1. General Introduction

As the global community confronts the multifaceted challenges of climate change, the need for innovative and inclusive decision-making becomes ever more pressing. Within the ambit of the CLIMAS Horizon project, this methodological guide offers a comprehensive approach to setting up and facilitating Climate Assemblies. Our primary contribution lies in shifting the focus from broad issues to specific policy dilemmas in a more balanced process. By doing so, we aim to foster deliberations that are both deeper and more targeted, ensuring they are rooted in a value-based approach. This guide provides a detailed framework on how to design all the assembly phases to achieve this specific deliberative focus, ensuring that participants’ contributions can have a meaningful impact on climate action. Before delving deeper, one might ask: What exactly is a Climate Assembly? Why is it deemed valuable, and how can its effectiveness be enhanced?

1.1. What is a Climate Citizens’ Assembly?

Citizens' assemblies are an attempt to involve everyday people in political decision-making and are one of the most popular practical expressions of deliberative democracy today[1]. Deliberative democracy is an inclusive form of governance in which decision-making is achieved through discussion and dialogue, with the aim of developing and moving individual preferences towards a collective agreement (Dryzek, 2001, p. 1). Climate Assemblies are therefore an attempt to address the challenge of climate change using a deliberative democracy approach.

Citizens’ Assemblies bring together a group of individuals, recruited through random and stratified selection to broadly reflect the wider population with respect to key demographics. The assembly deliberates on the basis of the value-based and experiential knowledge of the participants and the information provided by different perspectives from knowledge holders, which leads to the production of a set of recommendations with the aim to inform decision making[2].

1.2. Why a Climate Assembly?

Climate Assemblies are a response to a double crisis: On the one hand the crisis of Western democracy and classical political representation, and on the other hand the climate crisis. Some civil society groups, experts, and political leaders argue that short election cycles do not allow political representatives to address the long-term challenge of climate change[3]. In contrast, a group of citizens selected by lottery could approach it with intrinsic surplus value by creating a strong social mandate, focusing on long-term solutions, co-designing actionable plans, building public trust, and ensuring social justice through a socially-inclusive approach (Howarth et al., 2020, p. 1112-1113). In this spirit, and building, in particular, on the experience of the Irish Citizens' Assembly, the Extinction Rebellion movement in the UK and around the world began to call for the use of Citizens' Assemblies to politically address climate change, thus increasing the demand for these assembly formats[4]. The European Climate Foundation and many governments have also stepped up to the plate, and have promoted and financed the implementation of various climate assemblies across Europe. In reality, both public bodies and civil society institutions have established climate assemblies at different levels of governance, triggered by a variety of factors (KNOCA, 2022) and with a variety of outcomes.

1.3. How to improve Climate Assemblies?

Climate Assemblies have become a revitalising force for climate action and democratic innovation at the local, national, and global level. However, it is not entirely clear what the real impact of their recommendations on public policy has been (Thorman & Capstick, 2022). Are the recommendations lacking in specificity or depth? Do governments base their acceptance or rejection of recommendations depending on their own agendas? Do the benefits of implementing the recommendations take account of the associated trade-offs? Do they lack adequate technical detail? Is the assembly perceived as being monopolised by a particular political group or section of society, thereby generating suspicion among other groups? The methodology presented here is an attempt to address some of the problems related to these questions through a practical and specific approach.

First, we believe that a problem that cuts across the vast majority of Climate Assemblies is related to the fact that climate change is not one single issue[5], but is composed of multiple issues[6]. Mitigating and adapting to climate change will require society to make significant changes to the socio-economic system on which it is based. It is necessary to change the way we live, eat, move, use energy, etc. In this context, an assembly process with a limited duration may not have the necessary bandwidth to delve into such complexity with the required depth.

Second, what does it mean to discuss an issue in depth? As we will see in the following chapters, our assumption is that the most effective way for citizens to delve deeper into an issue or topic is to focus on the policy or value-based dilemmas[7] associated with it.

1.4. The Keys to The Methodology

This methodology guide is based on the hypothesis that the potential impact of Climate Assemblies' recommendations on public policy could be significantly increased by improving the design of the assembly process itself, including both agenda setting and follow-up. Building on the OECD's good practice principles for deliberative processes (OECD, 2020, p. 130-147)[8], this guide seeks to develop the principle of integrity through the concept of balance and the principle of purpose through the policy dilemma approach. It also proposes that assemblies should be permanently institutionalised to tackle the long-term challenges of climate change by breaking them down into manageable tasks.

1.4.1. Balanced Diversity

In discussions around the management and coordination of Citizens’ Assemblies, 'independence' and 'neutrality' are often emphasised [9]. While independence remains important and will be highlighted in chapter 3, especially regarding the role of an independent organisation experienced in delivering such processes, it is not the sole factor to consider. The commissioners who activate the assembly inherently have political affiliations. This makes it difficult to abstain from influencing key aspects of the assembly’s design[10]. They cannot merely entrust the neutrality of the process to contracted managers or operators with a clientelist relationship. Moreover, the pretence of neutrality can inadvertently strip the process of its democratic essence by removing conflicts and debates[11]. In contrast, this guide elevates 'balance' as the main core aspect, applied to the governance of the process, the content provision, and the body of participating citizens. We propose this shift towards balance as a more tangible and actionable principle.

Achieving balance in governance could be contemplated through the inclusion of stakeholders, which would ostensibly enrich the process with a variety of perspectives and values. However, there are challenges to this approach. Direct stakeholder involvement without a clear objective framework could introduce bias and undermine the perceived legitimacy of the assembly. This leads to the pivotal question of who decides how and to what extent stakeholder perspectives should be taken into account.

In practice, this methodology proposes political parties as the above-mentioned objective framework that can be seen as legitimate by the general public and interest groups. The diversity of values held by different parties serves as a good enough proxy for the distribution of values in society, allowing a wider range of stakeholder perspectives to be mapped. Consequently, the basis of process balance is achieved through the formation of a broad oversight group that includes representatives of all political parties, reflecting the composition of Parliament, with party representation based on the number of seats they hold. The oversight group, once constituted, nominates experts representing different stakeholders perspectives relevant to the political dilemma at hand. Ideally, this list of experts should be approved by randomly selected citizens[12]. The content group will be entrusted with curating a tapestry of information that is as rich and multifaceted as the societal debate itself. By weaving together these varied strands of perspectives, the content provided to the Assembly will embody the balance we seek, with the selection of experts being a transparent process that enhances the group’s legitimacy and the Assembly’s integrity.

Similarly, for the body of participating citizens, balance is operationalized through a combination of random selection and careful stratification. Existing academic attitude surveys serve as a foundational element in this process, providing an additional criteria regarding the policy dilemma under consideration. The assembly thus becomes a microcosm of society, both in demographic composition and in the diversity of perspectives, achieving a nuanced form of balance.

By seeking balance in these specific ways, this guide offers a pragmatic and democratic blueprint for setting up and facilitating citizens' assemblies. This approach embraces the complexities and diversities inherent in society across all phases of the assembly.

1.4.2. From Issues or Topics to Policy Dilemmas

Focusing the deliberation on a topic or issue in isolation presents two distinct but related challenges. The first is the risk of falling into an ‘epistemic trap’[13]. In this scenario, discussions veer toward technical nuances and evidence-based solutions, leaving little to no space to consider the valuebased trade-offs that are often the cornerstone of political debates. This focus on technicalities may set citizens up for a task that is typically the domain of experts, which may lead to frustration and missed opportunities for meaningful citizen deliberation.

The second challenge arises when generic questions are posed for deliberation. These questions often result in wish lists[14] that overlap with existing or planned policies, making it challenging for citizens to distinguish any new contributions. This is exacerbated by the fact that complexities of policy nuances are not easily grasped. Like the epistemic trap, this approach also lacks the time or space for considering the value-based trade-offs involved, which could be a basis for rejecting these proposals.

In short, the topic-based approach may lead to either overly technical discussions or overly broad wish lists, both of which lack the potential for significant policy impact. The key contribution of randomly selected citizens in the deliberations does not arise from their individual expertise or competences, but rather from their diverse value-based perspectives[15] and collective ability to reach agreement on policy issues. Thus, the objective should not be to turn lay citizens into experts, which will not happen in the limited time available for deliberation. Instead, the focus should be on creating a setting where their diverse perspectives are guided by a value-based approach, offering a pathway to navigate through the complexities of a political dilemma while minimising the technical aspects. This dilemma approach to deliberation will also help to keep a balanced perspective throughout the process.

1.4.3. Tackling Manageable Tasks in a Permanent Process

To avoid the creation of wish lists, it is important that the task of the Assembly is both specific and manageable, not made up of multiple dilemmas. Addressing the issues in depth sidesteps recommendations that may not be actionable, thereby making a meaningful difference to other political processes. In this way, it is proposed to break down the challenges of climate change into parts and avoid treating the issue in a generic way. Not all countries or regions face the same challenges. While some countries have a lot of solar radiation, others have a lot of wind. While in some countries the main threat is drought, in others it is flooding. This approach allows different regions to focus on different dilemmas and not replicate the same processes, complementing and enriching a global and more coherent deliberation with many assemblies in different places focusing on different challenges. There is no need for a region with a similar dilemma and social reality to replicate the same process, it can simply address a new mandate and learn from the recommendations of previous assemblies in neighbouring cities, regions, or countries.

With this in mind, we propose a continuous process to address each issue in a differentiated manner. Because of the climate emergency, some people might think that societies should address all issues as soon as possible. In practice, given the limited resources available for holding citizens' assemblies, the frequency of repetition of each deliberative process for each theme could vary according to the challenges of each place. This pragmatic approach will allow our societies to adapt and address the most relevant and urgent challenges at any given time with Assemblies, while dealing with the rest through classical decision-making processes.

1.5. How Do We Create This Methodology?

The initial draft of our methodology is based on our experience as mini-public practitioners. The experience of Deliberativa, the partner leading this guide, draws on local processes such as the Barcelona Youth Forum or the Madrid City Observatory, regional processes such as the Besaya Citizens' Jury, the Valencia Mental Health Assembly or the Gipuzkoa Climate Assembly, global processes such as the Global Assembly on Climate Emergency and European processes such as the Conference on the Future of Europe or the Citizens' Panels of the European Commission, the latter shared with another organisation of the consortium: Ifok. This second Ifok partner also has extensive experience in Germany in local, regional and national processes and is continuously contributing to the revision and improvement of the methodology.

The initial draft of the methodology aims to be implemented in the three processes led by the different partners of the consortium: The Generalitat de Catalunya leading the Catalan Climate Change Assembly, Green Liberty leading the Riga Assembly and Ifok leading the Edermunde Climate Assembly. In all these processes, Deliberativa will play a supporting role, with particular attention to the Catalan Assembly.

It is important to recognise that this methodology is very detailed and exhaustive and its implementation will depend on the political will and resources available in each scenario.

While the main driver of the methodology, the dilemma-based approach, will cut across all three cases, we expect that not all other elements of the methodology can be developed in the same way. In a distributed manner, more specific aspects of the methodology may be tested in one assembly but perhaps not in the other, and vice versa. The aim of these tests will be to calibrate the initial proposal in a kind of virtuous circle in order to generate a final proposal in 2025.

In addition, other climate or similar assemblies in which Deliberativa and Ifok are involved during these years will also serve to test and develop different parts of the methodology. In 2023 this will be the case of the design of the Permanent Assembly of the Basque Country, which Deliberativa is developing together with other actors and the Basque Government, or the National Citizens' Assembly on Nutrition in the German Bundestag, led by Ifok and other external partners to the consortium. Both cases will be informally added as learning cases that will also feed into this guide.

By 2024, it is possible that the Climate Assembly of the island of La Palma will also join the series of experiences.

2. Political and Budgetary Feasibility

Any Citizens’ Assembly that aspires to a strong follow-up in policy-making needs to be established by the appropriate authority, i.e. a public decision-making power. Ideally, in order to maximise the impact of the assembly, this public authority, institution, or organisation, hereinafter the commissioner[16],should be the one to establish the assembly process.

Establishing such an assembly requires a major commitment. The aim of this chapter is to understand the extent of this commitment and to help the commissioner decide whether it is feasible to activate the Citizens' Assembly process.

Firstly, we consider the question of permanence, followed by the political feasibility assessment, addressing not only the commitment to the outcome in the form of citizens’ recommendations, but also the role assigned to the commissioner, who will have an important voice in the process but will not control it. We then consider the question of budget feasibility: The commissioner needs to be aware of the scale of the effort, both in terms of external work packages to be contracted and internal services to be taken on. It is the commissioner who has to take responsibility for putting in place the resources, both financial and human, to enable the assembly to take place. Finally, we consider the possibility of activating an assembly that is not established by a public decision-making body.

2.1. Assessing Permanence

We noted in the previous chapter that making assemblies permanent is a key aspect of the innovation proposed in this guide. Thinking about permanence from the outset will be very useful for any commissioner who has led Citizens' Assembly projects in the past or who is very clear about what they want. For a commissioner who is approaching a Citizens' Assembly process for the first time and is not yet clear about how it works and what is wanted, a more pragmatic approach would be to treat the first assembly cycle as a pilot. In the latter case, a pilot assembly provides an opportunity to understand the mechanics and benefits of a Citizens' Assembly. Once the pilot is complete, the commissioner can evaluate it and decide whether to set up a permanent structure to look at other issues.

While this guide places emphasis on the establishment of a permanent process, it is equally equipped to guide those considering a one-time assembly or pilot assembly. Commissioners can effectively utilise the principles and guidelines presented here without the commitment to permanence.

2.2. Political Feasibility Assessment

In the absence of specific ad hoc legislation, the proposals and final decisions of the Citizens' Assembly will be called 'recommendations' and will not be binding. The commissioner will have the final say. However, the commissioner might want to consider that part of the decision-making power is delegated to the citizens participating in the assembly. Thinking and acting as if the process was binding, even if it is not and is based on the political commitment of the elected official, will help to make it more rigorous and effective. In this way we can define the first commitments to be made:

• At the end of the process, the members of the assembly will recommend what measures or legislation should be implemented to tackle the policy dilemma. It must be also assumed that, as a commissioner, it will be its duty to do all it can to implement these recommendations and, if it does not, to explain why.

Accepting the importance of these recommendations entails concern about how citizens arrive at them and how their implementation is followed up. The function of overseeing the good execution of the process implies accepting the governance of it.

• The process will be governed by giving centrality to a group of citizens who will participate in a Citizens' Council[17]. This council, being deliberative in nature, will consist of citizens chosen through a democratic lottery and its members will rotate periodically. Its role will be to set the agenda and oversee the functioning of the Assembly, as opposed to the role of the Citizens' Assembly[18], which will deal with a specific policy dilemma. This citizen body will work closely with the oversight group, which will prepare, together with the project team, the information and facilitation for the decision making of this body. These key actors are defined in the next chapter.

In practice, setting up a Citizens' Council means running two citizens' bodies in parallel[19]. Given the leap in complexity and cost this entails, we consider it wise to activate this second body after the first assembly pilot, once the commissioner has taken the step towards permanence.

The oversight body will be able to take final decisions provisionally until the Council is established.

As a consequence of this last commitment, the commissioner must accept that they will not have direct guidance over the assembly process. Instead, they will be part of a larger balanced group, from now on the oversight group, that will help to inform and oversee the work of the Citizens' Council and the project team to maintain balance. This includes the agenda setting that will define the policy dilemma that will create the mandate for the Citizens’ Assembly.

2.3. Budget Feasibility Assessment

A commissioner approaching the idea of a Citizens' Assembly for the first time is usually inclined to underestimate the scale of the project. It is therefore very important to understand beforehand the costs and the commitment of human resources required by their own administration in order to be able to assess its feasibility.

Here are some outlines to understand the scale of the effort for the first Assembly cycle, excluding the Council. This will be a 12-18 month single cycle process involving a dedicated project team of 23 people working almost full-time. Another team of about 6-8 people will work part-time, coordinated by the project team, and a large team of staff will be mobilised for the Assembly sessions, which could involve more than 30 people. Between 24 and 100 citizens will need to be informed, facilitated, fed, and accommodated during the Assembly sessions. Assembly working time is crucial: Whereas a minimum of 30-40 hours[20] is usually recommended for generalist assemblies, the dilemma-based method proposed here requires around 55 hours[21]. These hours are spread over about 4 to 6 weekends during 3 to 4 months.

Figure 1 shows a tree map of the different budget packages for a 55 hour assembly with 100 members. The core players are the team working throughout the year. The work of the operators will be more concentrated during the Assembly events. Logistics includes all costs related to the organisation of session events: Venues, catering, travel, accommodation, internet, sound equipment, projectors, screens, materials, etc. Facilitation includes the people who present, moderate, and facilitate the citizens' work. Recruitment includes sorting, contacting by letter and telephone, stipends for citizens, stratification and management of citizens. The project team coordinates all the work and supports the oversight and the content group. The content group prepares the content to be shared with citizens for informed and balanced deliberation. The communication team is responsible for communicating the process to the wider public.

Figure 1 - Permanent Citizens’ Assembly Budget 

We will consider two cost scenarios, the local one, where it is not necessary to pay for long travel and accommodation for participants, and the non-local one, where it is necessary to include these costs. Figures 2 and 3 show some estimates for monolingual meetings with 55 hours of deliberation:

Figure 2 - Estimates for 55 Hours of Deliberation in Non-Local Assembly

Figure 3 - Estimates for 55 Hours of Deliberation in Local Assembly

2.3.1. Size of the Assembly

An assembly typically ranges from a minimum of 24 to 32 members to as many as 200 members with sufficient resources. It is an attempt to represent[22]. As with any representative sample, the larger the sample, the more representative it will be. Ideally, larger assemblies could be held to improve representativeness, but the reality is that as the number of participants increases, the complexity and cost of the process increases strikingly. In figure 4 we show how the budget increases exponentially with the number of participants:

Figure 4 - Exponential Budget Increase Relative to Participant Numbers

This non-linear increase in the budget is due to the growing complexity of the event sessions, while the budget of the core players remains largely unchanged.

The budget of the core players and the number of hours of assembly work, which are crucial for the quality of the process, should not be limited by the ambition to have a very large assembly. A small assembly guarantees a simpler process in which its deliberations can be more in-depth and agreements can be more effective, compensating for the less representative nature of the sample. In practice, we are looking for a reflection of the demographics of the population, which can actually have political representativeness, not a statistical one.

2.3.2. Are Costs Reduced by Internalising Services?

It is possible to find specialised staff within the administration, or even use existing framework contracts to internalise costs. It is important to identify these services or staff and engage them at the right time. This strategy can reduce direct costs by around 25%. See figures 5, 6, and 7 for details.

If training of administrative staff is required, it is important to remember that this will always be more expensive as it requires the time of external consultants. However, this will lead to mediumterm savings on similar projects in the future. Commissioners intending to institutionalise[23] deliberative practices should engage in progressive training of their staff.

Figure 5 - The Path towards Institutionalisation

Figure 6 - Budget Estimation for Local Assembly with 25% Indirect Costs Reduction

Figure 7 - Cost Reduction Strategies in Administration and Their Impact

2.3.3. Feasibility of a Permanent Assembly Budget

Those commissioners who are willing and motivated to make the process permanent should also assess the feasibility of a permanent budget flow, which will inevitably be required. The costs should cover one assembly for each cycle of deliberation. Each assembly will have different citizen participants and address different dilemmas.

The Citizens’ Council should become active in a permanent process. This second body could increase the cost of each cycle by 10-20% due to the events of the Council, with a minimum of 24 persons, as shown in figure 9.

Figure 8 - Budget for Permanent Citizens' Assembly with Council Involvement

Figure 9 - Budget Estimates for Non-Local Assembly and Council Involvement

Each cycle, the commissioner will be able to internalise, systematise, and streamline implementation. In short, the cost of the process will be significantly reduced within a few years. Although there is not much experience of institutionalisation, the experts who have contributed to this guide suggest that after a few cycles, costs could be reduced and internalised, cutting out-ofpocket expenditure by around 25 to 35%24]. Another way to reduce costs could be to repeat cycles less frequently, for example twice a year instead of annually.

2.3.4. Navigating Financial Constraints

These budgets are based on experience in European countries and are indicative. For those local authorities who are looking for the cheapest price and still find that the figures are out of their budget: It is important to understand that the design should not go below 55 hours and 24 citizen members, and that all packages are essential for a quality deliberation. It is still possible to get the project off the ground if the commissioner has motivated and available staff. When we explore possible strategies to meet this challenge, several key approaches stand out:

• Shift to indirect costs: The commissioner will need to strengthen its internal team or make better use of existing supplier contracts. The costs will not disappear but making them indirect can facilitate the internal justification of the project.
• Seek external funding: Consider the possibility of obtaining funding from organisations that are willing to support the initiative without directly influencing its course.
• Form strategic collaborations: Partnering with institutions such as universities can be beneficial. For example, a scholar could take on the role of 'evaluator' as part of their academic journey.
• Engage volunteers, interns and experts: Harnessing the enthusiasm of volunteers, students, or experts who are passionate about the subject can be a cost-effective way to strengthen the team. Local government or educational institutions may have resources or programmes to facilitate this strategy.

By adopting these approaches and with a proactive leadership of the project team, commissioners could more effectively address the financial challenges.

2.4. An Assembly that is Not Established by a Public Decision-Making Body?

Although not common, there are cases of assemblies that are not activated by a commissioner[25]. It could be a community, a university, a private foundation, or any other body that has no decisionmaking power over the public sector. It is always possible to activate an Assembly in this way, however, as has been seen, it is often the case that such processes have great difficulty in passing on their recommendations to a commissioner[26]. Budgetary feasibility can be assessed in the same way as if one were a commissioner, but this is not the case for political feasibility.

First of all, the most advisable strategy is to convince a commissioner to hold the Assembly. If funds are available, it is more likely that a commissioner will be able to accept the political commitment.

In this case, this methodology will be followed and the organisation that has convinced a commissioner will have to provide funding without seeking to influence the process. Providing funds should not imply becoming the project team of the Assembly, although this decision could be taken by the oversight group if it considers that it has the independence and professional experience to do so.

If a commissioner does not agree to hold the Assembly, it may still be interesting to run the process in such a way that both the commissioner and other political parties can participate informally or as observers. The Assembly becomes a living example that can help political forces gain confidence in these deliberative processes and consider establishing their own assembly in the future. In this case, the organisation activating the process should behave like a commissioner in providing funds, but should not have a seat on the oversight group, which should be given equally to the different political parties or, failing that, to like-minded stakeholders nominated by them.

Past cases show that even when the Assembly is held by a commissioner, it is difficult to achieve adequate impact of the recommendations[27]. It is possible that a well-designed process, following a good methodology, can have more impact than a poorly designed Assembly held by a commissioner[28]. If stakeholders and public opinion perceive the process as legitimate and intelligent, there is nothing to stop the recommendations being implemented. Thus, an Assembly, ‘from the outside’, can also serve to show how best to use or design these kinds of processes.

3. Activating Core Players

Once the commissioner has secured the political and budgetary commitment, it is time to activate the project. In this chapter, we explain how to set up the core players who will be responsible for running a fair and balanced assembly process.

As a first step, it is essential to have an oversight group with sufficient political balanced diversity. This group will need to approve the selection of the evaluator, the project team[29], and the content experts. It will also help the Citizens’ Council to set the agenda by identifying policy dilemmas to be worked on. Once the project team has been recruited, it will be responsible for facilitating the work of the oversight group. The project team can also start to define the tender for the operator(s). The evaluator will start to monitor the process at the same time as the project team is activated. The evaluator will start to monitor the process at the same time as the project team is activated. Figure 10 shows the relationship between the working groups, figure 11 shows the relationship of these groups to the citizen bodies and figure 12 shows the temporal order of activation of the different groups and bodies.

Figure 10 - Core Players and Working Group Relationships11

Figure 11 - Core Players and Working Group Relationships

Figure 12 - Core Players and Working Group Relationships

3.1. Setting Up a Balanced Governance

As mentioned in the opening chapter, the commissioner must establish a balanced oversight group that will help to steer the process. This implies the inclusion of other political actors, particularly the main parties, but ideally all parties with parliamentary representation. This step is essential in this methodology, as the lack of agreement between the different stakeholders will be the lever that helps to identify the main dilemmas. In addition, addressing such dilemmas involves long and medium term policies and it will be crucial that changing parties in power are involved and committed to the process to ensure that citizens' recommendations are implemented.

The oversight group could be informed by advisors representing different stakeholders from institutions, civil society organisations, business, or academia. These experts can provide insight into the range of potential dilemmas and issues involved. The weight of these advisors will always be proportional to the party that includes them and the party may even delegate its participation in the group to this person. The people in this group may change if the parties so decide, as long as this does not interfere with the working schedule.

The main responsibilities of the oversight group are related to all major decisions that require balance and independence. In order to avoid deadlocks, broad majority decisions of 80% can be accepted for which parties have a weight proportional to their parliamentary representation. Validating the project team. The project team, both its in-house members and external experts, must be trusted by all parties for their track record and professionalism.

•  Validating the process evaluator. The evaluator will be the one who monitors the work of the project team and reports to the oversight group. It must be a trusted and professional agent.
•  Helping to prioritise the dilemmas. This task will be developed in the next agenda chapter.
• Helping to select the content group. Content strongly influences citizens when it comes to positioning themselves. It must be carefully balanced. This task will be developed in the chapter of content provision.
• Validating the editor of the content group. This task will be developed in the chapter of content provision.
• Validating the stratification, paying particular attention to the attitudinal criteria. This task will be developed in the chapter of democratic lottery.
• Collaborating with the commissioner to identify the appropriate government departments to receive the recommendations of the citizens’ assembly and setting a timetable for their implementation.
• Ensuring the progress and effectiveness of the implementation of citizens' recommendations. Political diversity will allow the oversight group to be critical of the rejection, delay, or modification of the original recommendations.

In this chapter, we will introduce tasks one and two leaving the rest for later chapters.

3.1.1. Permanent Citizens’ Council

As described in the previous chapter, it is advisable to set up a Citizens' Council, which could take over, validate, and make final decisions on some of the tasks of the oversight group[30].

The citizenmembersofthecouncilwillexperienceapartialrotationineachcycle[31], withone- third of the participants rotating out, while the members of the oversight group will rotate withtheelectoralcycle.Themostimportanttaskforthiscouncilwillbetodecidetheagenda, selecting the more pressing policy dilemma.

Figure 13 - Cycles of the Permanent Citizens Council

3.2. Selecting the Project Team

Who will lead and coordinate the deliberation? This is a very sensitive issue as there are many small coordination decisions that will affect the process. Therefore, the oversight group should validate[32] the selection of an experienced and independent organisation that inspires confidence in its ability to run a fair and balanced deliberative process[33] that meets international quality standards. As these processes are lengthy and time-consuming, it is important that this organisation or experts are adequately resourced and have sufficient time to devote to the project, which will last at least one year.

The project team should also include in-house experts representing the commissioner. They should be officials who understand this kind of process, who have sufficient authority within the government and who are also trusted by the opposition parties because of their professionalism; there may be an area of open government or citizen participation and that is where this profile could be found.

Figure 14 - Composition of the Project Team

The constitution of this group is a meta-decision, i.e. it has to be taken as soon as possible since the incorporation of this group is crucial for the very facilitation of the oversight group itself[34]. Once the project team is active, it should start organising the tender process for the operator(s), starting with the communications team. The operator(s) should also provide the citizen lottery service, logistics and facilitation. For reasons of simplicity, it is recommended to minimise the number of tenders and to encourage competitors to form consortia. Once the operator(s) have been selected, a member of their team could be included in the project team.

3.2.1. Selecting the Communications Team

The communications team will be a strategic member of the project team throughout the process and will require a specialised organisation, so it is usually recommended that they be included in a separate tender. They will ensure the transparency of the process by communicating each step of each phase to the public. They will also be responsible for a strategy to extend the deliberation to the general public. The aim of communication will therefore be not only to increase the visibility and impact of the whole assembly process, but also to translate the reflections, conflicts and agreements that take place in both the preparatory and assembly meetings.

The communications team needs a senior strategic public communications lead, supported by a team of multimodal documenters and creatives. A specific procurement process may be required.

3.2.2. Selecting Other Operators

Operators require a specific procurement. Depending on the ecosystem of providers in the region, different work packages could be split or merged to ensure that a minimum number of providers or consortia can present their proposals. If the project team has sufficient expertise or can recruit another organisation of advisory experts, it could activate the project without the operators. If the operator or one of the operators is needed in the project team from the beginning, the activation of the project could be delayed until the main operator is contracted.

3.3. Selecting the Evaluator

The evaluator should be an organisation or individual with experience in deliberative processes[35]. Ultimately, it will depend on the oversight group, which will also be responsible for its selection. This is crucial because the evaluator, like the project team, should be independent of both the commissioner and the project team. It should have access to all working meetings from the beginning of the process, including meetings of the oversight group itself. This is also a metadecision, like that of the project team, and should be taken as soon as possible.

The ultimate goal of the evaluation is to improve deliberative assemblies, so the evaluator's work should be proactive, not just critical. He/she should work together with the project team. If they have doubts about any aspect of the design, they should discuss it with the project team to try to avoid methodological errors. If problems still occur, the evaluator must interpret why they occur and suggest how to avoid them. The evaluator may disagree with the project team, in which case the data and arguments should be presented and the project team should be given the opportunity to attach a note prior to publication of the final report. If the problem is critical and relates to a failure of standard good practice or an imbalance in the process, and the project team does not correct the problem, the evaluator may report to the oversight group, which may intervene. The project team should thus have access to the draft evaluation, which it may discuss or annotate.

4.  Agenda Setting  

This chapter introduces the agenda-setting process. To understand this key task, we explore the concept of policy dilemmas and their role in citizens' assemblies. The complexity of dilemmas is explored using the example of climate change. Understanding the importance of aligning these dilemmas with the policy agenda, it discusses how to select dilemmas in such a way as to ensure that the dilemmas selected are impactful and relevant.

4.1. Policy Dilemmas

The population over which the commissioner has jurisdiction faces different challenges; there are different views on which of these challenges are most relevant. A variety of political parties, stakeholders, and experts will prioritise these challenges differently. Many citizens also have something to say about these issues. How does the Citizens’ Council, helped by the oversight group and the project team, frame the process by choosing the right policy dilemma for the Climate Assembly? But first, what exactly do we mean by a policy dilemma?

The democratic framework, and to a greater extent the framework of deliberative democracy, assumes citizens have the capacity to participate in political decisions. The usual critique of democracy from an elitist perspective argues that the issues of governance are too complex for the people to make good decisions (Jacquet, Niessen & Reuchamps, 2022, p. 297). While we will not enter into this debate in this methodological guide, we do acknowledge that political decisions usually require a certain degree of specialised or technical knowledge (Leino et al., 2022, p. 429). However, given their role in regulating the function of society, it is worth considering that such decisions inherently incorporate a subjective element that is connected to people’s values and interests.

Figure 15 - Political Decision

The conventional method of framing political issues often sets the stage for either an ‘epistemic trap’ or the creation of generic ‘wish lists’. For example, if we start by saying "How do you solve X?" we presume that X must be solved and the challenge can be quite technical. Questions such as "How do we reduce food waste?" or "How do we reduce greenhouse emissions?" make it probable that citizens will agree on the need for a solution. However, the lack of a clear understanding of how to address these issues, or the awareness of obvious and simplistic solutions, leads Citizens' Assemblies to become venues primarily focused on listening to experts. In these settings, the emphasis is on understanding and prioritising expert proposals, without a comprehensive grasp of the value-based trade-offs involved. In the end, the government receives a list of good ideas that it already knows and that are less specific and coherent than they would have been if they had been drafted by a group of experts. Many of the Citizens' Assemblies have tackled an issue or problem in this way[36].

Our aim is to design a process with a values-based mandate rather than a technical one[37]. Such a mandate will reduce reliance on specialised knowledge and increase reliance on value- related knowledge, which is already part of everyone's cognitive backpack. It will increase the capacity of citizens to work, the added value they can bring and the impact of their contribution. A policy dilemma is a value-based problem for which there is no clear or simple answer. But how can we identify such a policy dilemma in practice?

There is a useful approach that answers this question: What aspects of the policy issue do experts representing different stakeholders disagree over? In answering this question we must avoid technical approaches; it is not a question of deciding what kind of technique will solve something, but of identifying what value issue the experts disagree on[39]. For example, there are doctors who believe that euthanasia should not be used and others who believe that it should be used in certain cases. These preferences are not related to their speciality or level of expertise, but to their own values and beliefs[40].

Good examples of policy dilemmas are: "Should euthanasia be allowed, and to what extent?" or "To what extent should we allow abortion?” The challenge here for a Citizens' Assembly faced with such a task will be to strike a balance in a situation involving trade-offs based largely on their values. While some technical information may be needed to support the deliberation, the core of the exercise is now value-based. The recommendations will be a short, balanced agreement on how to proceed, something that neither the experts nor the different political parties or interest groups can legitimately or practically come up with[41]. Citizens are involved in catalysing social agreement.

We started with these examples, even though they are not related to climate change, because they are easy to understand. The problem is that it is not always so clear what the policy dilemma is on any given issue. Identifying a dilemma often resists easy analysis and may remain invisible, especially in contexts of uncertainty and fragmented information (Janin, Acloque & El Nour, 2023, p. 337). In these cases, it is necessary to work together with experts in the field to understand where the dilemma lies.

4.1.1. Policy Dilemmas in the Context of Climate Change

When we start to break down the challenge of climate change, we realise that it touches almost every sector of our economy, politics, and society as a whole. Tackling climate change means transforming our societies across the board. In terms of mitigation alone, we have issues such as: energy, transport of people and goods, the food cycle, construction and housing, and many more. If we add adaptation, we have: Drought, floods, high temperatures, migration, ecosystem adaptation, primary sector adaptation, etc. With such a variety of issues, it is conceivable that each may have its own particular dilemmas, such as abortion or euthanasia, which are different in nature.

However, there is an overarching pattern that allows us to interconnect a multitude of these issues.

To understand this pattern, let us consider an example. Anyone who cares about the environment faces a dilemma every time they go shopping. Many people know that eating less meat reduces environmental impact and emissions, as do using organic and locally produced items. They find themselves in a situation that demands a double effort: On the one hand, they have to change their habits, such as eating less meat, and on the other, they have to spend more because organic or locally produced items tend to be more expensive. Some possible questions to unfold the dilemma could be: How much effort do I need to make for the common good? Is it fair for me to make this kind of effort, given the effort others make? But it becomes even more complex when one considers that the common good is not so tangible in the short term, and for many people will not even be tangible in their lifetime.

In this way, we can divide the dilemma into two parts. It embodies the classic dichotomy of individual versus collective welfare, while also encapsulating the conflict between present and future well-being. The latter aspect can be expressed more precisely as the tension between the welfare of my generation and that of future generations. To add complexity, these dilemmas go to the edge of our moral intuition. The collective is no longer limited to a specific social group, region, or nation, but has expanded to the interconnected globality. The future is not only our children and grandchildren, but future generations – the future of humanity.

These fundamental dilemmas give rise to a number of practical dilemmas that many future Climate Assemblies will need to address. There are also some more specific ones. Let us take some additional practical examples. The installation of renewable energy requires land use, which is related to subjective values such as the appreciation of the current rural and urban landscape or the clash with other sectors of the economy (especially agriculture). Thus, a dilemma could be ‘renewable energy production vs. impact on existing land use’, and take the form of a question: How can renewable energies be deployed while maintaining a balance with the impact that their installations may have on the territory? As another example, agriculture may be responsible for a lot of emissions, but it also feeds people. To what extent are people willing to change their habits and the economies of food production in order to reduce emissions[42]? Another example: How can we insulate homes on a large scale to reduce carbon emissions from heating, when not everyone can afford the cost of renovation in the short and long term, especially low-income households?

4.2. Fitting into the Political Agenda

Coming back to the questions posed in the introduction: How does the Citizens’ Council, helped by the oversight group and the project team, frame the process by choosing the right policy dilemma for the Climate Assembly? Let us now try to focus on the political problem this poses in the context of liberal democracies.

Placing the dilemma before the assembly means setting the agenda. Ultimately, setting the political agenda involves establishing the hierarchy of decisions. Prioritising issue A over B may lead to the activation of legislative processes for A, and perhaps later for B, or even result in B being forgotten. This is therefore a very sensitive phase of the process.

The political agenda is usually set by the governing party, often influenced by interest groups or public opinion. In some cases a parliamentary institution might set the agenda for a minority or coalition government. Although very rare, there are also direct and participatory democracy processes to set the agenda, for example, a citizens' initiative, which allows civil society or interest groups to collect signatures and trigger a referendum or, more relevant to our case study, a deliberative citizens’ assembly to work on agenda-setting[43] which in this methodology inspires the idea behind the Citizens' Council.

It is also important to consider that within a government there may be different ministries with different agendas to push forward, and as usual, governments have limited time, resources, and political credit to make reforms. Not all ministries may be able to take major decisions; this is often carefully coordinated within the executive so as not to lose public support or to gain it, as in a game of chess.

The definition of the policy dilemma of an assembly should have an impact and fit into the overall political agenda. Otherwise it means that the political community does not expect any real decisions from the process and we may end up with fake deliberation or participatory whitewashing.

4.3. Choosing the Right Dilemma

While the final decision rests with the Citizens' Council, different political actors, both outside and inside the government, should be involved in shaping the agenda. As noted in the previous chapter, the oversight group will be helping with this task and should include a variety of political parties including the governing party, in order to bring a diversity of perspectives and broader political support to the process. This act of agenda generation is fundamental, as the political and institutional establishment delegates the political problem or dilemma to the assembly, accepting and giving it the legitimacy to deal with it.

The project team must design an appropriate dynamic to identify and prioritise dilemmas with the oversight group, encouraging informed and consensual final decisions in the Permanent Citizens’ Council. The process will consist of several sessions, sufficient to achieve a satisfactory and agreed result.

It could be the case that the dilemma has one or more defining oppositional relationships. For example: Meat consumption vs. emissions from meat production. Furthermore, the dilemma should be materialised in one or multiple questions. The answer to those questions will be the tasks of the assembly. Here are some example questions related to this scenario:

• To what extent should we reduce meat consumption given the need to reduce emissions from meat production?
• To what extent should individuals change their dietary habits to reduce the environmental impact of meat production? What help do they need from the institutions to make the effort to change?
• In order to reduce greenhouse gas emissions, to what extent should governments implement policies to regulate meat production and consumption?

4.4. Agenda-Based Feasibility Assessment

It is very important to take some aspects into account when considering the possible dilemmas to be deliberated and before deciding whether the assembly should take place. It is possible that the oversight group may advise that deliberation is not feasible due to the political situation, not fitting the political agenda or not finding an appropriate dilemma. Here are some key issues to consider:

4.4.1. Scenarios for Convening a Climate Assembly

•  There are no plans for major solutions to the dilemma.
•  There is no majority support for the solutions on the table.
• The commissioner does not want to bear the political cost of trying to propose a solution to the dilemma.
•  Possible solutions imply a long-term policy, which the opposition party could reverse when the government changes. The parties that alternate in power should be committed to the Citizens' Assembly process through the oversight group and therefore to its recommendations.
•  There is public confusion and misinformation about the dilemma.
• All related ministries as well as the presidency of the government are aligned and committed to the assembly through the oversight group.
• Parliament approves and supports the activation of the deliberative process.

4.4.2. Scenarios to Pause a Climate Assembly[44]

•  The institution has no competence to implement possible solutions.
•   The agenda related to the dilemma is already set. Either solutions are already in place or they are already planned. The government is clear about the solution and has the time and resources to implement it.
•  There is no time to manoeuvre in this legislature and the opposition parties are not committed to the process.
• Elections will be held in the following 6 months[45].
•  Solutions to the dilemma involve long-term policies and the alternating parties in power are not all committed to the assembly's outcome.
• Different government ministries related to the dilemma are not aligned and committed to the assembly.
•  The government presidency is not committed to the assembly.

Once a dilemma has been selected, the content group has to be created, this group, much more specialised in the topic, will still have room to suggest improvements in the framework and the specific question to be addressed by the Assembly.

5. Content Provision

As described in the previous chapter, the remit of the Citizens’ Assembly is defined by a question related to a political dilemma. The citizens' task throughout the process will be to answer this question in the form of a series of recommendations.

This deliberative process needs to be informed and all participants should receive the same basic knowledge about the dilemma and potential models or scenarios for moving forward. The content provision methodology is responsible for equipping participants with this knowledge. The content group[46] is the team that applies the methodology in this phase[47].

5.1. Selecting the Content Group

The information that is presented and how it is presented is one of the critical points of a deliberative process, as it can powerfully influence the opinions of the participating citizens. Deciding who will be responsible for selecting the content is therefore a very sensitive decision. The group must have sufficient legitimacy in the eyes of public opinion and also the backing of the different interest groups affected in the process[48]. The oversight group has the objective political diversity to tackle this task in a legitimate way. In coordination with the project team, it should make proposals to be validated and calibrated by the Citizens' Council. This will be a crucial task if there is no agreement in the oversight group.

In order to establish a balanced group of experts on the issue for content selection, the choice should not be based solely on their specific areas of expertise. It is crucial to recognise that all experts, indeed all sources (including witnesses and contributors), have a measure of bias. A highly beneficial approach is to involve experts who represent stakeholder perspectives, thereby exposing participants to a wide range of viewpoints. These could be science communicators, people from think tanks, or scientists with a strong link to practical implementation. To work toward this ideal, it is necessary to identify and map stakeholders associated with the dilemma. The primary divergent perspectives on resolving the dilemma ought to be addressed, along with an exploration of the different policy options for the government.

Our balanced governance approach allows this mapping of stakeholders and corresponding experts to be done with the help of the different political parties in the oversight group; different parties with different ideologies will target different stakeholders and different experts to represent them.

5.2. Composition of the Content Group

The content group should be composed of four distinct categories: In-house content experts, external content experts, experts on deliberation, and experts in communication. Each category has specific roles and responsibilities, as detailed below:

•  In-house technical experts on the subject: At least one from each ministry related to the dilemma. Their responsibilities include:

 Providing information on what the government has done, is doing, or will do in relation to the issue beyond what the Assembly recommends.

Explaining the powers and limitations of the government or area/department.

Detailing the policy scope or potential impact of the recommendations.

Offering additional information related to their expertise on the issue.

•  External experts on the topic: Approximately four to five experts covering the different stakeholder perspectives on the dilemma, including at least one or two non-specialist with a panoramic view of the issue. The external experts should:

Cover as wide a range of disciplines as possible. o Be responsible for generating content.

• Editorial coordinator with expertise in communication/education to the general public[49]: Selected by the project team and validated by the oversight group. He/she will be responsible for:

Executive editing, which involves translating the content proposed by experts into structured and balanced documents of the deliberative curriculum in an orderly manner and ensuring timely content production. o Curating the speaker programme with the experts. o Moderating panel discussions during the learning phase.

Ensuring the information is presented in a way that can be understood and digested by all citizen members in the time and formats available during the process[50].

Optionally, this role can be expanded to include a broader team including experts in communication, outreach, and educational inclusion.

• Members of the project team. One or two deliberative experts who will:

Review the work of the editorial coordinator with content experts to guarantee transparency and balance.

In case of conflicts, decisions within the group will be facilitated by deliberative experts, striving at all times to reach consensual decisions. If consensus cannot be reached, balanced proposals will be made in which different perspectives are presented.

5.2.1. How Does the Content Group Work[51]?

The editorial coordinator will be primarily responsible for the group’s work, receiving assistance from the deliberation expert when undertaking this work for the first time. The coordinator’s role involves interviewing experts, convening meetings, and drafting and producing the necessary documents for the deliberative curriculum. It is a very demanding job, requiring about two to three days per week, with an increased commitment in the months leading up to and during the Assembly. The content experts, in contrast, are expected to contribute an average of one or two hours per week during the whole process

The content provision methodology has two main phases. The first one begins at least two months, ideally three to four months before the start of the Assembly's deliberations and involves the creation of a deliberative curriculum. The second phase takes place during and between the deliberations of the Assembly, and we will call it ongoing content provision.

5.3. The Deliberative Curriculum

The deliberative curriculum is not a conventional curriculum for a school, but a somewhat special one in many ways. Firstly, it is aimed at an audience of different ages, educational levels, and abilities. Secondly, it should focus on tackling a political dilemma.

The main components of the curriculum are:

• Content white paper. What specific information needs to be included to fulfil the objectives?
• Tools and resources. What pedagogical tools are used for each information block? Which human and logistical resources are needed to implement the curriculum?

5.3.1. Creating the Content White Paper

In creating this initial document, the emphasis should not be on simplicity. Instead, the document should aspire to the standards of a white paper, prioritising robustness, coherence, and comprehensive referencing. This approach is vital to lay a solid foundation for an informed and substantive debate.

In order to establish a protocol for decision-making among the content experts, we consider it useful to establish some definitions of types of content:

•  Basic content: This is information that has been agreed by the content experts. Everyone agrees that it is factual and should be presented.
•   Balanced content: This is where there is no agreement among the content experts on what is factual or even whether it should be presented, so other differing views or additional information must be added to reach agreement.

The very first task of the content group will be to develop the basic structure of the deliberative curriculum. To do so, the following key aspects should be considered:

•  Understand the dilemma. What content is needed to understand the dilemma?
• Grasp the constraints. What are the constraints on government action? This includes considering past actions, planned initiatives, competences, the legislative calendar, and other potential constraints. What is the scope of the citizens’ recommendations? And what is the government's political compromise?
•  Identify potential scenarios or models to move forward. What are the main different solutions for the dilemma? What information is needed to understand these solutions? Are all main potential solutions from key stakeholders included?

 Determine solution compatibility. Why are these solutions incompatible? What different legal or other actions can the government take in either direction? If the solutions are compatible, deliberation loses most of its impact, since both solutions can be listed as an expert-generated product.

•   Evaluate the trade-offs. What are the trade-offs associated with each solution or between solutions? What information is needed to understand these trade-offs?

Based on this initial task, the content group must make a crucial decision: What is the minimum content required to ensure a viable deliberation? The content group should also decide how many scenarios can be explored, bearing in mind the need to consider the trade- offs involved. Prioritisation could be crucial at this stage. Removing blocks of information could involve rearranging the balance of content.

The editorial coordinator's primary responsibility is to thoroughly explore the diverse perspectives of the contributing experts. This includes a deep understanding of their background information, their individual arguments, and the nuances of their viewpoints. It is essential to discern and articulate the various lines of debate, as well as to identify any conflicts or incompatibilities present in their proposals.

To finalise the document, the editorial coordinator must ensure that it receives the explicit approval of all participating experts, confirming that it accurately represents their views and contributions. This consensus is crucial for the integrity and credibility of the document.

5.3.2.  Pedagogical Tools and Resources

Once the content group has the technical document, it will need to design how this content is exposed to citizens. To do this, different tools for dissemination and communication should be considered. The tools will feed the pedagogical component[52] of the facilitation methodology for the Climate Assembly. An essential part of this process is assessing and prioritising the resources available to achieve an acceptable mix for information provision.

We need to consider here the following questions: Who presents the information and how? What resources are required? This will entail generating specific materials such as texts, videos, infographics, and interactive installations. It will also involve disseminators or other knowledge holders. An essential part of this process is assessing and prioritising the resources available to achieve an acceptable mix for information provision.

The first standard material that each Assembly produces is a written information kit on the dilemma. This document should be based on the white paper, but it should be an informative version of it, presented in plain language to ensure that all the contents are accessible and easily understood by everyone, regardless of their educational background. It is important to identify the parts that are more complex to translate into plain text, such as graphics or complex concepts. In these cases, appropriate infographics will be designed and integrated into the kit. A digital online format of the kit would also allow the addition of short explanatory videos.

It should be noted that not all citizens will read the kit. It is therefore important to consider how this kit is conveyed orally to citizens during the learning phase. For this, a programme of speakers will be produced, which should emulate a school class rather than an academic conference. This programme may have presentations, which should be curated by experts in plain language, panel discussions, or interactive dynamics between the audience and the speakers. Speakers could be not only experts but also other types of knowledge holders, such as witnesses or testimonies from affected people or interest groups.

Interactive activities can help participants to integrate complex concepts or ideas. These could be practical group exercises, games or interactive museum-style installations.

5.4. Ongoing Content Provision

The content and deliberation experts should be present during and between the deliberations in order to carry out the ongoing content provision[53]. The aim of this stage is to reduce the risk of lack of knowledge and false or misleading information. These are some of the tasks associated with this phase:

•   Answering participants’ queries: The content group should respond to participants’ questions with multiple, balanced answers, either during the meetings or in between.
•  Monitoring meetings: Observe the assembly discussions to identify lack of knowledge and false or misleading information.
•  Analysing session outcomes: Analyse the results of each session, pinpointing knowledge gaps and inaccuracies. Additionally, the content group can help with other tasks in between sessions, such as clustering content created by citizens.
• Balancing information: Provide balanced information to counteract false or misleading data.
•   Supplying additional resources: Provide citizens and the organisational team with extra knowledge or expert speakers as needed between sessions, either proactively or in response to citizens’ requests

6. Democratic Lottery and Inclusion

One of the key features of deliberative processes is that participants are selected through a democratic or civic lottery. This means that the participants do not come through an open call but are invited through a lottery. The basic aim is to create a descriptive sample of the population based on certain socio-demographic criteria. In this chapter, we will explain how the democratic lottery works and what the key features should be when dealing with a defined policy dilemma.

6.1. The Method

The process consists of two phases. In the first phase, a lottery is held to select those who are to receive an invitation to register their interest, ideally giving every adult an equal chance of being invited. In the second, a stratified random draw is carried out to compensate for voluntariness biases. This is explained below.

6.1.1. The First Draw

The first draw should be based on the democratic principle of equal selection: Every adult should have an equal chance of being selected to receive an invitation. This principle is key, as it is one of the main drivers of the legitimacy of the whole process and has been used since ancient times[54].

In order to comply with this principle of equality, an initial draw must be made from a database containing all the inhabitants. If such a database is available, a draw will be made on it to determine the initial number of citizens to be contacted (I) using this formula:

I = 200* N

Where N is the number of assembly members or size of the assembly. This formula is based on two fundamental facts based on experience in other processes:

•  A conservative response rate to these invitations is approximately 4%[55]. In practice it can vary widely from 2 to 12%.
• The required number of citizen pool to perform quality stratification is P = 8 * N.

Thus, if we equalise the two conditions: 8 * N = 4% * I. And clearing it, we obtain the formula mentioned above: I = 200 * N.

To illustrate this, imagine an organisation aiming to form an Assembly with 50 members. Using the provided formula, the organisation would need to contact 10,000 citizens initially. Given the expected 4% response rate, it is anticipated that approximately 400 individuals will respond positively. This number aligns with the requirement to have a pool that is 8 times the size of the desired Assembly for quality stratification. Additionally, to account for potential vacancies before the Assembly commences, the organisation would also select an extra 10% as replacements[56], which equates to 5 additional members. Thus, while the target size of the assembly is 50, a total of 55 individuals (50 members plus 5 replacements) would be finalised from the pool.

It is important to note that if we do not have a database to sort the 200*N guests, we should use an approximation. A commonly used approximation is a database of postal addresses, as almost all residents, except the homeless, have an address. A less reliable method, but a good approximation in the absence of an integrated database of residential postal addresses, is street selection at points drawn by lot according to a criterion proportional to population density[57]. Another method may be the generation of random telephone numbers[58].

6.1.2. Stratification Criteria

Once we have achieved an adequate pool of citizens, who have volunteered and accepted the assembly’s participation conditions, we have to carry out a second draw, this time stratified. Why is stratification necessary? While the first lottery guarantees diversity and independence, the usual 4% volunteer rate from the invited citizens introduces potential biases that we must try to compensate for. The stratification criteria aim to counterbalance these biases[59]. These are the most commonly used criteria[60]:

• Gender: There may be bias, and more men may participate; although in some countries this is no longer the case, in others, men may be more interested in political processes and more willing to get involved.
• Age: In general, participation peaks in the middle age 40-50 years and decreases in older and younger people.
• Place of residence: Certain rural or deprived areas may have lower levels of participation.
• Socio-economic status: Determined by factors like education, occupation, income, household, and neighbourhood resources (Rodríguez-Hernández, Cascallar, & Kyndt, 2020). Those from lower socio-economic levels tend to participate less. Given the sensitivity around income and its correlation with education (Mou, 2023, p. 164), educational level is often asked to infer socio-economic status.

 The political Criterion Related to the Dilemma: The Attitudinal Criteria

A very strong bias is often related to political sensitivity towards the subject of the assembly. This criterion is key to our dilemma-centred methodology. If the assembly sample is biassed in relation to the dilemma, the process can be severely delegitimised.

In some countries, such as the US, it is customary to ask about voting recall. This may be sufficient given that political ideologies often align with the polarisation in most dilemmas[61]. However, this requires asking a very sensitive question that many people may be suspicious of, especially if they distrust the convening government. In this methodology we propose better to ask an attitudinal question related to the dilemma. The attitudinal question is not as strongly perceived as an interrogation about your ideology as the vote recall or other more direct questions about ideology, but it is indirectly related.

The attitudinal question has to mirror a question that has been recently asked in a survey on the same population scale as the assembly sample[62]. This question should be roughly connected to the dilemma. A common question in the context of climate change is often:

How concerned are you about climate change? A lot, some, a little, not at all.

We could also come up with more specific questions that would be better suited to the dilemma itself. The attitudinal criteria should be proposed by the project team and validated by the oversight group.

6.1.3. The Second Draw

The stratification has to reflect the reality of the population, so we must have the data of the chosen criteria in the population that the assembly intends to represent. From our pool of volunteer citizens, we have to make the stratification that best approximates the population data. In practice, there is no one-size-fits-all approach. There are always several ways to get approximate samples from the same pool of citizens. How do we justify the choice of selecting one sample over another?

First, we must sort the different possible samples according to their deviation from the overall population, giving priority to those with the least deviation. Then, we will make a second draw from among these samples with a minimum similar deviation. This second lottery establishes a fair criterion of choosing citizens who best represent the broader population. There are now computer algorithms capable of doing this process in mere seconds (Flanigan et al., 2021).

6.2. Inclusion Strategies

In democratic innovations such as Citizens’ Assemblies, the dynamics of participation differ significantly from the straightforward selection of lottery balls. Whereas each lottery ball has an impartial and equal chance of being selected, human participants exhibit a distinct pattern due to the voluntary nature of their involvement, leading to what is termed self-selection bias (Harris, 2019, p. 49). This bias is very strong: Around 96% of all invitations are not accepted. Although we have already seen that we can try to correct this by stratifying, there are other strategies to improve the acceptance of the invitation and minimise the effects of self- selection bias. These strategies are essential because there may always be hidden biases that cannot be corrected by the limited criteria of stratification.

Inclusion strategies try to maximise acceptance of the invitation. Here are some of the most common ones:

•  Economic incentive: Many people who have no interest in the political process of participation may have an interest in earning money, especially the poorer social strata. This payment per session is recommended to be up to €100 per full day of work.
•  All covered: Participation should not entail additional costs for citizens. Travel, meals, and accommodation should be covered and managed by agencies offering a personalised service.
•  Distinguished treatment: Citizens should feel relevant and be treated like any other policy maker. Venues must be attractive, catering must be selected, and accommodation must be up to standard. The invitation must be signed by the highest government official.
•  Letters certifying their participation: It may be vital for some citizens to request to miss a day of work or to leave earlier or arrive later to have an official letter[63].
•  Help for carers: Many carers of young children, elderly, or sick people find it difficult to leave their homes for several days. Allocating additional financial support for caregivers or creating spaces in assemblies to include caregivers is essential to facilitate their participation.
• Inclusive language: Making the invitation easy to understand for anyone who is not used to reading complex texts and for people who understand different languages is essential. Special attention should be paid to this not only in the invitation but throughout the whole process.
•  Plural and committed political space: Citizens must not feel that the participatory space is biassed. It is important to reflect at all times that the process incorporates politicians or experts from different political orientations. Alongside this pluralistic approach, a strong commitment from public decision-makers to the initiative has been a key factor in cases with high response rates and low dropout rates (OECD, 2020, p. 93).
• Tacit knowledge: It is important to communicate that they "do not need" to know about a particular subject. Their life experience, whatever it may be, is what is required of them.
• Special policy for people dealing with homelessness or extreme poverty[64]: Collaborating with specialised NGOs, aids in reaching out to these populations. A minimum number of participants with the same characteristics should be selected to help them build a collective voice. Given the numerous challenges they have already faced, it is advisable to provide them with special accompaniment both before the whole process and before each meeting. Such accompaniment serves to facilitate their participation and prevent feelings of added failure.

6.3. Who Applies the Methodology?

It is common for the selection of participants through a democratic lottery to be delegated to an independent organisation outside the government. This ensures that no one in government has access to participants' data. If this is not the case, it is strongly recommended that the technical government staff having access to the data should not be political appointees.

7 Facilitation and Inclusion Criteria [65]

In general terms, facilitation is about helping a group of people engage in a process to move them towards an agreement or solution.

 Deliberative facilitation stands apart from other forms of facilitation due to specific challenges (White, Hunter & Greaves, 2022, p.93). These include the high costs associated with public deliberation, the constraints of limited time, and the paramount importance of maintaining the process’s integrity and legitimacy. Additionally, it involves managing a diverse group of participants, each with their own perspectives. These participants must comprehend the policy dilemma at hand to form informed judgements and think critically. The ultimate goal is for them to engage in the deliberations and cooperatively achieve an outcome[66].

 Facilitation is usually carried out by an external team of facilitators. The project team recruits a professional team of facilitators with proven experience in the sector. It is also common for the operator who is part of the core project team to be part of an organisation with the role of both facilitating the event and coordinating the process.

7.1 Times or Stages of the Facilitation Process

When thinking of facilitation in a deliberative process, one might picture a group of citizens gathered around a table, with a guiding figure, the facilitator, ensuring symmetry and respect in contributions. Yet, the significance of the facilitation team extends beyond just the deliberation phase.

 The involvement of the facilitation team typically begins with contributing to the design of the assembly process[67]. Weeks before the citizens’ deliberation, facilitators engage with the members of the project team. During these initial discussions, they outline the structure of the sessions and discuss the content for each. The aim is to ensure a balanced approach during the citizens deliberation, encompassing information intake, fostering trust, promoting reflexivity, and ultimately coming up with clear recommendations that are validated by the citizens as a whole.

 In the post-deliberation phase, facilitators remain actively involved through the follow-up commissions and the accompaniment of the citizens afterwards. They also gather feedback  on the deliberation process, providing the project team and oversight group with insights into the participants' experiences. This enables a reflective approach to the process, ensuring continuous improvement for future deliberative processes (White, Hunter & Greaves, 2022, p. 13).

 We could therefore say that facilitation is present before, during, and after an assembly, and its work is essential for the proper development of the deliberative processes.

7.2 Facilitation Approach: From Less to More ‘Intervention’

Two philosophies of facilitation have been used in Climate Assemblies: Directive table facilitation and citizen self-organisation. The choice between these styles sparks ongoing debate among practitioners and academics (Smith, 2022, p. 8-9).

 Directive table facilitation, favoured by assemblies in Ireland, Scotland, and the UK, involves small tables of members. These groups are guided by a facilitator who is responsible for task adherence and upholding fairness in proceedings, such as balancing speaking time among participants. To foster a diverse range of perspectives and discussions, members rotate between tables regularly.

 Conversely, Denmark and France emphasised citizen self-organisation. The groups were generally left to develop their own working patterns and share responsibilities for developing different recommendations. Facilitators intervened only when group dynamics became challenging. The Luxembourg’s Assembly blended both approaches. Starting with table facilitation, they later transitioned to self-organising groups. During this second phase, designated 'spokespersons' led the groups. Chosen from the members, these spokespersons guided discussions and ensured cohesive recommendations, striking a balance between structured guidance and member autonomy.

 Both facilitation styles are valuable and can be used separately or combined, depending on the nature of the group and the desired objectives. While self-organisation can provide a sense of freedom, there are instances where directive intervention is needed to keep the conversation focused on the task at hand or to adhere to the schedule. In our guidelines, given the precise framing of political dilemmas, we will emphasise a more directive style of facilitation.

7.3 The Goal of a ‘Good Facilitator’  

  Facilitation is recognized as a living and creative work, requiring a flexible and adaptable approach to guiding group discussions. There is no single, definitive method to navigate these intricacies; rather, a spectrum of criteria stands out as essential. These criteria, which will be detailed in the Decalogue, underscore the foundational qualities necessary for effective facilitation, ensuring that the process is not only efficient but also responsive to the evolving dynamics of deliberation.

7.3.1 Decalogue for a Good Facilitator 

A good facilitator has to combine a number of different qualities and virtues (White, Hunter & Greaves, 2022, p. 95): 

• Empathy: The capacity to understand the emotions and viewpoints of participants, fostering a supportive atmosphere. 
• Active listening: The skill to listen intently and engage comprehensively with the contributions of participants, ensuring every voice is heard. 
• Communication skills: The ability to communicate clearly and effectively, facilitating understanding and participation. 
• Humour and presence: The talent for using appropriate humour and maintaining a genuine demeanour to enhance group engagement. 
• Equitable participation: A commitment to a balanced and inclusive environment, promoting equal participation across gender minorities, and other group dynamics. 
• Neutral stance: The discipline to guide without personal influence, allowing the group's discourse and decisions to unfold without facilitator bias.
• Focus: The capacity to keep the group's attention anchored to the task at hand. 
• Time management: Effective allocation and utilisation of time, ensuring that discussions are productive and timely.
• Adaptability and flexibility: The ability to manage setbacks and conflicts with a flexible approach, adjusting facilitation strategies as necessary. 
• Subtle guidance: The skill of steering conversations with a light touch, ensuring that participants remain central to the deliberative process.

7.4 In the Room: Practical Issues

 The preceding chapters of this guide have delineated the extensive preparatory work to be done before convening the first day of a Citizens' Assembly. Facilitation teams, involved in these early stages, will find their main asset within the Assembly to be ‘taking action’.

7.4.1 Before Starting 

At the beginning of the assembly, the facilitator’s role can transform initial nervousness and confusion into a constructive starting point for the subsequent meetings. A facilitator’s opening remarks are more than a mere welcome; they are an invitation to each member to share a part of their story, fostering a sense of belonging and importance. It is essential for participants to feel that they are embarking on a significant journey, one where their collective contributions are valued and pivotal in shaping the future of public policies. 

Effective communication is crucial for the facilitator, ensuring participants understand the essence of their role and the reasons for their selection. This skill, highlighted in the decalogue of virtues, is also key when elucidating the structure of the Assembly, the number of sessions, their distribution, and the overarching purpose of each session.

7.4.2 Explaining Rules of the ‘Game’ 

In the continuum of challenges that facilitators navigate, articulating ‘the rules of the game’ is a critical task. Assembly members must share a common understanding of these rules, which may be reiterated at the start of each session for clarity and consistency. Although these rules may adapt to the session’s topic, the group dynamics, or the particular nature of the discussion, the following principles stand as the pillars for creating a safe and effective space for all participants[68]:

• Mandatory respect: Diversity of opinions and expressions must be respected. Each viewpoint holds equal value within the assembly's discussions. 
• Encouraged authenticity: Authentic self-representation and the expression of unique perspectives are encouraged, contributing to the richness of the dialogue.
• Required constructiveness: Building upon the ideas of others is required to foster a collaborative atmosphere. The collective wisdom is enhanced through such constructive engagement.
• Ordered speaking: It is obligatory for participants to speak in turn, ensuring orderly discourse and that all voices are heard without interruption. 
• Advancement of ideas: Participants are expected to avoid repetition and aim to contribute new insights, thereby advancing the discussion. 
• Inclusive participation: Inclusivity in engagement is essential. Equitable consideration of all perspectives must be ensured.
•  Emotional openness: Respectful sharing of emotions is permitted and valued, recognizing their role in the collective decision-making process.  

7.4.3 Space Disposition

 The configuration of the Assembly plays a silent yet critical role in the success of a deliberative process. A dynamic environment, characterised by movable chairs and tables, is advocated to facilitate fluid exchange of ideas and perspectives. During the Assembly, the format will alternate between plenary sessions for collective updates and smaller groups for in-depth exploration of policy dilemmas. The plenary sessions serve as a platform for disseminating information to all participants, while the breakout groups are designed to stimulate participation, foster dialogue, and begin the consensus building process.  

7.5 Work Phases

 7.5.1 Learning Phase

 The first learning phase is where citizens delve into relevant content, preparing them to tackle the policy dilemma and understand the trade-offs involved. Facilitators ensure this content is communicated clearly and understood by all Assembly members.

 The facilitation team presents the material in formats accessible and engaging for everyone. This approach guarantees that the language and presentation style are inclusive, accommodating the diverse backgrounds within the Assembly.

 Expert presentations are an integral part of this phase. To keep engagement high and maintain focus, these presentations should be brief, ideally not exceeding 10 to 15 minutes. Following each segment, a time for questions and interactive discussion allows citizens to engage directly with the experts, clarifying doubts and deepening their understanding. 

Facilitators play a dynamic role in the learning phase, actively gathering questions and concerns from participants and addressing them systematically. This is achieved through expert inputs or realtime clarifications, ensuring that Assembly members gain a comprehensive understanding of the subject matter. Concurrently, facilitators assist citizens in processing and organising the content presented. By employing interactive methods such as worksheets or group discussions, they help participants structure their arguments and ideas around the dilemma. This approach not only clarifies the content but also sets the stage for in-depth discussions and debates in the subsequent phases of the assembly.

7.5.2 Deliberation Phase 

In this pivotal second phase, we engage in the deliberation process, where various solutions, specifically tailored to address the policy dilemma at hand, are juxtaposed, and their trade offs are thoroughly discussed. A significant challenge lies in establishing overarching principles that can serve as a common foundation for the Assembly's members. These principles, envisioned as a set of broadly accepted values, are instrumental in evaluating action proposals and underpinning public policy applications in the relevant field. 

Examples of such principles might include advocating for fair and local trade, minimising food waste, and reducing reliance on imports. These principles should not only resonate with the Assembly's objectives but also reflect a consensus among its members. 

During the deliberations, it is imperative for facilitators to steer the discussion towards principles that have garnered widespread agreement within the Assembly. Progressing on principles that face substantial opposition is counterproductive, as they are unlikely to be incorporated into the final recommendations.

By aligning these facilitation techniques with the Assembly's goals, we can ensure a more effective and inclusive deliberation phase, directly addressing the policy dilemma with well considered scenarios.

7.5.3 Drafting Recommendations 

The drafting of the recommendations by the citizens represents the culmination of a process that has evolved throughout all previous meetings. In these sessions, the necessary foundations have been established to facilitate the decision phase in a fluid and consensual manner.

Assembly members follow a step-by-step process to compile a report on their proposed recommendations. They may opt for private deliberations, akin to a legal jury, to decide their verdict without the presence of facilitators, ensuring an autonomous decision-making process.[69] The resulting report will include key recommendations and the degree of support for each, along with comprehensive descriptions of the points raised during the Assembly. A recommendation is generally recognized as having broad support if it garners a consensus of more than 80%.  

7.6 Focus on Inclusivity

 Ideally, deliberative processes should bring together people with very different profiles (in terms of age, gender, social, cultural, and economic background, etc.) and ideologies. This is the great richness of a deliberative group, and diversity is what ensures a good representation of all points of view, but it is also one of the most delicate points for a facilitator. Creating an atmosphere of respect and cordiality within the group is one of the most important and complex tasks. Depending on the topic of the meeting, the climate of the sessions can be more or less ‘passionate’, and maintaining a good balance between passion and mutual respect is fundamental to reaching the consensus that will follow. Facilitators need to ensure that all group members are supported and have equal opportunities to participate; this involves taking into account human behaviour and group dynamics, including displays of power and status among participants.  

7.6.1 Facilitation with a Gender Perspective

 Facilitation from a gender perspective needs to take into account a number of aspects in order to be successful; on this occasion, we will focus on the "during" of the Assembly. 

If we can get a sufficiently diverse group of participants and facilitate that group from a gender – sensitive perspective, this will have an impact on the final outcomes, generating proposals that take gender issues into account, and ultimately also recommendations that will lead to public policies that take gender into account, creating a virtuous circle.  

• Equal opportunity to participate. In general, women tend to speak less often than men, even more so in mixed groups. This is even more true when other factors such as race, social class, educational level or age are taken into account. The consequence is more cautious contributions and more frequent apologies. Facilitation strategies such as predefining word order or control the contribution time can be used. 
• Underestimating women's participation. The rest of the participants may underestimate the contribution of women, especially if the topic is not in line with the role traditionally assigned to them. Distraction during women's interventions is also a form of underestimation. Pay particular attention to manterrupting - unnecessary interruptions of a woman's speech by a man, and bropropriating - appropriation of women's opinions and contributions by men. Facilitators should point out this behaviour when it occurs.
• Use of language. Language is highly gendered and needs to be addressed as a fundamental issue in the process, both in the facilitation in the rooms and in any written material shared with the participants. In Spanish, the language is masculine, so special attention should be paid to using neutral formulas (persona), doubling the gender o/a (todos y todas) and trying to use collective terms (ciudadanía) whenever possible. The same applies to written material, where stereotypes should be avoided (e.g. the female housewife or the male executive). 
• The map of actors involved. The Assembly must be an example and a reflection of what we want to communicate, so we must strive for a balance between men and women speakers, both among the experts and among the testimonies that may come to share their experiences. Particular care must also be taken to avoid stereotypes (for example, the male farmer and the female flight attendant). 
• Different issues of interest. The vision and interests of women may be different from those of men, and this only enriches the discourse and the proposals made to the Assembly, and we must not lose sight of this issue in order to take this sensitivity into account. Dynamic suggested: group forms of artistic expression, such as composing songs or making drawings as a group. 
• Conceptual vs. practical. Women have traditionally been excluded from public debate and relegated to more practical issues, which are generally relegated to the private sphere. Our society also favours technical and rational discourse over emotional or subjective discourse. Although this is beginning to change, this should be taken into account in the Assembly, as it is a public space that is guided by a more rational/technical discourse. Dynamic suggested: starting the contributions with the expression of feelings and experiences related to the use of the services, benefits and rules that are the subject of the dialogue. 
• Balancing leadership. Men are more likely to speak, volunteer or be proposed as spokespersons on traditionally male-dominated issues. In deliberative processes, it's important to ensure that the conclusions reflect everyone's input. Strategies that   might work include selecting the person to represent each sub-group in advance, or facilitating creative and collaborative dialogue rather than competitive debate.

7.6.2 Intersectionality 

We don't want to forget the concept of intersectionality, defined as the interconnected nature of social categorizations such as race, class, and gender as they apply to a given individual or group, regarded as creating overlapping and interdependent systems of discrimination or disadvantage.

 Bearing this complexity will help us realise that each person's identity is the result of a combination of circumstances, and that citizen participation must therefore be able to encompass and correctly address this hyper-diversity.

8 Assembly Work Phases (Coming in Future Versions) 

This chapter is currently under development and will be included in future versions of the guide.

9 Communication and Extension of the Deliberations (Coming in Future Versions)  

This chapter is currently under development and will be included in future versions of the guide.

10 Follow-up and Impact (Coming in Future Versions)   

This chapter is currently under development and will be included in future versions of the guide.

11 Conclusion

This methodological guide, crafted on the foundational principles of balance and a focused shift from broad climate issues to specific policy dilemmas, stands as a significant methodological proposal in the realm of democratic engagement and climate action. It is designed to amplify the impact of citizens' deliberations on public policy and to introduce more actionable principles into the process of setting-up and facilitating climate assemblies.

The guide's development has been an exercise in exploring the potential benefits of these principles. The concept of balance, when applied to governance, content provision, and participant selection, is expected to create a more inclusive, representative, and comprehensive deliberative process. Similarly, our shift to concentrate on policy dilemmas, underpinned by a value-based approach, is believed to deepen the relevance and impact of the discussions. These strategies are hypothesized to transform the nature of citizens’ deliberations, ensuring that the discourse meaningfully engages with the core issues of climate-related policy. 

The forthcoming application of this methodology, particularly under Task 4.1 of the project, marks an important step in validating and refining our approach. This will involve the deployment of our framework in three specific assemblies: the Catalan Climate Change Assembly, guided by the Generalitat de Catalunya; the Riga Assembly, led by Green Liberty; and the Edermünde Climate Assembly, orchestrated by Ifok. Deliberativa's role, especially in the Catalan Assembly, will be crucial in supporting and observing these applications. These assemblies will serve as practical environments to test our hypotheses and refine the methodology based on real-world experiences and feedback. 

In addition, the ongoing development of further chapters, as outlined in our executive summary, continues to be a key focus. These chapters will delve into various aspects of Assembly work phases, facilitation methods, evaluation processes, and the impact assessment of our approach. Integral to these future tasks is the incorporation of findings from the D2.2 Report, which provides a crucial mapping of value-based problems in climate policy. This will be complemented by roundtables with 7-10 participants, serving as practical tests in climate assemblies to further validate and refine our methodologies. The insights gained from the D2.2 Report, the roundtables and the practical applications in the assemblies will inform and shape the evolution of these chapters, ensuring that our methodology is continually updated, responsive, and relevant to the needs of climate assembly facilitation. 

In essence, this guide is not merely a set of instructions; it represents a thoughtful and strategic approach to enhancing the efficacy of climate assemblies in influencing public policy. As we move forward with its application and refinement, our commitment is to ensure that the methodology evolves to meet the challenges of climate action and democratic participation effectively.

12 References 

​ATD Quart Monde (2023). La participation des plus pauvres à la Convention citoyenne sur la f in de vie [Last accessed on 4/10/1993]. 

Blijleven, W., van Hulst, M., & Hendriks, F. (2019). Public Servants in Innovative Democratic Governance. In Elstub, S. & Escobar, O. (Eds). The Handbook of Democratic Innovation and Governance (pp. 209-2024). Cheltenham: Edward Elgar Publishing. 

Blok, A. (2007). Experts on public trial: On democratizing expertise through a Danish consensus conference. Public Understanding of Science, 16(2), 163-182. 

Bouricius, T. G. (2013). Democracy through multi-body sortition: Athenian lessons for the modern day. Journal of Deliberative Democracy, 9(1). 

British Government (2020). How to run a citizens’ assembly: A handbook for local authorities based on the Innovation in Democracy Programme. [Last accessed on 18/09/2023]. 

Bussu, S. & Fleuß, D. (2023). Citizens’ Assemblies: Tod-Down or Bottom-Up? - Both, Please! In Reuchamps, M., Vrydagh, J., & Welp, Y. (Eds). De Gruyter Handbook of Citizens’ Assemblies (pp. 141-154). Berlin: De Gruyter. 

Carson, L. (2019). How to Ensure Deliberation Within a Sortition Chamber. In Gastil, J., & Wright, E. O. (Eds). Legislature By Lot (pp. 205-227). London: Verso.  

DemocracyNext (2023). Assembling an Assembly Guide. [Last accessed on 19/09/2023].   

Elstub, S. & Khoban, Z. (2023). Citizens’ Assemblies: A Critical Perspective. In Reuchamps, M., Vrydagh, J., & Welp, Y. (Eds). De Gruyter Handbook of Citizens’ Assemblies (pp. 113-125). Berlin: De Gruyter. 

European Commission (2022). European Citizens’ Panel on Food Waste. Information Kit. [Last accessed on 18/09/2023]

European Commission (2023). European Citizens’ Panel on Food Waste. Final Recommendations. [Last accessed on 18/09/2023]

Eymard, L. (2020). From the French Citizens' Convention on Climate to the Conference on the Future of Europe: A participatory science and democracy perspective. European Law Journal, 26(1-2), 136-140.

FIDE Policy Research (n.d.). Organising a Democratic Lottery. [Last accessed on 25/09/2023].

Fishkin, J. S. (2018). Democracy When the People Are Thinking: Revitalizing Our Politics through Public Deliberation. Oxford: Oxford University Press.

Flanigan, B., Gölz, P., Gupta, A., Hennig, B., & Procaccia, A. D. (2021). Fair Algorithms for Selecting Citizens’ Assemblies. Nature, 596, 548-552.

Global Assembly Team (2022). Report of the 2021 Global Assembly on the Climate and Ecological Crisis. [Last accessed on 25/09/2023].

Gül, V. (2019). Representation in Minipublics. Representation, 55(1), 31-45.

Habermas, J. (2006). Time of Transitions. Cambridge: Polity. 

Harris, C. (2019). Mini-Publics: Design Choices and Legitimacy. In Elstub, S. & Escobar, O. (Eds). The Handbook of Democratic Innovation and Governance (pp. 45-69). Cheltenham: Edward Elgar Publishing. 

Jacquet, V. (2017). Explaining Non-Participation in Deliberative Mini-Publics. European Journal of Political Research, 56(3), 640-659. 

Jacquet, V., Niessen, C., & Reuchamps, M. (2022). Sortition, its Advocates and its Critics: An Empirical Analysis of Citizens’ and MPs’ Support for Random Selection as a Democratic Reform Proposal. International Political Science Review, 43(2), 295-316.

Janin, P., Acloque, D., & El Nour, S. (2023). Agricultural and Food Dilemmas in Times of Crisis. Revue canadienne d'études du développement/Canadian Journal of Development Studies, 44(3), 335373. 

Jensen, C. B. (2005). Citizen projects and consensus-building at the Danish Board of Technology: On experiments in democracy. Acta Sociologica, 48(3), 221-235. 

Leino, M., Kulha, K., Setälä, M, & Ylisalo, J. (2022). Expert Hearings in Mini Publics: How Does the Field of Expertise Influence Deliberation and its Outcomes? Policy Sciences, 55, 429 450. 

MacKenzie, M. (2023). Representation and Citizens’ Assemblies. In Reuchamps, M., Vrydagh, J., & Welp, Y. (Eds). De Gruyter Handbook of Citizens’ Assemblies (pp. 21-34). Berlin: De Gruyter. 

Manin, B. (1997). The Principles of Representative Government. Cambridge: Cambridge University Press. 

Montañés, P., Soriano, M. E., & Manzano-García, G. (2023). The Social Perspective of Euthanasia in Spain: Variables that Predict Attitudes towards Euthanasia. Annals of Psychology, 39(2), 287-293.

Mou, W. (2023). A Quantitative Analysis of the Relationship between Education Level and Income. Journal of Education, Humanities and Social Sciences, 12, 160-166. 

newDemocracy Foundation (2018). Enabling National Initiatives to Take Democracy beyond Elections. Australia: NewDemocracy Foundation. 

Ney, S. (2009). Resolving Messy Policy Problems: Handling Conflict in Environment, Transport, Health and Ageing Policy. London: Earthscan. 

Niessen, C. (2023). How do elected officials perceive deliberative citizens’ assemblies? In Reuchamps, M., Vrydagh, J., & Welp, Y. (Eds). De Gruyter Handbook of Citizens’ Assemblies (pp. 325-336). Berlin: De Gruyter.

OECD (2020). Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave. Paris: OECD 

OECD (2021a). Eight Ways to Institutionalise Deliberative Democracy. Paris: OECD Public Governance Policy Papers. 

OECD (2021b). Evaluation Guidelines for Representative Deliberative Processes. Paris: OECD Publishing. 

Osborne, D., Huang, Y., Overall, N.C., Sutton, R.M., Petterson, A., Douglas, K.M., Davies, P.G. and Sibley, C.G (2022). Abortion attitudes: An overview of demographic and ideological differences. Political Psychology, 43, 29-76.

Rodríguez-Hernández, C.F., Cascallar, E. & Kyndt, E. (2020). Socio-economic status and academic performance in higher education: A systematic review. Educational Research Review, 29 (100305).

Scottish Government (2022). Scotland's Climate Assembly. Process, Impact and Assembly Member Experience: Research Report. [Last accessed on 03/08/2023]. 

Smith, E. K., & Mayer, A. (2019). Anomalous Anglophones? Contours of Free Market Ideology, Political Polarization, and Climate Change Attitudes in English-Speaking Countries, Western European and Post-Communist States. Climatic Change, 152(1), 17-34. 

Smith, G. (2022). Climate Assemblies - Key features. KNOCA report. [Last accessed on 02/11/2023] 

Turner-Zwinkels, F. M., & Brandt, M. J. (2023). Ideology Strength Versus Party Identity Strength: Ideology Strength Is the Key Predictor of Attitude Stability. Personality and Social Psychology Bulletin, 01461672231189015.

Vrydagh, J. & Caluwaerts, D. (2023): How do Mini-publics Affect Public Policy? Disentangling the Influences of a Mini-public on Public Policy Using the Sequential Impact Matrix Framework. Representation, 59(1), 117-136. 

White, K., Hunter, N., & Greaves, K. (2022). Facilitating deliberation, a practical guide. MosaicLab

13 Annex 

1.Other names have been used for similar formats of practical deliberative democracy. Citizens' juries have a longer history and many have been implemented at local level with around 24-35 participants. While the Citizen's Jury was developed in the USA and is very popular in English-speaking countries, similar formats have developed in parallel in other Western countries. In Germany they were called planning cells and in Denmark consensus conferences. Citizens' Assemblies are a similar format to citizens' juries but with a larger number of participants, usually between 60 and 200, with the aim of representing larger populations, such as the scale of a region, a country or even a group of states like the EU (called EU Citizens' Panels in this case). The name "Citizens' Assembly" is so popular that it is already used for formats regardless of the number of participants. For a depiction of different types, the number of participants, duration of the deliberations, outputs, and relevant examples, see: Smith & Setälä (2018). 

2.Scottish Government (2022).

3.For an overview of why elections tend to incentivize short-term measures, see: Guerrero (2014). For a more specific discussion of why electoral democracies fail in handling climate change, see: Willis, Curato & Smith (2022). 

4.Extinction Rebellion (2019). 

5.In what follows, we move from the term 'issue' to 'policy dilemma'. We use the term " issue " here as a way of connecting with the current terminology in use today.

6.Similar to the approach advocated in this guide, the Irish Citizens' Assembly on biodiversity loss also recognized the need to focus on specific, actionable areas due to the complexity of climate-related issues (Irish Citizens’ Assembly, 2023, p. 102) 

7.Throughout this document, the term value-based dilemma or policy dilemma is used interchangeably. 

8.The principles of integrity and purpose are inherent within the themes of transparency, accountability, and clear mandate as outlined by the Committee of Ministers of the Council of Europe (2023). 

9.For instance, in the basic standards for organising Citizens' Assemblies developed by Marcin Gerwin, it is stated that "the coordinators should be impartial, e.g. not active politicians or direct stakeholders" (2018, p. 83). Additionally, the guide by DemocracyNext (2023, p. 19) also emphasises the importance of independence in the governance structure Citizens' Assemblies. 

10. Some successful models, as observed in Ontario and Iceland, have employed a multi-party committee or multi party compromises to prevent immediate politicisation and ensure the assembly’s success (Niessen, 2023, p.332)

11. For a further development of this idea, see the criticism raised by Oleart (2023) regarding the pretended neutrality in the case of the Conference on the Future of Europe. 

12. This group of citizens will be part of the Citizens’ Council. More information in the following chapter.  

13. In this context, we refer to an "epistemic trap" as a situation in which individuals or groups become trapped in a loop of misunderstanding or misinterpretation of information due to certain cognitive or structural barriers. The term 'epistemic trap' is borrowed from Ernesto Ganuza, a fellow member of Delib, the Iberian network of organisations and individuals promoting deliberative democracy. Ganuza introduced this concept during one of our internal meetings, and it has yet to be formally published. 

14. We define wish lists as a list of vague and already known policy proposals with no added value. As suggested by Kainuma et al. (2023): “To avoid wish lists, the first step is to explain the problem in an easy-to-understand manner and clarify trade-offs. One way to clarify trade-offs is to explain the policy measures and their consequences under various conditions using scenarios”. 

15. Hélène Landemore (2012, 2013) has hypothesised about the potential epistemic enrichment that arises from involving randomly selected citizens in deliberating and participating in decision-making processes. According to her argument, a cognitively diverse sample, composed of citizens chosen by lot, leads to better outcomes compared to a relatively homogeneous sample comprising only a few experts. This advantage comes from the multitude of perspectives and life experiences they bring to the discussion. Her work on the Icelandic case during the 2010-2013 Constitutional process (Landemore, 2020) provides additional evidence that partially verifies this hypothesis.  

16. We adopt the term ‘commissioner’ from DemocracyNext (2023, p. 20) but with a key distinction. Our definition requires public decision-making power, whereas DemocracyNext includes any initiating entity without this criterion.  

17. Unlike most Citizens' Assemblies, the Ostbelgien Model from the German-speaking community in Belgium uniquely features both a permanent structure and a Citizens' Council, aligning closely with the council proposed in this text. The Council's primary roles are to monitor the implementation of recommendations and to set the agenda by selecting issues for discussion. Formed through a democratic lottery, the Council consists of 24 members who rotate every six months. A separate Citizens' Panel is responsible for developing the actual collective recommendations, which are then subject to a minimum of two parliamentary debates (OECD, 2021, p. 10). 

18. In the Brussels-Capital Region's Citizen Assembly for Climate, citizens contribute in three ways. They participate in deliberations within the assembly to formulate recommendations. A quarter of the members are randomly selected to decide the next topic for discussion. Additionally, ten citizens are invited to monitor the government's year-long response to the assembly's recommendations. Hence, in this case, citizens are not only involved in deliberating but also in agenda-setting and monitoring the implementation of their recommendations. 

19. We believe it is important to use two bodies when implementing permanent structures along the lines of those proposed by Bouricius (2013) and the example of Ostbelgien.  

20. According to OECD’s good practice principles for deliberative processes, a minimum of four full days, or approximately 32 hours, is advised for participants. This time allocation is deemed crucial for fostering meaningful dialogue and crafting well-considered collective recommendations (OECD, 2020, p. 11). 

21. Approximate suggestion based on our own experience.

22. It is not a statistical representation but a descriptive one. This type of representativeness aims to reflect the demographic diversity of the larger public, and is considered a legitimate form of political representation. Unlike traditional elected assemblies, which often lack such diversity, Citizens' Assemblies are composed by a descriptive sample of randomly chosen citizens to capture a broad spectrum of societal viewpoints (MacKenzie, 2023, p. 25-28).

23. The OECD supports the institutionalisation of deliberative practices for sustainable political commitment, enhanced public trust, and operational efficiencies. Although initial costs for staff training may be elevated, the organisation highlights that these investments lead to long-term cost reductions. Institutionalisation is especially relevant for addressing enduring challenges like climate change (OECD, 2021, p. 5, 20, 23).  

24. Estimates of a 25% to 35% cost reduction are based on the practical outcomes shared at the ‘How to Improve the Ost-Belgium Model?’ workshop during the Democracy RnD October 2023 annual conference.  

25. As previously noted, our guide considers the commissioner as an entity with public decision-making power. 

26. The concept of assemblies initiated without a commissioner aligns with what is often termed 'bottom-up' Citizens' Assemblies. These are generally led by civil society and may aim for more radical reforms. However, they frequently face challenges in achieving tangible impact, primarily due to their looser or no connections with decision-making authorities (Bussu & Fleuß, 2023, p. 152).

27. See, for instance, the Citizens' Panel 'Make Your Brussels Mobility' initiated by the Parliament of the Brussels Region. To analyse the influence of the panel, Vrydagh and Caluwaerts (2020) took into account the initial preferences of policy-makers. Through this lens, they highlighted that about one-third of the recommendations didn't have any influence on public policy, while another third had only a continuous influence, aligning with and reinforcing the initial policy directions. 

28. While connections with a public authority are instrumental in maximising the impact (Elstub & Khoban, 2023, p. 118) they do not guarantee it automatically (Bussu & Fleuß, 2023, p. 152). The design of the entire assembly process is equally pivotal. For instance, as suggested by Elstub and Khoban, if the outcomes are detailed policy recommendations, they “make it easier to see if they have been implemented and thereby hold the relevant policymakers to account” (2023, p. 118).  

29. In other processes, this group is often referred to as a motor group, coordination group or steering group.  

30. This depends on the number of working hours of the Permanent Citizens' Council.

31. Cycles can be annual or biannual.

32. In practice, validation could be done by allowing an oversight group to veto an organisation bidding for a tender or competition. This should only be done if an organisation is perceived to have a defined political position. 

33. This role is usually hired or selected by the government administration, which could lead to mistrust of other parties out of power.

34. For a discussion on how the organisers influence the entire deliberative process, see Gül (2019, p. 7-9).    

35. According to the OECD Evaluation Guidelines for Representative Deliberative Processes (2021), evaluation is a key element of a successful deliberative process. It strengthens trust, demonstrates the quality and neutrality of a process, and creates opportunities for learning. Independent evaluations are recommended as the gold standard. These evaluations provide an objective and fair assessment, ensuring that the process meets high-quality standards.

36. As an illustration, consider the final recommendations proposed by the European Citizens' Panel on Food Waste. 

37. As suggested by the newDemocracy Foundation’s Handbook: “Ideal topics or issues focus on trade-off decisions that require a public to deliberate on and form a common ground view for what the community’s ethical or moral position should be, their preparedness to pay (and to what level) and thus their preferred intent for the direction of a public decision” (newDemocracy Foundation, 2018, p. 100). 

38. Over time, the term "dilemma" has come to be used more broadly to describe any situation that involves a choice between two or more unfavourable options, even if it does not involve a logical paradox. It has become a common term in everyday language to express the challenges and conflicts people face when making difficult decisions.

39. According to James Fishkin (2018, p. 75), when citizens possess an understanding of the points of disagreement among experts, they are better equipped to independently evaluate conflicting viewpoints, rather than merely relying on expert opinions or technical knowledge.

 40. When presenting an argument on a policy issue, experts are not solely reliant on the selection and interpretation of relevant data. Instead, they frame their judgement on their personal values and beliefs (Ney, 2009, p. 9).

41. As argued by Jürgen Habermas, "deficits in democratic legitimation arise whenever the group engaged in making democratic decisions does not align with the group affected by these decisions" (2006, p. 78).  

42. To further delve into the complexities of dilemmas surrounding agricultural and food systems, see the “dilemma cycle” graphic created by Janin, Acloque & El Nour (2023, p. 343). 

43. This is the case in Paris and Ostbelgien, where there are currently Citizens’ Assemblies assigned with this task (OECD, 2021).  

44. We adhere to the recommendations proposed by the newDemocracy Foundation's Handbook for cases when establishing a citizens' assembly is unsuitable or entirely unfeasible. Considering this is crucial, as in an unsuitable scenario, individuals might grow cynical or sceptical not only about the present citizens' assembly but also about any forthcoming deliberative processes (newDemocracy Foundation, 2018, p. 61-65). 

45. In line with the Ostbelgien’s decree of 25 February 2019 establishing a permanent Citizen Dialogue, no assemblies are convened within six months prior to elections (Article 3, § 1).

46. In the consensus conferences in Denmark, the Danish Board of Technology (DBT) played a role similar to the content group. During the content provision phase, the DBT had the "methodological responsibility" of ensuring the process's legitimacy by selecting a fairly balanced panel of experts (Blok, 2007, p. 171), and acted as facilitators between them and the citizens' panel members (Jensen, 2005, p. 225-226). This content group’s responsibility is also highlighted by the recently published Assembling an Assembly Guide by DemocracyNext. Balanced, clear, and accessible information is key not only in terms of ensuring the legitimacy of the process but also for effective deliberation (DemocracyNext, 2023, p. 21). 

47. In previous assemblies, this team was simply part of the organisational team or the governance/steering committee. See, for instance, Convention Citoyenne pour le Climat (the French Citizens’ Convention on Climate); and the Scotland's Climate Assembly. 

48. The lack of representation from affected interest groups in Citizens’ Assemblies has raised concerns, potentially resulting in co-opted processes instead of open and inclusive deliberations. To mitigate this, the establishment of balanced advisory boards becomes crucial, overseeing expert selection and information presentation (Elstub & Khoban, 2023, p. 177).  

49. A good profile, which has proved its worth in both the European Panels and the Catalan Climate Assembly, and which is also recommended by the Danish Board of Technology, is that of a science journalist. 

50. The OECD (2020:112) emphasises the importance of preparing comprehensive educational materials at the outset of deliberative processes. Such materials play a pivotal role in making information universally accessible to all participants, regardless of their backgrounds. To further enhance inclusivity, organisers are encouraged to offer information in diverse formats, including braille, enlarged print, and subtitled videos, among others. Consider the information kit provided in the European Citizens' Panel on food waste as an illustrative example of comprehensive educational materials.

 51. Estimation based on Deliberativa and Ifok experience on the EU Next Generation panels of the European Commission.

52. For guidance on presenting information in Citizens' Assemblies, refer to How to run a citizens’ assembly: A handbook for local authorities based on the Innovation in Democracy Programme (2020, p. 39-40), which details communication tools such as press releases, FAQ sheets, social media graphics, and live blogging and tweeting, emphasising thoroughness, balance, and inclusiveness. 

53. During the French Citizens Convention on Climate, there was a notable emphasis on fact-checking and ensuring that citizens received timely and accurate information. Responses to complex questions were often provided after sessions to ensure participants were well-informed (Eymard, 2020, p. 137-138).

54. Random selection was first employed in the 4th century BC in ancient Athens as the primary method for appointing individuals to public office. It is important to highlight that this mechanism was exclusively applied to what was deemed the body of citizens, excluding women, foreigners, and slaves (Manin, 1997). 

55. Flanigan et al. (2021) note that response rates for Citizens’ Assemblies generally fall within 2-5% range. It is worth noting that the 4% response rate used in our formula is a conservative estimate within this range, especially useful for teams running a process for the first time. Actual response rates can vary based on the context and location of the assembly.

56. The practice of selecting an additional 10% of the assembly size as replacements ensures continuity and full representation in the event of early dropouts (FIDE Policy Research, n.d., p. 25). 

57. This was the case for the Global Assembly on Climate and Ecological Crisis (2022, p. 53), where 100 participants were randomly selected based on a NASA database that maps human population density. 

58. In the European Citizens’ Panels, participant selection was based on randomly generated mobile and fixed line telephone numbers. 

59. “Participation follows the social unequal distribution pattern of political engagement. Participants are better educated, mostly men, and older than the average population” (Jacquet, 2023, p. 642). 

60. The stratification criteria we have outlined echo those stated by DemocracyNext (2023, p. 34-35) and newDemocracy Foundation (2018, p. 137). While these criteria serve as a foundation, they can be expanded based on the assembly’s framing. For instance, the Romsey Citizens’ Assembly in Test Valley (UK), which focused on improving access to, from, and within Romsey, also adopted travel frequency as a stratification criterion (British Government, 2020, p. 28).

61. Ideology and party affiliation often go hand in hand, influencing political attitudes significantly (Turner Zwinkels, 2023). For instance, several studies have highlighted that ideological variables can predict attitudes towards euthanasia (Montañés et al., 2023), abortion (Osborne et al., 2022), and climate change (Smith & Mayer, 2019).

62. In the Scottish Climate Assembly, participants were stratified based on key demographic and attitudinal criteria towards climate change. The Assembly utilised data from national surveys to gauge the broader sentiment around climate change in Scotland, and there was a general alignment between this public sentiment and the perspectives of the assembly members (Scottish Government, 2022, p. 8, 92, 105-106).  

63. “A next step would be for employers to provide paid leave to participate in a deliberative process, as is the case with criminal juries” (OECD, 2020, p. 43).  

64. Effective outreach to vulnerable populations may require specialised strategies, such as partnering with organisations representing these groups or tapping into community ambassadors (FIDE, n.d., p. 21). For instance, during the French Citizens' Convention on the End of Life, recognizing the difficulties of reaching those in precarious situations, collaboration was established with ATD Quart Monde, an NGO focused on supporting the most disadvantaged and ensuring their voices are heard in public debates. As a result, eight women, either activists with ATD Quart Monde or closely affiliated, were integrated as participants (ATD Quart Monde, 2023).

65. The discussions in this chapter are further complemented by Annex A, "Checklist - CLIMAS Guidelines for Inclusive Facilitation and Inclusive Climate Assemblies", which offers detailed guidelines and frameworks relevant to the topic.

66. As stated by Lyn Carson, “the facilitator must have a genuine curiosity, willingness, and ability to help the group surface differences and disagreements, while exploring them respectfully, to reach an outcome that works for all” (2019, p. 218). 

67. Based on recommendations from newDemocracy Foundation's Handbook (2018, p. 167), the facilitator is engaged from the outset of the process because they are directly responsible for implementing each aspect of it. Their unique insight into the feasibility of design decisions makes their early involvement essential for effective planning.  

68. The facilitator's approach to managing communication patterns and discourse within the assembly's chosen design significantly influences the group atmosphere, fostering a positive environment that contributes to effective deliberations (Blijleven, van Hulst & Hendriks, 2019, p. 215). 

69. The approach outlined here aligns with the strategies proposed by Citizens’ Democracy, a Swiss organisation focused on deliberative democracy initiatives.  13 Annex 

1.Other names have been used for similar formats of practical deliberative democracy. Citizens' juries have a longer history and many have been implemented at local level with around 24-35 participants. While the Citizen's Jury was developed in the USA and is very popular in English-speaking countries, similar formats have developed in parallel in other Western countries. In Germany they were called planning cells and in Denmark consensus conferences. Citizens' Assemblies are a similar format to citizens' juries but with a larger number of participants, usually between 60 and 200, with the aim of representing larger populations, such as the scale of a region, a country or even a group of states like the EU (called EU Citizens' Panels in this case). The name "Citizens' Assembly" is so popular that it is already used for formats regardless of the number of participants. For a depiction of different types, the number of participants, duration of the deliberations, outputs, and relevant examples, see: Smith & Setälä (2018). 

2.Scottish Government (2022).

3.For an overview of why elections tend to incentivize short-term measures, see: Guerrero (2014). For a more specific discussion of why electoral democracies fail in handling climate change, see: Willis, Curato & Smith (2022). 

4.Extinction Rebellion (2019). 

5.In what follows, we move from the term 'issue' to 'policy dilemma'. We use the term " issue " here as a way of connecting with the current terminology in use today.

6.Similar to the approach advocated in this guide, the Irish Citizens' Assembly on biodiversity loss also recognized the need to focus on specific, actionable areas due to the complexity of climate-related issues (Irish Citizens’ Assembly, 2023, p. 102) 

7.Throughout this document, the term value-based dilemma or policy dilemma is used interchangeably. 

8.The principles of integrity and purpose are inherent within the themes of transparency, accountability, and clear mandate as outlined by the Committee of Ministers of the Council of Europe (2023). 

9.For instance, in the basic standards for organising Citizens' Assemblies developed by Marcin Gerwin, it is stated that "the coordinators should be impartial, e.g. not active politicians or direct stakeholders" (2018, p. 83). Additionally, the guide by DemocracyNext (2023, p. 19) also emphasises the importance of independence in the governance structure Citizens' Assemblies. 

10. Some successful models, as observed in Ontario and Iceland, have employed a multi-party committee or multi party compromises to prevent immediate politicisation and ensure the assembly’s success (Niessen, 2023, p.332)

11. For a further development of this idea, see the criticism raised by Oleart (2023) regarding the pretended neutrality in the case of the Conference on the Future of Europe. 

12. This group of citizens will be part of the Citizens’ Council. More information in the following chapter.  

13. In this context, we refer to an "epistemic trap" as a situation in which individuals or groups become trapped in a loop of misunderstanding or misinterpretation of information due to certain cognitive or structural barriers. The term 'epistemic trap' is borrowed from Ernesto Ganuza, a fellow member of Delib, the Iberian network of organisations and individuals promoting deliberative democracy. Ganuza introduced this concept during one of our internal meetings, and it has yet to be formally published. 

14. We define wish lists as a list of vague and already known policy proposals with no added value. As suggested by Kainuma et al. (2023): “To avoid wish lists, the first step is to explain the problem in an easy-to-understand manner and clarify trade-offs. One way to clarify trade-offs is to explain the policy measures and their consequences under various conditions using scenarios”. 

15. Hélène Landemore (2012, 2013) has hypothesised about the potential epistemic enrichment that arises from involving randomly selected citizens in deliberating and participating in decision-making processes. According to her argument, a cognitively diverse sample, composed of citizens chosen by lot, leads to better outcomes compared to a relatively homogeneous sample comprising only a few experts. This advantage comes from the multitude of perspectives and life experiences they bring to the discussion. Her work on the Icelandic case during the 2010-2013 Constitutional process (Landemore, 2020) provides additional evidence that partially verifies this hypothesis.  

16. We adopt the term ‘commissioner’ from DemocracyNext (2023, p. 20) but with a key distinction. Our definition requires public decision-making power, whereas DemocracyNext includes any initiating entity without this criterion.  

17. Unlike most Citizens' Assemblies, the Ostbelgien Model from the German-speaking community in Belgium uniquely features both a permanent structure and a Citizens' Council, aligning closely with the council proposed in this text. The Council's primary roles are to monitor the implementation of recommendations and to set the agenda by selecting issues for discussion. Formed through a democratic lottery, the Council consists of 24 members who rotate every six months. A separate Citizens' Panel is responsible for developing the actual collective recommendations, which are then subject to a minimum of two parliamentary debates (OECD, 2021, p. 10). 

18. In the Brussels-Capital Region's Citizen Assembly for Climate, citizens contribute in three ways. They participate in deliberations within the assembly to formulate recommendations. A quarter of the members are randomly selected to decide the next topic for discussion. Additionally, ten citizens are invited to monitor the government's year-long response to the assembly's recommendations. Hence, in this case, citizens are not only involved in deliberating but also in agenda-setting and monitoring the implementation of their recommendations. 

19. We believe it is important to use two bodies when implementing permanent structures along the lines of those proposed by Bouricius (2013) and the example of Ostbelgien.  

20. According to OECD’s good practice principles for deliberative processes, a minimum of four full days, or approximately 32 hours, is advised for participants. This time allocation is deemed crucial for fostering meaningful dialogue and crafting well-considered collective recommendations (OECD, 2020, p. 11). 

21. Approximate suggestion based on our own experience.

22. It is not a statistical representation but a descriptive one. This type of representativeness aims to reflect the demographic diversity of the larger public, and is considered a legitimate form of political representation. Unlike traditional elected assemblies, which often lack such diversity, Citizens' Assemblies are composed by a descriptive sample of randomly chosen citizens to capture a broad spectrum of societal viewpoints (MacKenzie, 2023, p. 25-28).

23. The OECD supports the institutionalisation of deliberative practices for sustainable political commitment, enhanced public trust, and operational efficiencies. Although initial costs for staff training may be elevated, the organisation highlights that these investments lead to long-term cost reductions. Institutionalisation is especially relevant for addressing enduring challenges like climate change (OECD, 2021, p. 5, 20, 23).  

24. Estimates of a 25% to 35% cost reduction are based on the practical outcomes shared at the ‘How to Improve the Ost-Belgium Model?’ workshop during the Democracy RnD October 2023 annual conference.  

25. As previously noted, our guide considers the commissioner as an entity with public decision-making power. 

26. The concept of assemblies initiated without a commissioner aligns with what is often termed 'bottom-up' Citizens' Assemblies. These are generally led by civil society and may aim for more radical reforms. However, they frequently face challenges in achieving tangible impact, primarily due to their looser or no connections with decision-making authorities (Bussu & Fleuß, 2023, p. 152).

27. See, for instance, the Citizens' Panel 'Make Your Brussels Mobility' initiated by the Parliament of the Brussels Region. To analyse the influence of the panel, Vrydagh and Caluwaerts (2020) took into account the initial preferences of policy-makers. Through this lens, they highlighted that about one-third of the recommendations didn't have any influence on public policy, while another third had only a continuous influence, aligning with and reinforcing the initial policy directions. 

28. While connections with a public authority are instrumental in maximising the impact (Elstub & Khoban, 2023, p. 118) they do not guarantee it automatically (Bussu & Fleuß, 2023, p. 152). The design of the entire assembly process is equally pivotal. For instance, as suggested by Elstub and Khoban, if the outcomes are detailed policy recommendations, they “make it easier to see if they have been implemented and thereby hold the relevant policymakers to account” (2023, p. 118).  

29. In other processes, this group is often referred to as a motor group, coordination group or steering group.  

30. This depends on the number of working hours of the Permanent Citizens' Council.

31. Cycles can be annual or biannual.

32. In practice, validation could be done by allowing an oversight group to veto an organisation bidding for a tender or competition. This should only be done if an organisation is perceived to have a defined political position. 

33. This role is usually hired or selected by the government administration, which could lead to mistrust of other parties out of power.

34. For a discussion on how the organisers influence the entire deliberative process, see Gül (2019, p. 7-9).    

35. According to the OECD Evaluation Guidelines for Representative Deliberative Processes (2021), evaluation is a key element of a successful deliberative process. It strengthens trust, demonstrates the quality and neutrality of a process, and creates opportunities for learning. Independent evaluations are recommended as the gold standard. These evaluations provide an objective and fair assessment, ensuring that the process meets high-quality standards.

36. As an illustration, consider the final recommendations proposed by the European Citizens' Panel on Food Waste. 

37. As suggested by the newDemocracy Foundation’s Handbook: “Ideal topics or issues focus on trade-off decisions that require a public to deliberate on and form a common ground view for what the community’s ethical or moral position should be, their preparedness to pay (and to what level) and thus their preferred intent for the direction of a public decision” (newDemocracy Foundation, 2018, p. 100). 

38. Over time, the term "dilemma" has come to be used more broadly to describe any situation that involves a choice between two or more unfavourable options, even if it does not involve a logical paradox. It has become a common term in everyday language to express the challenges and conflicts people face when making difficult decisions.

39. According to James Fishkin (2018, p. 75), when citizens possess an understanding of the points of disagreement among experts, they are better equipped to independently evaluate conflicting viewpoints, rather than merely relying on expert opinions or technical knowledge.

 40. When presenting an argument on a policy issue, experts are not solely reliant on the selection and interpretation of relevant data. Instead, they frame their judgement on their personal values and beliefs (Ney, 2009, p. 9).

41. As argued by Jürgen Habermas, "deficits in democratic legitimation arise whenever the group engaged in making democratic decisions does not align with the group affected by these decisions" (2006, p. 78).  

42. To further delve into the complexities of dilemmas surrounding agricultural and food systems, see the “dilemma cycle” graphic created by Janin, Acloque & El Nour (2023, p. 343). 

43. This is the case in Paris and Ostbelgien, where there are currently Citizens’ Assemblies assigned with this task (OECD, 2021).  

44. We adhere to the recommendations proposed by the newDemocracy Foundation's Handbook for cases when establishing a citizens' assembly is unsuitable or entirely unfeasible. Considering this is crucial, as in an unsuitable scenario, individuals might grow cynical or sceptical not only about the present citizens' assembly but also about any forthcoming deliberative processes (newDemocracy Foundation, 2018, p. 61-65). 

45. In line with the Ostbelgien’s decree of 25 February 2019 establishing a permanent Citizen Dialogue, no assemblies are convened within six months prior to elections (Article 3, § 1).

46. In the consensus conferences in Denmark, the Danish Board of Technology (DBT) played a role similar to the content group. During the content provision phase, the DBT had the "methodological responsibility" of ensuring the process's legitimacy by selecting a fairly balanced panel of experts (Blok, 2007, p. 171), and acted as facilitators between them and the citizens' panel members (Jensen, 2005, p. 225-226). This content group’s responsibility is also highlighted by the recently published Assembling an Assembly Guide by DemocracyNext. Balanced, clear, and accessible information is key not only in terms of ensuring the legitimacy of the process but also for effective deliberation (DemocracyNext, 2023, p. 21). 

47. In previous assemblies, this team was simply part of the organisational team or the governance/steering committee. See, for instance, Convention Citoyenne pour le Climat (the French Citizens’ Convention on Climate); and the Scotland's Climate Assembly. 

48. The lack of representation from affected interest groups in Citizens’ Assemblies has raised concerns, potentially resulting in co-opted processes instead of open and inclusive deliberations. To mitigate this, the establishment of balanced advisory boards becomes crucial, overseeing expert selection and information presentation (Elstub & Khoban, 2023, p. 177).  

49. A good profile, which has proved its worth in both the European Panels and the Catalan Climate Assembly, and which is also recommended by the Danish Board of Technology, is that of a science journalist. 

50. The OECD (2020:112) emphasises the importance of preparing comprehensive educational materials at the outset of deliberative processes. Such materials play a pivotal role in making information universally accessible to all participants, regardless of their backgrounds. To further enhance inclusivity, organisers are encouraged to offer information in diverse formats, including braille, enlarged print, and subtitled videos, among others. Consider the information kit provided in the European Citizens' Panel on food waste as an illustrative example of comprehensive educational materials.

 51. Estimation based on Deliberativa and Ifok experience on the EU Next Generation panels of the European Commission.

52. For guidance on presenting information in Citizens' Assemblies, refer to How to run a citizens’ assembly: A handbook for local authorities based on the Innovation in Democracy Programme (2020, p. 39-40), which details communication tools such as press releases, FAQ sheets, social media graphics, and live blogging and tweeting, emphasising thoroughness, balance, and inclusiveness. 

53. During the French Citizens Convention on Climate, there was a notable emphasis on fact-checking and ensuring that citizens received timely and accurate information. Responses to complex questions were often provided after sessions to ensure participants were well-informed (Eymard, 2020, p. 137-138).

54. Random selection was first employed in the 4th century BC in ancient Athens as the primary method for appointing individuals to public office. It is important to highlight that this mechanism was exclusively applied to what was deemed the body of citizens, excluding women, foreigners, and slaves (Manin, 1997). 

55. Flanigan et al. (2021) note that response rates for Citizens’ Assemblies generally fall within 2-5% range. It is worth noting that the 4% response rate used in our formula is a conservative estimate within this range, especially useful for teams running a process for the first time. Actual response rates can vary based on the context and location of the assembly.

56. The practice of selecting an additional 10% of the assembly size as replacements ensures continuity and full representation in the event of early dropouts (FIDE Policy Research, n.d., p. 25). 

57. This was the case for the Global Assembly on Climate and Ecological Crisis (2022, p. 53), where 100 participants were randomly selected based on a NASA database that maps human population density. 

58. In the European Citizens’ Panels, participant selection was based on randomly generated mobile and fixed line telephone numbers. 

59. “Participation follows the social unequal distribution pattern of political engagement. Participants are better educated, mostly men, and older than the average population” (Jacquet, 2023, p. 642). 

60. The stratification criteria we have outlined echo those stated by DemocracyNext (2023, p. 34-35) and newDemocracy Foundation (2018, p. 137). While these criteria serve as a foundation, they can be expanded based on the assembly’s framing. For instance, the Romsey Citizens’ Assembly in Test Valley (UK), which focused on improving access to, from, and within Romsey, also adopted travel frequency as a stratification criterion (British Government, 2020, p. 28).

61. Ideology and party affiliation often go hand in hand, influencing political attitudes significantly (Turner Zwinkels, 2023). For instance, several studies have highlighted that ideological variables can predict attitudes towards euthanasia (Montañés et al., 2023), abortion (Osborne et al., 2022), and climate change (Smith & Mayer, 2019).

62. In the Scottish Climate Assembly, participants were stratified based on key demographic and attitudinal criteria towards climate change. The Assembly utilised data from national surveys to gauge the broader sentiment around climate change in Scotland, and there was a general alignment between this public sentiment and the perspectives of the assembly members (Scottish Government, 2022, p. 8, 92, 105-106).  

63. “A next step would be for employers to provide paid leave to participate in a deliberative process, as is the case with criminal juries” (OECD, 2020, p. 43).  

64. Effective outreach to vulnerable populations may require specialised strategies, such as partnering with organisations representing these groups or tapping into community ambassadors (FIDE, n.d., p. 21). For instance, during the French Citizens' Convention on the End of Life, recognizing the difficulties of reaching those in precarious situations, collaboration was established with ATD Quart Monde, an NGO focused on supporting the most disadvantaged and ensuring their voices are heard in public debates. As a result, eight women, either activists with ATD Quart Monde or closely affiliated, were integrated as participants (ATD Quart Monde, 2023).

65. The discussions in this chapter are further complemented by Annex A, "Checklist - CLIMAS Guidelines for Inclusive Facilitation and Inclusive Climate Assemblies", which offers detailed guidelines and frameworks relevant to the topic.

66. As stated by Lyn Carson, “the facilitator must have a genuine curiosity, willingness, and ability to help the group surface differences and disagreements, while exploring them respectfully, to reach an outcome that works for all” (2019, p. 218). 

67. Based on recommendations from newDemocracy Foundation's Handbook (2018, p. 167), the facilitator is engaged from the outset of the process because they are directly responsible for implementing each aspect of it. Their unique insight into the feasibility of design decisions makes their early involvement essential for effective planning.  

68. The facilitator's approach to managing communication patterns and discourse within the assembly's chosen design significantly influences the group atmosphere, fostering a positive environment that contributes to effective deliberations (Blijleven, van Hulst & Hendriks, 2019, p. 215). 

69. The approach outlined here aligns with the strategies proposed by Citizens’ Democracy, a Swiss organisation focused on deliberative democracy initiatives. 

14 Annex A. Checklist - CLIMAS Guidelines for Inclusive Facilitation and Inclusive Climate Assemblies

 14.1  Before the Assembly

14.1.1 Context

 …

14.1.2 Aspects to be checked before the assembly

Before the assembly:

14.2 During the Assembly

14.2.1 Context

 Three fundamental phases in all citizen assembly processes:

• A learning phase where citizens are confronted with relevant information to be able to face the dilemma and understand potential solutions. 
• A deliberative phase, in which the different solutions are confronted and trade-offs are addressed. 
• A final phase in which agreements are reached in order to draft the final recommendations.

 Distinct facilitation spaces: 

• Plenary settings, e.g., to set the scene at the beginning versus smaller roundtables 
• General sessions versus those where the actual deliberation to confront different solutions or approaches and their trade-offs takes place 
• In expert groups versus groups of general public 
• Virtual versus face-to-face settings
• Etc.

14.2.2 Aspects to be checked during the assembly 

 During the assembly: 

14.3 After the Assembly

14.3.1 Context 

…

14.3.2 Aspects to be checked after the assembly

After the assembly:

14.4 References

Wenger, Etienne, McDermott, Richard and Snyder, William. Cultivating Communities of Practice. Boston: Harvard Business School Press. 2002.

INDIMO. D3.2: Communities of practice report (Draft). Available at: https://www.indimoproject.eu/resource/d3-2-communities-of-practice-report-draft/. 2021. 

Gov.uk and university of leeds: Inclusive Language Guidance - Equality and Inclusion Unit (leeds.ac.uk), available at: https://equality.leeds.ac.uk/support-and-resources/inclusive languageguidance/#:~:text=Do%20use%20language%20that%3A%201%20Acknowledges%2 0diversity%20a nd,when%20they%20feel%20comfortable%20doing%20so.%20More%20items. 

Kimbra White, Nicole Hunter, Keith Greaves. Facilitating Deliberation, A practical guide, Mosaic Lab. 

NHS Institute for Innovation and Improvement. The Facilitator’s Toolkit: Tools, techniques and tips for effective facilitation. 2009