Structural Change Through ‘Collective Action as Democratic Practice’: Linking Grassroots Democracy With Social Justice

Abstract

Iris Marion Young argued that tackling structural injustices that arise from complex interactions requires collective, political action. Nevertheless, it remains unclear what kind of collective action leads to social justice and how. This article aims to fill this gap by developing the idea of ‘collective action as democratic practice’ that emphasises crafting democratic institutions from the bottom-up to re-orient structural processes in ways that resist oppression and domination. This departs from Young’s own account of collective action as engaging in communicative action to pressure powerful agents. It also departs from previous attempts to conceptualise structural change through the cumulative, rather than collective, actions of individuals changing their behaviours independently of one another. I argue that my approach is a more promising way to theorise collective action from a structural perspective. This approach has wider implications for understanding the outward-looking political potential of self-organised projects by citizens.

Introduction

In recent years, there has been a surge in discussions of structural injustice, a concept that has attracted renewed interest among contemporary political theorists since the groundbreaking work of Iris Marion Young. Since the publication of a series of articles by Young (2003, 2004, 2006a) on this subject and her posthumously published Responsibility for Justice (Young, 2011), there has been a growing literature that either engages critically with her work, or that applies her theory to empirical cases. Young presented her ‘social connection model’ of responsibility, wherein people who participate in structural processes that produce injustice have a political responsibility to ‘join with others who share that responsibility in order to transform the structural processes to make their outcomes less unjust’ (Young, 2011: 96). Following Young’s own focus, much of the subsequent discussion on structural injustice focusses on analysing the social-connection model of responsibility and to whom and how to assign responsibility for them. ¹

Nevertheless, an account of how agents should discharge their responsibility for correcting structural injustice is not as well developed. Although Young argued that tackling structural injustices that arise from complex interactions requires collective, political action, it remains unclear what kind of collective action leads to social justice and how. ² This essay seeks to fill this gap and to advance Young’s theory by offering an account of the kind of collective action needed for structural change towards social justice. Specifically, I develop an account of ‘collective action as democratic practice’ that pushes the boundaries of democratic decision-making in ways that resist oppression and domination. In contrast to the dominant idea of collective action as voicing protests to injustices to pressure powerful agents, collective action as democratic practice resists dominant structural processes by directly enacting democratic processes in governing social and economic activities. I aim neither to refute the significance of political advocacy, nor to argue that my account is the only way to correct structural injustices. However, I want to argue that this type of collective action deserves more attention from a structural perspective because it seeks to restructure social relations through people’s practices; that is, through the social processes that produce structural relations themselves.

This essay proceeds as follows. In the second section, I review Young’s structural approach, which underscores her relational conception of justice, and explore her argument that changing structural processes requires concerted collective action. The third section goes on to examine existing arguments regarding how to theorise collective action for structural change in the context of Young’s theory. Specifically, I consider Robin Zheng’s (2018) ‘role-ideal model’ of responsibility, which holds that structural change can occur through the interactive effects of individuals extending their social roles, and Robert Goodin and Christian Barry’s (2021) argument that structural change may occur through the cumulative effects of people individually ‘taking a stand’ against injustice. I argue that both accounts take an individualistic and, ultimately, aggregative view of social change while failing to provide a model of collective action towards social justice. In the fourth section, I present my account of ‘collective action as democratic practice’ as a better way to conceptualise collective action from within a structural approach. Instead of pressuring powerful agents to reform existing institutions, this type of collective action involves citizens directly ‘crafting institutions’ from the bottom-up to re-orient people’s practices. Moreover, I argue that the democratic organisation of such institutions is the key to social justice understood in terms of relational equality. Finally, the fifth section considers the objections that these types of collective practices would not necessarily translate into society-wide justice because they fail to cultivate a sense of solidarity beyond members, or because they can support the status quo by absorbing discontent. By taking the case of worker cooperatives as a concrete and paradigmatic example of collective action as democratic practice, I argue that we can find resources within a structural approach to address both objections.

Young’s Structural Approach and Relational Conception of Social Justice

Young distinguishes between harms that occur as a result of individual interactions and structural injustice. The latter concerns the ways in which complex interactions by masses of people create patterns in how people are socially situated (Young, 2011: 70–71). Individuals’ specific actions or particular policy interventions are neither just nor unjust by themselves since justice involves how people’s interactions mediated by social institutions work to position people in relation to others. Structural injustice occurs when these social processes place certain groups of people in relationships of systematic oppression and domination by others (cf. Young, 2006a: 114, 2011: 54).

Young’s focus on structure as the subject of justice resonates with John Rawls’s influential institutional theory of justice, which sees questions of justice as matters of society’s holistic institutional arrangements rather than those of individual ethics or particular policies. ³ Nevertheless, she formulates structure in a more dynamic way than Rawls’s basic structure; that is, not in terms of a specific set of ‘major’ institutions that can be separated from others, but as a recursive and evolving process between the actions of individual actors mediated by social institutions on one hand, and objectively experienced patterns of social relations and positions produced by those interactions on the other (Young, 2000: 95, 2006b, 2011: 64–74). According to Young (2011: 70–71),

Social structures are not a part of the society; instead they involve, or become visible in a certain way of looking at the whole society, one that sees patterns in relations among people and the positions they occupy relative to one another . . . we take a structural point of view on social relations when we try to see how the actions of masses of people within a large number of institutions converge in their effects to produce such patterns and positioning.

Thus, in contrast to Rawls’s ideal theory that presupposes a ‘moral division of labour’ ⁴ between a just basic structure and individuals’ pursuing their respective life plans within it, for Young, societal actors themselves are implicated in the creation and transformation of the background structure. This means that, while we need to distinguish between the evaluation of people’s actions at the level of individual interaction on one hand and that of structural justice on the other, the latter is not independent from people’s choices and actions (Young, 2011: 73). In her non-ideal theory, how societal actors behave matter for social justice; they can act in ways that either promote or inhibit social justice. ⁵

Young’s formulation of structure in this way has the following implications. First, it indicates her relational egalitarian conception of social justice. Young (1990: 9, 37, 39) negatively defines social justice as the absence of institutional conditions of oppression and domination, two social conditions that she describes as constituting injustice. While oppression consists in ‘systematic institutional processes which prevent some people from learning and using satisfying and expansive skills in socially recognized settings’, domination refers to ‘institutional conditions which inhibit or prevent people from participating in determining their actions or the conditions of their actions’ (Young, 1990: 38). This understanding of justice as the institutional conditions that enable self-development, recognition and self-determination for everyone presupposes the idea of people’s equal status as moral agents, capable of forming their own perspectives and of shaping shared conditions on an equal footing with others. ⁶ Social justice, then, involves how structural processes position people in different ways in terms of their opportunity to realise these moral capacities. ⁷

A second implication of Young’s approach is that it accounts for people’s agency and responsibility in relation to social injustice while, at the same time, avoiding the individualisation of responsibility for their social situation that is produced through the complex interactions of masses of people. In other words, it avoids victim-blaming those who are disadvantaged by structural processes, while making room for people to exercise their agency in relation to unjust structures. Young develops this argument in terms of her ‘social connection model’ of responsibility in contrast to the backward-looking ‘liability model’, which seeks to pin blame for identifiable perpetrators of harms. According to the social-connection model, people who participate in structural processes that produce injustice bear a forward-looking responsibility to ‘join with others who share that responsibility to transform the structural processes to make their outcomes less unjust’ (Young, 2011: 96). While no individual can personally be blamed for causing structural injustice, people nevertheless have a political responsibility to engage in collective action to transform them because their actions, taken together as a whole, are implicated in producing and reproducing those structures. Both Young’s own argument and much of the subsequent literature have focussed on analysing the social-connection model of responsibility, and to whom and how to assign responsibility for them (see McKeown, 2021). While this has produced a rich literature that develops and applies her model to different empirical cases, her conception of collective action and how exactly it leads to social justice is not as fully spelled out.

A good starting point is to examine what Young herself says. Young characterises intervening in structural processes through collective action as a ‘political responsibility’ in two senses. The first is that ‘political responsibilities derive from the social and economic structures’ in which people ‘act and mutually affect one another’ (Young, 2004: 376). That is, its subject matter is not the direct well-being of particular others, but the justice of the background structure that condition people’s actions (McKeown, 2018). Second, this responsibility is political in the way that it should be discharged:

by politics or the political I am referring to the activity in which people organize collectively to regulate or transform some aspect of their shared conditions, along with the communicative activities in which they try to persuade one another to join such collective action or decide what direction they wish to take (Young, 2004: 377).

Thus, the first step for transformative political action involves critically reflecting on our social positions and acknowledging that one participates, along with many other people, in social processes that produce unjust outcomes. Second, discharging that responsibility involves bringing attention to this fact through communicative action and persuading others to enjoin in collective action to transform them (Young, 2004: 380, 2006a: 123, 2011: 112). But as others have pointed out, Young leaves us hanging here without fully specifying how this might be done (cf. Goodin and Barry, 2021; McKeown, 2021; Zheng, 2018). ⁸ The question, then, is what kind of collective action should we take with others and how does it lead to more just outcomes?

Collective Action for Structural Change

The purpose of this section is to examine some of the ways in which Young’s idea of collective action has thus far been theorised to come to a better understanding of what kind of collective action would lead to social justice and how.

One conception of collective action involves pressuring powerful agents to transform social institutions that mediate structural processes. This appears to be the main interpretation that Young herself advances. According to Young (2011: 150),

Social change requires first taking special efforts to make a break in [structural] processes, by engaging in public discussions that reflect on their workings, publicizing the harms that come to persons who are disadvantaged by them, and criticizing powerful agents who encourage the injustices or at least allow them to happen. ⁹

Young mentions on several occasions that political intervention in background structures often involves working through state institutions, which wield considerable power over the conditions within which societal actors conduct their activities (cf. Young, 2006a: 123, 2011: 112, 151). Yet, influential institutional actors are not limited to the state. Her well-known example of the anti-sweatshop movement involves putting pressure on institutional actors like ‘local manufacturers and large multinationals’ responsible for their production (Young, 2011: 133–134), as well as ‘institutions that purchase them in bulk’ (Young, 2011: 125). According to this interpretation, collective action involves making claims on influential institutional agents to reform how they structure people’s interactions.

While this is an important strategy, Robin Zheng (2018: 876) has argued that, since powerful agents also face structural constraints, these actors have limited room for implementing structural change on their own. ¹⁰ Young (2011: 151) herself stresses that it is often difficult for government or other powerful agents to change structural processes because ‘the rules and practices of these institutions are more aligned with the powers and processes that produce or perpetuate injustice than with those that seek to undermine it’. Moreover, she argues that government action requires the active support of societal actors to be effective. Thus, structural change requires an approach that is more embedded within structural processes themselves.

As an alternative, Zheng (2018) proposes her ‘role-ideal model’ of responsibility. For Zheng, structure is sustained by people performing their respective social roles, which she defines as the set of predictive and normative expectations that apply to individuals in virtue of their relationships with others. Although role performance is irreducible to specific acts and allows for people’s agency, that agency is constrained or enabled by the bundle of expectations that accompany certain relationships – as a parent, colleague, employee, citizen and so on. According to this model, people are responsible for structural injustice because unjust structures are constituted by people’s role performance: ‘it is through performing a social role that an individual (together with others) enacts structure’ (Zheng, 2018: 874). Since transforming social structure requires concerted action rather than the modification of any individual’s action taken by itself, Zheng proposes changing the ways in which people perform their social roles to conceptualise how people could exercise agency for social change. According to Zheng (2018: 877), ‘structural change is made possible when all individuals throughout the entire system push the boundaries of their social roles . . .’. This means that each individual should act in ways that go beyond the normative expectations attached to their social role, as teacher, for example, by ‘influencing the way that others carry out their roles (as students, deans, etc.)’. Through modifying interlinked social roles that constitute a stable structure, people engage in transforming that structure itself.

While Zheng importantly takes us forward in stressing how effective structural change must be rooted in people’s practices rather than relying only on their claims, her approach has been criticised for being overly individualistic and ‘straying too far into the arena of privatized virtue ethics’ (McKeown, 2021: 9). Although pushing the boundaries of social roles ‘throughout the entire system’ involves the interactive actions of multiple people, it diverges from the idea of collective action in the sense of coordinated action among a group of people to achieve a common objective. Rather, it involves the cumulative effects of people modifying their behaviours individually. This approach is similar to Goodin and Barry’s (2021) argument that if enough people ‘take a stand’ against injustice, it may well lead to just outcomes. Although no individual can hope to make a difference individually, if enough people take a stand for justice, their actions taken together may have real consequences. As they put it,

Irrational though it may be for any of them to aim at those consequences intentionally when performing their own isolated individual actions, those good consequences might nonetheless be achieved as the by-product of many people performing individually inconsequential actions for other purposes (Goodin and Barry, 2021: 346).

Yet, this quote shows that these actions are essentially ‘individual actions’ having a cumulative effect as a ‘by-product’, rather than as the result of coordinated collective action. Moreover, both on Zheng’s and Goodin and Barry’s accounts, it would only be by chance that the cumulative effects of individuals’ interlinked but uncoordinated actions would lead to more justice, even if, unintentionally, they could have widespread, societal consequences.

The arguments above show us that collective action towards structural justice should at once be rooted in people’s practices, while at the same time being ‘collective’ action rather than the aggregate of individuals’ uncoordinated activities. Before considering what this might look like, let me consider an objection to this type of intentional and coordinated collective action aimed at achieving social justice. Goodin and Barry (2021: 344) have argued that in the same way that the processes that produce injustice are too complex to disentangle, so the collective responses that would correct these injustices – and each individuals’ role within them – are also impossible to identify. Even if organising collectively is the solution as Young claims, it is still not clear how the organiser (or collective body) could pre-determine how each individual within that collective body should act to produce the right outcome, given all the unforeseen consequences that complex interactions involve (Goodin and Barry, 2021: 345). Indeed, this is why they argue that the most we can hope for is that people’s uncoordinated actions taken as a whole may lead to more just outcomes, even if we should not expect that to result.

My view is that this analysis misconceives the way we should approach collective action from a structural perspective. Goodin and Barry’s (2021) account conceives of collective action as a function of what individuals do within collectively organised groups. But recall that structural injustice revolves around people’s social positionings in terms of the resources and opportunities open to them for action (Young, 2011: 54–55). Likewise, collective action in relation to structural injustice should not be understood in terms of any particular course of action – such as donating to a particular charity or supporting a particular political party – by individuals or groups. From this perspective, it would indeed be impossible to disentangle the right course of action in a web of interactive processes. By contrast, structural justice, which concerns how people are positioned in relation to others, involves re-organising social relations. This also shows us why changing people’s behaviour individually is insufficient for realising justice and why a collective approach is necessary. For making virtuous individuals, by itself, does not necessarily alter people’s power relationships and social positions.

To sum up, justice-promoting collective action that is internally consistent with Young’s structural approach should be rooted in people’s practices, be collective in the sense of coordinated activities to shape shared conditions, and revolve around structuring social relations rather than people’s specific acts. How, then, can social relations be altered? As Young (1990: 15–38) has previously argued in her critique of what she calls the ‘distributive paradigm’ of justice, social justice in terms of the assignment of non-material, relational goods like power, opportunity and self-respect, depends on institutionalised processes rather than on distributive patterns as such. In other words, relational equality involves how collective decisions are arrived at rather than so much what those decisions are. Institutional arrangements condition structural processes, which in turn, shape social relations. Thus, a relational egalitarian conception of social justice requires re-organising institutional arrangements to regulate social processes in ways that uphold people’s relational equality. ¹¹ In the next section, I present a model of collective action that centres on crafting democratic institutions from the bottom-up to institutionalise more just structural processes. Rather than pushing the boundaries of people’s social roles, what we should push are the boundaries of participation and democratic decision-making to counter unequal social relations.

Collective Action as Democratic Practice: Crafting Institutions to Re-orient Structural Processes

What kind of collective action leads to more egalitarian social relations? The answer to this question that I now want to develop involves what I shall call ‘collective action as democratic practice’. This, in short, refers to institutionalising democratic decision-making processes in everyday social and economic affairs. Democratic decision-making procedures have both instrumental and intrinsic values for resisting oppression and domination (Young, 1990: 92). ¹² Instrumentally, the question of who has the power to influence collective outcomes and who does not affect collective outcomes, including those that condition people’s ability to develop and exercise capacities. Moreover, democratic decision-making procedures have expressive value for people’s social standing as moral agents who are equally entitled to make claims on one another concerning common rules (Anderson, 1999, 2009; Rostbøll, 2015; Waldron, 2013). Democratic decision-making procedures are thus key components of altering structures of domination where a certain sub-set of people wield arbitrary decision-making power over others’ actions and the conditions of their actions.

As discussed above, the main model of collective action advanced by Young focusses on pressuring powerful agents. However, she also touches upon an alternative form of collective action. Social change may occur, not only via the state or other powerful agent, but rather ‘through collective action in civil society independent of, or as a supplement to, state policies and programs’ (Young, 2011: 112). This type of collective action is more practice-based than making claims on the government and can be traced back to her earlier discussion of ‘insurgent social movements’ that seek to ‘carve out new social spaces’ for alternative institutional forms within existing structures of welfare capitalism (Young, 1990: 82). Examples include urban land trust movements and tenants’ organisations that provide affordable housing, women’s collectives that provide health services or shelters, autonomously organised services provided by racial minorities aimed at their empowerment and capacity building (Young, 1990: 85–86). Moreover, these movements ‘focus on broad issues of decisionmaking power and political participation’ (Young, 1990: 83). While Young herself does not develop a robust account of how such practice-based collective action transforms unjust structural processes, my contention is that they can be reformulated as direct enactments of practices that push back against oppression and domination.

Let me now formalise the two different strategies of collective action as ‘collective action as democratic voice’ and ‘collective action as democratic practice’, respectively. The first strategy focusses on organising to pressure powerful actors. This is the strategy of lobby groups and labour unions that exercise political power collectively to draw out concessions from powerful actors like the state or capital owners with the aim to establish fairer rules for common social institutions. While this approach is important, especially for giving voice to victims of injustice and for raising public awareness about existing injustices, voice alone may not be sufficient for transforming existing structural processes in which the powerful are most entrenched (Young, 2011: 151; Zheng, 2018: 876). The second strategy focusses on reshaping existing structural patterns by directly expanding grass-roots democratic practices. These democratic practices can be expanded in terms of the populations they include, as well as in terms of the social spheres they encompass. By expanding the scope of participation to previously excluded groups, increasing their influence in collective decision-making procedures, and extending the spheres that come under democratic control, this approach directly enacts institutional arrangements that give rise to structural processes that resist domination and oppression.

Collective action as democratic voice aims to improve public deliberation and influence public opinion in relation to the institutions of representative democracy. This is the strategy that Young (2011: 151) refers to when she states that ‘political struggle about state policy must involve vocal criticism, organized contestation, a measure of indignation, and concerted public pressure’. By contrast, collective action as democratic practice resonates with what James Tully (2013: 224) has called ‘cooperative democracy’ where ‘the people cooperate the relationships that govern their various activities by bringing them under their democratic control’. The emphasis here is not on increasing representation and influence in the public sphere, but on acting democratically in governing common social and economic activities. As Tully (2013: 230) writes,

According to democracy as cooperation it is not sufficient to advance public reasons against injustice and for justice and then return to the private sphere and participate in those activities. To be just, one ought to ‘walk the talk’: bring one’s ethical way of being in the world in accord with the principles of justice one espouses in public and, therefore, cooperate democratically in activities that embody and realize the just response.

Tully’s idea of cooperative democracy is inspired by Elinor Ostrom’s (1998) account of people’s capacity to engage in self-government over common social and economic activities. This has been described as the power of citizens to ‘craft institutions’ to achieve collective aims in cooperation with others (Kashwan et al., 2019: 139). Eric Olin Wright’s (2010: 323) discussion of ‘interstitial strategies’ for social change also involves bottom-up institution-building within the ‘cracks’ of dominant social structures. ¹³ While these projects are usually set up to solve practical economic and social issues, such as social or environmental protection, they combine aspects of social movements since they have a deliberate political purpose to transform existing structures that produce unjust relationships. As a form of collective action for the sake of more just social relations, my account of collective action as democratic practice emphasises the institutional design of such collective projects, or how participants coordinate their activities for their common objectives. More precisely, these institutional arrangements should be designed democratically to counter relationships of domination and oppression.

Now, in her Inclusion and Democracy, Young herself distinguishes two functions of associative activities in civil society. The first aspect is conceived as ‘self-organization’, whereas the second aspect is referred to as ‘the activity of the public sphere’ (Young, 2000: 163). Self-organising activities ‘enable people who believe that their sorts of experiences, interests, and needs are socially and politically marginalized to find one another and develop their social voices’. Activities producing public spheres, on the other hand, deepen democracy by ‘multiplying fora and aspects in which people are represented in public discussion, and by activities that make public officials and powerful private actors accountable’ (Young, 2000: 164). These two functions of civil society appear roughly to correspond to my account of collective action as democratic practice and as voice. While I agree that there are many overlaps between my account of collective action as democratic practice and Young’s account of self-organising activities, I contend that the latter does not emphasise as much the institution-building aspect of associational activities that I want to highlight as significant for generating more just structural processes.

Young highlights three functions of civil society as self-organisation: providing a voice for the excluded, social innovation, and providing goods and services. Self-organisation enables marginalised groups to ‘find each other’ and aids the ‘articulation of group consciousness’ (Young, 2000: 165). In other words, it promotes empowerment by enabling members to form and reclaim their group identity. The examples given for social innovation include alternative practices, such as ‘organic farming, herbal healing, evangelical religious worship, or car pooling’, which, in some cases could come to be adopted by the wider society (Young, 2000: 166). Finally, self-organising activities often have a practical purpose of providing goods and services that improve the lives of participants. Surely, associations that engage in collective action as democratic practice could, and often would, have all three of these functions. Nevertheless, Young’s discussion of self-organisation here, I suggest, is not strongly embedded within a structural approach. That is, it does not describe in detail the mechanism by which these functions interact with, and alter, structural processes.

Young (2000: 179) essentially sees civil society as fora of communicative action ‘by which the society communicates to itself about its needs, problems, and creative ideas for how to solve them’. I do not deny that this way in which civil society allows for the appearance of plural perspectives and projects is important for enhancing democracy. My argument, however, takes a more practice-based – as opposed to communicative – approach to social change: because background structure is produced as a result of the actions of masses of people mediated by existing institutions, crafting alternative institutional arrangements is a way to re-orient existing practices that produce those structures. As discussed previously, decision-making processes play a central part in structuring social relations (cf. Young, 1990: 22–23, 2006b). Moreover, democratic procedures have instrumental and expressive value for respecting people’s equal social standing. Thus, the democratic cooperation of everyday social and economic affairs can be theorised as a kind of collective action among plural individuals with the political aim to realise egalitarian social relations. In other words, it is a kind of collective action whose end is at least partially constituted by the social relations embedded in the democratic process itself. As I will argue in the next section, worker cooperatives are paradigmatic examples of democratic practices whose purpose – in addition to mutual self-help – is to realise egalitarian social relations in the productive sphere. Collective action as democratic practice, then, induces social justice understood in this relational manner by generating structural processes that are conducive to egalitarian social relations via self-organised democratic institutional arrangements. While Young’s account of self-organisation describes the communicative functions of civil society as a realm of experimentation for collective problem-solving, in my account, the institutionalisation of egalitarian social relations through democratic practice is a key aspect of coordinated collective action for re-orienting structural processes to become more relationally just.

Actual instances of this type of bottom-up collective action often involve people who have been marginalised within the dominant system as they are organised for the very purpose of improving their disadvantaged social positions (cf. Deveaux, 2021; Hendriks and Dzur, 2022). ¹⁴ They may include the unemployed or precarious workers, the disabled, the poor, the elderly, women, racial minorities and immigrants, among others. While it may be difficult directly through protest to change the behaviour of powerful agents whose interests are embedded in the system, marginalised communities can be better placed to implement alternative institutional arrangements that counter it (Deveaux, 2021). ¹⁵ Because their projects revolve around fulfilling shared needs, they focus on issues of practical concern, such as the organisation of work and the provision of care. ¹⁶ Thus, these activities have the potential to push the boundaries of democratic control to populations and spheres – economic production, care and so on – which are often the loci and subjects of structural injustice. The goal for this type of collective action then, would be to enhance and spread their democratic institutional arrangements concerning practical matters of life and work so as to push back against systemic structural processes.

Let me now take stock of the arguments above. In contrast to collective action as democratic voice, which focusses on making claims against powerful agents to change institutional arrangements that produce structural injustice, collective action as democratic practice seeks directly to craft bottom-up democratic institutions for governing everyday social and economic affairs. Democratic decision-making procedures are important for countering structural processes that produce injustice because decision-making procedures have both practical and expressive values for people’s social standing in relation to others. Crafting democratic institutional arrangements, then, is a type of collective action that aims to alter shared conditions that structure social relations. In this sense, it differs from the unpredictable and aggregative effects of individuals changing their behaviours independently of one another (Goodin and Barry, 2021). It also differs from Young’s account of self-organising civil society in terms of the emphasis it places on institutionalising democratic procedures for re-orienting structural processes as opposed to changing public discourse. My aim is not to diminish the importance of exercising voice to shape public opinion as a key political strategy. Rather, the point I want to make is that my approach is more aligned with a structural approach since it is rooted in people’s practices themselves. Whereas Zheng (2018: 870) argued that social ‘roles are the site where structure meets agency’, my argument is that institutions constitute this meeting point. By institutionalising democratic decision-making procedures at the grass-roots, collective action as democratic practice aims to restructure people’s interactions in ways that generate more just social relations.

Linking Collective Action as Democratic Practice and Social Justice: The Case of Worker Cooperatives

This section considers in more detail how collective action as democratic practice interacts with social structure to realise social justice. One objection to my argument so far might be that the impact of such projects, even if they benefit participants and communities, would be circumscribed and local, with limited implications for society at large. In this sense, some might object that although I have argued that my approach is distinct from existing individualistic and cumulative approaches, the difference between them is not as great as I claim. That is, only the unit of actors has changed from individuals to collective bodies. Yet, just as virtuous acts by individuals might not lead to society-wide justice, independent bottom-up initiatives – even if they are collective ones – may only have limited or contingent effects on a system-wide level.

To consider this objection, in this section, I will refer to the case of worker cooperatives within a capitalist society as a concrete example of democratic collective practice aimed at transforming dominant background structures. According to Wright (2010: 327–328), ‘worker-owned cooperatives are the quintessential form of interstitial organization’, which seek directly to implement an alternative reality within the existing society. Despite variations in size and scale, the structure of democratic decision-making (e.g. direct or representative), or the sectors within which they operate, the core features of worker cooperatives are (1) voting on a one-person, one-vote basis by everyone who works in it and (2) allocation of profits or residual according to an agreed schedule based on factors, such as hours worked or amount of work produced (see, e.g. Ellerman, 1990: 96; Miller, 1989: 83). The distinguishing feature of a cooperative is that capital rights are separated from control rights and profit rights that are distributed on the basis of membership rather than share ownership (Ellerman, 1990). Thus, worker cooperatives are real-life examples of democratically governed productive activities, which are purposefully designed to counter capitalistic social relations (Cheney et al., 2014; Dufays et al., 2020).

Despite this, the relationship between the internal organisation of worker cooperatives and their society-wide impact has long been a point of contention. These can roughly be divided into two types of objections. The first is that democracy within the workplace would not necessarily translate into society-wide solidarity or egalitarian relations. The second is that well-meaning bottom-up projects by citizens can work to depoliticise and deflect from structural inequalities, even while benefitting participants and communities. Let me now examine these objections in turn.

On the one hand, a long-standing and prominent argument in support of workplace democracy has been that it serves as a school of democracy (Cohen, 1989; Mill, 2008 [1848]; Pateman, 1970; Young, 1990). ¹⁷ According to this argument, the justification for workplace democracy is that experiences of participation and deliberation about issues of common concern in private associations foster civic virtues like active character and a sense of the common good necessary to make democratic politics a substantive ideal. On the other hand, others point to the possibility that worker cooperatives may be subject to ‘collective egoism’ which counters an effective sense of justice (see, e.g. White, 2014: 142). David Miller (1989: 333) argues that ‘small solidaristic communities would not necessarily support a general scheme of redistribution’, and that ‘market socialism is unlikely to remain stable if people’s only active involvement is in the running of their own enterprises’. These arguments problematise the potential of worker-managed firms to promote inward-looking tendencies and pose the question of how to translate the sense of solidarity within these organisations to society at large. In either case, the conclusion seems to be that the relationship between workplace democracy and people’s sense of justice is empirically contingent: democratic workplaces may or may not cultivate society-wide solidarity and civic virtues. ¹⁸

From a structural perspective, the problem with this type of argument is that it focusses on individuals’ character rather than social structure; it focusses on creating virtuous individuals as a route to positive social change. But recall how social structure is the result of the often unintended consequences of the actions of masses of people (Young, 2011: 62–64). These structural processes interactively construct the parameters of possibility for people’s actions, sometimes despite, or even against, their internally held beliefs. As an example, take the case of worker cooperatives in a capitalist market system. Despite some well-known and successful cases, cooperatives remain marginal players in most developed economies (Dufays et al., 2020; Malleson, 2014). Against the view that this simply reflects workers’ higher preferences for goods other than self-government in the workplace, such as higher monetary reward or more leisure time, Miller (1989) has argued that the market is not neutral towards different types of productive organisations. According to Miller (1989: 93), ‘The market has a structure and a logic, and it pays to go along with them’. According to this logic, people who have non-commodity based conceptions of the good like cooperative relationships will be ‘handicapped in their pursuit of the good life, not because their conceptions are naturally expensive but because of the institutional framework within which that pursuit occurs’. Others have also argued that existing market environments are embedded within artificially constructed financial and legal institutions that favour capitalist firms (Anderson, 2015; Malleson, 2014). ¹⁹ For example, Malleson (2014: 109–110) argues that legal structures, such as limited liability, which allows unlimited numbers of people to combine their finances with limited risk while gaining ownership and control rights, have given significant advantage to capitalist corporations for raising capital. By contrast, cooperatives have been charged with problems of underinvestment because they cannot sell shares to external investors without losing internal control (Jacob and Neuhäuser, 2018: 941; Malleson, 2014: 84). In addition, while financial institutions like the corporate banking sector and bond market favour capitalist corporations, conventional banks may be unwilling to extend loans to cooperatives, which are seen as reluctant to pay them back to maximise collective wages (Dow, 2003; Miller, 1989). Moreover, capitalist firms already have a competitive advantage when social investment in knowledge and training are geared towards preparing people to work in hierarchical organisations (Bowles and Gintis, 1993; Malleson, 2014). Young’s structural approach adds to these institutional arguments by emphasising how people rationally pursuing their interests within such constraints aggregately reproduce the existing system. Regardless of their values and preferences, most people will behave in ways that conform to existing institutional rules because these rules reward or facilitate such behaviour while sanctioning or excluding others. Yet, as Young (2011: 63) argues, ‘Many large-scale social processes in which masses of individuals believe they are following the rules, minding their own business, and trying to accomplish their legitimate goals can be seen to result in undesirable unintended consequences when seen structurally’.

The point I want to make from this example is that in searching for a way to link worker cooperatives with social justice, we should look beyond people’s internal preferences and values. This is not to say that a psychological argument is unimportant for democracy and social justice. The significance of Young’s structural approach is that people’s actions are implicated in the production and reproduction of social structure. Structural change requires people to be motivated to reflect on their social situation and to act in relation to it. However, a structural approach equally recognises how people’s actions are constrained by existing structures. As the example above shows, these may prevent people from realising certain values, even if they have the internal disposition and skills to do so. From this point of view, the question for democratically managed firms, when we consider their society-wide impact, should not be reduced to how well they cultivate people’s civic virtue, which may or may not be actualised within existing constraints. Instead, we should equally consider how such a way of organising economic practices reconfigures people’s opportunity structures to act in new and different ways.

The second objection against worker cooperatives from the perspective of social justice comes from a more structural perspective. A prominent objection towards well-meaning bottom-up projects by citizens is that they can work to depoliticise and deflect from structural inequalities, even while benefitting participants and communities. Some Marxists have argued that alternative systems within a dominant capitalist environment could actually work to strengthen the latter ‘by siphoning off discontent and creating the illusion that if people are unhappy with the dominant institutions they can and should just go off and live their lives in alternative settings’ (Wright, 2010: 326). This resonates with how theorists of power have come to criticise the idea of ‘empowerment’ as a depoliticised concept that focusses on individual fulfilment, ‘disconnected from the underlying causes of people’s powerlessness’ (Gaventa, 2021: 110–111). Others have argued that citizen-led projects for practical problem-solving ‘can distract citizens by keeping them busy reconciling gaps in the market or state, when they should be contesting underlying structural causes of policy problems’ (Hendriks and Dzur, 2022: 693). As I have already noted, what I am calling collective action as democratic practice overlaps with what Wright (2010: 324) calls an ‘interstitial strategy’ with a deliberate political purpose to alter social relations in contrast simply to collective problem-solving to fill gaps in public services. Nevertheless, going back to our example, the question remains as to how islands of worker cooperatives within a sea of capitalism could possibly challenge it.

Although the impacts of interstitial strategies and how successful they are in overcoming corrosive pressures will differ from case to case, I believe that there are resources within a structural approach to offset this charge of depoliticisation. Contrasting Rawls’s institutional theory of justice with Young’s structural approach, again, is helpful for understanding the potential link between such interstitial strategies and social justice. Rawls’s theory draws a clear division between the public and private realms. For Rawls, associational actors are free to pursue their particular plans within the basic structure, whose role is to maintain background justice. According to this division of labour, private associations are realms of individual freedom and freedom of association, unburdened with the task of maintaining background justice directly (Rawls, 1993: 265–269, 1999: 456–464). Note how the charge that interstitial strategies can be depoliticising also shares this assumption of an institutional division between ‘public’ and ‘private’ realms. That is, self-organised projects by grass-roots citizens are seen as private activities, which can be institutionally separated from matters of justice. By contrast, Young’s structural approach suggests a more fluid relationship between actors in civil society on the one hand, and the background structure that conditions individual interactions on the other. This is because peoples’ cumulative actions mediated by associations in civil society are implicated in the construction of shared conditions that structure social relations. Social structure, in other words, is realised through people’s practices. According to this understanding, associations are not only settings in which people can pursue their respective conceptions of the good, but also the settings through which they actively shape the background structure.

Moreover, as a form of collective action, crafting institutions is a way to channel the behaviours of multiple people. Institutions shape incentive structures and build new opportunities for action (Kashwan et al., 2019: 139). This type of self-organising is thus distinct from people simply pursuing their various ends in civil society that may result in cumulative outcomes as a by-product. Rather, it aspires consciously and collectively to re-orient institutionalised practices that together recursively reproduce existing structures. As an interstitial strategy, this type of social movement would seek to expand and fortify the impact of their alternative institutional arrangements. One way to do this would be to form networks and alliances between and among movements (Ackerly, 2018; Deveaux, 2021). ²⁰ Cooperatives have historically established alliances at local, national, and international levels (Dufays et al., 2020). These alliances are not only important for enlarging the political ‘voice’ of cooperatives, enabling them to increase their leverage in lobbying for political infrastructure and resources, but also for enabling them to scale up their activities by pooling economic and human resources and knowledge (Malleson, 2014, 87). By combining voice and practice, cooperative alliances employ a dual strategy that seeks to influence the overall institutional framework in ways that further their projects while extending democratic structural processes to new spheres and populations. By creating new opportunity structures that re-orient people’s actions, this kind of institution-building can have implications for the background structure, which is realised through people’s accumulated actions. Any social change that results from this would be incremental rather than revolutionary and the ability of such projects to resist corrosive pressures from the system would depend, among other things, on how successful they are in combining and embedding their alternative arrangements. Nevertheless, the idea of interactive structural processes provides us with the tools to conceptualise how enactments of grass-roots democracy can have implications, not only for the non-domination of the immediate participants of those projects, but also for creating a more just society.

Conclusion

The arguments in this essay contribute to advancing understanding about what kind of collective action is necessary for social change towards justice from a structural perspective. I have argued that what I have called collective action as democratic practice is a more promising way to theorise such collective action than existing accounts because it is kind of collective action that is rooted in people’s practices and centred on changing social relations.

My view departs from Young’s own account of collective action as engaging in communicative action to pressure powerful agents to reform existing institutions. Instead, collective action as democratic practice involves ‘crafting institutions’ from the grass-roots to fulfil citizens’ shared social and economic needs. While there are many types of self-organised activities for meeting needs, not all of which have directly political aims, I have argued that those aimed at structural change towards justice must be governed democratically since social justice from a structural perspective involves how people are positioned in relation to others. Democratic decision-making procedures are significant for structural justice because they have both consequential and expressive implications for people’s equal social standing. Since institutions channel people’s interactions, crafting democratic institutions is a kind of collective action that aims to realise just social relations through re-orienting structural processes that produce patterns in the ways people are socially positioned.

I should note that my account does nothing to undermine the importance of exercising voice to politicise structural injustice as a strategy for social change. Rather, my position is that political advocacy may be more effective when combined with direct enactments of democratically organised cooperative practices. This stance comes from the insight of a structural perspective wherein people’s agency – even those of politically conscious individuals – is constrained by existing structures. Raising awareness then, may not be sufficient for resisting dominant structural processes. While it is impossible to resist such processes individually, combining with others to institutionalise democratic arrangements directly through practice is a strategy to open up the parameters of possibility by creating new opportunities and incentive structures. Here, my view also departs from previous attempts to conceptualise structural change through the cumulative, rather than collective, actions of individuals changing their behaviours independently of one another.

Finally, the arguments in this essay have wider implications for understanding the outward-looking political potential of self-organised projects to meet citizens’ needs. Existing literature on self-governed initiatives by citizens for solving practical, collective problems sometimes problematise such initiatives as having only local and narrow scope, often allowing governments to off-load public responsibility on vulnerable citizens, while de-politicising structural problems, especially in a neoliberal context. ²¹ Broad, systemic change, by contrast, is associated with accessing policy windows that align with powerful policy-makers through political advocacy (cf. Hendriks and Dzur, 2022: 694). Existing arguments that highlight the political agency of the structurally oppressed also ultimately focus on how they could shift public opinion in ways that put pressure on powerful actors (Deveaux, 2021; Hayward, 2017). The arguments in this essay which situate self-organised projects within structural processes enables us to conceptualise how they could actively change the background structure through an alternative, practice-based route: by enacting, expanding and fortifying collective democratic practices that alter social power relationships and positionings. Combining this kind of collective practice with collective voice would help to embed just social relations through people’s everyday activities of life and work.