Mediating Structures: Their Organization in Civil Society

Abstract This paper explores a counter-weight to anti-democratic trends suggested by civil society theory: mediating structures (also known as ‘intermediate’ or ‘intermediary’ organizations). We look briefly at the range of ways in which the concept has been used in social science and then focus on literature which employs the concept specifically in the context of debates about sustainable democracy We discern from the theoretical literature four distinct functions said to be performed in democracies by mediating structures and we offer case examples of organizations which perform those functions in the contemporary real world. We conclude by providing pointers to how mediating structures might be identified, supported and sustained in the face of anti-democratic trends today.


Introduction
In both Europe and the United States, scholars and commentators are raising questions about the sustainability of traditional models of democratic and pluralist governance in the face of populist trends and their potentially anti-democratic, totalitarian impacts (for example , Albright 2018;Buller 2017;Etzioni 2018;Runciman 2018). It is timely, therefore, to consider what earlier theorists have suggested about counterweights to anti-democratic forces and how such theories might apply in the present day. 1 This paper, then, focuses on one powerful concept suggested by civil society theory; one which can help to sustain democracy: mediating structures (also sometimes referred to as 'intermediate' or 'intermediary' organizations). Broadly speaking, these are organizations positioned between local community activities on the one hand, and large-scale organizations of government operating at higher levels of population aggregation. Often but not always, these are organizations recognized in their respective countries as 'nonprofits' or 'third sector organizations'. Yet not all nonprofits are mediating structures and not all mediating structures are nonprofits; the two concepts (nonprofitness and mediating structures) have emerged from different theoretical traditions.
We look first in this paper at how the mediating or intermediate concept has been used in social science literature. We show how widely used the concept has been, not only in political analysis but also in sociology, economics, organization studies, development studies and social policy. We then move on to focus on literature which employs the concept more specifically in the context of debates about sustainable democracies.
We find that the literature talks in abstract terms about the importance of mediating structures for civil society and democratic governance but does not offer examples of what mediating structures are in a concrete, empirical sense. To enhance the potential applicability of the mediating structures concept to our present times, we draw out from the theoretical literature four distinct functions said to be performed in democracies by mediating structures. We then offer case examples that illustrate how these functions are performed in the current real-world context. We set out for each example organization, an outline of how they perform their distinctive functions, formally as well as 'behind the scenes'. We conclude by suggesting what are the organizational characteristics which civil society mediating structures have in common; thus providing pointers as to how they might be identified, supported and sustained in the face of anti-democratic trends.
Margaret Harris is the corresponding author.
© 2019 Harris and Milofsky, published by De Gruyter. This work is licensed under the Creative Commons Attribution 4.0 Public License.

The Ubiquitous Mediating Concept
As a preliminary to our conceptual and theoretical literature review, we note here that the 'mediating structures' term -and similar ones such as 'intermediary organization', 'intermediary structure' and 'intermediates' -are employed in some countries, some industries and some organizations with special meanings; they are national or professional jargon in effect . For example, translations of Chinese non-profit scholarship use the term 'intermediary organizations' as a synonym for the non-profit or third sector (Chan 1999) and in some countries (including the UK) national nonprofits and voluntary associations may refer to their regional organizational levels, those located between local groups and headquarters, as 'intermediary bodies'.
For the most part, these special terminological usages are not intended to carry any particular conceptual implication beyond the simple idea of being poised 'between'. In the remainder of this section, we identify literature that uses the terms to denote deeper conceptual distinctions and interpretations.

The Linking Idea
Although the metaphor of a 'network' is now most commonly used to denote the idea of webs of multiple ties between people, groups and organizations, some authors denote such complex systems of ties as 'intermediaries' or 'intermediates' (Ramu 1986). This reflects the seminal work of Granovetter (1973) in drawing attention to the nature (especially the strength or weakness) of the ties themselves and not simply the overall 'patterns' created by those ties. Grabowicz et al. (2012), for example, consider the differences in the interactions achieved by online and offline social groups and how these interactions can boost coordination of activities by social movements. On a broader level, scholars have looked at linkages created, within and across fields, in order to pursue policy advocacy activities (Lamping, Raab, and Kenis 2012;Mosley 2014;Wechsler and Friedrich 1997). Some have looked at structures whose prime purpose is to create links between industries and governments and others have looked at structures whose prime purpose is to create links between nonprofits and government (Tsukamoto and Nishimura 2006) or between individual nonprofits (Torry 2016).

Within-Industry Collaboration
In many industries, there are structures (also often called 'intermediary organizations') which draw together and link industry actors in order to coordinate action on matters of common interest, to provide support to small, local or developing organizations, to promote collectively the industry itself or/and to reduce competitive uncertainty (Kim and Jeong 2014)) . These may include professional associations or groups concerned with industry networking, promotion and lobbying. Some coordinate over regional, national and international geographical areas. Academics have studied examples in manufacturing industries (Hessels 2013), water management (Moss et al. 2009), research (Braun 1993;Suvinen, Konttinen, and Nieminen 2010) and in service-providing sectors. Casalino (2001) studied the advent of 'intermediate organizations' in US managed care systems and Anthony and Austin (2008) studied a single 'intermediary organization' in the social work field. The main point of note here is that these structures are linking and collaborative but they are primarily intended to serve the needs and interests of the industry itself.

Mediating Development Aid
In the field of international development the organizational challenges of 'mediating' between donor countries on the one hand and recipient governments and local NGOs on the other hand, are widely debated. The challenge of mediating here is tangible due to the need to bridge the geographical distance gap and ensure some degree of accountability for monies spent thousands of miles distant from donors. Thus a report on the reform of USAID policy in 2011 discussed available 'models' of 'Intermediate Support Organizations' which could potentially further the US Government's 'partnerships' with 'local non-profit organizations' in grantreceiving countries (Ashman et al. 2011). Sometimes the donors studied have themselves been consortia of country donors, as in Bruton and Williamson's (2005) study of 'intermediary nongovernmental organizations' delivering European Union grant funding in Northern Ireland.

Responding to Welfare Needs
Some non-profit scholars have conceptualized the body of voluntary and non-profit social welfare organizations in a country as performing an 'intermediary' function between a central or national government and local systems of welfare provision. Thus Bauer (1990), refers to the 'intermediary systems' of welfare provision established in Germany and the US, while Evers (1995) goes further and conceptualizes third sector organizations as constituting an interlinking 'intermediate area' 'intermeshing resources' of the state, business and household sectors in pluralist welfare systems. This conceptualization of an intermediary function has also been applied to health care systems (Lamping, Raab, and Kenis 2012).

Sustainable Democracies
The mediating or intermediary idea, then, has been widely used in a variety of ways in the social sciences. Yet a review of the literature reveals one further powerful application of the concept; one that is employed within debates about civil society and democracy: the idea that in sustainable democratic polities, there are entities which mediate between governments and other large institutions on the one hand and citizens at the grassroots on the other hand; "those institutions standing between the individual in his private life and the large institutions of public life" (Berger and Neuhaus 1977: 2;also Lomasky 2002). Mediating structures provide "a counterposition of state and society" (Selznick 1992: 505;also Calhoun 2011). 2 Although modern democratic states vary in the extent to which they encourage the development and survival of such intermediaries between themselves and their citizens (DiMaggio and Anheier 1990;Dubbink, Graafland, and van Liederkerke 2008), and although the size, nature and composition of the mediating layer may vary between countries (Kim and Kim 2018;Salamon, Sokolowski, and Haddock 2017;van Deth 1997), these structures are generally described in positive terms. They are seen to provide benefits by enabling information and viewpoints to be passed upwards and downwards between grassroots organizations and governmental agencies, by drawing together disparate viewpoints, filtering extreme ideas, reducing individual social isolation, and reducing the information deficits of governments and policy makers (Durkheim 1960: 28;Milofsky and Harris, 2017). This mediating or linking process, it is argued, helps to sustain liberal (or 'pluralist') democracies because it provides a bulwark against the formation of "mass" societies in which states and governments directly interact with, and ideologically control, anomic citizens (Skocpol 2011;Thomson 2005).
Within this broad idea about benefits for democracies of mediating structures, authors vary as to how specific they are about the organizational features of mediating structures. Some authors do specify particular categories of organizations which they believe uphold democracy (e. g. churches or fraternal organizations as in Skocpol 2011 or community activist organizations as in Boyte 2018;Kunreuther 2011). Others are less specific about the types of organizations they regard as intermediaries and refer more often to the functions that they expect mediating structures to perform to ensure healthy societies and sustainable democracies (Calhoun 2011).
Thus de Tocqueville (1838), who wrestled with the puzzle of how democracies could be sustainable in the face of tendencies for those with political power to draw increasing amounts of power to themselves and restrict the freedoms of citizens, thought that mediating structures -which he understood to be churches, schools, and fraternal and professional associations -served to provide alternative loyalty bases and alternative sources of information for citizens. Later, mass society theorists such as Kornhauser (1959) and Mannheim (1940) described the strong group ties and meaningful societal participation which they thought had historically been provided by church, clan, guild and local neighbourhood; ties which they saw as having been eroded by urbanization and industrialization. Berger and Neuhaus (1977: 34;also Glasser 1999) focused on 'mediating structures' which could provide security for individuals. They saw such structures as 'schools for democracy' and as including not only family, neighborhood and church but also 'voluntary associations' (which they understood as a body of people who had voluntarily organized themselves in pursuit of particular goals).
Social capital theorists (e. g. Coleman 1994;Putnam 2000), have tended to focus more on organizations which they believe can build trust, social cohesion and a sense of belonging, such as voluntary leisure associations and other community-based service-providing groups (Couto 1999;Nisbet 1969). Civil society theorists, in particular those who are 'neo Tocquevillians', also see the value of voluntary associations in democracies; especially for "curbing the power of centralizing institutions, protecting pluralism and nurturing constructive social norms, especially 'generalized trust and cooperation'" (Edwards 2009: 7) as well as for increased 'civic engagement' (Almond and Verba 1980;Lanero, Vazquez, and Gutierrez 2017). Putnam (2000), Skocpol (1999), and Hunter (1993), writing in a US context, are especially concerned about the apparent erosion of national organizations with regional and state chapters as well as grassroots participation. These latter are seen as the means through which matrices of social ties are developed horizontally and vertically within states and federations (Milofsky 2001(Milofsky , 2008bWarner and Lunt 1942). They link local associations and interest groups to national public policy making (Bushouse 2017). Their ability to do so has been enhanced in recent years by the internet and social media. As citizens participate in multiple organized activities, they inevitably experience value conflicts; experiences that are thought to educate them about the complexities of society, encourage toleration of diversity, facilitate critical evaluation of government policies and make them less likely to embrace extreme value positions (Coser 1956;Foley and Edwards 1996;Walzer 1997).
Our review in this section of the literature which links the mediating structure and sustainable democracies concepts, suggests then, that 'voluntary association' and engaged citizen interaction, are widely seen to be crucial elements in mediating structures (Ahrne 1994;van Deth 1997). Whatever organizational form these mediating structures might take, they are said to be embedded in, and to foster, active interaction between local citizens. They also orientate citizens to, and educate them about, participation in the public space (that is, the broader social milieu within which citizens exist). Yet writers rarely make explicit what kinds of 'voluntary associations' they have in mind nor do they offer exemplary organizations.
Some critics of Putnam (2000) (e. g. Thomson 2005) have pointed to the variety of organizational forms encompassed by the generic term 'voluntary association' and have speculated about which ones are most likely to provide the benefits for democracy which Putnam's work assumes; without 'meaningful' participation and multiple memberships, the assumed benefits for democracy-enhancing social capital accumulation may not accrue (Jochum and Brodie 2013;Wollebaeck and Selle 2002). Still, in general, the literature turns repeatedly to the idea of 'voluntary associations' as a key part of mediating structures without giving us a clear picture of what these important kinds of organizations are like in practice. The literature tends to tell us the functions that these organizations perform, rather than how they are built to enable them to fulfil their functions (Comas 2016).
Although largely silent about what mediating 'voluntary associations' might look like in practice, the literature does enable us to draw out a list of the functions which authors have seen to be necessary characteristics of mediating or intermediate organizations in democracies. Some authors suggest just one crucial defining function and others suggest combinations of two or more functions. Looking across the literature reviewed, we can identify the following four functional characteristics as distinctive to mediating structures.
-Type One: Organizations which link individuals and grassroots community activities to governmental agencies, public policy makers and national nongovernmental agencies across several organizational tiers, and between which information flows horizontally (across local spaces) and vertically (between levels of aggregation). Skocpol (2011) calls these 'fellowship organizations'.
-Type Two: Participatory organizations whose members are intensely involved in discourse and action about public issues but are impervious or resistant to external control from the state or other mediating organizations.
-Type Three: Organizations that foster public discourse and enable a focus on the common good; what Glasser (1999) has termed 'republican democracy'.
-Type Four: Organizations which frame a culture that tolerates diverse viewpoints, builds shared values, facilitates social ties, builds trust, draws on diverse organizational forms, and encourages broader civil engagement.
In the next section we offer some brief examples to illustrate how these four types of mediating structures (with organizational features identified in the theoretical literature) are currently manifested in practice; that is, we show how the four identified functions can be translated into real organizational forms. We present US examples and draw on current and past community research by one of the authors of this paper, Milofsky.

Mediating Structures in Practice: Some Examples
In this section we move beyond our review of earlier theoretical literature, to explore what mediating structures with the potential to help sustain democracies, can look like in the contemporary real world. We address separately each of the four functions identified above from the literature because we expect, drawing on organizational theory, that they might entail different organizational mechanisms. Our examples emphasize central tasks that must be carried out for identified functions to be enacted, although we recognize that in practice our four 'function types' may not be totally distinct from one another and that some mediating organizations may enact more than one of the four functions. All of our examples are built around formal organizations that serve as anchor institutions for voluntary, participatory activities.
Type One: Organizations which link individuals and grassroots community activities to governmental agencies, public policy makers and national nongovernmental agencies across several organizational tiers.

Example: Volunteer Fire Companies 3
Volunteer Fire Companies have a well-developed, autonomous presence at the local level and a significant number of participants who are actively involved on a regular basis. Members participate in maintenance activities at the weekend and will turn out to handle emergencies every day of the year. Each local company has a distinctive culture and style and is autonomous. At the same time, and as with the fraternal societies studied by Skocpol (1999), there are clear flows of information and interaction between members and companies, both vertically and horizontally, ensuring organizational integration. Companies in a community or region interact with each other to carry out training, emergency response, fundraising, and civic boosterism, enabling and reflecting ongoing horizontal flows of information. The local members who are most integrated into the regional and state organization may serve as skill instructors for new members and other participants in local companies. They may travel to other communities to provide instruction there. In addition, some local members are in (subordinate) mentoring relationships with people in distant companies who are particularly skilled. They work with these master trainers when they travel to provide instruction at geographically distant locations.
Leaders and members also participate in large regional and national conventions and meetings, enabling vertical information flows. They become educated about issues and perspectives being developed and advocated in other places. They also learn how to address government leaders about problems that have been identified at lower levels of the organization. These extra-local connections are important for passing along technical innovations and formulating new regulations and policy requirements that are codified at the state level as well as in the requirements and practices of fire insurance companies.
While the linkages are partly a way to carry out changes in state and national training requirements, transmission of those changes must go through the volunteer system (teachers and trainers are not paid) and are thus incorporated into the local organizational culture in a participatory way.
Type Two: Organizations which are impervious or resistant to external control from the state or other mediating organizations.

Example Women's Shelters 4
In contrast with Type One organizations, which actively create horizontal and vertical links and accept some degree of influence from other organizations, Type Two organizations, are generally resistant to the influence of governmental or other mediating organizations. Yet in order to achieve their goals, they must nevertheless remain connected to local communities and the wider society. They reconcile the apparently contradictory pulls of autonomy and connection through having a stable internal culture that is meaningful to their members but that also reaches out to the wider world -without intimidation or cooption (Arnold 1995).
Such organizations may have a limited life span if they are formed around charismatic leaders or single issues which become less popular over time (Rodgers and Knight 2011). Equally, small-scale movement organizations may be so successful that they are able to force political adjustments by the regional or the national government so that laws are passed to support their issue and public funds are allocated to them. The challenge for local-level social movement-linked organizations, such as our example women's shelters, is to remain organizationally active and politically critical, as well as effectively engaged with both national and regional organizations (Staggenborg and Taylor 2005). 5 Women's shelters originated in the early 1970s. Feminist theorists argued that threats of male violence were one reason male dominance persevered (Brownmiller 1975). Activists created community groups that provided shelter to women who were abused by their husbands. By re-analyzing emergency room records, others showed that family violence was more common than most had thought (Stark and Flitcraft 1996). The shelters grew in number at the local level and as a broader policy movement. By the mid 1980s legislative reforms were enacted that gave funding to women's shelters and rape prevention organizations, and legitimacy to shelter leaders in things like training police to handle family violence calls more effectively (Marshall 1971).
Following a radical feminist ideology, early shelters took the position that when women came for protection they must leave their abusive partner and learn to build a new, autonomous life for themselves and their children (Loseke 1992;Wharton 1987). This clashed with values held by main stream funders like the United Way that argued all abused women needed protection even if they did not want to leave their husbands. 'Pro-choice' values advocated by shelters ran afoul of conservative members of the community who would support organizations in their efforts to protect families from violence but who were 'pro-life' (Milofsky and Morrison, 1994). Government funding once obtained also required that shelters keep careful records, give extensive training to volunteers, and accept a more mainstream ideology that would allow them to work effectively with the courts and the police. Agencies seemed to succumb to social pressures that have been noted as accompanying success and leading activist political movements to become more centrist (Milofsky 2008a;Piven and Cloward 1979).
A difference between the feminist movement and other activist movements, like the HIV/AIDS movement (Chambre 1989) is that issues around the oppression of women are never "solved". Issues that lead to activism continually arise and produce new energy. So women's shelters may have become more like mainstream social agencies, but they continue to be part of the activism that opposes sexual abuse in the workplace and in nonmarital social relationships such as the "Me Too" movement or in conservative political regimes that encourage misogynistic speech (van Natta 2010). They continue to be radical and oppositional to government despite their maturity as organizations and despite the maturity of the social movements (see footnote 5) to which they are allied (Staggenborg and Lecomte 2009).
Type Three: Organizations which frame and enable a focus on the common good rather than sectional special interests.

Example: The Little League 6
Like Type One and Two Organizations, this category includes organizations that link on both the horizontal and vertical planes. The links may be with others carrying out similar activities and operating with a similar values-set aimed at building mutual support and voluntary action outside of the political realm; and links may extend to international contacts. Our example for this category is Little League Baseball and Softball. This is the largest association of youth sports leagues in the world. It involves boys (and some girls) between the ages of 9-12 throughout the world in organized baseball leagues .7 While Little League Baseball is a world-wide association, its control is concentrated among residents of the town where it was founded, Williamsport, PA. With total revenues in 2017 of over $30 million against expenses of $29 million the organization achieved a profit of slightly over $1 million which was added to net assets of about $82 million. It has five regional centers in the U.S. and four other regional centers covering different parts of the world. Each regional center is supported by a system of local leagues which themselves are joined in a hierarchical regional structure of elected leaders.
There are several core office employees but all other leaders in the organization are volunteers and a strong ethic of voluntarism is promoted throughout. Simply to operate, a large number of adults must give their time to coach and instruct young people, to referee games and to manage the paperwork aspects of organizing teams, collecting personal records on individual players, and reporting competition results. Participation is active, focused, and tightly organized from the local to the international level. Yet none of this activity is oriented towards political activity or to promotion of special interests.
Little League is an important mediating organizational force in that it supports active, autonomous volunteering across age groups in a large number of communities (Fine 1979). It integrates these local organizations through a set of rules that have evolved from practice over decades and that local organizations are required to accept. Like Skocpol's (1999) fraternal organizations, local-level organizations send representatives to governing bodies at sequentially higher levels of aggregation until the world level is reached. The world organization provides administrative and instructional support. Through these means voluntarism and participation are tightly integrated, helping to create a national and international social system without relating to government agencies.
Type Four: Organizations which frame a culture which tolerates diverse viewpoints, builds shared values, facilitates social ties, builds trust and encourages broader civil engagement.

Example: The Kettering Foundation 8
The Kettering Foundation is an operating foundation committed to fostering public dialog among civically engaged citizens, discussing and developing concepts of democracy, and working to support individuals and institutions to solve problems -using tools of democracy (Rosen, Merritt, and Austin 1997). Kettering's activities give focus to activities that foster public dialog. 9 It is explicitly oriented to 'deliberative democracy' where participants represent clashing interests and values but nonetheless join together to discuss how to resolve social and community problems (Matthews, McAfee, and Charity 2002). It creates settings where this kind of discussion can happen at the local, regional, national, and international levels.
Although over its 90-year history Kettering has supported research, and although one of its programs brings democracy thinkers to the Foundation in Dayton, Ohio, for residencies that may last up to two years (Boyte 2009), much of the foundation's activity is focused on "deliberative democracy" workshops. These bring citizens from around the world to Dayton to discuss how communities collectively can identify their pressing problems and work to resolve them (Milofsky 2001(Milofsky , 2008b. While much of its work involves building, supporting, and teaching methods for fostering democracy at the local level, the Foundation also has supported international initiatives to resolve national and international conflicts. Thus it has sponsored deliberative democracy meetings where younger leaders of factions from within a country, or from countries that have contesting relations with each other, come together over a period of years to develop perspectives on peace-making. The focus is less on trying to resolve contested issues than to create a neutral environment where present and future leaders can get to know each other, discover shared values, and consider together what it would take to achieve peace in their situations. The hope is that as time passes and these younger leaders become advocates for their own group, they will have strong, constructive relationships with their opposite numbers in other groups. This approach has had some striking successes, such as with U.S./Soviet Union resolution of conflict in the early 1990s (Saunders 1999).

Discussion
We began this paper by talking about the variety of ways in which the term "mediating structures" has been used in the social science literature. We then focused specifically on literature addressing the role of mediating structures in helping to sustain pluralist democracies. Such structures have been seen as not only important in their own right, but also as performing a crucial role within civil society. Without their integrating and participatory activities within social and political systems, society would be atomized; sectors and levels of the political system would not be linked or tied together. Mediating structures foster and support civic activity and participation at all levels and make links at aggregate levels to people at the grassroots.
As well as establishing the theoretical significance of mediating structures in sustainable democracies, we have moved in this paper beyond the theory, to indicate how these structures can work in practice. From the literature we were able to discern four key functions that mediating structures are said to perform in democracies, giving us the foundation of a typology of mediating structures. Then, drawing on Milofsky's current and recent research in the US, we provided an example of a real-world organization for each of the four functions of mediating structures we had identified from the theoretical literature. For each example organization, we set out an outline of how they perform their distinctive functions, formally and 'behind the scenes'.
We chose each example because it appeared to us to conform closely with the literature-identified 'type'. Each is organizationally distinct in terms of the functions it performs in civil society and we can also see that each type has a different organizational form. The first type links active local organizations to regional and national organizations where issues, interests, techniques, and skills are shared in a reciprocal fashion up and down the hierarchy. New ways of seeing and doing tasks are integrated with governmental and industry requirements even though local cultures and practices retain vitality and are not subordinate to the regional and national culture.
The second type of organization also produces a local to national dialog but relationships with other organizations and government institutions are more oppositional. This type may be close to social movements and organizationally vulnerable if public interest in their issue wanes or if the problem they were set up to address is resolved. The third type of organization has a system of volunteer participation that is separate from the political culture and without any major input from governments or their agents. The fourth type of organization focuses on fostering active citizenship as a complex and meaningful process in itself. As with Type Two organizations, there is recognition that citizens often become active because a social problem has emerged in their community; they enter the public space in a quest to figure out what to do about the problem. These organizations foster public dialog events and encourage people with different values and different political points to speak out.
Although the four example organizations are distinct in terms of their organizational mechanisms, we can also see that they have some common characteristics. Most obviously they all create and consolidate linkages between individuals and groups. Moreover, the linkages are enacted on the two planes identified by Skocpol (1999); that is both vertically and horizontally. Additionally, in each example organization we see that the vertical and horizontal linkages happen not only within the organization itself but also across the organization's formal boundaries, extending out to other individuals and groupings. These organizations, then, are 'tied together' internally, creating networks of participation and they also become part of broader networks as they reach out to create linkages with other organizations.
The multiplicity and complexity of the ways in which our four example organizations create and are integrated into organizational networks seems to distinguish them from the kinds of organizations identified in the first part of our literature review where we set out social science literature in which 'mediating' terminology is used but is not related to the concept of sustainable democratic structures. For example, organizations that tie NGOs to donors and recipients perform comparatively simple functions that do little more than connect.
So we can now tentatively suggest that the kinds of mediating structures which are important for sustaining civil society and pluralist democracy are ones which make connections between individuals and groups on horizontal and vertical organizational planes; and ones which do so both internally and externally; that is, their linking functions are enacted across their formal organizational boundaries. Looking across our four examples we can additionally identify some further apparent common characteristics.
First, it seems that civil society mediating structures enable and facilitate changes in policies and culture as information and ideas pass between different actors in different directions. Second, grassroots community participation is an integral part of their operation and is built into their organizational design and organizational values. Third, civil society mediating organizations are tied to aggregate level organizations in meaningful ways. Local participants attend regional, national, or international meetings and events with a substantive focus, support for and enhancement of the local organizational culture. There is also a series of network connections that move up and down the vertical dimension of social life. These higher level meetings are influenced in terms of meanings, political connections, and reinterpretation of issues by the perspectives, actions, and interests of lower level organizational members. Meanwhile local participants are given legitimacy, cultural support, technical education, and ongoing network relationships with the higher level organizations. The relationships, cultural meanings, and quality of organizational productivity entailed by these vertical relationships create strongly integrated organizations and communities; they are, in theory and practice, 'mediating civil society organizations'.

Conclusion
This paper has been researched and written in response to widespread concern about threats to democracy; concern that democratic dialog and participation in civil society are declining in Western polities (Milofsky and Harris 2007). While theory tells us that mediating organizations are important for democratic participation, there has been little discussion of what these organizations actually are, how they operate, what makes them energetic and vital, and what might be done to support them. This is the contribution we hope our paper has made; to show what mediating structures look like in today's real world. The paper is, then, a first effort to identify types of mediating organizations and supply examples that provide context and richness to those types. With more empirical research we might learn whether all mediating structures have the potential to sustain democracies, whether such mediating organizations are at risk and what steps might be taken to strengthen them organizationally.

Notes
1 There is, of course, an extensive literature on what is meant by the concept of democracy. For the purposes of this paper, we are using the term broadly to refer to forms of 'liberal democracy' traditionally valued in Western Europe and North America (Runciman 2018 Armony (2004) has critiqued the assumption that there is necessarily an association between democratic polities and civic engagement and liberal democracies, although his analysis is largely based on emergent rather than established political systems. 3 Material for this section was collected by Milofsky in Summer of 2017 during an ethnographic field study of volunteer fire companies in a Pennsylvania small town. Results have not been published but there are taped interviews that support statements contained here. 4 Much of the material for this section is based on ethnographic research by Milofsky and his students that has been published and is documented in the text. Most of the data collection was done as part of classes over a period of decades. Our work with women's shelters in Pennsylvania small towns has continued in the form of new student projects. Some of the more recent information we present in this section has not been documented in formal notes but has come out of informal conversations with directors and out of student papers. This methodology has been described as 'slow sociology' (Hunter 2014). 5 We distinguish here and throughout this paper between 'social movements' and 'organizations', concepts grounded in their own theoretical frameworks and literatures. Although some organizations may spring from, or be embedded within, social movements (such as the womens movement or the public journalism movement) the two concepts are distinct.
6 Most of the information for this section was collected from the Little League website (littleleague.org). However, this is a central Pennsylvania local organization that Milofsky and his students have encountered and studied over the years. One of the important contributions of the Little League is that it encourages and interfaces with teams in communities in Pennsylvania small towns and elsewhere in a way that is encountered frequently in community research. 7 It should be noted that Little League is different from the 'sports clubs' which have been studied by nonprofit scholars (e.g Balduck, van Rossem, and Buelens 2010;Doherty and Hoye 2011) in that they are organizations which are both embedded in their local communities AND linked to each other through national and international networks. Nonprofit sports clubs serving only adults and their local communities can face stiff competition from for-profit gyms. Little Leagues also arise from a community-oriented ethos but they differ in that their main activists are parents who want to create and participate in activities with their children. 8 Much of the information for this section was gathered from the Foundation website at Kettering.org. However, Milofsky had direct experience with the foundation (documented in Milofsky 2001) that involved a trip to the foundation, many contacts with a close colleague who worked at the Foundation and explained its programs, and many contacts with diplomat Harold Saunders (Saunders 1999) who worked with the foundation and who explained its international peace programs to us. 9 We focus in this paper on just one example of a nonprofit organization that has a long history of fostering public deliberation events. We would note, however, that fostering citizen engagement through public dialog is not an unusual activity. Thus the goal of the 'public journalism' movement (Rosen, Merritt, and Austin 1997) is to educate about public issues without taking a partisan political stance. What began as a somewhat academic, philosophical discussion has become more immediate and active with the growth of social media which has allowed blogging, direct news reporting and reflections on public and philosophical issues (Rosenberry and Burton III 2010). Again, service learning in schools and colleges promotes public dialog by embedding young people in situations where they directly view and experience challenges facing communities and society (Boyte 2018). In this way they learn about the public sphere and how to act within it, but in a way that is politically nonpartisan.