Can agonism be institutionalised? Can institutions be agonised? Prospects for democratic design

Contestation

Following Schmitt’s (2008) concerns about the dangers of depoliticisation, agonists promote contestation in order to prevent democratic apathy, keep citizens engaged in democratic processes and prevent them from turning towards non-democratic conflict. Mouffe (2000: 7) argues that when there is a lack of contending political positions to represent citizens, they may seek non-democratic forms of identification. She celebrates the growth of ‘left populism’ in Latin America, the United States and Europe, seeing it as capable of bringing together different democratic struggles in opposition to a common adversary (Mouffe, 2016). Contestation is not only perceived by agonists as important to political identification, it is essential to the creation of identity. According to agonists, it is through collective contestation that citizens gain autonomy. In Arendt’s (2013: 38–44) account, just as the ancient Greek polis required citizens to distinguish themselves as the best (according to the principle of aien aristeuein or ‘ever to excel’), agonistic contestation promotes competition and excellence in the presence of others. Drawing on Nietzsche, agonists argue that such development of citizens’ capacities improves the well-being of democratic society (Owen, 1995: 139). Contestation also provides a tool for exposing the domination of minority values and citizens in democratic society. As Foucault (2003: 221–226) argues, it is the ‘permanent provocation’ between power and freedom that ‘make visible those fundamental phenomena of “domination” which are present in a large number of human societies’.

Contingency

Agonists reject Platonic claims to philosophical truths, perceiving societal values as necessarily open to continual contestation. Following Nietzsche, agonists affirm that all values are situated in a particular community (Nietzsche et al., 2006: 114), contrary to liberal claims of neutrality (Rawls, 1973: 10), rationality (Elster, 1998: 8) and universality (Held and McGrew, 2007: 62). Agonists echo Foucault’s concern that ‘truth isn’t outside of power, or lacking in power’ (Foucault and Gordon, 1980: 72), perceiving claims to truth to be entangled in power relations, a reflection of society’s ‘regime of truth’ (Connolly, 1991: 73) and a threat to diversity. Agonists are concerned that when one group in a pluralist society claims to possess truth or universality, it ‘cannot recognise and respect any plurality of narratives, traditions or civilisations as equal yet different’ (Connolly, 2000: 149). It also renders its values ‘incontestable’ (Connolly, 1995: 123), preventing them from giving way to new concepts. Contingency is fundamental to overcoming democratic exclusion and promoting pluralism (Connolly, 1995: 93). Agonists take up Nietzsche’s notion of ‘untruth’ and Foucault’s ‘games of truth’, requiring citizens to call into question the societal values intrinsic to their democracy (Nietzsche et al., 2006: 8). While deliberative democrats are concerned that exclusion will simply be replaced with ‘decisionism’ or ‘arbitrary and dangerous’ forms of power (Honig, 2007: 2), contingency combats this threat in two ways. First, it ensures the continual challenge of all dominant expressions of power; second, it works alongside agonistic contestation to provide a direct form of democracy that holds ideas up for scrutiny and critique. Contemporary debates about ‘post truth’ and ‘fake news’, associated particularly with the Trump presidency in the United States, reveal the contestability of facts and values, and the ways in which truth claims contribute to domination within democratic societies.

Interdependence

Agonists place strong emphasis on politics as a collective process in which citizens develop their identities relationally. For Nietzsche and Foucault, understanding a concept lies not in the concept itself, but in its comparison with related concepts (Foucault, 1971: 335; Nietzsche et al., 2006: 7). For agonists, identity is ‘specified by comparison to a variety of the thing I am not’ (Connolly, 1991: xiv). All citizens in a democracy are necessarily interconnected, creating and recreating themselves accordingly. The fabric of society thus consists of a diversity of threads – a ‘crazy quilt’ (Tully, 1995: 197). Mouffe’s adversarial account sees diverse identities less as a pluralistic web of interconnectedness and more as oppositional groups, formed in relation to one another (Mouffe, 1993: 2). The construction of an ‘us’ and ‘them’ is necessary for a left populist challenge to what she calls the ‘post democratic’ political establishment (Mouffe, 2016). However, whether identity is perceived as constituted by pluralistic individuals or oppositional groups, agonists converge in affirming that citizens do not fully exist without others. Drawing on Arendt’s (2013: 23) view that no human life is possible without others, agonists claim that ‘humans are incomplete outside of social form’ (Connolly, 2000: 153), rendering direct political participation fundamental to becoming a citizen (Tully, 2008a: 172). Agonism is a collective process dependent on democratic engagement between diverse citizens. Agonism can overcome domination, challenging power relations by weaving an entangled web of interdependent groups (Connolly, 1995: xxviii).

Perfectionist agonism: Preference ranking of diverse values

A term attributed to the work of David Owen (1995), perfectionist agonism includes a ‘commitment to the cultivation and continuous improvement of citizens’ virtues and capacities’ (Fossen, 2008: 378). Drawing on Nietzsche, perfectionists see values as necessarily perspectival and communal (Owen, 1995: 138). Democratic contestation involves a public, collective ranking of values, during which citizens enhance their capacities for ‘self-mastery’ (Owen, 1995: 118). There is a resonance with processes of participatory budgeting (PB) as pioneered in Porto Allegre (Brazil), which involve citizens in collectively ranking budgetary resources (Smith, 2009: 33–39; Wampler, 2007). First, as the name suggests, this institution requires the direct participation of community members. Second, citizens have a direct input in establishing the rules of the game, reflecting Owen’s calls for autonomy, contingency and contestability, and a move away from aspirations of consensus. Perfectionist self-mastery is a tool to reflect on past atrocities, to strive towards nobility and ‘to pursue the goal of a humanity characterised by nobility’ (Owen, 1995: 118–119).

In deciding which perspectives ‘win’, citizens are required to come ‘to an honest judgment concerning the degree to which they satisfy the interests (exhibits the virtues) of the practice’ (Owen, 1995: 143). Citizens ask themselves whether a judgement fulfils the principle of eternal recurrence; whether they ‘desire this once more and innumerable times more?’ (Owen, 1995: 118). Perfectionist contest also makes use of ancient Greek competition, encouraging citizens to strive to surpass one another’s perspectives, arriving at new standards of self-mastery (Owen, 1995: 139). By promoting new standards, this agonistic model aims to ensure and enhance the well-being of the state; it could help (for instance) with establishing public spending priorities in the face of austerity, as in the citizen engagement processes in Iceland following the 2008 financial crisis. Preference ranking among citizens could also provide a means for discussing complex issues, such as fracking, where economic, environmental and health considerations all need to be considered (Beebeejaun, 2013: 18–21). Owen’s concept of eternal recurrence has the potential to restore citizens’ passions, enable them to reach better decisions and promote solidarity. It is not shared values that unite diverse citizens but participation in a shared process (Owen, 1995: 163). Evidence from PB in Porto Alegre demonstrates a range of education and wealth among participants, and a rough gender balance (Souza, 2001: 168–169). The binding nature of PB transcends political discussion, and turns citizen input into action, resonating with Owen’s calls to engage citizens in a ‘quest’. The PB experience demonstrates the possibility of institutionalising principles from perfectionist agonism. The goal is to design democratic practices that are more participatory, enable citizens a greater input in the formulation of the democratic process, allow for diverse participation and ensure that citizens’ voices are turned into action.

Adversarial agonism: Decision-making in the context of plural values

Predominantly associated with Chantal Mouffe, this model is inspired by Schmitt’s (2008: 28) contention that conflict is fundamental to politics. For adversarial agonists, a blurring of political boundaries is dangerous because it can impede identification with a particular position, leading to democratic apathy (Mouffe, 2007: 5–7) or affiliation with fundamentalist identities (Mouffe, 1993: 6). The adversarial approach aims to provide democratic spaces in which citizens can express conflict, preventing displacement of conflict into other areas. In the adversarial model, agonistic contestation comprises three groups: friends – those who agree on the interpretation and implementation of liberty and equality; adversaries – those who share allegiance to liberty and equality but dispute their interpretation and/or implementation; and enemies – those who either do not share these values or who threaten the democratic process (Mouffe, 2000: 13–20). Adversarialism harnesses the passions of citizens to keep the democratic contest alive. It also draws on the differences between collectivities to promote unity within each group. According to Mouffe (2000: 13), collective identities can only be established via ‘an us/them’ distinction. Yet there is a significant difference between the adversary and the enemy; the former constitutes a worthy opponent, while the latter is to be excluded (albeit never permanently) from the democratic contest (Mouffe, 2000: 102, 2005: 20). The goal of democratic politics is not to eradicate conflict but rather to transform antagonism (enemies) into agonism (adversaries), in order that citizens can engage in legitimate contestation (Mouffe, 2000: 103). She discusses the case of abortion policy, where this concept could help citizens to perceive others as having the right to defend their views (in spite of their substantive disagreement). Such commonality could prevent antagonism from emerging, allowing instead the expression of passionate conflict.

The concept of legitimacy could be useful to mediate other polarised discussions, for instance, constitutional issues in the United Kingdom like Scottish independence or relationships with the European Union. Referenda are potentially an arena for institutionalising adversarial agonism since citizens are required to take a yes/no stance, thereby harnessing passions and encouraging agonistic contest. However, effectiveness is dependent on (1) the quality of information that is provided to each camp, not just about their own arguments but those of the other side, and (2) the establishment of arenas for respectful contestation that could help to combat antagonism and demonstrate the legitimacy of all arguments, without seeking to remove such conflict. Such characteristics were notably absent in the 2016 Brexit referendum in the United Kingdom, which left a powerful legacy of antagonism. A Home Office report found that hate crimes increased by 41% in the month after the referendum (Forster, 2016).

Another potential institutional format is the citizens’ assembly, such as that in British Columbia, Canada (2004), where 161 citizens came together to discuss electoral reform (Warren, 2007). The process entailed three stages: information, deliberation and voting. Citizens’ assemblies have been used in many other contexts, including Iceland (as noted above) and, more recently, in testing arguments for devolution to English city regions (Flinders et al., 2016). Adversarial agonistic institutions could draw from both referenda and citizens’ assemblies to promote a passionate arena of contest in which opposing citizens learn to perceive one another as legitimate adversaries. Given the focus these initiatives place on deliberation, it is also important to note that Mouffe is critical of the deliberative tendency towards consensus and rationality. Instead she affirms the importance of ‘recogniz[ing] those forms of exclusion for what they are and the violence that they signify’ (Mouffe, 1993: 145). For Mouffe, a non-conflictual consensus is impossible. Adversarial agonistic institutions would need to recognise and be transparent about the voices that were not represented by the discussion and/or the decision. In addition, it would be necessary to transform deliberation into contestation, shifting the focus from education to impassioning by creating space for personal, emotive accounts and relying less on experts and technical judgements. Scaled up, adversarialism could construct a ‘collective will’ (connecting different popular struggles) capable of challenging ‘post-democratic’ elites, institutionalised through hybrid ‘movement-parties’ like Syriza and Podemos (Mouffe and Errejon, 2016). In supporting the growth in left populism in the United States associated with Bernie Sanders’ 2016 presidential campaign, Mouffe (2016) argues for the ongoing, intentional construction of such a ‘collective will’ in opposition to the common adversary of ‘Wall Street, the political establishment and the oligarchy’. Such a collective will makes no claims to universalism but rather reflects a convergence of plural political identities in opposition to a common adversary. Mouffe argues that populism is reawakening engagement in politics in many countries (in the face of previously declining electoral turnouts), but an agonistic rather than an antagonistic sensibility is required if populism is to be associated with progressive social change.

Inclusive agonism: Engagement in the context of diverse identities

This model is primarily associated with William Connolly and James Tully who employ agonistic engagement as a means of enhancing inclusion in pluralist society. Influenced by insights from Nietzsche, Foucault and Arendt, inclusive agonism perceives citizens as necessarily interconnected, reliant upon others to be complete (Connolly, 1991: xiv; Tully, 2008a: 3). Whereas Mouffe sees the inter-related character of identity as a provocation to engage in democratic contest, Connolly and Tully see it as a tool to render democratic society more inclusive. This approach also reflects their understanding of interdependency as a plurality of interconnected individuals, compared with the Mouffe understanding of oppositional group identities. Endorsing political engagements that increase the role of citizens (such as norm formulation and grassroots politics), the focus is on identifying agonistic behaviours that render society more inclusive. Principles of agonistic respect, critical responsiveness and contestability can help to overcome ressentiment in pluralist contests about, for instance, gay rights and religious values. Connolly (2005: 123–127) claims that his agonistic respect extends beyond liberal toleration since it flows in all directions, comprising a ‘more ambiguous relation of interdependence and strife’ (1993: 382); minority toleration is not simply reliant upon the decision of the dominant majority. Similarly, Tully’s (1995: 23) mutual recognition calls on citizens to avoid comprehending others through prior judgements, instead understanding them as they wish to be understood. Inclusive agonism could prevent gay rights advocates from being tolerated as the inferior minority, encouraging both sides to respect one another, while still disagreeing fundamentally. Such an approach could be applied within the Church of England, in seeking to recognise and value a role for openly gay clergy.

In addition, Connolly (2005: 127) calls for a critical responsiveness that involves ‘careful listening and presumptive generosity’ towards novel positions, which could subsequently ‘engender a shift in criteria of judgment, justice, identity, or legitimacy’. Critical responsiveness could address feelings of ressentiment by enabling both sides to express their differences, and to develop some (even partial) understanding of the other side. For Tully (1995: 207, 2008b: 110), audi alteram partem (‘always listen to the other side’) encourages citizens to pay particular attention to minority voices and to adopt a critical attitude to one’s own culture and those of others. Drawing on Foucault, Tully’s (1995: 115) account of ‘non-consensuality’ calls on citizens to ‘act differently’, demonstrating alternative possibilities and exposing what people ignore of their own situation. As Tully argues, these principles could enable Quebecois voices to be heard in Canada, preventing legislators from assuming that they understand the Quebecois culture and instead requiring them to debate directly with Quebecois citizens. Through abandoning calls for deliberative discussion and instead promoting a diversity of interactions (based, for instance, on narratives and story-telling), a wider range of citizens could be encouraged to participate, with the promise of opening up dialogues between diverse cultures, and enhancing cross-cultural understanding.

There have been attempts to institutionalise such principles in the form of ‘truth and reconciliation commissions’ in a range of countries that have experienced civil war, internal terror and even genocide. The focus is on providing people with the opportunity both to speak and hear personal narratives from different sides, with an emphasis on giving space and respect, rather than eradicating difference or building consensus. Inclusive agonism could offer opportunities to challenge the domination of marginalised groups within more settled environments: for instance, enabling Syrian refugees in Europe to be understood on their own terms rather than conflated with Islamic extremists or economic migrants. Furthermore, Connolly’s concept of contestability requires citizens to acknowledge that conflicting others may not share their view. As a result, citizens ‘connect positively through reciprocal confession that those in each group confront doubts, forgetfulness, or uncertainties in themselves that may invert those confronted by others’ (Connolly, 2005: 125). Through acknowledging and exhibiting the contestability of their position, agonistic behaviours can help different groups to perceive one another as less threatening to their own identity, enabling multiple identities to coexist in democratic spaces. A challenge exists to develop such spaces via the Internet and social media. These channels have great potential for taking inclusive agonism further, given their capacity to open up interactive and explorative communication between diverse citizens without the condition of a dominant majority granting toleration to a minority. However, an absence of respect and critical responsiveness characterises many web-based exchanges involving minority or marginalised identities (Citron, 2014). At the same time, research shows the importance of the web for expanding knowledge of, and respect for, ‘new’ identities associated with trans and non-binary sexualities (Alexander, 2002).

The three approaches – perfectionism, adversarialism and inclusivism – differ significantly, but it is necessary to consider all of them together when examining how agonistic principles could be institutionalised, in order to promote a more collective, impassioned and inclusive democratic arena. Agonistic interventions could help politics to become more inclusive by addressing feelings of ressentiment and overcoming domination and marginalisation, thereby opening up dialogue and encouraging understanding. They could also provide conflicting citizens with a space to express their passionate differences, engaging them in democratic politics, thereby preventing shifts to non-democratic outlets. The collective capacities of citizens could be enhanced, allowing better decisions to be reached, while promoting solidarity through a common process. Agonism could help to overcome pluralist conflict by encouraging a greater diversity of democratic participation, a revived political sphere, improved decision-making and increased social unity.

Processual

Informed especially by adversarial and inclusive approaches, the process of agonistic democratic design can be seen as important as the outcome. Attempts at institutional reform provide an opportunity for ‘the discovery, clarification and elaboration’ of the values that undergird existing and alternative institutional arrangements (March and Olsen, 1989: 90). Agonists reject the idea that a universally accepted ‘good society’ will emerge from (or guide) such a process. Rather, agonistic design reflects a deep pluralism, whereby diverse actors express perspectives that need to be recognised rather than reconciled. Agonists do not concern themselves with the outcome of the pluralistic contest since there is no universal truth to be arrived at. Rather, they focus on the process of discussion in which citizens constitute their values against those of others, and in which the emergence of new values leads to a reconstitution of others. Griggs et al. (2014: 30) argue that democracy should avoid proceduralism and promote a ‘practical ethos of questioning and collective problem solving’. This notion reflects Owen’s commitment to a collective cultivation of virtues, as well as Connolly’s emphasis on overcoming ressentiment through passionate yet respectful engagement. Thus, citizen participation in democratic design should transcend the usual boundaries of the ‘political’, being involved in the process of norm formulation via a wide range of educational, cultural and social media channels.

Collective

Agonistic democratic design engages a multiplicity of actors in the active (yet open-ended) construction of a collective will. It is a design process of un-doing as well as doing, exposing and interrogating the embedded meanings within extant institutions as well as elaborating new ones through respectful contestation. Rather than seeking to fix rules and invisibilise their underlying values, agonistic democratic design asks questions about the ‘rules of the game’ and opens up spaces in which alternatives can be elaborated and contested. Wagenaar (2014: 248) argues that the ‘inherent conflict’ between citizens’ diverse subjectivities and ‘ensembles of administrative belief’ can be made productive. Through a ‘process of enacted interdependence’, it is possible for diverse actors to come together to form governance networks that facilitate collective learning and action, as well as transforming relations of antagonism and resentment into those of agonistic respect and mutual recognition. Such a process encourages deep pluralism, diverse participation and interdependency through measures such as random selection, honorariums and rewards, and network facilitation – measures that actively promote more diverse attendance and active participation. It is important to design institutions that reflect both adversarial accounts of identity (Mouffe) and more pluralistic ones (Connolly, Tully and Owen). A combination of ‘yes/no’ referenda-style topics and debate-like fora on one hand, with ‘preferenda’ (where citizens rank a series of options in order) and pluralistic view-sharing on the other, can encourage impassioned democratic engagement while also transforming relations between agonistic citizens. Agonistic design creates opportunities for citizens to be involved in crafting and adapting specific institutional arrangements, and in holding politicians accountable for previous decisions on institutional design. Such a design resonates with Connolly and Tully’s calls to expose and overcome domination, and reflects Owen’s emphasis on the autonomy of citizens.

Contextual

Agonistic democratic design acknowledges and values the necessarily embedded nature of democratic institutions. Such a process reclaims parochialism, seeing the distinctive character of local agency and environments as an important resource (Lowndes and Wilson, 2001). Agonistic design facilitates bricolage rather than seeking to stamp out actors’ institutional tweaks and subversions. Especially in contexts of resource shortage and/or inter-group enmity, ‘refurbished’ or ‘recombinant’ institutional arrangements represent a contingent settlement that enables collective action (Crouch, 2005; Lowndes and McCaughie, 2013). Acknowledging the contextually embedded and indeterminate character of institutional boundaries makes a democratic virtue of Tully’s ‘crazy quilt’. An excess (or surplus) of institutional opportunities for democratic design allows some innovations to thrive and others to wither, in the absence of any expectation of design perfection or permanence (opposing the a-contextual liberal mantras of ‘best practice’ and ‘good governance’). A ‘fine mess’ rather than a homeostatic model is the aim of agonistic democratic design (Cleaver and De Koning, 2015: 12). This design mirrors Mouffe’s claim that all consensus is necessarily exclusionary and that society’s goal should not be to resolve such exclusion, but to acknowledge and maintain contingency. A prime example would be the way in which PB in Porto Alegre is not codified into municipal law, allowing it to be restructured or even abandoned.

Contestable

Agonistic democratic design involves contesting existing institutions and also making new arrangements open to challenge. Designing for agonism means valuing conflict, and making spaces for agonistic contests, which allow for passionate expression of differences. Such passionate conflict is reflected in contestable and contingent institutions, which are always subject to challenge: for instance, through the provision of short election cycles, frequent rotation of those leading democratic institutions, and spaces for challenge and critique. It is also inherent in institutions that harness citizens’ passions and sustain engagement by tackling controversial ‘value’ topics (like abortion), using facilitators who actively encourage citizens to bring in personal accounts and stories, and enabling citizens to make binding decisions. There is a link here to the embedded aspect of agonistic institutional design, in which the flow of conversation across formal and informal institutions (and politics and wider society) upholds opportunities for continual contest. Following Tully (2008a), agonistic design has to embody ‘the freedom of speaking and acting differently’. The ‘rules of the game’ are contestable and democratic institutions are designed to be permeable. In addition to speaking to Mouffe’s commitment to passionate democratic engagement, this design also resonates with Owen’s virtue cultivation and Connolly’s drives towards future pluralisation.

Provisional

Agonistic democratic design rejects an endpoint. There is no expectation of arriving at a juncture where the political process is deemed perfect. Instead, democratic design is understood as necessarily incomplete, involving institutions whose ‘gappiness’ is not just inevitable (given the impossibility of shutting down interpretive spaces) but also productive (as a potential locus of creativity among political actors). Provisional (or emergent) institutional design seeks to harness, rather than suppress, open-ended processes of political reflection (Andres, 2013; Lowndes, 2005). It acknowledges the contingent character of political environments, which destabilise modernist attempts to ‘fix’ institutional settlements across time and space. Institutions are no more than forms of contestation that are ‘contingently instituted, sometimes deeply sedimented, but always recontestable’ (Griggs et al., 2014: 28). Types of rule are only ever ‘more or less institutionalised’ and are always subject to contestation and contingency. Hence agonistic design promotes temporary and experimental democratic forms that allow for the practical elaboration of contesting perspectives (rather than being ‘pilots’ for the ‘roll out’ of pre-determined schemes). This contingency seeks to address Foucauldian concerns about the institutional tendency towards self-preservation, emphasising the unending and imperfect democratic contest promoted by Owen, Mouffe, Connolly and Tully. Internal institutional differentiation is encouraged, even ‘the deliberate creation of institutional irritants’ (Goodin, 1996: 39).

In summary, we have argued that an agonistic perspective alerts us to potentially positive aspects of political institutions that have traditionally been regarded in a negative light. Conventional institutional design seeks to close the gaps in democratic institutions, resolve the ambiguities, settle the power disputes, and establish ‘fitness for purpose’ and ‘good practice’ across a range of contexts. Instead, we argue that the inherent incompleteness of all political institutions could be built upon to open up agonistic spaces. Rethinking democratic institutions in this way may be as important as developing agonism-specific democratic devices.