Introduction

For the best part of three decades, citizens have been changing the way that they interact with their democracies. While voter turnout and engagement with traditional membership-based groups like political parties, trade unions and even churches has been declining, citizens have begun engaging with ‘other areas of social and political behaviour’ (Mair 2005, p. 8) across civil society, in ways that have embraced new technologies, professionalised what was once primarily volunteer, and organised social, economic and environmental interests in new ways (e.g. Cameron 2020; Halpin 2014; Han 2009; Jordan & Maloney 2007; Mair 2013; Putnam 1995; Skocpol 2003; Vromen 2017). Nevertheless, or perhaps even as a result of these changes, the democratic expectations on third sector, or civil society organisations (CSOs), are high. This includes CSOs as vehicles for pushing back on the perils of populism or government restrictions on civil liberties (Bolleyer 2021a; Mouffe 2019), monitoring government activity and aiding transparency and accountability (Keane 2009; 2022), and generally acting as transmission belts between the collective interests of people and communities, their elected governments and accompanying democratic institutions (Albareda 2018; Jordan & Maloney 2007). Whether CSOs are able meet these expectations is a live and open debate, particularly in the context of the aforementioned shifts in the socio-political landscape. In general, from a healthy democracy point of view, there is clear interest in how CSOs operate as well as in the impacts of their actions.

In this article, we take a common factor in these discussions, the connections that CSOs have with people and communities, and create a framework based on how power and relationships are mediated and negotiated between CSOs and their constituents—the Loci of Power and Connection (LOPAC) framework. In doing so, we take an expansive view of civil society and CSOs, aligning with Bolleyer’s (2021b, p. 498–9) definition of ‘collective action organisations’ which have political goals and which ‘attempt to influence public policy, directly or indirectly’. This excludes state and market bodies and institutions, which have a primary focus on determining (and implementing) public policy or generating private profit, respectively. It includes a diverse range of CSO organisational types, including interest groups, service providers, philanthropic foundations, community centres, universities, churches, political parties, think thanks, and various other non-government, not (primarily) for profit organisations. CSOs typically organise around collective goals and engage in activities which include, but are not limited to, attempting to influence public policy, partnering with governments and other funders on charitable service delivery, providing education and training, as well as attempting to secure and maintain funding for these organisational activities. Too many excellent scholars to name specialise in different subsets of CSO—including whether CSOs are membership-based and aim to be representative of those members or not. Unlike many of these scholars, we are less interested in organisational forms or structures, or the ways in which regulators, funders or researchers delineate between CSOs. Instead, this paper focusses on understanding the democratic impact of different activities across civil society. As Schlozman et al. note: ‘membership associations and memberless organisations mobilise the same kinds of techniques in the pursuit of policy influence’ (2015, p. 1018).

The LOPAC framework aims to deepen understanding of the implications of the choices CSOs make. An interplay of policy interests, priorities and actions, within the limitations of time, context and available resources, mean that CSOs have a limited carrying capacity—a limited number of issues they can feasibly act upon and particular ways they can act depending on their contexts (Halpin & Fraussen 2021). In this paper, we suggest that these decisions have democratic implications and thus offer a framework for scholars, practitioners and policymakers to explore these. The next section positions democratic practice and impact within the roles for CSOs in a pluralist democratic system. This is followed by a close examination of different types of ways which the democratic connections between people and CSOs can be conceptualised, focussing on scales of decision-making and the intensity and purpose of the connection. After building the framework out of these theoretical discussions, the resulting quadrants of CSO activity types—Bare Essentials, Crowd Control, Honeycomb, and Closed Doors—are related to common CSO activities such as systemic advocacy, service provision and capacity building. Finally, the framework is examined through key theoretical considerations and trends, including the questions raised in this introduction.

Democratic roles of civil society organisations

From a normative point of view, Fung (2003) offers six democratic functions for CSOs: the intrinsic good of association and freedom to associate; civic socialisation and political education; resistance and checking power; interest representation; public deliberation and the public sphere; and direct governance. These represent ideals for the variety of democratic impacts CSOs can have, without assuming that they necessarily do either individually or collectively. As noted earlier, one way to conceptualise these functions collectively is to view CSOs as transmission belts in the broadest possible sense, with CSOs acting as an organised intermediary between a group of people and some form of consolidated power. As Jordan and Maloney note ‘The first sort of democratic virtue bestowed on groups lies in their provision of a responsive and direct form of ‘particularised’ linkage—groups are key democratic transmission belts’ (2007, p. 7). Not all CSOs perform each of Fung’s six functions, although many perform more than one of them, depending on contextual ‘democratic circumstances,’ such as how established ‘basic democratic procedures’ like parliaments and electoral systems are, or levels of ‘human and economic development’ (Fung 2003, pp. 516–18). An inclusive view of CSOs as transmission belts sees a ‘mixed ecology’ of CSO types (p. 536) where it is reasonably simple for CSOs of all shapes and sizes to claim, ‘they are acting on behalf of a wider community, or that the wider community will benefit from the changes, or goods and benefits they seek’ (Jordan & Maloney 2007, p. 24). Thus, CSOs can perform a variety of democratic roles; whether they do so effectively is another matter entirely.

One end of the transmission belt relates in particular to the ways that CSOs engage with their constituencies—members, supporters, beneficiaries, stakeholders or otherwise. These connect to the more associational roles of CSOs, the impacts of people interacting with each other within or facilitated by CSOs. This is the practice of Alexis de Tocqueville’s little schools of democracy (1835/2003); the development of social capital, particularly in membership-oriented associations, that provided a counterweight to centralised power in democracies like the United States of America (USA) (Putnam 1995; Skocpol 2003); and the idyllic practice of conflict resolution through deliberative democracy (e.g. Dryzek 2002; Fung 2005), whereby ‘participants should try to establish the common good through discussion’ (Jordan & Maloney 2007, p. 13). From a representative standpoint, these CSO practices can provide a form of input legitimacy through democratic participation and responsiveness (e.g. Schmidt 2020) and engender pluralism across a wide and diverse political landscape, acting as a safeguard against overreach by centralised power (Dahl 1978; Jordan & Maloney 2007). It has been the changes to this form of CSO activity over time—reduced engagement from citizens in the internal workings of CSOs—that has led to concerns about declining social capital and democratic deficits creating disconnection between everyday citizens and those who contest and decide public policy, including CSOs (Mair 2013; Putnam 1995; Skocpol 2003). That said, the ‘cumbersome consultation processes’ that involve regular citizens or members, have been observed to ‘hamper the capacity of CSOs to efficiently respond to policy demands in changing environments’ (Albareda 2018, p. 1220). Many parts of the sector have professionalised over a similar time period, and increased public and philanthropic funding into the sector has been accompanied by economic rationalist expectations around reporting and accountability that can on the surface appear contrary to the expectations of citizens (e.g. Goodwin & Phillips 2015).

The other end of the transmission belt focusses more squarely on the outcomes or impacts that CSOs achieve on behalf of people—the results of interest representation. To reframe Schmidt (2020), if input legitimacy is democracy of the people, then output legitimacy is government for the people. The first is analogous to CSOs being responsive democratic actors; the second to them being responsible ones. As we have suggested, when it comes to a CSO’s democratic practice and impact, neither is necessarily contingent on the other. Fung observes that many important democratic outcomes are achieved by CSOs that “are large lobbying organisations whose members (emphasis mine) do little more than contribute financial resources” (2003, p. 524). Similarly, the bureaucracies delivering a variety of essential human services at scale across numerous democracies mirror more the impenetrable bureaucracies of Soviet rule, satirised by Frank Kafka in The Castle and The Trial, than they do the democratic “ideal” of deliberating fraternal associations (e.g. Cottam 2019). This is an acknowledgement that those striving for democratic outcomes, despite holding and attempting to practice democratic values, often find themselves confronted by entrenched structures of power that need to be overcome by more than reasoned debate. These activities are incorporated into most conceptions of democratic pluralism, and in particular those that regard a degree of opposition or agonism as essential democratic features (Honig & Stears 2011; Medearis 2015; Mouffe 1999). As Fung notes ‘When persuasion fails, muscular political mobilisation—by social movements, political parties, or interest groups—is warranted,’ (2005, pp. 415–6). At their best, these endeavours provide output legitimacy for CSOs through their democratic impact, though as we have seen, doing so effectively can come at the expense of more deliberative and associational activities.

We are beginning to see that CSOs can perform a variety of democratic roles in a variety of ways with a variety of impacts. Systemic advocacy can involve actions as diverse as lobbying behind closed doors and participating in protest movements, each of which can shift the direction of public policy. Citizens can be deeply involved through community organising activities, or by simply donating a small amount to help a CSO place content in the public domain and contribute to public debate. Charitable service delivery can involve largescale exercises conducted by some of the biggest CSOs in the world helping thousands of people daily, or self-organised community groups where passionate volunteers look out for community members. Research can be conducted at arms-length from participants or by working alongside them. Fundraising can be done via intimate dinners or at the click of a button. All of these artificially binary activities can involve democratic contributions and significance, via practice or impact, input or output, through CSOs working with people or acting on their behalf. At the same time, they can just as easily not. Lobbying can just as easily further the interests of elites. Community-based activities can struggle to have impact at scale. One kind of activity can impact a CSO’s ability to perform another. As transmission belts, CSOs are clearly conduits and mediators of people’s interests and aspirations, conducting activities with, for and alongside people. But to understand these implications of their actions, we now need to look a little deeper at how this plays out.

Ways of conceptualising community connection

The framework at the centre of this paper has been created through an examination of some of the ways that scholars have explored how organisations interact with people and communities, and what this might mean in terms of their democratic function. Arnstein (1969) and Lawson (1980, 2005), for example, establish scales of community connection based on degrees of constituent control. Arnstein’s Ladder of Citizen Engagement, created through an examination of US Federal Government programs and policies, focusses on ‘the extent of citizens’ power in determining the plan and/or program’ (1969, p. 216). Arnstein suggests that the typology could just as easily be applied to organisations like churches, and separates the eight rungs of the ladder into three categories: degrees of citizen power, degrees of tokenism, and non-participation, all of which can be mapped onto a scale from democratic to undemocratic. Within this scale, Arnstein distinguishes between connections that are partnerships, that ‘share planning and decision-making responsibilities’ (1969, p. 221); between tokenistic connections like surveys and public consultations, which offer ‘no assurance that citizen concerns and ideas will be taken into account’ (p. 219); and then connections like ‘rubberstamp advisory committees or advisory boards’ (p. 218) that Arnstein suggests, ‘signifies the distortion of participation into a public relations vehicle by powerholders’ (p. 218).

Lawson establishes a similar scale through her concept of linkage between political parties and citizens (1980, 2005), noting the fundamental importance of the ‘links between citizens and policymakers’ (1980, p. 3) and that linkage ‘has often been used as a metaphor for democracy and even as a synonym for representation’ (1980, p. 8). At one end of Lawson’s scale, participatory linkages are described as democratic, while coercive ones are undemocratic. The items in the centre of the scale—the more commonly practiced coercive, clientelistic and responsive linkages—can either advance or inhibit democracy, depending on how these activities are conducted, based on whether the links serve to centralise or decentralise a political party’s power (2005). Lawson’s concept of linkage highlights ‘the protean adaptability of parties,’ (Merkl 2005, p. 7) a description that can easily be extended beyond political parties to other types of CSO. Do they ‘really help us with the tasks of building democracy?’ Lawson asks (1980, p. 5). The answer, similar to the connections explored in Arnstein’s Ladder of Citizen Engagement, is “sometimes”.

Both Arnstein’s and Lawson’s scales ascribe democratic impact via participatory mechanisms that are related to decision-making power that citizens have. Essentially, they are talking about democratic practice. Bolleyer & Correa (2022) make a comparative distinction between members as an ‘organisational resource’ and ‘member influence on intra-organisational decision-making [that] allows members to hold organisational leaders to account’ (p.520). A blunt assertion of this may be: ‘If democracy is interpreted as rule by the people then the more participation there is in decisions, the more democracy there is,’ (Verba & Nie 1972, as cited in Jordan & Maloney 2007, p. 5). However, as we have explored above, these forms of democratic practice can inhibit democratic impact—democratic struggles are not always ‘fully theorised as relations among citizens’ (Medearis 2015, p. 24). We therefore need to take a more expansive view of how CSOs connect and interact with people and communities that take in notions of democratic impact.

In this regard, both Halpin (2006) and Han (2014) explore the connections between CSOs and people in terms of how they balance different types of connection in order to achieve their desired impact. Halpin explores the democratic legitimacy of interest groups by distinguishing between groups pursuing representation and those acting in solidarity (2006). This distinction implies different types of connection between an organisation and their affiliates—those who make up the group—and their beneficiaries or constituency: ‘those whose interests the group’s advocacy is aiming to advance’ (p. 925). Because CSOs acting in solidarity are not claiming to be representative—they might for example be an environmental or animal rights group, for which meaningful representation is not practical; or an organisation wanting to act as allies to populations they are not a part of, such as First Nations communities—the test of democratic legitimacy is not necessarily the extent of decision-making power that affiliates of these CSOs have. Instead, it is whether the outcomes of their actions shift power-relations in a way that meets the democratic goals of their constituency.

While Halpin looks at the structural aims of different organisations, Han (2014) explores different types of activities conducted by singular organisations. Han’s focus is on what leads to ‘higher levels of activism’ and engagement within membership-based civic associations (p. 156). In doing so, they distinguish between transactional mobilising, CSO activities aiming to connect with or activate a maximal number of people as quickly as possible, and transformational organising, deeper connections that develop relationships between people and the organisation and develop leadership capabilities. In line with the democratic concerns around trends towards professionalisation and disengagement with membership-oriented CSOs noted above (Mair 2013; Putnam 1995; Skocpol 2003, etc.), Han observes that CSOs ‘face the temptation to focus exclusively on mobilising because of the ease with which mobilising can now be done’ (p. 124). However, Han concludes that high-engagement CSOs consciously balance both types of activity, ‘not only to reach the broadest possible pool of potential activists but also to invest deeply in a subset of those activists and transform them into civic leaders’ (p. 156).

Clearly, understanding the democratic relationships between people and CSOs is a multidimensional task. If we consider only of the democratic roles or potential of CSOs along a single dimension—such as how much decision-making power people have over the organisation, or whether they are membership or non-membership focussed—we miss much of the nuance of how CSOs actually operate and contribute. In practice, all CSOs balance the tension of their need to interact with people in their constituency with their need exert influence of some form over more powerful interests. Examining only one aspect of this tension cannot tell the full story.

Key dimensions: power and connection

Two common threads emerging from the work of the scholars discussed above are (1) variations of decision-making power and (2) the kind of connection a CSO has with people. Decentralised decision-making, that hallmark of participatory and deliberative democracy in practice, is not only a redistributive mechanism—in terms of redistributing control over common resources within CSOs and other institutions—it also helps individuals ‘distinguish between (their) own impulses and desires,’ and learn ‘to be a public as well as a private citizen’ (Pateman 1970, p. 25). Yet, as we have established, the reluctance of those with power to cede it requires the periodic mobilisation or centralisation of people power. As Wolin notes, ‘Democracy was born in transgressive acts, for the demos could not participate in power without shattering the class, status, and value systems by which it was excluded’ (2018, p. 106). Thus, both decentralised and centralised decision-making have democratic utility—the former redistributes power and allows new civic leaders to emerge, while the latter allows for the concentration of collective resources towards strategic goals such as shifting structural inequities.

We can observe CSOs of all kinds practicing both centralised and decentralised decision-making. A heavily centralised CSO in effect operates like a non-publicly listed corporation, with decision-making and power transfer (i.e. succession) controlled at the top. This may have great utility in terms of being able to make quick decisions and direct resources towards strategically defined activities, like public advertising campaigns, yet it replicates simultaneously the elite power structures which CSO activity often aims to disrupt. In contrast to this kind of CSO, a deeply decentralised CSO would be run for and by the members, structurally resembling an idealised civic association or political party, federated into multiple layers of representation of similarly membership-run groups at a state, regional or community level.

Another pervasive factor we have seen is the inevitable connections between CSOs and people. The idea of relational connections leans into normative ideals of participatory democracy (Arnstein 1969; Lawson 2005), with some seeing the ‘web of relations housed in an association’ having ‘tremendous value, greater than the goals of the association’ (Kateb 1998, p. 37). Han observes that: ‘Creating relational conditions, then, refers to the organisation's attempt to create a larger context within which people feel like the social relationships they desire (with each other and with the organisation) are more likely to emerge’ (Han 2016, p. 298). And while reduced participation in civic associations, churches and political parties, has been observed to reduce the relationality of these CSOs (Putnam 1995; Skocpol 2003), the importance of nurturing relationships based on trust remains a core feature of an array of CSO activities, including community development, community organising and community-led service delivery (Chambers 2020; Cottam 2019). In this way, relational CSO activities seek to generate impact through connections with others, whether at a deep grassroots level or within the halls of power.

At the other end of this connection spectrum, transactional CSO connections and activities attempt to generate impact through scale. They perceive the connection as an ‘exchange relationship’ (Han 2014, p. 94) with the idea to make the transaction as quick as possible. This includes the use of digital communications technologies to crowdsource donations and signatures on e-petitions, resources which CSOs can then direct to other strategic activities such as developing materials and public campaigning (Vromen et al. 2022). Though the connection between the CSO and the person is fleeting and transactional, often literally, scholars have begun to explore these activities as ‘new modes of citizen engagement’ which offer different but still democratically significant alternatives to falling rates of deeper more direct participation in civic associations (p. 5). Transactional connections can also describe the way that CSOs interact with people through charitable service delivery. The primary focus of many free services that CSOs deliver, particularly those that are funded by governments through welfare states like Australia or the United Kingdom, is to get people out of crisis and away from the service provider as quickly as possible. However, while provisions like free legal assistance, transitionary housing and primary health care undoubtedly help hundreds of thousands of people on the brink every year, critics point to the lack of flexibility within these heavily bureaucratised systems, as well as the lack of attention to dealing with the systemic factors that push people into crisis in the first place (e.g. Cottam 2019; McKnight 1995).

Returning to some of the variations in CSO activities explored earlier, we can make a distinction between relational and transactional connections. Where fundraising via an intimate dinner would involve relational connections, the same amount of money could be raised through an email appeal where the only connection involves a handful of clicks. Lobbying politicians face-to-face is a relational advocacy activity whereas generating signatures for an online petition is transactional. Most service delivery at scale is transactional—the aim is to get the person out of contact with the system as quickly and efficiently as possible, to get to the inevitable next person in the queue. On the other hand, effective case management or youth work can require building deeper relational connections between a CSO worker and a community member. Decision-making itself can be relational or transaction—with deliberative democratic practice at the relational end of the spectrum and expressing individual preferences through anonymous voting being at the other.

The Loci of Power and Connection

There appear no straightforward answers when it comes to CSOs balancing decision-making power and the intensity of the connections they have with people and communities. Involving people in decision-making develops critical democratic skills, social capital and provides input legitimacy at one end of the transmission belt, yet challenging entrenched power successfully often necessitates strategically targeted actions developed and controlled by a smaller few. Similarly, while deep relational connections can ensure that the interests of people and communities are at the forefront of what CSOs choose to do, transactional connections can allow organisations to generate the resources required to operate at scale, including taking on and competing successfully with entrenched corporate and political power at the belt’s other end. In relation to this dilemma, the Loci of Power and Connection (LOPAC) framework visualised in Table 1 offers a tool for CSOs and those who study them to understand the potential democratic ramifications of the choices they make.

Table 1 The Loci of Power and Connection (LOPAC) framework for CSO activities
Full size table

The LOPAC framework counterbalances the scales of decision-making power and connection intensity discussed in the previous section. The horizontal x-axis relates to decision-making power and whether it is decentralised or centralised; the vertical y-axis represents intensity or authenticity of the connection, whether relational below the x-axis or transactional above it. This creates four quadrants representing distinct types of CSO activity, which have been named to signify figurative aspects of the activity type. These are: (1) Bare Essentials: transactional connection and decentralised decision-making; (2) Crowd Control: transactional connection and centralised decision-making; (3) Honeycomb: relational connection and decentralised decision-making; and (4) Closed Doors: relational connection and centralised decision-making. These are explored in more detail below, including how common CSO activities might align with different activity types.

Bare Essentials activities

Bare Essential activities involve transactional connections between a CSO and those outside it, along with decentralised decision-making in terms of how people engage with the CSO. The defining characteristic in terms of engagement is that elements of the CSO are there for people to engage with, but whether people do or not is entirely optional. The connection involves a quick exchange. This might include a CSO having some core infrastructure like a website, social media platform or an accessible physical location that people can visit or click on. It can include meeting legislative requirements such as having a constitution and producing annual reports that are made available for download. Bare Essential activities can also include fundraising staples like charity stores, donation boxes on the counters of retail stores and even selling raffle tickets in shopping malls. These kinds of activities suggest no particular effort on behalf of a CSO to build a meaningful relationship with people, regardless of what the transaction might contribute to the overall aims of an organisation. Through these activities, people can choose whether or not they interact with the organisation, but that interaction is unlikely to affect the substance or direction of the organisation, or the person involved. A CSO’s use of social media advertising or a petition on a website to harvest email addresses fits squarely in this category—there is no further purpose for the interaction other than building the contact database of the CSO.

Although they are in themselves shallow, Bare Essential activities and strategies can form the crucial backbone or scaffolding for other CSO connections and activities. People in databases can be mobilised for activities such as attending events and making microdonations. Yet while these Bare Essentials can provide a springboard for more impactful activities, CSOs that only focus on the Bare Essentials are essentially bare. These activities are unlikely to make any positive democratic contribution, despite the amount of time and resources some CSOs put into them. Social media platforms can clearly be an excellent base infrastructure for other more impactful activities (Murthy 2018); however, this mode of engagement has also been criticised for fostering weaker forms of activism (Gladwell 2010). Like many technologies—the outcome depends on what you do with it. On their own, Bare Essential activities are unlikely to address systemic barriers that people face, develop civic leaders or provide essential services that help people in crisis. CSOs with limited time and resources might well question the purpose served by their Bare Essential activities, and whether those resources would be better targeted elsewhere.

Crowd Control activities

Crowd Control activities involve transactional connections between CSOs and people, while the decision-making for how engagement occurs, and for what purpose, is heavily centralised. These activities and strategies are some of the most common and public forms of CSO activity—largescale service delivery, micro donations and even mobilising mass protests. Similar to state-run processes like elections or a welfare state, these activities are highly organised and data-optimised, aimed at connecting with or activating as many people as possible in the shortest achievable timeframe. Effective service delivery by charities or mobilisations by interest groups or social movement organisations provide these quick and transactional connections, which ideally have a very low barrier to participation. People should be able to make online donations through as few clicks as possible, and ideally on a recurring basis. Someone who needs advice or assistance should get it through the first service provider they call, rather than be referred around from waiting list to waiting list. High density protests can be mobilised through carefully crafted alerts via email, phone or social networking sites. The aim of Crowd Control activities is defined and measurable but, compared to more relational CSO activities, impact is generated purely by virtue of the number of people involved.

The democratic potential of Crowd Control activities is clear. Being able to provide effective and timely services and support at scale offers a societal safety net for those in crisis. The provision of free services such as health, education and housing are similarly regarded as important to ensuring equal opportunity. Crowd Control activities are essential when it comes to advocating for social and political change and ensuring that voices which are collectively marginalised can be heard through participation in public discourse and influencing decision-makers.

However, despite these potential outcomes, the centralisation and bureaucratisation of Crowd Control activities also creates a clear hierarchical power relationship between the organisation and the people involved, somewhat antithetical to the democratic legitimacy of CSOs. A tightly controlled feedback process into a CSO’s policy platform, or a post-service client survey, might be considered Crowd Control activities—the type of connection which Arnstein identifies as consultation or placation, ‘a higher-level tokenism because the ground rules allow have-nots to advise but retain for the powerholders the continued right to decide,’ (1969, p. 217). If untempered by more relational and decentralised factors, CSOs specialising in Crowd Control activities may find themselves reinforcing elite power structures. CSOs need to ask whether the services they offer are defined by what their funders want or by what communities want; interest groups need to ask whether their advocacy aims for systemic change or merely replacing one set of elite actors with another.

Honeycomb activities

Honeycomb activities involve relational connections between CSOs and people, along with decentralised modes of decision-making and engagement. These kinds of activities are the fertile breeding grounds for the trust, relationships and social capital that are regarded as essential to CSO democratic activity. People engaged in Honeycomb activities build sustained and deep connections, where they have high degrees of autonomy over what they are doing and the collective direction of the CSOs involved. A quintessential Honeycomb-focussed CSO might be the local sporting club run by volunteers, or a small local charity established by a group of friends. Decision-making is collectivised and largely done by consensus. The strength of the group comes down to the strength of the relationships between those involved and a sense of mutual obligation and service between them—which likely involves various other self-organised and relational activities such as providing food at games, or recycling sporting equipment through different age groups. We can see Honeycomb activities in operation throughout the broader CSO landscape, such as self-organising mutual aid networks, reading groups and capacity building practices like training workshops, consensus decision-making and community organising. These activities highlight the interactions between people as being crucial to the outcomes that are generated.

Further, the strength of relationships, trust and commitment built through Honeycomb activities can become the foundational launchpad for more centralised or transactional activities. The relationships built through regular social picnics or barbecues provided by CSO can encourage previously sceptical people to access available services or help the service providers understand what kinds of support people actually need. Time spent building relationships between activists helps build the solidarity and commitment required to run long and intense advocacy campaigns. Decentralising decision-making and mentoring within an organisation can help build commitment as well as developing the organisational leaders of tomorrow.

Despite these benefits, conducting Honeycomb activities can be time and resource intensive, and the outcomes are not nearly as easily quantified or measured as more transactional pursuits. Making decisions through deliberation and consensus can also be impractical in times of crisis, when quick and decisive responses are required. As noted earlier, Honeycomb activities can be viewed as ‘cumbersome consultations’ which get in the way of CSOs achieving impact (Albareda 2018, p. 1220), particularly in the professionalised high-pressure quick-turnaround operating environment inhabited by the modern major CSO. Together, these factors may cause CSOs to drift away from Honeycomb activities, despite those in leadership positions in organisations inevitably acknowledging how important these activities are.

Closed Doors activities

Closed Doors activities are also relational; however, the concentration of power and decision-making is highly centralised. These activities see the leadership of CSOs planning and developing relationships with key selected individuals in order to advance the goals of the organisation, typically at an elite or leadership level. The quintessential Closed Doors activity would be the CEO of a CSO regularly lobbying public officials, including making them guests of honour at fundraising dinners, conferences and similar events. Closed Doors activities include building relationships with major donors, whether HNWIs (high net worth individuals) or philanthropic foundations, or the leaders of other CSOs. They include the development of alumni networks and the curation of leaders’ summits and closed-door workshops. Unlike Honeycomb activities, Closed Doors activities are carefully curated and controlled—they are categorised by their exclusivity. Participants are in a sense encouraged to see value in the activity or relationship by the virtue of participating at the exclusion of others. The voice of the broader membership or supporter base of a CSO may be represented by a leadership figure within the CSO. This could be the CEO, or in some organisations an appointed representative of a broader group, such as CSOs which have designated “Lived Experience Leads”.

Closed Doors activities can be extremely important for negotiating and procuring democratic outcomes. CSOs often pursue Closed Doors relationships to build the resources, including personnel and funding, that are needed to conduct other activities. There are times when a handful of CSO leaders, particularly those representing larger organisations or institutions, can benefit from sitting down face-to-face and trying to resolve differences or negotiate agreements with a degree of privacy or confidentiality. There are times when courses of action need to be strategically agreed and plotted behind closed doors, rather than in the public domain. Even modern campaigns that embrace elements of decentralisation and grassroots leadership require the sharp focus of a narrower and more selectively centralised coordinating group (Mogus & Liacus 2016; Tattersall 2010).

Of course, it should be fairly obvious that an overreliance on Closed Doors activities can severely undermine the democratic claims to legitimacy of CSOs. We can see this in the calls for transparency around who donates to political parties and what they might expect in return (Ratcliff & Halpin 2021); or when CSO leaders, inadvertently or otherwise, develop ‘interest in creating and maintaining a particular set of images that support policy-making arrangements that reflect dominant ideas’ rather than pursuing activities which might substantively change the conditions faced by their constituents (Jordan & Maloney 2007, p.20). All CSOs are likely to engage in Closed Doors activities in some form or another, whether these are the volunteer executive committee meetings of a sporting club, or major campaign planning workshops attended by CEOs. CSO leaders engaging in these activities may need to consider how their views and actions are authorised, and whether the outcomes from these meetings have material and structural significance for the people and communities they aim to represent.

LOPAC trends and challenges

We have now outlined four distinct types of CSO activity, each of which has clear democratic utility and drawbacks, and connected them to a variety of commonplace CSO activities across a broad spectrum of CSO types. Clearly, different CSOs will perform these different activity types to different extents, depending on their context, along similar lines to the concept of CSO hybridity (e.g. Minkoff 2002). The degrees to which these activities may be combined lends itself to future empirical application of the framework, within CSOs themselves and across the broader CSO ecosystem, where organisations frequently collaborate in order to generate impact. Pre-empting these potential applications, this section connects the LOPAC framework to some of the open CSO questions and challenges foreshadowed earlier in the paper.

Earlier, we distinguished democratic practice from democratic impact as what CSOs do compared to the outcomes of their actions. As such, given the discussion around those agonistic, activist or unfortunately oppositional aspects of democracy (Fung 2005; Honig & Stears 2011; Medearis 2015; Wolin 2018) there may be the temptation to conceptualise democratic practice along the lines of decentralised decision-making equalling an idealised form of democratic practice, with centralised decision-making being the reality required for democratic impact. However, both Honeycomb (decentralised decision-making) and Crowd Control (centralised) activities exhibit distinct modes of both practice and impact. The transactional mobilising and transformational organising identified by Han (2014) as essential ingredients of impactful organisations in a sense map neatly onto the Crowd Control and Honeycomb activity types—the more relational activity of organising, cultivating agency and leadership amongst participants, builds the connections, commitment, social capital and trust that can then be directed in a controlled fashion towards a particular outcome. The honey can be scraped, bottled and put to good use.

Han’s work suggests an important interrelationship between Honeycomb and Crowd Control activities—'people power the associations’ and ‘associations power the people’ (p.28)—that warrants deeper investigation for other types of CSOs besides the membership-based organisations which Han observes. What role, for example, might Honeycomb activities play in interest groups that focus almost exclusively on more transactional forms of engagement such online petitions or digital fundraising? Vromen (2017) has suggested that CSOs in this situation—facing as they do the claim of diminished democratic practice through a lack of relational connections with the people in their databases—have begun employing storytelling as a strategy to ‘successfully elicit an effective response among members and targets, helping them to feel connected with a campaign and a movement’ (p. 226). And yet, put into the LOPAC framework, storytelling of this kind likely remains a Crowd Control activity. While personal narratives may generate an emotional response, which in turn creates a sense of connection which makes people more likely to contribute to the CSO's donation requests, they do not have the hallmarks of Honeycomb activities, particularly the relational aspects. Narrative storytelling may in this sense provide only a slick veneer of connection between a CSO and the communities they represent.

These Crowd Control and Honeycomb activities can be expected to resonate similarly with a CSOs ability to operate as a transmission belt between the interests of citizens and the state. Bare Essentials and Closed Doors activities can similarly contribute towards a CSO’s input and output legitimacy. For example, many CSOs use their websites as tools for displaying key documentation and other regulatory requirements, which form some of the minimum expectations for input legitimacy that allow CSOs to operate and continue to secure financial support from various sources. These include annual financial reports, governing documents like constitutions, and tax status. These signifying CSO documents also build legitimacy through less formal means, providing the names and credentials of the governing boards and leadership teams of CSOs, creating the impression for observers that the organisation has the experience and personnel required to deliver outcomes. Other Bare Essentials activities like establishing social media platforms and email distribution lists perform a similar function, allowing CSOs to broadcast a curated impression of their activities to their key primary audiences. In this way, Bare Essentials activities can enable CSOs to keep in regular communication with their constituencies at both ends of the transmission belt, generating a public image of passive authorisation to continue operating unchallenged. Whether this legitimacy is “earned” in terms of democratic impact is a different matter entirely; an analysis of these Bare Essentials activities as “essentially bare” when it comes to democratic connection suggests that it is not.

Examining Closed Doors activities highlights a similar tension within CSO legitimacy. As noted previously, activities like strategically coordinating activities with coalition partners or lobbying elected officials and other decision-makers are necessary parts of a CSO’s toolkit. These kinds of activities generate impact through trust, relationships and negotiation rather than more public forms of conflict resolution we might see via Crowd Control activities. Indeed, a CSO’s supporters likely expect the organisation to have these kinds of relationship and exert influence in this manner, particularly those CSOs that combine government-funded service delivery with systemic advocacy. However, when CSO’s lean too far into these kinds of relationships, they can run into accusations that they are unrepresentative and elitist, or that they are perpetuating some of the issues about power imbalance and exclusion raised by within criticism of deliberative democracy (Dryzek 2002; Fung 2005). It may thus be important for CSOs to balance Closed Doors activities with other kinds of LOPAC activities, in order to avoid generating negative democratic outcomes and a lack of community-level authorisation over time.

Finally, the LOPAC framework offers a lens through which to view the democratic impacts of the professionalisation and marketisation of CSOs. The concerns of scholars who have observed a decline in engagement with member-based associations (Mair 2013; Putnam 1995; Skocpol 2003) can in part be understood as a shift over time from CSOs favouring Honeycomb activities to instead favouring Crowd Control ones. At a base level, Skocpol (2003) describes this as technological advances enabling leaders to build power and influence without having to first build a social base. Another way to frame this is that groups that desire to challenge systemic oppression and barriers to participation take what they perceive to be the most efficient path to achieving these goals. Mass association and relationships via Honeycomb activities were clearly required to address many of the post-feudal and post-industrial concerns of the 19th and 20th Centuries; the success of these activities along with technological development paved the way for the prominent agenda-setting role of today’s highly professionalised interest groups. Similarly, the successes of Honeycomb oriented service provision at a community level led to large scale public and philanthropic funding for charitable service delivery, which over time became less and less relational, particularly through the impacts of marketisation and other staples of neoliberal economic governance (e.g. Goodwin & Phillips 2015). Where the political aspirants of yesteryear proudly declared their credentials as members of federated associations with autonomous local branches (Skocpol 2003), today’s hopefuls may be more likely to trumpet leadership roles of large charities and interest groups that specialise in Crowd Control activities.

The potential dangers of an over-reliance on Crowd Control and Closed Doors activities can be observed by revisiting the potential legitimacy for CSOs that rely on Closed Doors activities. The historic trend away from Honeycomb activities—as a result of increased professionalisation and the subsequent stronger relational bonds between CSO, political and corporate leaders—means that over time the participation of everyday citizens in the decision-making around and delivery of public policy becomes less and less necessary, feasible or convenient, and the disconnection between people and the state begins to be mirrored by a disconnection between people and organised civil society. One conclusion about this democratic disconnection is that it can be exploited by populist political actors, who often incorporate calls for fairer and more representative democracy into their pitch for popular support (Mouffe 2019).

At the same time, it is possible to observe a resurgence of CSOs embracing Honeycomb activities, through practices such as community organising and community-led collective impact. The underlying principles of community organising are quintessentially Honeycomb, focussing on supporting the building of relationships around areas of common concern and then advocating collectively on them. The Alinsky-school of community organising gained particular prominence following the 2008 US presidential election, which connected community-level organising with mass-scale politics, and has since been expanding across the USA, the UK and Australia, including via the Black Lives Matter movement (Mundt et al. 2018). In comparison, collective impact practices aim to address systemic social issues through government, industry and communities working together around common goals (e.g. Kania and Kramer 2011). Initial reviews of collective action projects in action have noted that, in order for these collaborations to work, open discussions about power differentials in decision-making need to be addressed, particularly in ways that endeavour to centre community-level leadership in activities that aim to solve community-level problems (Graham, Skelton & Yettica-Paulson; Salignac et al. 2018; Smart 2017).

Applying the LOPAC framework to these ongoing debates about the practice and impact of CSOs offers different yet clearly interconnected ways to understand the democratic implications of CSO choices. While all types of LOPAC activities clearly have some utility, the brief analysis above suggests a crucial historic role for Honeycomb activities, which connect CSOs to communities through forms of relational connection and decentralised decision-making. If Honeycomb activities are forms of “Gold Standard” for democratic connection, CSOs looking to achieve democratic impact may want to consider rebalancing their suite of activities. The LOPAC framework explicitly describes types of engagement activities rather than types of organisations to allow this form of analysis. Some of the more binary or linear scales outlined at the beginning of this paper, which were used to generate the LOPAC framework, can lead us into over-simplifying whether a CSO or other institution is democratic or not. The LOPAC framework has the potential to sit within the very large grey areas of “it depends how you do it” which dominate both Arnstein’s Ladder of Citizen Participation and Lawson’s conceptualisation of linkage between political parties and citizens, and offer a tool through which CSOs and their observers can understand some of consequences of CSO activity. Finally, while some types of CSOs may tend towards one type of LOPAC activity or another, the reality is that CSOs are likely to utilise and combine multiple of these activity types in their day-to-day work.

Conclusion

This paper has proposed a framework for understanding the activities of civil society organisations (CSO) in relation to their democratic connections with people and communities. This was done by juxtaposing whether the connection is transactional or relational with whether engagement is centralised or decentralised. The LOPAC framework posits four distinct types of CSO connection—Bare Essentials, Crowd Control, Honeycomb and Closed Doors—each of which has the potential to make important contributions towards democratic outcomes. In this way, the framework aligns with democratic models which are pluralist, representative and oppositional (Dahl 1978; Honig & Stears 2011; Medearis 2015; Wolin 2018). The framework aligns with a democratic desire for a multifaceted civil society where people freely build and create connections and organisations which help resolve power struggles through reasoned deliberation and negotiation. It also acknowledges that there are moments when this preference for relational connection and collective decision-making needs to be temporarily put aside in order to create structural responses that can overcome entrenched opposition and deliver outcomes for a broad section of the population.

An initial application of the framework through the lens of historic trends in CSO structures and activities suggests that the professionalisation and centralisation of CSO activity have occurred at the expense of the Honeycomb activities, which themselves represent core associational democratic values in action. Relatedly, while Bare Essentials, Crowd Control and Closed Doors activities all have some democratic utility, without the temperance of Honeycomb activities, the transactional connections and centralisation of power make these potentially undemocratic activities for CSOs to pursue in the long run. The framework lends itself towards a variety of methods of empirical testing and analysis, which may serve to demonstrate its potential utility for third sector scholars, funders and CSOs themselves. In particular, the framework offers CSOs a way of understanding the democratic implications and contributions of CSO activities through how they connect with the people and communities they endeavour to represent and serve.