselves, access information, as well as form social relationships over long distances, the limits of this have recently become apparent. Sophisticated repressive governments can firstly curtail the reach of the internet. Second, social media provide an avenue for the powerful to anonymously influence public opinion via the dissemination of false information, fake Twitter accounts, bots and paid trolls that harass people whose voices the powerful find inconvenient. Social media are also an extremely profitable and sophisticated marketing channel for both commercial and political messages, with the ability to fine tune messages based upon detailed information about individuals that companies such as Facebook and Google have accumulated. Such media have also not escaped the more general power relationships in the form of gender, race and class that shape all forms of social interaction, as can be seen in the online harassment of female and minority voices, as well as unequal access due to resource limitations.

Macro actors manage group solidarities

Unlike the previous headings, where the impact of macro actors is mostly unintended, in this section I discuss deliberate attempts by macro actors to affect social capital. Macro actors can attempt to instil broader identities, or suppress narrowly defined identities. The prominent case of this is nation building by a centralised state, but this strategy can also be applied by non-state actors. Such a strategy may, under ideal circumstances, improve the capacity of smaller groups to cooperate. An example of this was the radical white South African students in the 1970s and 1980s, who played a major role in starting and organising anti-apartheid trade unions and community organisations. This is an example of what Woolcock (1998) calls linking social capital. Industrial unions that organise workers on a class rather than a racial basis can reduce racial tensions (Greenberg, 1980). In contrast to this, macro actors can follow a strategy of divide and rule which, if successful, will worsen cooperation among micro and meso actors. Surveillance efforts by security agencies, as well as the use of informers, may reduce trust among micro actors. Clientelistic relations with politicians reduce the autonomy of local organisations, and increase the competition among them for the favour of the leader.

Under this heading one should also consider the rise of the religious right in the United States in the 1980s and 1990s. This was the result of the mobilisation of especially evangelical Christians by politicians, televangelists and media organisations such as Fox News and numerous talk radio stations, often supported by wealthy donors. Out of this a large number of voluntary organisations were born which protested against things such as abortion or gay marriage.

The social capital of macro actors impacts the social capital of micro actors

Because the wealthy and the powerful tend to socialise with, and thus help, one another, individual members of these networks can mobilise large amounts of the resources of the collective to further their own projects. Micro actors may find that their own projects and their ability to cooperate among themselves will be affected by the manoeuvrings of powerful actors, sometimes in unexpected ways due to the often hidden nature of the social connections of the latter. This was demonstrated by the leaks of Democratic National Committee (DNC) emails in the United States. From this it became clear how elite Democrats were mobilising their social connections to advantage themselves and their family members, often outside of the public eye. Leaks have also revealed how the 2016 Clinton campaign outmanoeuvred grassroots organisations though their hidden control over the funding mechanisms of the DNC.

Of course, micro actors can also organise themselves, and mobilise to protect their interests. The relationship between micro and macro actors is therefore not just a top down one, and macro actors also have to take into account the responses of micro actors. One can see this in the pendulum swings in American politics, where periods of elite control are often alternated by grassroots mobilisation attempting to wrest control back from elite actors. Macro actors also engage in competitive and cooperative interactions among themselves, and, like micro actors, need to consider the unfolding interaction situation as well as institutional structures. They do, however, have more autonomy in dealing with these than micro actors, as pointed out above.

Macro actors and the rise of religiously diverse social networks and religious tolerance

In this section, I return to the critique that the connection posited by theorists of cooperation between bridging social capital and positive social outcomes is largely spurious. This is because they ignore the extent to which both these variables have been produced by deeper social changes. Armed with the previous discussion of the role of macro actors, I argue that Putnam’s discussion of religion is problematic because both diverse religious networks and religious tolerance are the result of deeper social factors. In order to make my case, I provide a brief summary of a few social factors relating to the role of macro actors and macro institutions that may serve as a point of departure for such an explanation. This is merely a provisional list – a proper explanation of these phenomena would entail a thorough historical analysis.

Religion in the United States no longer seems to be associated with class divisions to the same extent as a century ago. Declining class differences reduced the antagonism between religious groups. I have already referred to the role of Federal programmes such as the G.I. Bill in this transformation. They have allowed, inter alia, poor white Catholics to escape from ethnic ghettos and join White Anglo Saxon Protestants (WASPS) in the suburbs. Another result of these programmes was increasing levels of education. This increased the status of the laity relative to the religious leadership, and reduced the authority of the latter. I would suspect that this facilitated a change whereby churchgoers became more open about the existence of alternative religious truths. Increased levels of education also allowed poor whites to transcend manual occupations, and meet members of other religious traditions as equals in the world of work. This, coupled with the sheer availability of people from alternative religious traditions in the suburbs and workplaces with whom to socialise, increased the chances of bonding with them as friends and possible marriage partners.

This is the kind of analysis that becomes possible if the role of macro actors in influencing the nature of networks is considered. Religious networks do not pull themselves up by their bootstraps, but are fundamentally shaped by social forces that transcend them.

Interdependencies between micro situations

In the preceding paragraphs I have made an argument for the importance of the actions of powerful actors in explaining the rise of both religiously diverse social networks and increasing tolerance of religious diversity. The same argument, with the same caveats about its preliminary nature, can be made for the importance of horizontal relations between groups. Particular historical struggles, such as the Democratic Party’s campaign in the 1960s to get JF Kennedy elected as a Roman Catholic president, and the need to convince the electorate that fears of the papacy were unfounded, need to be considered. Similarly, one needs to reference struggles against WASP dominance of universities and businesses throughout much of the 20th century. All of these involved the mobilisation of large numbers of people into groups for the purpose of collective struggle.

As far as the current role of religiously diverse social networks is concerned, rather than their historical emergence, it can once again not be divorced from their relationship to other groups. In this respect one needs to know, for example, how religiously diverse neighbourhoods will interact with other identities, such as class or race. It is impossible to do this while abstracting from the dynamics of actual groups.

Interdependence and the aggregation problem

It is therefore essential to accommodate the role of individual and collective agency as well as the historical interaction of micro, meso and macro actors in accounting for the rise of bridging social capital and its supposed positive social effects. This is true not only for the case of religion discussed here, but also more generally. This has another, related, implication for the social capital literature. This concerns how the collective action problem is posed and solved in social capital theories deriving from the rational choice school.

Authors such as Putnam (2000) and Uslaner (2009) who apply the social capital concept to societies rather than small groups, tend to think that the nature of social relations in general or in small groups can be extrapolated to society as a whole. This reflects what Mouzelis describes as a logic of aggregation. I have already discussed Mouzelis’s critique of this: individual social capital situations cannot be aggregated because they are interdependent, and not independent. They are interdependent firstly due to powerful actors subsuming them. They are, in addition, interdependent because of the nature of their interconnections on the horizontal level.

Of course, theorists such as Putnam do not study small groups as such. They instead focus on data about groups that are derived from individual responses on survey questions (Ponthieux, 2004). They therefore do not focus on any particular groups, but rather on how the typical individual will respond in a typical group. Despite this, they still face the problem of generalising from the behaviour of people in typical small groups to society as a whole. Their approach is to follow a logic of aggregation, as Mouzelis claims. Because actual groups are not independent from one another and macro actors, this does not work. This is demonstrated below.

Their point of departure, as I said previously, is the collective action problematic. For them, this problem is merely a thought experiment: how cooperation is possible in general. Because they abstract from the dynamics of actual groups, it does not refer to the collective action problem of any group in particular. However, once one considers the collective action problem from this perspective, it becomes clear that it is connected to the interests of that group, which may be opposed (or aligned) to the interests of other groups. If a criminal gang manages to solve their collective action problem, and becomes more effective as a result, it will invite retaliation from other groups in society.

Our view of the collective action problematic should therefore shift from a singular to a compound one to avoid presenting a false view of collective action problems on the macro scale. To solve the latter, it is essential to consider the large number of smaller competing, and aligned, collective action problems and their relationship. In addition, any individual collective action problematic is embedded in a wider context that determines whether it is going to be solved in a cooperative or competitive direction.

One illustration of this is the way in which the trust and solidarity that grow within groups can intensify the conflict between these groups in the broader society. As a result, the solution of the collective action problem on the level of these groups impedes the solution of this problem on the level of society. This point has been made by both Granovetter (1985) and Sennett (2013). In such cases, the whole is less than the sum of the parts, as in-group solidarity reduces between-group solidarity, with the result that it detracts from the total amount of social capital in the society. Related to this is the fact that groups who are more successful at working together will tend to out-compete those that fail in this respect. The solidarity within groups is thus an essential element in their collective mobilisation to advance their interests in a competitive struggle.

What this demonstrates is that even if norms and networks solve the collective action problem for the typical small group (they do so only in a restricted way), it is fallacious to think that they will also solve it on the scale of society as a whole. The same argument applies to the supposed advantages of bridging social capital for society. It generally does not transcend the boundaries of the typical small group. And, as we saw above, because of its lack of autonomy from macro actors and macro institutions it is any way quite limited as a solution to the collective action problem on the micro level.

Conclusion

This article is devoted to a discussion of Robert Putnam’s work as the most prominent representative of what I call the cooperation school of social capital. I focus particularly on his work on religion in America, and his claim that bridging social capital in the form of diverse social networks can reduce communal tensions in diverse societies. My argument is that religiously diverse networks cannot be a primary solution to problems of religious intolerance, because both of these phenomena are the result of more fundamental social changes. Bridging social capital represents less the solution to problems of communal mistrust than an indication that this problem has already been partially solved. I therefore argue that the criticism that Putnam’s work is troubled by endogeneity is justified. This weakness of Putnam is widely shared by the rest of the cooperation school of social capital.

The social changes that are not accommodated in Putnam’s theory can be analysed with reference to Mouzelis’s notion of macro actors and horizontal interdependencies. Macro actors influence the rise and functioning of the social networks (and thus the social capital) of the less powerful. They reduce or enhance the capacity of the less powerful to cooperate more generally. This article has highlighted a number of such effects of macro actors: regulation of group solidarities and access to resources, as well as affecting the settings of interaction and norm enforcement. All of these influence the stability of social networks and trust levels within these networks.

In addition, a number of implications of the horizontal relations between groups were highlighted. For example, the cooperation within groups can intensify the competition and conflict between them. Local solidarity on the meso level can thus reduce solidarity on the macro level. Once one transcends the confines of particular groups, and broadens the analytical focus to relationships between groups, it becomes clear that there is not a single collective action problematic, but a multitude of such problematics. These need to be reconciled for an understanding of the possibilities for cooperation on the macro level. The horizontal and vertical interaction of numerous groups thus impacts the preconditions of the social capital of individual groups.

Consequently, the conditions for cooperation on the macro level cannot be derived from the behaviour of people in small groups. Micro social capital situations cannot be treated in isolation, but need to be reconciled to see how they affect one another, and are impacted by macro actors. This helps us to see that social capital is often quite fragile and subject to disruption, and thus subject to forces beyond the control of the local group. This insight is an important correction for policies that intend bootstrapping local solidarities into a basis for self-reliance and development.

This article, finally, indicates the limits of a theory of social capital that focuses either on the conditions for cooperation or on the attempts by individuals or groups to advantage themselves relative to others. Both the competitive and cooperative dimension need to be accommodated in a good theory of social capital. This is the case for two reasons. A consideration of the role of macro actors and macro institutions in social interaction will show that, depending on the social context, either of these dimensions will come to the fore. This leads to the second point: when the focus shifts to the relationships both within and between groups, it becomes apparent that cooperation and competition often entail one another. This aspect has not yet received enough attention in social capital theory. Analysing the relationship between the two is therefore one of the tasks that social capital has to engage with in the future.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Anthias

(2007). Ethnic ties: Social capital and the question of mobilisability. The Sociological Review, 55, 788–805.

Bourdieu

(1997). The forms of capital. In Halsey

Lauder

Brown

Wells

(Eds.), Education, culture, economy, society (pp. 46–58). Oxford, UK: Oxford University Press.

Burt

R. S.

(2004). Structural holes and good ideas. American Journal of Sociology, 110, 349–399.

Coleman

J. S.

(1988). Social capital in the creation of human capital. American Journal of Sociology, 94 (Supplement), S95–S120.

Denord

Hjelbrekke

Korsnes

Lebaron

Le Roux

(2011). Social capital in the field of power: The case of Norway. The Sociological Review, 59, 86–108.

Desmond

(2016). Evicted: Poverty and profit in the American city. London, UK: Penguin.

De Wet

. (1995). Moving together: Drifting apart. Betterment planning and villagisation in a South African homeland. Johannesburg, South Africa: Witwatersrand University Press.

Fine

(2010). Theories of social capital: Researchers behaving badly. London, UK: Pluto Press.

Fukuyama

(1995). Trust: The social virtues and the creation of prosperity. London, UK: Hamish Hamilton.

10.

Granovetter

(1985). Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91, 481–510.

11.

Greenberg

(1980). Race and state in capitalist development: South Africa in comparative perspective. New Haven, CT: Yale University Press.

12.

Hooghe

(2008). Voluntary associations and socialization. In Castiglione

Van Deth

Wolleb

(Eds.), Handbook of social capital (pp. 568–593). Oxford, UK: Oxford University Press.

13.

Julien

(2015). Bourdieu, social capital and online interaction. Sociology, 49, 356–373.

14.

Lin

(1999). Social networks and status attainment. Annual Review of Sociology, 25, 467–487.

15.

Lin

(2008). A network theory of social capital. In Castiglione

Van Deth

Wolleb

(Eds.), The handbook of social capital (pp. 50–78). Oxford, UK: Oxford University Press.

16.

Morrow

(1999). Conceptualising social capital in relation to the well-being of children and young people: A critical review. The Sociological Review, 47, 744–765.

17.

Mouzelis

(1974). Social and system integration: Some reflections on a fundamental distinction. British Journal of Sociology, 25, 395–409.

18.

Mouzelis

(1995). Sociological theory: What went wrong? Diagnosis and remedies. London, UK: Routledge.

19.

Mouzelis

(2008). Modern and postmodern social theorizing: Bridging the divide. Cambridge, UK: Cambridge University Press.

20.

Ponthieux

(2004, January 21–23). The concept of social capital: A critical review. Paper presented at 10th ACN Conference, Paris.

21.

Portes

Landolt

(2000). Social capital: Promise and pitfalls of its role in development. Journal of Latin American Studies, 30, 529–549.

22.

Putnam

R. D.

(1993). Making democracy work: Civic traditions in modern Italy. Princeton, NJ: Princeton University Press.

23.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon and Schuster.

24.

Putnam

R. D.

(2007). E pluribus unum: Diversity and community in the twenty-first century. The 2006 Johan Skytte Prize Lecture. Scandinavian Political Studies, 30, 137–174.

25.

Putnam

R. D.

Campbell

D. E.

(2010). American grace: How religion divides and unites us. New York, NY: Simon & Schuster.

26.

Putnam

R. D.

Goss

K. D.

(2002). Introduction. In Putnam

R. D.

(Ed.), Democracies in flux: The evolution of social capital in contemporary society (pp. 1–25). Oxford, UK: Oxford University Press.

27.

Rossteutscher

(2008). Social capital and civic engagement: A comparative perspective. In Castiglione

Van Deth

Wolleb

(Eds.), Handbook of social capital (pp. 208–240). Oxford, UK: Oxford University Press.

28.

Rothstein

Stolle

(2008). Political institutions and generalized trust. In Castiglione

Van Deth

Wolleb

(Eds.), The handbook of social capital (pp. 273–302). Oxford, UK: Oxford University Press.

29.

Sampson

R. J.

(1991). Linking the micro- and macrolevel dimensions of community social organization. Social Forces, 70, 43–64.

30.

Sennett

(2013). Together: The rituals, pleasures and politics of co-operation. London, UK: Penguin.

31.

Siissiäinen

(2000, July 5–8). Two concepts of social capital: Bourdieu vs Putnam. Paper presented at ISTR Fourth International Conference ‘The Third Sector: For What and for Whom?’, Trinity College, Dublin, Ireland.

32.

Smith

S. S.

Kulynych

(2002). It may be social, but why is it capital? The social construction of social capital and the politics of language. Politics & Society, 30, 149–186.

33.

Uslaner

(2009). Corruption. In Svendsen

G. T.

Svendsen

G. L. H.

(Eds.), Handbook of social capital: The troika of sociology, political science and economics (pp. 101–121). Cheltenham, UK: Edward Elgar.

34.

Williamson

O. E.

(1985). The economic institutions of capitalism: Firms, markets, relational contracting. New York, NY: Free Press.

35.

Woolcock

(1998). Social capital and economic development: Toward a theoretical synthesis and policy framework. Theory and Society, 273, 151–298.

36.

Woolcock

(2004). Why and how planners should take social capital seriously. Journal of the American Planning Association, 70, 183–189.