Fishkin part 3: The limits and future of deliberative democracy

Alongside the possibilities of representative, informed and moderated citizen deliberation, it is worth reviewing the limits of implementation as well. The closing part of the series discusses the questions of participatory inequality, scalability and resource demand, then the new models and the possibilities for application in Central Europe (OECD, 2020).

01 Critical approaches and further developments

Limits

In the course of the practical application of deliberative democracy, the structural and methodological limits soon come to the surface. One of the most serious is participatory inequality: people's social position, cultural capital and communicative ability can amplify or suppress their voice. Alice Siu's essay Deliberation & the Challenge of Inequality examines this criticism. According to its empirical results, in carefully designed deliberation that ensures more equal conditions of participation, differences of status and skill do not necessarily become inequalities of influence. The risk of participatory inequality is especially important for traditionally under-represented groups (communities struggling with language barriers, the under-educated population, minorities), which is why inclusive design and equitable moderation are not merely a technical question (Siu, 2017).

The second limit is scalability: as the size of a deliberative forum grows, maintaining quality becomes ever harder. The study by Nicole Curato, John S. Dryzek and their co-authors, Twelve Key Findings in Deliberative Democracy Research, discusses the smaller forums as part of a wider deliberative system. The question of scaling is thus also how the mini-publics connect to the wider public. A newer trend is automated deliberation or digital scaling: Hélène Landemore's lecture given at Stanford examines to what extent processes can be scaled with artificial intelligence in such a way that the depth of argument is not lost — while elements of human interaction such as non-verbal gesture, empathy or conflict management demand particular attention. Along the five dimensions of scaling (content, accessibility, time, infrastructure and political embedding) the main challenge is building up the "civic infrastructure" — the institutional and social networks that sustain deliberation authentically and continuously (Curato et al., 2017; Landemore, 2022).

The third limit is resource demand: deliberative processes require high financial, organisational, administrative, research and moderation capacity. Although deliberative polls are often referred to as a "gold standard", many events can be sustained only from research funds, and not every political community is able to maintain the model in the long term. A frequent criticism is that many experiments operate as a temporary "project method", do not become a permanent institutional element, and do not examine long-term sustainability. A study comparing fifteen Citizens' Initiative Review processes (2010–2018) found a connection between the quality of the participatory experience, satisfaction and the sense of learning; the connection with the change of opinion was weaker. The quality of the process and the change of opinion are therefore not the same measure of success (OECD, 2021; Knobloch and Gastil, 2022).

02 New trends: mini-publics and hybrid models

The most recent wave of deliberative democracy is defined by two mutually reinforcing directions: the rapid spread of mini-publics (small, representative citizen bodies selected by lot), and the hybrid assembly models that link deep, small-group deliberation with wider, often digital, involvement. Mini-publics have by now become a distinct democratic "genre" at municipal, national and EU level, and all of them promise to fit Fishkin-style deep, informed deliberation into decision-making. According to a Swiss study, the support for mini-publics is also connected to trust in one's fellow citizens and to dissatisfaction with the existing direct-democratic institutions. At the same time, respondents did not wish these forums to replace the existing institutions. That is, the social legitimacy of mini-publics is not homogeneous, and it is worth adapting the design to the given political culture (OECD, 2020; Kübler et al., 2025).

Hybrid models — especially as a lesson of the pandemic — are "hybrid" in three senses: an online–offline combination (in-person sessions and digital interim phases); the linking of deep deliberation and broad involvement (alongside the mini-public, open collection of opinion, community platforms); and the institutional joining of citizen recommendation and decision-maker responsibility (a formal obligation to respond, follow-up). The online component, with appropriate design, can increase access and cost-effectiveness, while the in-person phases preserve the elements of trust-based and empathic learning — good practice deliberately alternates between the two (OECD, 2022).

The "textbook example" of the hybrid models is Ireland, where the citizens' assemblies became the antechamber of national-level reforms. The process of the abortion reform shows well how deliberation by lot and in depth connects to a broad social decision: after long deliberation underpinned by experts, the citizens' assembly made recommendations, which parliament put on the agenda, and which a referendum then legitimated. The lesson is not that every issue must end in a referendum, but that the recommendations of a mini-public gain real weight when their path to a formal decision is clear and institutional feedback exists. More and more "hybrid democratic innovations" link concentrated, small-group deliberation with a large-scale voting or preference-ordering procedure, in order to unite depth and breadth in a single process. For practical implementation, detailed handbooks and "toolboxes" are now available, giving guidance for every phase of hybrid design — recruitment, informing, moderation, ethics, evaluation — with particular attention to the inclusive design of the online phases (Citizens' Assembly, 2017; Side, 2022; Hendriks and Michels, 2024; OECD, 2022).

03 The future of deliberative governance in Central Europe

The following thoughts weigh the possibilities for applying the methodological and institutional considerations presented here in Central Europe.

The future of deliberative governance in Central Europe depends on whether the countries of the region will be able to transform their traditional conception of democracy: from an election-based, vertical representative system towards a model in which decision-making becomes a process of community learning. The deliberative outlook does not demand that citizens take part directly in every decision, but that they have the chance to reflect before decisions — to understand, to weigh, to ask questions, and so to contribute more meaningfully to the shaping of public policies.

Because of the historical peculiarity of the region — the centralised administrative tradition and the dominance of central power — deliberative governance here is not simply a new method but a cultural breakthrough. Municipal decision-making has so far typically proceeded along a formalised, expert and legal logic, while citizens' participation was mostly limited to expressing an opinion or to protest. Deliberative practice, by contrast, creates a communicative space where the debate is ordered not by the hierarchy of power but by the force of arguments. If this becomes built into the workings of local government, it is not only a democratic innovation but also increases the efficiency of decision-making — since well-informed, jointly prepared decisions are easier to implement and to have accepted.

The future may be defined by three interrelated directions. The first is cultural change: the acceptance that decision-making is not the exclusive terrain of power but a community responsibility — this requires openness, curiosity and patience on the part of local leaders. The second is capacity-building: without trained moderators, a professional background group, data, publicity and documented feedback, deliberation remains merely symbolic. The third is technological and organisational innovation, which, by linking in-person and digital participation, is able to create scalable yet human processes. The deliberative local government of the future will probably work in a hybrid form: local forums, online consultations and community assemblies, reinforcing one another, form a decision ecosystem in a transparent structure (OECD, 2021; OECD, 2022).

The deliberative turn does not take the place of elected representation but makes up for its missing dimension: dialogue before decisions. The Central European cities and districts that are able to introduce such processes will become not only more democratic but also smarter — because they make local knowledge, experience and community sensitivity part of politics. In this, the Fishkin model is not a recipe but a framework: proof that democracy is capable of reinterpreting itself if it returns to the ideal of the thinking citizen. Democratic quality is ultimately measured not in the speed of decisions but in the depth of the shared understanding behind them.

04 Closing thoughts

Deliberative democracy does not promise a new system but a new spirit within democracy. Its essence lies in the recognition that human communities seek not merely governance but meaningful cooperation. In modern political systems, where the quantity of information has grown to an unprecedented degree, the real challenge of decision-making is no longer how to divide power but how to divide attention and understanding.

Deliberation is a form of community thinking. There, where opinions do not merely compete but enter into dialogue with one another, democracy becomes more than an electoral mechanism: it becomes shared reflection. This is the point where Fishkin's kind of thinking brings a new quality — placing at the end of the process not the decision but the learning. Deliberation does not abolish differences but gives them meaning: it shows that democracy is not about identical opinions but about thinking together.

The legacy of Fishkin's thought is not a methodology but a principle: that social peace is not the result of unanimity but the fruit of the courage for open, argued debate. Deliberation does not promise a perfect democracy, only a more thinking one — and that is precisely enough to make politics human again.

Sources

• Citizens’ Assembly (2017): First Report and Recommendations of the Citizens’ Assembly: The Eighth Amendment of the Constitution.
• Curato, N., Dryzek, J. S., Ercan, S. A., Hendriks, C. M. and Niemeyer, S. (2017): Twelve Key Findings in Deliberative Democracy Research. Dædalus, 146(3), 28–38.
• Hendriks, F. and Michels, A. (2024): Exploring the Democratic Merits of Hybrid Democratic Innovation: Combining Deliberation and Voting in Participatory Budgeting New Style. International Journal of Public Administration. Online publication: 19 August 2024.
• Knobloch, K. and Gastil, J. (2022): How Deliberative Experiences Shape Subjective Outcomes: A Study of Fifteen Minipublics from 2010–2018. Journal of Deliberative Democracy, 18(1).
• Kübler, D., Gut, R., Heimann, A., Stojanovic, N. and Colombo, C. (2025): Who Supports Deliberative Mini-Publics in a Context of Direct Democracy? The Role of Trust and Dissatisfaction. International Review of Administrative Sciences, 91(3).
• Landemore, H. (2022): Can AI Bring Deliberation to the Masses? Lecture, Stanford Institute for Human-Centered Artificial Intelligence, 19 October 2022.
• OECD (2020): Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave. OECD Publishing.
• OECD (2021): Eight Ways to Institutionalise Deliberative Democracy. OECD Public Governance Policy Papers, 12. OECD Publishing.
• OECD (2022): OECD Guidelines for Citizen Participation Processes. OECD Public Governance Reviews. OECD Publishing.
• Side, K. (2022): “Changed Utterly”: The Citizens’ Assembly on the 8th Amendment. Feminist Encounters, 6(1), 05.
• Siu, A. (2017): Deliberation & the Challenge of Inequality. Dædalus, 146(3), 119–128.