II. Democracy Beyond Elections: Why Deliberative Mini-Publics?
Traditional representative democracy faces significant challenges. Scholars point to increasing affective polarisation undermining social cohesion and trust (Reiljan 2020; Rudolph & Hetherington 2021), declining voter turnout (Kostelka & Blais 2021), and increased mis- and disinformation (Farkas & Schou 2019). Scholars and practitioners advocate for augmenting traditional democratic structures with practice of deliberation. These approaches aim to foster rational discourse on public issues and build consensus around the general interest, echoing Habermasian aims (1975). Deliberative mini-publics (DMPs) are one approach for augmenting decision-making processes in a way that encourages discourse and building a shared understanding. The handbooks coalesce on three key benefits of DMPs to democratic systems: (1) encouraging citizens to take ownership of and engage with complex policy issues; (2) addressing inequities in traditional democratic structures; and (3) helping authorities make difficult decisions.
First, DMPs may help citizens to take ownership and to engage with complex policy decisions. For the authors of Enabling National Initiatives, the primary issue in current democratic systems is that public opinion is valued more than public judgement, which echoes the claims of Fishkin (2018) among others. DMPs, they argue, instead capture a response ‘after having access to diverse sources of information, critical thinking and deliberation with other diverse members of their community’ (45). This, the authors continue, counters apathy and helps more of the population to own their decisions, rather than deferring to elected officials. Likewise, DMPs create spaces for people to grapple with complex policy issues (Assembling an Assembly).
Second, the handbooks argue that DMPs are more representative than traditional models of democratic engagement, addressing both inequalities in power and non-participation. DMPs encourage broader participation than standard public engagement methods, such as Town Halls, due to their recruitment mechanism. Sortition, or a two-stage, random stratified sample model, is used to form a diverse, inclusive, and broadly representative group of participants, and avoids corruption (Enabling National Initiatives: 25; Innovative Citizen Participation: 87–9, Facilitating Deliberation: 24; Assembling an Assembly). Demographic criteria such as age, gender, and location, are used to stratify panellists such that they are broadly representative of the target population – although given the size of the panel, minority voices will be small, and organizers may choose to increase their quota such that they comprise a larger panel population than general population (Enabling National Initiatives, 120; see also Farrell & Stone 2020; Dryzek & Niemeyer 2008). The management of minority voices notwithstanding, due to the recruitment mechanism, DMPs can claim to be broadly representative, or at least address disproportionality in recruitment for traditional consultations (Innovative Citizen Participation: 87). These recruitment claims offer the deliberative process both input legitimacy, in that no group is systematically excluded (Farrell & Stone 2020; Fishkin 2018; Smith 2009), and throughput legitimacy, in that the diversity of voices avoids co-option and increases quality of deliberation (Bohman 2007; Landemore 2012).
Finally, DMPs offer benefits not just to participants. While a DMP may not always lead to policy outcomes, it can aid in policy implementation, in agenda setting, and in institutional learning (Goodin & Dryzek 2006; Russell 2017). These avenues for influence – and benefit to the policymaker – are echoed in these handbooks. DMPs have helped public authorities take difficult decisions and provide realistic solutions (Innovative Citizen Participation: 25; Facilitating Deliberation: 24, Assembling an Assembly), they gather a broad range of ideas (Enabling National Initiatives: 24), and can build trust in government and democratic institutions (Assembling an Assembly).
DMPs are particularly beneficial for complex issues that affect a broad community; where there are no ‘right’ answers, or answers are values-driven; where there is appetite for creative solutions; and when decision-makers are willing to acknowledge or accept recommendations (Enabling National Initiatives: 62–65; Facilitating Deliberation: 26; Innovative Citizen Participation: box 4.1). In addition, there are red flags for situations in which a DMP is not suitable: for example, not having sufficient buy-in from senior staff or having insufficient time, capacity, or interest to run a process effectively (Facilitating Deliberation: 26, 34–5; Enabling National Initiatives: Ch. 2). It should provide comfort to those interested in coupling DMPs with democratic systems (e.g., Curato & Böker 2016; Smith & Setälä 2018) that these handbooks describe DMPs as tools to complement existing models of engagement as part of a suite of engagement processes (Innovative Citizen Participation: 111; Facilitating Deliberation: 41).
III. Limitations
These handbooks raise questions about the connection between theory and practice in the study of deliberative democracy. Limitations includes the lack of coherence about the principles of good deliberation; the role of civic lotteries in ensuring representation, and especially what representation means; and the relationship between different level of government and DMPs.
A. Conflation of deliberation and deliberative process
The activity that these handbooks explore is not deliberation writ large, but a deliberative mini public: a ‘carefully designed forum where a representative subset of the wider population come together to engage in open, inclusive, informed and consequential discussions on one or more issues’ (Curato et al. 2021: 3). Both DemocracyNext’s Assembling an Assembly Guide and Enabling National Initiatives also define a Citizens’ Assembly, and the authors of Innovative Citizen Participation state that they use deliberative processes as shorthand for representative deliberative processes, a term again used interchangeably with deliberative mini public (OECD 2020: 10–11).
However, there are notable theoretical differences between the definitions of deliberation used in these contexts. In Innovative Citizen Participation (11), the term deliberation refers to a public or group deliberation that emphasizes finding common ground. In Facilitating Deliberation, it is a process of long and careful consideration and discussion, where everyday people gather to review relevant information and eventually make group decisions or recommendations (15). Enabling National Initiatives defines deliberation as ‘participants in a discussion having equal chance to speak and contribute, balanced by the broadest access to sources of information’ (53). These definitions are sufficient for the average user but fail to draw out the nuance of deliberative theory. Habermas conceptualises deliberation as a key part of democracy, in which participants with equal rights engage in rational discourse and aspire for mutual understanding (1975; 2006). Gutmann and Thompson define deliberative democracy as an ideal in which ‘free and equal citizens, justify decisions in a process in which they give one another reasons that are mutually acceptable and generally accessible, with the aim of reaching conclusions that are binding in the present on all citizens but open to challenges in the future’ (2004: 7). One outcome of the scholarly turn towards deliberative systems (e.g., Parkinson 2018; Parkinson & Mansbridge 2012) is recognising that both formal and informal deliberations are valuable, and that a single process is insufficient to bring about the deliberative ideal.
Highlighting the conflation of deliberation, DMPs, and deliberative mini publics may seem overly fastidious to the average policymaker seeking to just engage people better – but it risks exacerbating existing concerns that too much attention is paid to small-scale, micro-deliberative forums or deliberative services, rather than facilitating mass deliberation in the public sphere (Chambers 2009; Hendriks & Carson 2008).
B. Requirements for representation
The handbooks emphasise the importance of ensuring the recruitment process enables inclusive participation. Four guides discuss how to run a Civic Lottery, a method of random stratified sampling. The OECD, Facilitating Deliberation, and Assembling an Assembly point to MASS LBP’s invaluable How to run a Civic Lottery (2017) for further direction. Scholars may disagree on the precise nature of representation, but there is consensus that there is some level of random selection in DMPs for both claims of representation and ensuring a diversity of voices (Bohman 2012; Brown 2018; Curato et al. 2021; Saward 2010).
Notably absent in these guides is discussion about whether a Civic Lottery is required – or just beneficial – for a deliberative ‘mini-public’, as well as discussion on how these civic lotteries may confer legitimacy to the process. The relationship between a civic lottery and legitimacy is unclear. While Pow (2021) finds that sortition (marginally) increases legitimacy scores, Courant (2022: 164) argues that sortition produces ‘weak representatives’, as they have no binding authority. Lafont (2019) goes further, arguing that through the process of taking part in a DMP, the participants lose their claim to being representative.
Furthermore, potential barriers to participation, such as socioeconomic status, are not fully explored. Both Assembling an Assembly and Enabling National Initiatives suggest ways to minimise such barriers, such as offering honoraria, but do not clarify why such steps are important. To claim that these processes are representative, organisers should be open about the variables on which they are and are not selecting participants. Further, they should be open about how design decisions may affect the participation of certain groups, especially historically disadvantaged groups.
C. Challenges of institutionalisation
Third, these handbooks do not explore the different implications of running DMPs with different levels and different entities of public authorities. Assembling an Assembly, Facilitating Deliberation, and Enabling National Initiatives all provide practical guidance for diverse audiences, including project teams, managers, and facilitators; however, they do not explore the differences that may occur at or with different levels of government. Innovative Citizen Participation explores trends in DMPs, pointing to their plurality in local governments (compared to regional or national), but only briefly hypothesize why this may be the case (69–70). Not all levels of government engage equally, and future handbooks may be strengthened by exploring the competing values that these different levels face: for example, economic constraints that affect engagement at a national level (Dryzek 1996) or the opportunities provided by federal structures as being laboratories of democracy (Volden 2006). Although some work exists about the effects of governance structures on citizen assemblies (e.g., Boswell, Dean & Smith 2022), deliberative democracy would benefit from understanding the effects of the public administration organisational structures (Opitz 2024).
Eight Ways covers why institutionalisation is important – to allow public decision makers to take more hard decisions better; to enhance public trust; to make representative deliberative processes easier and less expensive (a point supported by a footnote on p. 212 of Facilitating Deliberation, that when processes are not embedded, they risk losing institutional memory); and to strengthen democratic fitness (9). Yet these arguments are perhaps not compelling for those currently unconvinced about the value of DMPs.
One-off deliberative consultation processes may not be enough to make meaningful change, while a growing sector of democratic innovations consultants can also be problematic (Hendriks & Carson 2008). Institutionalisation remains elusive, and mini-publics currently lack the design, place, and power in the political system to be defined as proper institutions (Courant 2022). Yet the OECD especially conceptualises institutionalisation as introducing permanent DMPs. These standing DMPs represent the structural components of institutionalisation, but not institutionalisation itself. The handbooks would benefit from also advocating for embedded DMPs, or processes which (1) sit in a productive relation to other democratic institutions and (2) are difficult to abolish or bypass (Bussu et al. 2022). For those seeking to institutionalise or embed DMPs, conceptual clarity is essential to effectively channel efforts. Fostering a supportive political culture that values ongoing collaboration between the various democratic entities will further enhance the potential for DMPs.
D. Going beyond the Global North in deliberative democracy
The organisations constructing these guides are predominately based in – and thus reflect – a Global North perspective. The OECD and DemocracyNext both include examples from Latin America, Africa, and India, but these are positioned as interesting ideas, rather than comparable cases from which to draw learnings. In doing so, the handbooks risk further contributing to a critique of deliberation as too grounded in Western democratic theory, and as a tool for ongoing colonization (See Asenbaum et al. 2024 and Morán & Curato 2022). Not all values or approaches will work universally, and understanding how they function in different contexts is important for the long-term stability and uptake of the DMP project. Despite calls for recognising and legitimising non-Western epistemologies in deliberative decision-making (Ibhawoh 2024), the most highly cited scholarship is still largely centred in the Global North. Incorporating Indigenous approaches should not be challenging in countries like Canada and Australia – and is valuable for addressing contemporary crises in democracy – and its absence is notable. However, there have been positive steps in the field to recognising voices from the Global South. Projects such as Demo.Reset and work from International IDEA has kickstarted the conversation (Demo.Reset 2022; Curato et al. 2024); the next step is to ensure this work is meaningfully incorporated into our understandings of deliberative democracy.
IV. Conclusions
These guides provide a rich set of materials that together describe the practice of deliberation in a novel and thoughtful manner. However, they contain a wealth of information that at times becomes unwieldy – there are lists of principles of deliberation, principles of facilitation, facilitation tools – and while they are presented in clear, plain language, for a reader who is not embedded in the deliberative landscape it is challenging to know where to begin. Likewise, some of these principles are less principles, and more a mix of normative statements, questions to consider, and critiques of the current system. In the absence of alignment between all guides, greater clarity on how the principles were developed – and moving away from an assumption that these values should be universal – would be valuable.
These handbooks suggest that scholarship and practice still largely do not speak to each other. There are three parts to this: (1) data collection; (2) research goals; and (3) theoretical claims and language. On data collection, practitioners have a wealth of experience of what works and what doesn’t, and are on the front line of this work, but their studies largely sit outside of academic structures. Similarly, there is a groundswell of experiments in deliberative democracy (Gastil 2018), but summaries focus on peer-reviewed studies. Academic journals can better incorporate experience from practice, such as the opportunities within the Journal of Deliberative Democracy to include practice-based reflection. Second, there may be a disconnect between practice and scholarly research goals: what practitioners want to know, and what academics are researching. For example, there may be overemphasis on some aspects of a deliberative process such as pre-post panel survey research (Theuwis, Van Ham & Jacobs 2024), and insufficient research on the surrounding structures and contexts of a DMP. Finally, ensuring consistent language and claims, and cross-referencing academic literature and practice, would help build a more comprehensive narrative. In previous research, I found incongruence between theoretical claims of deliberation and why policymakers choose deliberative activities in practice (Massie 2023). To move toward standardisation of deliberative principles and processes (see Parry 2023), bringing together academia and practice is critical. In this third wave of deliberative scholarship, we recognise strengths of, but do not blindly defer to, DMPs (Bächtiger & Goldberg 2020; Lafont 2019). These handbooks are an excellent starting point, both for developing research questions, and highlighting the need for greater discourse between academia and practice.