Citizens' assembly

A citizens' assembly is a group of people selected by lottery from the general population to deliberate on important public questions so as to exert an influence. Other names and variations include citizens' jury, citizens' panel, people's panel, mini-publics, people's jury, policy jury, consensus conference and citizens' convention.

A citizens' assembly uses elements of a jury to create public policy. Its members form a representative cross-section of the public, and are provided with time, resources and a broad range of viewpoints to learn deeply about an issue. Through skilled facilitation, the assembly members weigh trade-offs and work to find common ground on a shared set of recommendations. Citizens' assemblies can be more representative and deliberative than public engagement, polls, legislatures or ballot initiatives. They seek quality of participation over quantity.

With Athenian democracy as the most famous government to use sortition, theorists and politicians have used citizens' assemblies and other forms of deliberative democracy in a variety of modern contexts. The OECD has found their use increasing since 2010.

Membership
Achieving a sufficiently inclusive and representative group of everyday people helps ensure that the assembly reflects political equality and the diversity of a community. Some of the components are described below.

Selection
Assembly members are most often selected through a two-stage process called sortition. In a first instance, a large number of invitations are sent from the convening authority at random (often around 10,000-30,000). The principle is that everybody should have an equal chance of being selected in the first place. Amongst everybody who responds positively to this invitation, there is a second lottery process, this time ensuring that the final group broadly reflects the community in regards to certain criteria such as gender, age, geographic, and socio-economic status, amongst others. This is called stratification – a technique that is also used in opinion polls.

Random selection in governance (known as sortition) has historical significance and the earliest known instances include the Athenian democracy and various European communities.

Size
The size of a citizens' assembly should ideally be large enough to capture a representative cross-section of the population. The size depends on the purpose, demographics, and population size of the community. Assemblies typically consist of between 50 and 200 citizens.

Turnover
Regular turnover of participants is common. This can help to maintain viewpoint diversity in the long term and avoid sorting the assembly into in-groups and out-groups that could bias the result, become homogenous or get captured by private interests.

Functions
In general, the purpose is to have an influence on public decision making. The function of a citizen's assembly has no a priori limits. Though assemblies are sometimes limited in scope, the purpose of an assembly can vary widely. Modern assemblies have tended to propose rather than directly enact public policy changes due to constrictions in place by most constitutions. Assembly proposals in those systems are then enacted (or not) by the corresponding authority. Sometimes a proposal is sent to the general electorate as a referendum.

Deliberation
A key component of assemblies is their deliberative nature. Deliberation allows members to reflect on their values and weigh new information in dialogue with subject-matter experts and their peers. By incorporating the views, information and arguments of experts and then asking the participants to engage in collaborative discussion, assemblies aim to enable the participants to educate themselves and produce a vote or result representative of the public interest.

Parkinson argues that the intent of deliberation is to "replace power plays and political tantrums with 'the mild voice of reason. Deliberation attempts to marry procedural effectiveness with substantive outcomes. Parkinson continues that the process reframes "political legitimacy" as involving "not just doing things right, but doing the right things". This view contrasts with the purely procedural account of legitimacy, of which Rawls says "there is a correct or fair procedure such that the outcome is likewise correct or fair, whatever it is, provided the procedure has been followed properly." While deliberation is itself a procedure, it deliberately incorporates factual information, and thus broadens the consideration of legitimacy.

Agenda-setting
Agenda-setting refers to establishing a plan for the substantive issues that the assembly is to consider. In major examples of assemblies, such as those in British Columbia and Ontario, the legislature set the agenda before the assemblies were convened. However, Dahl asserts that final control over agendas is an essential component of an ideal democracy: "the body of citizens...should have the exclusive control to determine what matters are or are not to be decided." Today, agenda-setting is a component of the ongoing citizens' assemblies in Ostbelgien (the Germany-Speaking Community of Belgium), Paris, and the Brussels Citizens' Assembly on Climate.

Briefing Materials
Briefing materials should be balanced, diverse and accurate. One approach is to have an advisory committee of randomly selected from the population to set the rules and procedures without undue influence by government officials.

Guardrails
Especially when juries or assemblies have more than advisory powers, the checks and balances grow to ensure that those participating can't make unilateral decisions or concentrate power. In Athenian democracy, for example, this meant a complex array of carrots and sticks as guardrails that successfully blunted the temptation of corruption. Étienne Chouard argues that in large part because elected politicians wrote the constitutions, that governments that use elections have far fewer guardrails in place than those based largely on sortition.

Some worry that Assemblies might not provide the same accountability as elections to prevent members from engaging in inappropriate behavior. Pierre Étienne Vandamme points to other methods of accountability (including from separate Citizens' assemblies) and the benefits of being able to vote one's conscious and not be subject to the same external pressures as elected politicians. Assemblies can also provide a check on elected officials by setting and enforcing the rules governing them, instead of politicians self-policing.

Decision
At some point, the assembly concludes its deliberations and offers conclusions and recommendations. This is typically done in a voting process such as through the use of secret ballots to help keep citizens from becoming public figures and able to vote their conscience.

Application
Etienne Chouard argues that elected officials have a conflict-of-interest when it comes to creating the rules by which power is distributed in a democracy, such as in drafting constitutions. He argues for sortition (e.g. Citizens' assemblies) as ideal for this type of decision-making.

Andrew Anthony believes Citizens' Assemblies would be useful for specific cases, but worries that with more complex issues that juries (or in this case assemblies) would not outperform elected officials.

Modern examples
Assemblies in modern times mostly send recommendations to politicians or voters for approval. For randomly-selected bodies with decision-making power, some examples can be found in ancient Athens, the Republic of Venice, the Republic of Florence and juries in the United States. The OECD identified 733 assemblies from 1979-2023. The OECD also documented the growth of permanent assemblies starting in 2010. Claudia Chwalisz discusses eight ways that deliberative democracy has been institutionalised.

Global Assembly
The Global Assembly was organized in 2021 to coincide with the COP26 in Glasgow in October–November 2021. It is the first body that has attempted to make a claim to represent the democratic wishes of the global population as a whole.

Belgium
The G1000 is a donation-funded initiative launched in 2011 by David Van Reybrouck with an online survey to identify issues. 700 participants came together for a day to deliberate in Brussels.

First modern permanent assembly
In September 2019, the German-speaking region in Belgium launched the first ongoing citizens' assembly since the renaissance. The body shares powers in the legislature with an elected body. The citizens' assembly has two bodies: the Bürgerrat meets monthly and has 24 members serving for 18 months and they set the agenda for another randomly selected body of less than 50 people who decide on issues during 3 weekends over 3 months. Brussels and Wallonia have since created permanent advisory citizens' assemblies.

Canada
Pioneering citizens' assemblies proposed changes to the electoral systems of British Columbia in 2004 and Ontario in 2006. While the recommendations of these assemblies did not garner the 60% of votes necessary in follow-up referenda, they inspired more deliberative assemblies in Canada and around the world, even helping to popularize the term "Citizens' Assembly."

Denmark
Johs Grundahl discusses Denmark as a major hotspot of consensus conferences in the 1980s and one of the earliest attempts by policymakers to include the lay public's opinions in their decision-making through public engagement. The purpose of consensus conferences is to "qualify people's attitudes, inasmuch as they are given all the information they require until they are ready to assess a given technology" and the resulting product may look different from that of other types of assemblies due to the need to reach consensus. Consensus conferences are generally deemed suitable for topics that are socially relevant and/or that require public support.

Participants are randomly selected from a group of citizens who are invited to apply. Invitees are members of the lay public who have no specific knowledge of the issue. The resulting panel attempts to be demographically representative.

Panel members participate in two preparatory weekends and are given material prepared by a communicator to gain a basic understanding of the topic. The panel then participates in a 4-day conference. The panel participates in a Q&A session with experts, where they hear opposing views. Members then prepare a final document summarizing their views and recommendations. On the final day, the panel then discusses their final document with policy- and decision-makers.

France
France hosted a Citizens Convention on Climate in 2019 and 2020, where 150 randomly selected citizens made recommendations to elected officials on environmental policies.

France hosted another convention on the end of life concerning assisted suicide and euthanasia in 2022 and 2023 to advise the French parliament in coordination with a consultative legislative assembly, which some worry will dilute the process. This process had 170 participants.

Germany
Since the 1980s, local and regional governments in Germany increasingly experimented with consultative bodies drawn from randomly-selected citizens. One of the variations that originated in Germany and has inspired similar experiments elsewhere is known as planning cells, where one or more cohorts of randomly-selected citizens go through a process of hearing from speakers and deliberating on an issue in order to efficiently get more representative and deliberative input from a population.

Ireland
After the Irish financial crisis beginning in 2008, an assembly was among various proposals for political reform. The subsequent Fine Gael–Labour government's programme included a "Constitutional Convention" comprising a chairperson nominated by the Taoiseach, 33 legislators nominated by political parties, and 67 citizens selected to be demographically representative. It met from 2012 to 2014, discussing six issues specified by the government and then two assembly-selected issues. It issued nine reports, recommending constitutional amendments and other changes to statute law and legislative practice. The government put two of the eighteen proposals that needed a referendum on the 2015 ballot, while some of the others were able to be implemented. The 2015 Irish Constitutional Referendums saw gay marriage legalized.

The Fine Gael–independent minority government formed after the 2016 general election established an assembly in July 2016 "without participation by politicians, and with a mandate to look at a limited number of key issues over an extended time period."

Netherlands
Held in 2006 and composed of 143 randomly-selected Dutch citizens, the Burgerforum Kiesstelsel was tasked with examining options for electoral reform. On December 14, 2006, the Burgerforum presented its final report to a minister of the outgoing People's Party (VVD). A response to the report was delivered in April 2008, when it was rejected by the government of the then ruling coalition.

Tegen Verkiezingen has maintained a list of news articles related to citizens' assemblies and sortition in the Netherlands since 2018.

Poland
Beginning in July 2016 after the municipal response to flooding was deemed inadequate by many citizens, Gdańsk assemblies comprising approximately 60 randomly-selected residents from the city's voter rolls made binding decisions to address problems. The membership was then balanced according to factors such as education-level, age, sex and district. For example, the assembly has the same percentage of senior citizens as the city. The assembly met for several days, heard testimony from experts, asked questions and deliberated in small groups before rendering its decisions. Assembly meetings were described by some as calm and enjoyable.

United Kingdom
The People's Parliament was a program run on TV in the UK in 1994 that showed randomly-selected regular folks debating policy topics.

In 2019 the British government announced the UK Climate Assembly, with 108 citizens aiming to deliberate over how to reach net-zero emissions by 2050. Meetings were delayed due to the COVID-19 pandemic and took place over six weekends between January and May 2020, with a report published in September 2020.

The Citizens' Assembly of Scotland met in 2019 and 2020 to discuss very broad themes, which some academics said lead to less tangible outcomes than other assemblies but an impressive report and equally successful coordination online. In 2021, Newham London Borough Council became the first local authority to introduce a citizens' assembly as a permanent body, with rotating members.

In a 2019 survey conducted of British citizens by the Royal Society of Arts, 57% of those surveyed thought that a citizens assembly would not be sufficiently democratic because it was not large enough. Northern Ireland had the highest support for an assembly, whichthe authors speculated was perhaps due to the use of assemblies in the Republic of Ireland.

The Beyond Politics party has called for assemblies to be utilized for direct decision-making in the UK. In 2023, the Labour Party said it was drawing up plans to introduce citizens assemblies inspired by the assemblies in the Republic of Ireland should it enter government. It has been suggested by the party that the assemblies could give the public a say on issues such as devolution, assisted dying, house building and constitutional reform, among others. The party has also suggested that these assemblies could override the government on issues for which they come to a decision.

Oregon
A Citizens' Initiative Review (CIR) is a panel that deliberates on a ballot initiative or referendum to be decided in an upcoming election in order to produce a useful summary for voters. The panelists are chosen through means such as random sampling and stratified sampling to be demographically representative. This often involves paying for the time and travel of the roughly two dozen participants. While not quite a citizens' assembly according to John Rountree and Nicole Curato, they note it shares many of the same characteristics.

The state of Oregon created the first permanent Citizens' Initiative Review in 2010, while pilots have been run in places including Colorado, Arizona, Massachusetts, and Sion (Switzerland).

Texas
In the late 1990s, Texas power companies tapped then University of Texas professor James Fishkin to run an intensive form of poll (known as a deliberative poll), where representative samples of ratepayers (customers) were selected to join eight sessions where they could learn about and deliberate on whether they would want to, for example, pay $2 to $5 per month more for their energy bills in order to get more energy from wind power and improve energy efficiency. The dramatic rise in ratepayers willing to pay more after the eight sessions has received much credit for motivating Texas' wind power boom.

Common interest
Electoral reform, redistricting, campaign finance law, and the regulation of political speech are often claimed to be unsuitable for management by self-interested politicians. Assemblies have repeatedly been deployed to replace such political judgments. Fearon and separately Nino support the idea that deliberative democratic models tend to generate conditions of impartiality, rationality and knowledge, increasing the likelihood that the decisions reached are morally correct.

Peter Stone, Oliver Dowlen and Gil Delannoi assert that selection by sortition prevents disproportionate influence by "special interests". Term limits could further reduce the opportunities for special interests to influence assemblies.

Deliberation
Deliberative democracy aims to harness the benefits of deliberation to produce better understanding and resolution of important issues. Assemblies are intended to stimulate deliberation, in which the participants can less easily be captured by special interest. Deliberative polling advocate Fishkin claimed that deliberation promotes better problem-solving by educating and actively engaging participants. Deliberation is claimed to lessen faction by emphasizing resolution over partisanship. Additionally, citizens who were not selected tend to perceive those chosen as both technical experts and as "ordinary" citizens like themselves. As happened in British Columbia, these features encouraged voter comfort with the actions of the assembly. For example, a study comparing the debate quality of an Irish Citizens' Assembly and an Irish parliamentary committee found that citizens showed a deeper cognitive grasp of the subject matter at stake (abortion).

The possibility of group polarization concerns Cass Sunstein who argued in 1999 that deliberation tends to move groups, and the individuals who compose them, toward a more extreme point in the direction indicated by their own predeliberation judgments. Consensus conferences also have the potential to make individuals tend to the extreme in their opinions, i.e. citizens essentially rally around their own views in the presence of opposing views. However, Fishkin responded that this depends on how the assembly is structured. Resources such as briefing materials and expert testimony are meant to ameliorate extreme views by supplying information and correcting misinformation/misunderstanding.

Representative and inclusive
Random lotteries have been explored as election alternatives on grounds that it allows for more accurate representation and inclusivity. A truly randomly selected group can embody the "median voter". Participants are supposed to represent the common person. Selection by lot can correct the unrepresentativeness of many elections. Successful political candidates typically require access to education, money and connections. Though elected legislators generally have more experience, they are likely to focus on their supporters rather than the larger population. Representative democracies have been criticized as not representative at all. The lack of female and minority representation in the US Congress is often cited as an example. While others lament the importance of branding in electing candidates (with recognizable last names, for example, fueling political dynasties).

Money is argued to have an outsized role in election outcomes. Lessig argued that elections are dominated by money. When random selection is used alongside statistical analysis, accurate representation can be attained, although in practice, a large number of citizens' assemblies do not reach a sample size large enough to achieve statistical representativeness. Overlaying quotas on the initial random selection corrects for disproportionate ability/willingness across various groups, improving representativeness.

Cognitive diversity
Assemblies allow for increased cognitive diversity, understood as a diversity of problem-solving methods or ways of interpreting the world. Quasi-random selection does not filter out cognitive diversity as elections are alleged to do. Similarly, the process does not attempt to select the best-performing or most skilled agents.

Some studies report that cognitively diverse groups produce better results than homogenous groups, a phenomenon commonly referred to as the wisdom of the crowd. Lu and Page claim that cognitive diversity is valuable for effective problem solving. They selected two problem-solving teams from a diverse population of intelligent agents: the randomly selected team outperformed the "best-performing" agents. Unique perspectives and interpretations generally enhance analysis. These results imply that it may be more important to maximize cognitive diversity over individual competence. Landemore argued that random selection results in increased efficacy, diversity and inclusivity. In fact, Mill famously argued that governing assemblies should be a "fair sample of every grade of intellect among the people" over "a selection of the greatest political minds." This analysis contrasts with those concerned about competence of individuals selected.

Representativeness
While sortition can enable a more representative group and outcome than through elections, that outcome is not guaranteed. James Fishkin argues that some important ways to improve representativeness include reducing time burdens, requiring employers to give employees time off if-needed, and adding remote options while increasing the benefits in terms of pay and power that is given to an assembly.

Dietram Scheufele worried in 2010 that the selected individuals with the time and interest to join civic meetings like consensus conferences often results in an unrepresentative survey sample, especially if most of those invited do not choose to participate. :16-19 He also cites concern around participant group dynamics and how personalities have played an important role in producing different outcomes of discussions in experiments in the 1990s. :19, 24 In a 2023 blog post, political scientists Tiago C. Peixoto and Paolo Spada say that claims of representativeness vary in their assertiveness and accuracy, and that the three main challenges to representativeness include getting a minimum sample size, sampling error, and non-response.

Efficiency
Instead of asking all citizens to deliberate deeply on every issue every election, assemblies/juries might save voters time by only asking for short bursts of their time and attention on one specific issue instead of more frequent elections or long ballots associated with voter fatigue. The biggest potential for cost-savings stems from the wisdom of the crowd that could be less susceptible to forms of influence by special interests seeking narrow benefits at the expense of the rest. John Burnheim critiques representative democracy as requiring citizens to vote for a large package of policies and preferences bundled together in one representative or party, much of which a voter might not want. He argues that this does not efficiently translate voter preferences as well as sortition, where a group of people have the time and the ability to focus on a single issue.

Participation
Compared to elections, assemblies lack frequent mass participation, as the assembly involves a tiny minority of the public. When people vote, they interact with the government and with the law. Elections and voting represent for some an important element of sovereignty, even if the vote makes little difference. Eliminating elections undermines the consultation process that allows those voters to feel like a more involved citizen in a representative democracy.

Lafont, for example, argues that assemblies undermine deliberation. She argues that this is because assemblies asking the public to accept the results of their deliberation is akin to an elite democracy. While she clarifies that "this variety differs from the standard elite model to the extent that it does not ask citizens to blindly defer to the deliberations of a consolidated political elite.... [it] blindly defer to the deliberations of a few selected citizens." Fishkin argues in turn that this model is not elite because it uses ordinary citizens who are representative of the population. Lafont rejects this characterization, arguing that people are "subjected to a filter of deliberative experience" which makes them "no longer a representative sample of the citizenry at large."

Landemore responds to Lafont by arguing that while her concerns are valid, large-scale discourse is simply impossible, never mind superior. Landemore recommends making assemblies "as 'open' to the larger public as possible." For example, their decisions could be validated via a referendum.

Fishkin notes a trilemma among the ideas of political equality, deliberation, and participation. In a body such as an assembly, political equality is achieved through a random and ideally representative selection process, while deliberation is achieved in the actions of the assembly. However, since the body is made up of a subset of the population, it does not achieve the goal of participation on a broad scale.

Fishkin attempts to solve that trilemma by considering an entire deliberative society, which would constitute a deliberative macrocosm. He sees assemblies as experiments on how to realize macro-scale deliberation later on.

Warren and Gastil claim, in the British Columbia case, that other citizens should have been able to "treat it as a facilitative trustee (a trusted information and decision proxy)." Participants essentially became informal experts, allowing them to act as an extension of the larger public.

The introduction of the assembly, according to John Parkinson, undermined the trust and power that British Columbia political parties and advocacy groups had gained. It could also "undermine the epistemic, ethical, and democratic functions of the whole".