Citizen’s Climate Convention : some lessons for the future

Introduction

In France, from October 2019 to June 2020, 150 randomly selected citizens worked collectively and developed proposals to enable the country to reduce its greenhouse gas emissions. This is the Convention Citoyenne pour le Climat, or the Citizens’ Climate Convention (CCC). President Emmanuel Macron addressed the Convention on June 29, 2020, and the government is currently drafting legislation based on these proposals. The mandate of the Convention provides for the 150 citizens to meet one last time to respond to the government regarding both the timetable for the bill’s implementation and its substance. Work is therefore still in progress.

Despite this, other citizens’ conventions have already been announced by presidents of regions, mayors, and other local actors. A reform of the national Economic, Social and Environmental Council (EESC)[1], currently being examined by Parliament, would enable it to organize more citizens’ conventions. Similar initiatives have been taken abroad in the wake of the French experience (in the United Kingdom, Scotland and Spain, among others). Companies are also beginning to take interest in this type of exercise in order to foster deliberation among their employees or to initiate a dialogue with some of their stakeholders.

This movement is taking place within a broader context: in a recent report[2], the OECD listed 25 citizens’ consultations around the world in 2019 and 30 to 40 either underway or announced since October 2019. However, a closer look at these different experiences reveals that the words « convention » or « conference » cover very uneven practices.

If future conventions are not to be distorted or turned into mere communication tools by sprinkling citizen deliberation on decision-making processes that will not take it into account, and if, as is being suggested in France, future conventions are to be based on the CCC, it is essential to highlight its early lessons. This is the purpose of this paper.

We first wish to underline the main singularities of the French experience. While it is a continuation of past initiatives, it also differs from them by the nature of its mandate and the form of its governance, two characteristics that we believe should be emulated in the future.

We then address some of the issues it has had to resolve in order to achieve its goals. These often took the form of dilemmas, which we can reasonably expect will affect future experiences. We are not sure that the solutions chosen for the CCC can simply be duplicated in the future, but we think it is useful to keep them in mind.

Finally, we offer some general principles that we think should imperatively guide future conventions.

We believe that they should be made into law so that they may be invoked in the future against both organizers and mandators before the competent courts. While it may be legitimate to experiment before codifying, since innovation requires certain freedoms, it would be more difficult to accept future exercises being carried out in the absence of any legal framework, especially if citizens’ conventions are expected to flourish and produce political and normative outcomes, even indirectly. However, this legal framework remains to be built, as several observers of the Convention have pointed out.

Before continuing, we note that this first assessment is far from being exhaustive. Certain questions have been deliberately left aside, such as everything relating to the publicity of the Convention’s debates and its limits, the potential wider involvement of society in this exercise, and the relations and complementarities between elected representation and citizen deliberation. These questions will only be touched upon here and will have to be analyzed at a later stage. Similarly, we will not comment the ongoing process between the convention and the Government. Indeed, following the publication of the CCC’s proposals in June 2020, the executive branch decided on a follow-up process in which the members of the Convention were invited to join along the Ministry of Ecological transition and various actors. Finally, this paper does not seek to determine whether the government and the President have honored their commitments: it is simply too early to judge.  At the time of writing, a bill is under consideration and will be presented to the Council of Ministers in February, the last step before submission to Parliament. The Convention’s mandate provides for a « response to the response » from citizens allowing them to express their views on the Government’s response.

1. The singularity of the French citizens’ convention 

1.1. Crises, random drawing and deliberation

Citizen’s conventions differ from other instruments of participatory democracy by their highly deliberative dimension. They are participatory, of course, in that they involve the direct participation of citizens drawn by lot. However, they are even more so deliberative, because they are fundamentally based on open discussion between these citizens on the basis of quality information and with the aim of formulating proposals[3].

Indeed, the aim here is not to gather the citizens’ spontaneous opinions on a given issue but rather to ensure that the citizens, in light of plural and diversified information, debate together and agree on proposals to be submitted to public decision-makers. Like consensus conferences, these conventions seek to answer a question that suffrage cannot resolve: rather than the majority opinion on a problem, what is the opinion that results from informed judgement? In other words, what do citizens propose when they have taken the time and trouble to inform themselves and deliberate together? The ballot is individual and deliberately blind to voters’ motivations; the outcome of citizen deliberation is instead the result of a process of collective investigation of the issues.

From this point of view, the CCC is not a radical innovation. It has notable precedents abroad, in a wide variety of formats and on a wide range of topics, as will be discussed below. These include the Canadian experiences (in British Columbia in 2004 and Ontario in 2007) and the Irish examples, which have often served as a reference for the initiators of the CCC[4]. Even in France, the model is not entirely new: in addition to the consensus conferences that have been held since 1998 (on GMOs, vaccination…) in a rather different form, citizens’ conventions at the regional level were also organized in March 2019 as part of the Grand Débat national[5]  (comparable in principle but very different in terms of method and outcome.)

What are the reasons that push politicians to solicit randomly drawn “citizen-deliberators” to collectively elaborate recommendations on a given subject? Because it is indeed political actors, at least in the cases just cited, who have taken the initiative to convene such temporary assemblies (in general, the government or Parliament, or, as was the case in France, the head of state[6]).

Circumstances play a part in this motivation, of course. As in the Icelandic experience of 2011 (the 'pots and pans revolution’), the Irish assemblies followed the financial crisis of 2008 and the political upheavals it led to. Similarly, in France, the CCC initiative[7] took shape following the Yellow Vests movement, which arose from a rejection of tax raises on motor fuel (a carbon tax) but was prolonged by strong criticism of the representative system (with the demand for the Citizens’ Initiative Referendum in particular.) In each of these cases, a major crisis shook the political balances in place, or even the foundations of the regime, pushing those in charge to seek new sources of legitimacy or new paths to consensus.

But circumstances do not explain everything. And, in a way, they only highlight certain structural weaknesses of representative democracies: the people in their diversity have often been contrasted to their elected representatives, who are regularly described in moments of crisis as a small elite turned in on itself and essentially concerned about its perpetuation. The idea then developed that representation only very imperfectly fulfilled the democratic promise of a government « of the people, by the people and for the people » (Article 2 of the Constitution), or even betrayed it. In response to this dissatisfaction, the search, if not for a “more effective self-government” but for a “representative government more attentive to society”, more permeable to citizens’ “demands of involvement and intervention”, was regularly promoted[8].

Citizens’ conventions give shape to this quest[9]. They give voice to those who are rarely heard in representative institutions, or who are confined to the sole role of being voters. The use of random drawing is the determining factor in this ambition: unlike elections, it guarantees that anyone can access these assemblies, including citizens who are usually far removed from the public arena. At the same time, it revives the idea that the power of the people is « the power of those who have no more right to govern than to be governed[10].”  It also recalls the promise contained in Article 6 of the Declaration of the Rights of Man and of the Citizen[11]:  »All citizens have the right to participate personally, or through their representatives, in the formation of the law."

By calling for informed deliberation, however, these conventions go beyond a purely participatory ambition. They seek to welcome citizens’ reasoning and judgement, considering that they are likely to complement those of elected representatives: indeed, citizens evaluate each option not only in light of general information, expertise and vested interests, but also in light of their practical and personal experiences, highlighting difficulties and expectations that are too often overlooked by representatives and the executive. It is both collective thinking and skills of experience that intersect in citizens’ exercise of deliberative judgement.

The French experience conducted between July 2019 and June 2020 (including nine months of work during which citizens met seven times between October and June[12]) is exemplary of the aforementioned motivations and characteristics, which can also be found in most foreign examples. However, the French experience differs from them by the nature of its mandate and the organization of its governance. These are what make it truly innovative.

1.2. Le mandat 

In his engagement letter, the Prime Minister mandates the CCC to « define structural measures to achieve, in a spirit of social justice, a reduction in greenhouse gas emissions of at least 40% by 2030 compared to 1990. » He goes on to say: “At the end of its work, it will publicly send a report to the government and to the President of the Republic on its discussions and on all the legislative and regulatory measures it has deemed necessary to achieve the objective[13]”.

The President announced on 25 April 2019: “What comes out of this agreement, I commit to do, [and] will be submitted without filtering either to a vote in Parliament or to a referendum or through regulations with direct application.[14]”

To sum up, the members of the Convention were therefore asked:

• to take up the fight against global warming as a whole and not sectoral issues or controversies;
• to formulate structuring proposals and not a general vision;
• to work in a spirit of social justice and not solely with the focal point of the ecological transition;
• and to propose measures with a completed legal form (« legislative or regulatory measures ») and not mere recommendations.

A final feature of the contract between the mandator and the CCC members was that the former undertook to submit the proposals resulting from their work to legitimate democratic bodies without distorting them (« without filtering »).

None of these characteristics are completely new, but their close combination marks the originality and singular ambition of the French experience. Let us begin with political commitment. Because it provided citizens with a guarantee of their usefulness, this promise of the 'unfiltered’ measures was certainly one of the keys to their mobilization, both during recruitment and in the subsequent process (it is worth noting the remarkable assiduity with which its members followed the debates[15]). CCC members felt a sense of mission: they internalized a sense of responsibility. The presidential pledge (reiterated and further elaborated upon during his participation in one of the CCC’ sessions in January 2019[16] ) was also decisive in terms of the public attention given to the Convention’s work by the media, stakeholders and the general public.

In foreign experiences, the mandator had also made commitments to citizens’ assemblies, but these have often taken quite different and, above all, vaguer forms. Only the Citizens’ Assembly on Electoral Reform, held in British Columbia (Canada) in 2004, had a horizon comparable to that of the CCC: its 160 citizens were to propose electoral reform that the provincial government committed to deliver to voters without filter through a referendum[17]. In Ireland, for the second assembly, the government committed[18] to respond to each of the citizens’ recommendations and to publish a timetable for the implementation of a referendum, subject to the government’s own approval of the recommendations. There was therefore a “filter.”

The other essential component of the contract with citizens is the question put forth to them and the nature of the response expected from them. In most previous experiences, whether in France or abroad, members were asked either to settle an established controversy (GMOs, right to abortion…), to express a general vision on a subject or to formulate the main principles likely to guide a policy or a constitutional revision. It is much rarer to ask citizens drawn by lot to define the ways and means of a public policy as a whole. Rarer still is it for them to be asked to submit proposals in a completed legal form, meaning they are practically put in a position to « pre-legislate. » Only the Icelandic Constituent Assembly of 2010–2013 seems to have borne such an open and ambitious mandate, even though the project of providing a political community with a new Constitution may seem both more fundamental and less technical than that of defining a country’s climate policy. In other cases, such as in Ireland, the second convention had to answer multiple questions on subjects that were not directly related to each other: the legalization of abortion, the ageing of the population, climate change, and referendum and parliamentary reforms[19]. As we can see, some of these questions were very broad and, given the often very limited time available, required answers of a fairly general nature. Others, on the contrary, were very specific and at the heart of relatively structured societal controversies: these did not call for special effort in terms of proposals and political imagination, but rather arbitration conducted by the citizens.

The French experience shows that an assembly of citizens drawn by lot and asked to present normative proposals aimed at reforming a public policy can do so in a fairly precise manner. Whatever one may think of the CCC’s proposals, its final report attests that, in nine months and seven working sessions[20], the « 150 » handed in a certified copy of a large part of their work conforming to the term of their mandate[21]. Contrary to what is still a widely held view, it is therefore neither necessary nor desirable to confine this type of initiative to « small societal issues », as it is sometimes conceived, or to a pure exercise in arbitration on established controversies. If deliberative democracy is to be taken seriously, citizens must be shown that their collective intelligence is being solicited on issues that matter and on which the construction of a collective consensus is by no means self-evident. For this, too, it is essential that the mandator make clear commitments to the outcomes of the exercise. Nothing would be more alien to these lessons than to replicate the exercise on issues of lesser importance and within the framework of a contract that does not clearly specify its political outcome.

1.3. Governance 

The second singularity of the CCC lies in the organization of its governance. While most foreign experiences are characterized by relatively light governance, the CCC was endowed by its initiators with a large governance committee. In both Scotland and Ireland, a qualified person chaired the Convention with the support of a secretariat to organize its work. This person was sometimes a full member of the Convention, as in British Columbia, in which the Chair was the 161st member of the assembly. In Ireland, the President, who was also a judge on the Supreme Court and appointed by the government, was the 100th member. She was supported by a ‘Steering Group’ with a secretariat and other members of the Convention elected to serve on it[24]. This group met once a month to oversee the work program and validate the speakers. In addition, the Chair actively participated in drafting the proposals by interpreting the will of the citizens, with the final result requiring a consensus between her and the assembly.

In the French case, not only were the Convention’s debates not chaired by a single outside personality, but they were not chaired at all: it was the facilitators[25] who were responsible for ensuring that the working sessions ran smoothly. The general organization of the Convention was overseen by a Governance Committee which did not take part in the debates, but merely introduced and concluded the sessions through its co-presidents and the general rapporteur[26]. This committee brought together 17 personalities[27] : two co-presidents, a general rapporteur, experts on climate, participatory democracy, and the economic and social field, two representatives of the Ministry of Ecological and Solidary Transition and two members of the Convention drawn by lot from among the volunteers[28].

Appointed intuitu personae, the members of this committee sat and spoke in their personal capacity and not as part of an organization. "Autonomous in carrying out its tasks[29], the Committee was independent both of the political authorities and of the EESC, which hosted the Convention. At weekly meetings, the Committee was responsible for steering the Convention, drawing up its program and working methods, and protecting the independence and respect for the will of the Convention.

This plural composition ensured that none of its components were in a situation of hegemony; controversy was present within the committee. Moreover, decisions were taken by consensus and in a collegial manner[30], obliging each member to argue, to convince the others, and the co-presidents to seek, in case of disagreement, a path of compromise. It was quickly decided to involve the representatives of the consortium of facilitators in the meetings, but without allowing them to take part in the decisions, since their role was to implement the objectives set by the Governance Committee. Nevertheless, they contributed greatly to the development of the work programs and provided high-level professional expertise.

Finally, the dialogue between the Governance Committee and the members of the Convention was intended to be constant. Two Convention members participated in all the meetings, and an Internet platform reserved for them enabled them to question the Committee.

This mechanism was reinforced by a college of three guarantors[31] appointed by the presidents of the National Assembly, the Senate and the EESC. Their role was to « monitor » the Governance Committee, ensure compliance with the main principles governing the exercise, certify and proclaim the results of the votes and act as an appeal body for citizens in the event of a contested decision. They also acted as advisors to the Committee when it encountered ethical concerns.

This organization can of course be open to criticism. For example, some observers emphasized an excess of formalism and rules, where they saw a mark of the « French spirit » as opposed to freer Anglo-Saxon practices that left more room for spontaneous debate. The fact remains that these rules have been quite effective in guarding against various risks of influence: the influence of governance on citizens (by rigorously separating the governing body from the deliberative body) and external influences on governance (the diversity of its members and the numerical balance of its composition protecting the whole.) Indeed, some may have been surprised that the organizers encouraged citizens to go outside the CCC framework and meet with stakeholders. In contrast, this was forbidden in certain foreign experiments. The governance of the CCC has also led to conscious and deliberate debate about methodological and procedural choices that elsewhere were sometimes left to the discretion of facilitators and/or personal chairmanship, even though these choices may have significant consequences for the course of the exercise. This is why we believe it is imperative that future Citizens’ Conventions have a comparable governance, both in its structure and its principles.

2. Problems, dilemmas and solutions

In addition to this organic singularity, methodological choices in the face of various problems are what characterize the French experience. For each of these choices, several options were possible with a considerable impact on the final result. We present four of them here: the composition of the panel, the structuring of the work, the link with expertise and the use of voting. One or more dilemmas correspond to each of them, which we try to highlight. For we are convinced that while some of the answers to these difficulties are specific to the CCC, these difficulties will arise again in the future.

2.1. The composition of the panel

On 25 April 2019, the President of the Republic announces that the CCC will bring together 150 citizens drawn by lot. Two months later, the Prime Minister specifies in his engagement letter that these citizens must be both drawn by lot and « representative of the diversity of French society. » The executive branch wishes to combine the virtues of chance and representativeness. Each of these two orders raises particular difficulties: on what basis should the draw be organized? How should criteria of representativeness be defined? Above all, their combination is not self-evident: how can the random draw alone produce a representative assembly? How can the legitimacy tied to chance and the legitimacy tied to representativeness be articulated? To overcome these dilemmas, several difficulties had to be resolved.

First of all, it was decided that the drawing of lots would be carried out based on telephone numbers and not on electoral lists, as is the case for juries. The guarantors of the national Grand Debate had made the same choice for the regional citizens’ conferences a few months earlier[32]. Electoral lists have the disadvantage of excluding people who are incorrectly registered or not registered at all, or more than 9.5 million French citizens who thus don’t express themselves during elections, based on data from 201433. The random generation of telephone numbers is more inclusive and offers the advantage of quick and easy contact. Between August and September 2019, Harris Interactive[33] drew on a randomly constructed directory of 300,000 telephone numbers[34] to make calls and offer those with faltering political engagement to participate in the CCC.

This drawing of lots met the president’s first requirement of randomness but did not ensure that the group would be representative of the diversity of French society. It was therefore necessary to define selection criteria by which the participation of certain volunteer citizens drawn by lot could ultimately be refused.

It should be emphasized that this sought representativeness could not be of a statistical nature: in relation to a population of 67 million individuals, a sample of 150[35] is much too small. However, since the aim was to reflect social diversity and not majority of opinion, it was possible to seek representativeness of a descriptive, if not figurative, nature, i.e., to reproduce in the assembly the main socio-demographic balances observed in the general population. To achieve this, it was still necessary to determine the indicators deemed to be the most structuring to describe French society today. The question rested on how to determine what best characterizes an individual’s social position.

The Governance Committee finally retained six criteria: gender, age[36], level of diploma, socio-professional category, type of territory (urban centers, isolated municipalities, impoverished so-called “priority districts”, etc.) and region of residence (Ile-de-France, Occitanie…) In doing so, the Committee hypothesized that the social position of an individual could not be defined by two or three simple indicators (age, gender and socio-professional category, for example), but by a multiplicity of factors, among which the level of education (a fairly good reader of many other inequalities) and location in the territory (identified by two different criteria) also had to play a role. It would have been tempting to add others such as income, wealth or type of social mobility if the size of the final sample had not quickly limited the exercise. The multiplicity of criteria and the discussions that accompanied their definition testify to the difficulty of establishing a description of today’s society, particularly in the wake of a movement such as the Yellow Vests, which had sparked intense debates between a fairly classic approach to social inequalities and a more geographical and territorial approach.

For each of the categories selected, the Governance Committee set itself an objective corresponding to the known balances in the general population. The CCC thus brought together 26% of citizens without a diploma or with only a primary school or middle school certificate, 8% of inhabitants of the Grand Est region, 11% of inhabitants of the aforementioned “priority districts”, etc.[37] On arrival, the citizens initially drawn by lot formed a 'France in miniature, ' combining selection by chance and by criteria of representativity.

The representation of certain categories was, however, an exception to the rule. This was the case for overseas citizens, those living in very precarious circumstances and farmers, or a total of eleven members. For citizens overseas, [38]  the aim was to alleviate certain practical difficulties (moving people from French Guyana or Mayotte for five three-day sessions might deter more than one.) Concerning the second group, it was hard to imagine being able to use telephone calls to reach out to and convince men and women who were homeless and/or living in extremely difficult conditions[39]. In regard to farmers, these are professions for which the commitment to regularly come to Paris could be problematic, and the probability of contacting a farmer via the random draw was low[40].

Conversely, the Governance Committee did not wish to exclude actors from sectors closely tied to greenhouse gas emissions (lorry drivers, airline pilots, etc.) It thus made the choice not to demand a form of axiological neutrality from such citizens and to acknowledge where they came from, thus sometimes bearing the interests of their profession, and possibly holding political convictions.

This question of political attitude arises more specifically, however, in relation to the main subject under discussion. How could it be ensured that the plenum would not be made up of an overwhelming majority of individuals committed to the fight against global warming, given that participation in the Convention was based on volunteerism and was therefore more likely to motivate those who were already devoted to the cause? In order to overcome this difficulty, an additional criterion could have been included: prior examination of participants’ positions on the issue of global warming. In the United Kingdom, this was used as a recruitment criterion and, in the end, 17% of the members recruited said they were not very or not at all concerned about the issue, a proportion in line with what was observed in the general population. In France, although some citizens declared themselves to be “climate change sceptics” during the first session, this was not a selection criterion. A study carried out by researchers from the Paris School of Economics compared the positions of CCC members (during the first working session) with those of a representative sample of 1,003 French people[41]. To the question « Say how important environmental protection is to you, on a scale of 1 to 10, » 25% answered 10 in the national sample versus 45% in the Convention, a significant difference. However, there were broadly similar results on the perceived seriousness of climate change: 76% of CCC members considered that its consequences will be extremely painful compared to 71% of the French population, a difference that falls within the margin of error inherent in a sample as small as the CCC[42]. Moreover, support for certain measures is quite similar in the two groups: 87% of CCC members consider it very or somewhat desirable to promote the use of low emission or shared vehicles, compared with 79% of the French population. And 71% of the French say they would vote yes in a referendum on banning the advertising of polluting products, one of the CCC’s flagship measures.

Incorporating a criterion of attitude towards climate issues, as the British did, would allow the selection bias to be circumvented. However, this would raise a number of methodological questions, including which survey or poll to use as a basis for selecting candidates. Another way to get around the selection bias in volunteerism would be to make participation in the convention compulsory, as is the case of jury duty. Such an obligation does not exist anywhere abroad and raises other issues: how would a dynamic and productive deliberation be held with people who do not wish to be there and who potentially challenge the process itself? And how do we measure the appetite for debate, for participation? Wouldn’t participation be distorted if it were imposed on people who reject it? Moreover, it seems to us that this could only be done if the conventions were perfectly embedded in the institutional framework and framed by law.

This question of prior positioning as opposed to volunteerism is crucial. On an issue such as global warming, there is not such a high level of conflict in France for it to be impossible to assemble a properly representative panel absent more precautions; the surveys of the last twelve months even underline the growing concern of the French on this subject. But what about a more technical or much more divisive subject? How can it be ensured that the vast majority of participants are not supporters of one side or the other, or that only the most qualified people feel legitimate enough to deal with a particularly complex subject? The aim here is to ensure that the recruited audience is not only geographically and sociologically but also ideologically representative.

What lessons can be learned from these initial choices? Firstly, it seems to us that drawing lots through the random generation of telephone numbers is a robust and more inclusive method in any case than drawing lots from electoral lists. Secondly, if we want to control for potential bias linked to volunteerism, we think it is useful to add a measure of attitudes on the chosen subject to the selection criteria. Thirdly, if citizens are to produce a normative consensus that can later take the form of legislation, it is important that they be representative of social diversity. To this end, volunteers drawn by lot must be selected on the basis of relevant socio-demographic criteria, yet without such criteria being fixed by law.

2.2. The structure of the work

The Convention was initially scheduled to deliver its conclusions in early February 2020, in four months and five sessions. Due first to the strikes in December 2019 and then to the lockdown in mid-March, its work was finally spread over nine months, with seven three-day sessions and multiple activities planned in between sessions. Nevertheless, the issue remained delicate: how would 150 non-expert citizens be enabled in record time to acquire the necessary knowledge to make proposals to the executive a few months later for « legislative and regulatory measures » likely to reduce our greenhouse gas emissions « by at least 40% » over the decade, and all this « in a spirit of social justice »? This implied a powerful process of raising the level of citizens’ competencies.

Not only did everyone have to familiarize themselves with the mechanisms of the greenhouse effect and the factors of greenhouse gas emissions, but also had to bear in mind the situation of public policies on the issue, the obstacles they might encounter, the weight of the contributions required from each party (the State, companies, households, etc.), the implications of a social justice requirement in the distribution of efforts, etc. Moreover, in order to formulate proposals of a legislative or regulatory nature, CCC members had to familiarize themselves with the levers and constraints of public action.

A first step therefore consisted in offering a common information base, [43] followed by a segmentation of the general topic into five major themes[44] corresponding both to the main sources of greenhouse emissions and to facets of common day-to-day life through which each member could assert its experience. Two options were considered on this basis. The first was to study the five themes one after the other, using a so-called sequential method, which had been favored by the Irish conventions. The whole plenum would then have worked on the same subjects and at the same pace: one theme per session with, each time, a sequence of information, a sequence of deliberation and then a sequence of proposals. This was the most egalitarian option, the one that ensured that all citizens covered all the subjects. It also had the advantage of bringing the 150 together from start to finish.

However, it came up against two difficulties. The first was that it did not allow the time for non-expert citizens to accumulate enough information to deal with the complexity of the issues at stake on each of the subjects; it was even feared that, as the acquisition of information was cumulative, the first subject would not be treated as well as the second, the second as well as the third, etc. The second: how could one ensure that the decisions taken on the first weekend would be consistent with those of the last weekend without the latter’s replacing the former’s?

To overcome these difficulties, the Governance Committee chose to structure the work simultaneously. The 150 were divided (by drawing lots) into five working groups (« eating », « travelling », « housing », « consuming » and « working & producing »), each of which was responsible for specializing in its theme, hearing the experts and exploring the issues at stake before formulating proposals and submitting them to the plenum.

However, this method had symmetric drawbacks to the previous one. Because it risked splitting the Convention into five parallel conventions[45], this division of work called for compensations in order for the group of 150 to exist as a whole. It was thus decided that at each session the working groups would share their thoughts and progress among themselves, that together they would hear a number persons from outside the CCC and that, in all, between one third and one half of the time would be devoted to work in plenary. This made it possible to detect inconsistencies or redundancies between the various groups’ proposals, but above all it made it possible for the final decision that would be taken by 150 people to reflect the Convention’s will. Thus, during the last sessions, the work was essentially done in plenary sessions to allow everyone to take ownership of the proposals before they were voted on as a whole.

In addition, a sixth group with representatives from each of the other groups was to be set up to deal with cross-cutting issues such as the possible revision of the Constitution or questions of funding. But it was quickly abolished, at the request of some members who refused to allow « several speeds » within the assembly itself. Eventually, volunteers drew up proposals on constitutional issues. They were worked on by all members before being submitted for adoption[46]. As for the financing of the proposed measures, it should be reminded that this was not a prerequisite of the CCC’s initial written mandate, even though the Prime Minister Édouard Philippe, in his introductory speech on October 4, 2019, had insisted on the importance of this dimension. It seemed to us, however, that it was particularly through funding issues that citizens could determine the spirit of social justice that they intended to propose to the executive: « who should be called upon? » and « to what extent? » are indeed crucial questions of equity. However, the Governance Committee was very hesitant about how to work on this issue. Between expressing preferences, working on the basis of scenarios developed by experts and a dedicated working group, it never really found the right option. It seems to us that, on this issue, it would have been necessary to proceed sequentially, not simultaneously, and to devote a weekend, at the end of the process, to these questions[47]. Nevertheless, it should be noted that the citizens have formulated guidelines on financial matters in their final report.

Despite its limitations, its procedural requirements and the risks of fragmentation it posed to the Convention, the simultaneous method proved to be quite appropriate for the scope of the mandate, the interdependence of the subjects, and the level of competence it required from citizens. By organizing a division of labor for the examination of the various themes, it also enabled citizens to go quite far in the level of detail in their proposals. However, it proved to be unsuitable for the management of issues that cut across different themes.

2.3. The link with expertise

Contrary to what a certain democratic spontaneity suggests, citizen deliberation does not in itself convey any revealed truth. If it is not enriched by any input of external information, it can quickly lead to generalities that are of little interest, or even turn into a “café du commerce” discussion[48].  In order to rise to the challenge of issues as technical as global warming, citizens must be able to sharpen their judgement through contact with multiple sources of information and expertise. However, these external inputs also expose them to all kinds of influences. This is the third problem that the Governance Committee has faced: offering Convention members a wide range of expertise while ensuring that they preserve their autonomy.

Although very broad, the scope of information useful for informed deliberation was narrowed by a few boundaries. The executive’s mandate, the terms of which had been communicated to citizens as soon as they were recruited, did not, for example, call for a discussion on the adaptation to climate change, but on its limitation, and it pointed to national, rather than global or European, policies. Above all, it did not question the reality of the phenomenon or its anthropogenic causes, but rather set them as a starting point. The aim was therefore not to present members with the opinions of climate sceptics, but first and foremost to present the main conclusions of the international scientific consensus.

It is these conclusions that the members of the convention appropriated, first by familiarizing themselves with the « information base » and then by hearing from climate experts (climatologists, economists, etc.). These various contributions also enabled them to assess the state of climate policies in France and the instruments that public authorities have adopted in this area, such as definitions, key figures, etc. Many people still talk today of a « slap in the face » to describe what they felt when they realized, during this first stage, the seriousness of the issues at stake and the task that was theirs.

In addition to this introductory sequence, the Governance Committee wanted to enable citizens to maintain this link with scientific and factual expertise throughout their work. Thus, and this is another peculiarity of the French Convention, a group of fact-checkers[49] composed of volunteer academics from several institutions and specialized in different fields had the task of responding, as quickly as possible, to all the members’ questions and verification requests[50]. They were mobilized to verify statements read here and there or those said at a hearing. In this sense, they served as a kind of internal counterweight within the CCC itself to the expertise that was offered elsewhere.

This was all the more important given that citizens then met or heard from nearly 130 speakers from the worlds of universities, business, NGOs, local authorities, trade unions, administration[51]… These speakers were solicited partly at the citizens’ request, partly by the Governance Committee to ensure pluralism of information. Indeed, the aim here was to enrich the members of the Convention with a panorama, not only of the scientific knowledge available, but also of the sticking points, proposals and existing initiatives. These meetings took various forms: hearings before the plenum or in working groups, cross-examinations, “speed dating” during the third session[52]… Finally, another information channel completed this system: a platform for external contributions, which collected several thousand contributions from legal or natural persons outside the CCC, contributions that were regularly summarized[53].

In light of these various contributions, CCC members gradually developed and formalized their proposals. To advise them in this process, the Governance Committee made a « support group » available to them from November onwards. Appointed intuitu personae, these fourteen experts[54] were chosen for the diversity and complementarity of their fields of expertise. At the end of each session, the citizens could instruct them to test their ideas, quantify their impact, make international comparisons, etc. The experts’ answers were then submitted to the citizens, who decided whether or not to take them into account. This technique allowed this support group’s work to be rigorously framed, as it was acting at the citizens’ request and was not to deviate from it. In order not to expose personal convictions, they were asked to work in a collegial manner and, if necessary, to explain their differences.

Finally, the Prime Minister’s engagement letter stated that « technical and legal support would be provided to ensure the legal transcription of the proposals. » A so-called « legistic committee » of six jurists[55] was therefore set up. It first helped the citizens familiarize themselves with the main principles of law: hierarchy of norms, drafting constraints, law-making, competent institutions, etc. Next, it transcribed the will expressed by the citizens as faithfully as possible in law terms, again in a collegial manner. These transcripts were then submitted to the citizens, who had the option of rejecting those they found did not conform to their intention. In the final report, the validated transcripts accompany the initial text as written by the citizens (the reader can thus distinguish between the citizens’ own production, on the one hand, and, on the other hand, the legal transcripts that were deemed faithful but were not written by the citizens themselves.) The aim of this committee was therefore not to provide the citizens with legal competence. These transcripts do, however, enable the members of the Convention to engage in dialogue with the government one day and with parliamentarians the next; they are, in a way, a guarantee of the « unfiltered » proposals.

Expertise was thus closely involved in the process, in particular with the creation of the « support group. » Experts were made to serve the citizens, not as holders of any kind of truth, but as a cabinet of advisers like the cabinets of elected representatives or politicians. Thus, the citizens, in the face of the stakeholders, the support group, the drafting committee or even the Governance Committee, remained the ultimate decision-makers, and the experts their subordinates.

Far from the image of people who are poorly informed and easily influenced, the members of the Convention have regularly asserted their full sovereignty, including by refusing or even dismissing certain expert views. The issue of the « carbon tax » is interesting in this respect: three external speakers have successively presented it as a major tool in the fight against global warming, while proposing to reform it to make it fairer and to ensure that its revenues are more directly allocated to the ecological transition. The members of the Convention refused to enter into this debate, believing that it was not up to them to rehabilitate or reject a tool that had divided society a few months earlier[56]. This autonomy in the face of expertise has also enabled citizens to be sometimes freer and more persevering than experts and not to refrain a priori, in the name of legislative or financial constraints, from exploring certain proposals when experts may tend to be self-restrained.

All in all, the CCC will have been characterized by a stronger and more integrated use of experts than in most foreign examples. Its mandate required it to put citizens in a position to make informed judgements on very complex and diverse matters. The use of experts was not optional for the CCC, but a condition for its success. However, in order not to subject non-specialist citizens to their direct influence, it was necessary not only to ensure the pluralism of these external inputs, but also to methodically structure the relationship with experts. Firstly, by maintaining a strong hierarchy between the latter and the citizens to whom the procedures always gave the last word (cf. the vote of adoption or rejection on legal transcripts) and the means to reject a point of view (cf. the example of the carbon tax.) Secondly, by granting the possibility of recourse to third-party verification (cf. the fact-checkers) and the opportunities to confront different interpretations of a problem (cf. the twenty-three cross-examination hearings[57]). Finally, by strictly framing the work of the experts closest to the citizens through mandates formulated by the citizens themselves (cf. the support group.)

2.4. Use of the vote

Citizens’ deliberation exercises are unique in that they favor conditions of informed, free and open exchange, where everyone is likely to change their mind in light of the arguments and information received. This is why arbitration tools, such as voting, are often kept at a distance. Indeed, voting « serrates » debates: it creates winners and losers, if not antagonizes « blocks » and gives rise to campaign strategies. Its regular use risks increasing the conflictual nature of exchanges, reducing the mobility of opinions and, ultimately, corrupting the spirit of deliberation.

Nevertheless, voting also has virtues that can be very useful. A secret ballot gives a voice to the silent, to those who rarely speak in public. It requires no reasoned justification and remains blind to individuals’ motivations. In this way, it serves as a counterweight to those who speak « loudly » by giving a synthetic representation of the assembly where « everyone counts and everyone counts as one », to use Bentham’s formula. Above all, the vote makes it possible to decide, to formally adopt positions and thus to certify a collective or at least a majority will. For the CCC, this information was all the more important from the perspective of the promise of the « unfiltered » submission of proposals, as it allowed the proposal’s support to be weighed and its level to be clarified.

The dilemma rested therefore on how to use voting without stifling the deliberation. To overcome it, several choices were made: a) not to hold a vote at the beginning of the work, b) to announce from the beginning that a vote would be held at the end to validate the proposals and the report as a whole, while making room for dissenting opinions, c) to structure an amendment procedure so as not to impose the choice between adoption and rejection.

The Governance Committee announced early on that a vote would be held at the end of the process to adopt the CCC’s proposals and its report as a whole.

It also announced that those who were in the minority in these votes could draft dissenting opinions (referred to as « minority opinions »), which would be appended to the report[58]. This proposal was guided by the concern not to leave minorities in a position of mere « losers » by recognizing the value of their arguments and making them known to the public. In this way, it was hoped that the vote would not brutally devalue the implicit principle of the deliberative process. Moreover, it allowed everyone, especially the legislator, to know all the positions of this « France in miniature. »

These two announcements had several practical consequences that became apparent as the process unfolded. The first was of reassuring those who were in disagreement with the dominant opinion in the working groups and who did not always have the necessary audacity to express it[59]. They knew that, when the time came, they would be able to distance themselves from the majority and put forward their arguments. In a way, even though there were ultimately very few minority opinions appended to the report[60], this possibility acknowledged the value of everyone’s arguments, regardless of their « popularity. »

The other consequence was that, from the first session onwards, CCC members knew that, at the end of the work and via their ballots, they could sort through the proposals and possibly reject some of them. Therefore, it seemed less necessary for them to prioritize them before the vote. Thus, during the third session, the Governance Committee offered the members of the different working groups to prioritize their proposals by giving them a rating. The aim was to identify those that needed to be examined in greater depth in order to allocate more time to them. This procedure did not produce the expected results, with some considering that it was too early to decide, some simply refusing to participate because they felt that all avenues had to be kept, and others getting around the problem by grouping the proposals into categories so broad that they allowed almost all of them to be included! The CCC members in fact refused to enter into this logic of funneling proposals.

As a result, between this refusal and a vote organized at the very end of the process, they could defer the need to prioritize their work and slightly relax their efforts to find a consensus in the course of the discussion. They could thus devote themselves more freely to exploring a large number of proposals.

Nevertheless, these proposals drawn up in the working groups had to be discussed by all before being submitted to the plenum for a vote. Of course, time for debates was provided, but it would not have allowed the various positions to be taken into account and the proposals to be further enriched in the absence of an amendment procedure. Thus, after the sixth session, this procedure enabled everyone to propose modifications, alternative methods or add elements (for example, many amendments were aimed at clarifying the application of the proposals in overseas territories.) Under the condition of being deemed admissible and supported by at least twenty members[61], these proposed amendments were then put to a vote by the full CCC membership. But, here again, to ensure that the vote did not come down to the opposition of camps, conciliation procedures were put in place prior to the vote to allow the working group carrying the proposal and citizens wishing to modify it to find common ground before voting on the substance, or to allow the bearers of contradictory amendments to seek a common solution. Out of 57 amendments deemed admissible, 53 were adopted by the plenum.

All of these choices made it possible for deliberation not to be stifled. On the contrary, they contributed to the richness and diversity of the CCC’s final report. Conversely, they partly explain why CCC members were not very selective. The results clearly show a very strong consensus on the proposals submitted to the plenum: 147 out of 150 proposals were adopted with approval rates between 85% and 100%; two were adopted at less than 60% (59.7% for reducing speed limits on highways and 58% for amending the preamble to the Constitution); only one was rejected (the 28-hour work week.) Some commentators have inferred that Convention members were « manipulated » or uniformly « militant, » or both. This ignores the dynamics not only of the deliberation but also of the plenum. And it is to ignore the various possible meanings of the “yes” vote.

The high level of agreement reached in the final vote is primarily the result of the deliberation and the consensus it produced. It would have been surprising if the citizens had perfunctorily adopted or even rejected their own proposals, which were the result of a long process of adjustment among themselves and amended according to their wishes. It should be recalled here that, unlike parliamentary assemblies, the Convention was not structured from the outset by rival groups, and its members were not assigned to partisan factions whose power relations are generally reflected in the outcome of votes. The citizens’ « yes » means first of all « yes, we agree with ourselves » (and perhaps even for some « we do not want to call into question the Convention and its work. »)

Two other factors may have played a role in this lower selectivity and strong final consensus. The first: some citizens may have felt legitimate to express an opinion and make proposals but not necessarily to take a firm and final decision. Their « yes » could then mean « yes, we want the Parliament to examine this proposal, » and not « yes, we want the Parliament to adopt this proposal. » In a way, this attitude was consistent with the 'no filter’ promise, meaning transmission and not a priori adoption (in any case, the executive is not the legislature!) There was nothing to prevent a wide range of proposals from being transmitted and leaving it up to the legislature to sort them out.

The second factor: the rejection of selectivity may have resulted in part from the simultaneous structuring of the work (see above) and the specialization of working groups. Indeed, during the vote, despite the efforts of the facilitators to enable the pooling of proposals, these remained identified with groups. Thus, the approval of citizens who were not members of the working group can also be interpreted as a sign of confidence in the group that had been working for several months on a measure. The « yes » can then mean « yes, we rely on those of us who have specialized in this issue. » Only the proposal on the reduction of working time provoked a broad debate within the Convention during its final session before being rejected by the plenum: different visions of society confronted each other with a high quality of argumentation.

Voting was also sometimes used to resolve problems linked to exceptional circumstances (reorganization of the agenda, expression of preferences on dates, etc.) or to validate unforeseen collective choices. This was the case when, in April 2020, in the middle of the first lockdown, the CCC convened remotely to discuss a collective contribution to the government’s recovery strategy. There were concerns that the accelerated timetable for policy decisions would not allow the executive to take the CCC’s proposals into account, which, due to the circumstances, had not yet been formally adopted. At the same time, the CCC’s members felt that some of their proposals combined ecological, economic and health benefits that were particularly valuable in this context. The remote meeting on 3–4 April was therefore devoted to a number of questions: Should the CCC be involved in this debate? If so, should it do so by communicating some of its as of yet unadopted proposals directly to the government or through a more general public statement? Finally, if it wished to impart some of its proposals, which ones should be put forward, in what form, and should they be made public?

The various votes that were organized finally led it to communicate some of its proposals to the executive without, however, making them public, and to publish a more general text on the role of climate policies in the context of the unfolding crisis[62]. This episode was a source of friction between both the members of the Convention and the Governance Committee and among the members of the Convention themselves. Quite a number of people took the floor to oppose some of the votes or the form of the proposed choices (whether the proposals should be summarized or whether they should be described in detail.) Nevertheless, in total, the majorities that emerged from the ballot boxes were clear and unambiguous (with the exception of the vote on the publicity of the proposals[63]). In this way, the vote opportunely demonstrated the gap between a sometimes very vocal minority and a more discreet majority, despite the many efforts made by the facilitators to make everyone’s voices heard.

3. the main principles of the citizen’s conventions

As we have seen, the Citizens’ Climate Convention was based on a number of procedural choices that were decisive for the outcome of its work. Moreover, the mandator’ commitment – the future of which will tell whether or not it has been honored – suggests that this exercise may have political consequences on the subsequent choices of the government and legislature: at the time of writing, several CCC proposals have already given rise to numerous political decisions or public announcements[64], and a bill is being prepared.

However, so far there is no clear legal framework to refer to in the event that the process is challenged[65]. This absence, which can be understandable at this stage of initial experimentation, exposes future conventions to all sorts of distortions, even instrumentalizations. There is therefore a real risk that their legitimacy will suffer as a result. What could be tolerable in the case of a first experiment of this scale will be much less tolerable the second time and even less so the third time.

What normative framework should guide these exercises? This is not just a question of their « institutionalization »: whether they should be entrusted to the EESC or to another institution is secondary. The main question is what minimum rules they must comply with in order to be valid, whatever the framework. It is not a question of codifying their methodology. It must remain flexible and adaptable. The problems that the CCC has had to overcome will arise in different ways in the context of another mandate. It is a question of identifying the main principles that should guide future conventions and be enshrined in law. We present five of them.

1. The mandator: we have seen that the status and commitment of the mandator are key in mobilizing citizens, as they determine the usefulness and seriousness of the exercise in their eyes as well as in the eyes of the public. For this reason, it is essential that the mandator be seen as the embodiment of the general interest and that it makes a verifiable and realistic commitment on what will be done with the proposals resulting from the convention. With this in mind, potential mandators are the executive, the Parliament or local executives concerning exercises that are themselves local: presidents of regions, departments or mayors; in general, those who have the competence to implement proposals that may emerge from the convention. In light of this, it follows that the EESC, which is a consultative assembly placed under the authority of public authorities, which does not express the general will and which has no normative power of its own, could not initiate future Conventions but could only act in the context of a delegation by the government, parliament or, where appropriate, local authorities (unless there is a citizens’ initiative[66]).
2. Representativeness: the law should at least impose a principle of public representativeness to which random drawing may be combined, as in the case of the CCC. The legitimacy of the proposals depends closely on this representativeness. Too many exercises in participatory democracy suffer from primarily mobilizing citizens who are already the most mobilized by public affairs or the most attentive to participatory processes, i.e., a section of society that already knows how to make itself heard and assert its preferences and interests. In our opinion, this representativeness must be both sociological and geographical and be based on criteria which, at a minimum, must include gender, age, level of qualification and territory of residence. It would be useful if it also took into account the attitude to the subject matter. In the absence of an obligation to participate, this criterion would make it possible to control the bias associated with volunteering, even in the event of a random draw. This requirement of representativeness can only be fully effective if everyone is given the material means to participate. This implies not only paying for accommodation and travel costs, but also providing for allowances, compensation for lost wages and possibly childcare reimbursements as foreseen in the CCC.
3. The impartiality and independence of the organizers: by organizer, we mean the convention’s governing body (governance committee or steering committee), which is responsible for organizing the work and the main methodological choices. We have seen that, in the French convention, the methodological choices made by the governance committee were instrumental. It must therefore be impartial with regard to the substance of the subject and remain so both in the formal and informal context of the convention. It is, however, illusory to demand from each member of the governance committee the axiological neutrality that can be expected from a judge, especially since it is useful to include people with proven experience and expertise on the subjects discussed within the convention, as was the case with the CCC. In the absence of the statutory guarantees surrounding the members of a court, distance from the convention members is more easily achieved here through the pluralism of round table discussion. In the case of the CCC’s governance committee, it was this pluralism that ensured that no party could be in a position of hegemony. The organizer must also be independent of the mandator. Once the mandate is over, it must have decision-making autonomy in the fulfillment of its mission, meaning it must be able to make decisions based on the interests of the Convention and not on the mandator’s preferences. Thus, the mandator cannot be the organizer and should not even be able to appoint the members of the governance committee. At the very least, it should share this power of appointment with other institutions, political opposition groups or be appointed by an independent committee. In the case of future conventions organized by the EESC, the principle of impartiality raises some difficulties. The EESC will certainly argue that plurality already prevails within its assembly, due to the great diversity of organizations represented (83 in total.) However, their political weight in the EESC and its bodies varies greatly. It is therefore by no means guaranteed that the domination of the strongest organizations will not have an impact on the choice of governance that will reflect the internal balance of this assembly. Consequently, it would seem prudent to us to ensure that at least 50% of the seats on the governance committee are occupied by persons from outside the EESC and that its presidency is itself entrusted to an outside person or shared with him/her in the event of a co-presidency. This precaution would also make it possible to more peacefully address subjects of prime interest to the organizations represented at the EESC, particularly to its social partners: work, income, social protection, etc. These are all issues on which the citizens of this country would undoubtedly have a lot to say. It is worth pointing out here, with regard to impartiality, that it must be both objective and subjective. In other words, impartiality cannot be based solely on the criterion of good faith: every effort must also be made to ensure that it is not doubted by the public. To this end, it is important that the governance of the Convention be permanently under the control of a college of guarantors who are the guardians of the principles proposed here and who serve as a recourse to the members of the Convention or to external actors in the event of a challenge. To ensure their independence, it is important that the guarantors be appointed neither by the mandator nor by the organizer. We suggest that they should be appointed by three courts with numerous statutory guarantees of independence: the Cour de Cassation and the Conseil d’Etat, the Cour des comptes or the Constitutional Council. [67]
4. Access to information and its pluralism: participants must have access to information that is accurate, reliable and, if not neutral, at least sufficiently plural and diversified to foster free and informed judgment. This requires the mobilization of the expertise of both scientists and actors in the concerned sectors (NGOs, companies, public authorities, etc.), as well as constant concern for balanced external intervention. We believe it is essential for citizens to be able to rely on a network of experts who accompany them throughout their work, either to check the information they receive or to help them quantify it, compare their proposals, understand their impact, etc. It would therefore be a mistake to pit citizens and experts against each other.  One of the lessons we must learn from the French Convention is, on the contrary, that a) the contribution of expertise is a condition for the exercise’s success, b) relations between experts and citizens must be finely tuned, c) under these conditions, trust can be established and citizens’ rise in competence significantly accelerated. Furthermore, it must be possible for the expertise to be spontaneous and to come from legal or natural persons not solicited by the organizer and/or citizens. The opening of external contribution platforms can respond to this concern for plural information while contributing to the inclusion of a wider public. Finally, if the EESC were to become the operator of citizens’ conventions in the future, the principle of pluralism of information would naturally require that the EESC’s outputs (particularly its previous opinions on the subject and their authors) should not be over-represented in the information base communicated to convention members. This output is already the reflection of a compromise between the representatives of several organizations within the institution; it does not reflect clear-cut or even opposing positions, which would make it possible, through their plurality, to fully understand the issues at stake and to measure their impact. Moreover, this compromise does not give it such neutrality for it to be considered impartial. To pretend otherwise would amount to considerably biasing the deliberative exercise.
5. Transparency: the aim here is to respond to two issues, namely public scrutiny of the principles listed and the general public’s appropriation of the convention debates. The first involves making all of the information received by the members of the convention (released documents, hearings, lists of experts, etc.) available to the public and giving access to the entire system to observers and researchers specializing in issues of citizen participation. The second involves extensive media coverage of the conventions and transparency about the citizens themselves – on their journey, their social and geographical situation, but also on their debates – so that the entire population can grasp the issues at stake, reflect on them and form its own opinion. However, this transparency has two limits. The first concerns the protection of the convention members’ private lives, i.e., women and men who are in a situation of temporary representation without having been candidates and who have not broken with the codes and constraints of public life. For example, in the framework of the CCC, the organizers did not publicly communicate Convention members’ civil status[68], and some citizens explicitly asked not to be filmed or photographed or wished their surname not to appear in the final report. It was also a question of protecting them from the lobbies whose interference we feared. However, traditional lobbies, representatives of companies or certain economic sectors, kept a certain distance from the citizens, out of ignorance of the process, by underestimating its importance or perhaps simply because they did not know exactly how to approach the citizens; unlike some NGOs, which contacted them regularly and without many ethical scruples. The second limit to transparency lies in the serenity of the debates. Some work sequences were closed to the public and the press in order to ensure freer speeches, especially from the most reserved citizens, and to allow them to exchange views with each other at the most decisive moments without fearing the immediate judgment of the press. In the same logic, the votes were organized by secret ballot in order to guarantee the expression of all. All in all, transparency must remain the rule, but can be tempered in the light of these two considerations: the protection of the privacy of individuals who are not destined for public life, on the one hand, and the protection of the serenity of the debates, on the other.

***

What can we learn from CCC’s experience for future conventions?

In the course of the above developments, we have identified seven lessons.

1. The mandate : The CCC experience has shown that a lot can be asked of citizens. They are able, through procedures and methodology that we have shown can be complex, to respond to a widely open question and to draw up a fairly sophisticated form of legislation. There is therefore nothing to suggest that the exercise should be confined to more binary questions, to the mere expression of general visions or to issues deemed to be « simpler, » as is still all too often the case.
2. Governance : The CCC’s experience has shown, in our view, that it is desirable to provide citizens’ conventions with its own governance that is structured, plural in its composition and autonomous in its decisions. There is nothing natural about the democratic engineering of this type of model, and it is not prudent to leave it to the usual practices of facilitators or to the preferences of an isolated presidency, as we have seen in several foreign experiences. Moreover, it makes sense to make clear separations between governance and debate facilitation on the one hand, and between governance and the mandator on the other.
3. The composition of the assembly : the combination of drawing lots (from randomly selected telephone numbers) and recruitment on the basis of socio-demographic criteria of representativeness make it possible to combine the advantages of these two methods. These are, on the one hand, the possibility given to anyone to participate, and on the other, the assurance that the plenum correctly represents social diversity and, at the same time, that it possesses sufficiently varied experiences, points of view and interests. In these selection criteria, we believe it is important to include the citizens’ prior ideological position on the question.
4. Structure of the work : a citizens’ convention faced with a mandate as broad and open as that of the CCC immediately raises the difficulty of increasing the competence of its members. Between a sequential method allowing everyone to become educated on the entire subject and a simultaneous method allowing groups of citizens to specialize in one part of the issues, the latter method seems preferable to us, bearing in mind the precautions we have outlined so as not to fragment the assembly into so many separate sub-groups.
5. Expertise :  a strong contribution of external expertise always runs the risk of experts holding sway over the citizens. We do not, however, believe that the need for such expertise is a matter of debate; it is better to run this risk than to run the risk of under-informed citizens whose conclusions then tend to list generalities of little interest. However, relations between citizens and experts and the different positions of expertise must be carefully structured to control this risk of influence and ensure that it is the citizens who speak and have the final say. We have outlined some of the procedural precautions that have been put in place for this purpose in the CCC.
6. Voting : The practice of voting is, in some respects, contradictory to the search for a widely inclusive consensus. Moreover, it limits the exercise of free and open deliberation and risks corrupting the spirit of the deliberations by creating antagonistic « blocks »… Nevertheless, because the vote is individual, secret and silent as to the individual’s motivations, voting makes it possible to give a voice to the most reserved. Moreover, it allows the collective will of the group to be decided and certified. For these different reasons, it should not be excluded from the Convention members’ work; it could even be used earlier and more often than was the case in the CCC. Especially if its use is accompanied – which is desirable – by procedures that give a voice to those in the minority opinion.
7. Principles : Finally, we believe that a legal framework is now needed to set out the general principles that should govern citizens’ conventions. The fact that a first large-scale experiment has taken place outside any legal framework of this nature is inherent to its innovative nature. But that a second, third or fourth experiment take place in the absence of such a framework is likely to raise great difficulties. We are convinced that the securement of future citizens’ conventions depends much less on their enrolment in an existing institution than on the establishment of a few major guiding principles which, once enshrined in law, will be enforceable against the mandator and future organizers. These are the status and commitment of the mandator, representativeness of the citizens drawn by lot, impartiality and independence of the organizers, access to and pluralism of information, and transparency of the process. Each of these principles may generate a great deal of discussion but, because they will provide a legal basis for possible disputes, they will curb possible deviations or abuses.

All of these recommendations only make sense if we want to give full power to this new exercise and if we are convinced that, far from destabilizing representative democracy, it enriches it. Today this opinion is far from unanimous. Some commentators have equated the Citizens’ Climate Convention to a form of direct democracy entirely at odds with representative democracy. It has even been described as "the staging of a form of dismissal of elected bodies in favor of 150 individuals drawn by lot”[69] . We would like to conclude by responding to this point.

Citizens’ conventions have nothing to do with direct democracy: here, it is not the people as a whole who are directly called upon to state the law, as in a consultative referendum, but simply a sample of citizens who are invited to deliberate and offer proposals. In this sense, citizens’ conventions are rather a new type of representative democracy, complementary in our view to the representative electoral system. Admittedly, the members of the Convention are not elected, but they do represent their fellow citizens in a certain way. For one, because their assembly is representative in the descriptive sense of the term, it resembles them. Second, because the members are temporarily entrusted with a normative role, they are in fact asked to anticipate, after deliberation, the norms, mechanisms and means likely to structure public policy. This function certainly brings them closer to parliamentarians. But proximity does not mean rivalry, less still antagonism. Indeed, elected representatives are the ones who have the final say when it comes to discussing and voting on the law, in accordance with the Constitution. This is, moreover, the meaning of one of the CCC’s conclusions. During the last session, its members had to decide whether or not to request the organization of one or more referendums on their proposals. On this occasion, three complementary arguments were heard against the referendum: 1) the fear of a plebiscitary type of vote, 2) the fear that the referendum campaign would not allow the French to be sufficiently informed both on substance and form (at least not as much as the convention members) to make an enlightened decision, 3) the willingness to transmit the vast majority of their proposals to Parliament so that parliamentarians « take responsibility » and confront potential opposition. In the end, the plenum considered that the French people should only be directly consulted on three proposals out of 149: the two proposals to revise the Constitution and the proposal to include the crime of ecocide in French law. For the rest, it left it up to the legislature.

[1] In French, the Conseil Economique Social et Environnemental or CESE. Constitutional consultative assembly representing key economic, social and environmental fields, promoting cooperation between different socio-professional interest groups and ensuring they are part of the process of shaping and reviewing public policy. The translated acronym is not to be confused with the European Economic and Social Committee, a consultative body of the European Union.

[2] Full report: http://www.oecd.org/fr/gov/innovative-citizen-participation-and-new-democratic-institutions- 339306da-en.htm

[3] On the difficulties of combining participation and deliberation, see Éric Buge’s analysis, « À la recherche du diamant de Micromégas. Quelques réflexions à propos du Grand Débat national et de la Convention citoyenne pour le climat, » Archives de philosophie du droit, tome 61, 2020, forthcoming.

[4] See Dimitri Courant’s commentary on the Irish experiences: https://laviedesidees.fr/Les-assemblees- citoyennes-en-Irlande.html

[5] In January 2019, President Macron launched a ‘Grand Débat’, or Great Debate, across France, which established numerous channels through which citizens could make their voices heard (including these regional conventions, as well as town hall and local meetings and online suggestions submitted to the Debate’s website, to name a few.). The Grand Debate’s regional conventions can be accessed here: https://granddebat.fr/pages/les-conferences-citoyennes-regionales

[6] In Australia and Belgium, the Citizens’ Convention are initiated by NGOs.

[7] See the interview with Léo Cohen, Revue XXI, number 52, Autumn 2020.

[8] The phrases in quotes are those of Pierre Rosanvallon, La Contre-démocratie. La politique à l’âge de la défiance, Paris, Seuil, 2006, p. 300.

[9] In a way, the efforts made to recognize the legitimacy of intermediary bodies and to bring about “social democracy” (notably with the Larcher law of 2003) also seek to make up for the shortcomings of representative democracy. But they are guided by the pursuit of a different end: to affirm the existence of intermediate interests between particular interests and the general interest, rather than to give a direct voice to citizens in their diversity. The idea of “organized civil society” that the Economic, Social and Environmental Council seeks to embody in practice refers to a set of organizations that aspire to represent distinct collective interests.

[10] Jacques Rancière, La Haine de la démocratie, Paris, La Fabrique, 2005, p. 54.

[11] Human civil rights document from the French Revolution, set by France’s National Constituent Assembly in 1789

[12] The CCC Governance Committee was installed on July 2, 2019. Citizen members were first convened in early October 2019. The President of the Republic responded to the CCC on June 29, 2020.

[13] The mandate : https://www.conventioncitoyennepourleclimat.fr/wp- content/uploads/2019/09/lettre-de-mission.pdf

[14] Press conference by the President of the Republic on 25 April 2019: https://www.elysee.fr/emmanuel- macron/2019/04/25/conférence-de-presse-grand-debat-national

[15] 8% of the initial panel dropped out or were disbarred, 94.4% of incumbents at the end of the CCC had attended all sessions or missed less than two out of seven sessions.

[16] A session is defined as a work sequence physically bringing together all the members of the Convention from Friday 2 p.m. to Sunday 3 p.m.

[17] In 2005, a first referendum was held, but did not reach the previously set threshold of 60% of the votes cast in favor (57.7%). In a second referendum in 2009, 60.9% of voters voted against.

[18] See the « Resolution of Dail Eireann approving establishment of the Citizens’ Assembly » ; https://2016– 2018.citizensassembly.ie/en/The-Eighth-Amendment-of-the-Constitution/Final-Report-on-the-Eighth-Amendment-of- the-Constitution/Final-Report-incl-Appendix-A-D.pdf

[19] Work of the Citizens’ Assembly 2016–2018: https://2016–2018.citizensassembly.ie/en/

[20] This working time on the same subject is quite higher than what is observed in most of the foreign examples. It is estimated that, on average, members spent more than 500 hours each on CCC (the equivalent of more than 14 FTE weeks.) This includes not only the sessions themselves, but also all the work done between sessions (technical or thematic webinars, etc.). See Gilles-Laurent Rayssac, « Quand le dialogue experts/non-experts permet de faire de la politique », https://www.metiseurope.eu/2020/09/20/quand-le- dialogue-experts-non-experts-permit-de-la-politique-example-of-the-convention-citizens-for-the-climate/

[21] See the CCC’s final report: https://propositions.conventioncitoyennepourleclimat.fr/pdf/ccc-rapport-final.pdf

[22] https://www.climateassembly.scot

[23] https://www.climateassembly.u

[24] First Report and recommendation of the Citizen’s Assembly – The eighth amendment of the constitution: https://2016–2018.citizensassembly.ie/en/The-Eighth-Amendment-of-the-Constitution/Final-Report-on-the-Eighth- Amendment-of-the-Constitution/Final-Report-incl-Appendix-A-D.pdf

[25] Professionals in the field of citizen dialogue engineering and facilitation, members of Eurogroup Consulting, Respublica and Missions Publiques

[26] One exception should be noted: during the introductory session, Laurence Tubiana, co-chair of the governance committee, presented the Paris Agreement to the citizens as a former ambassador of the negotiations for the Paris Conference (COP21).

[27] The detailed composition of the Governance Committee: https://www.conventioncitoyennepourleclimat.fr/en/comite-gourvernance/

[28] For each intersession, a man and a woman were drawn at random.

[29] These are the words of the Prime Minister’s previously quoted engagement letter.

[30] With the exception of less than 10 decisions for which a vote was held in the Governance Committee.

[31] https://www.conventioncitoyennepourleclimat.fr/les-garants/

[32] See press release of the Guarantors’ College of 5 February 2019: https://granddebat.fr/media/default/0001/01/3^e5fb4c95^e71727190a2ce1a46fcdf31211fc0f0.pdf

33 3 million unregistered and 6.5 million incorrectly registered, according to an information report by the National Assembly on the modalities of voter registration on electoral lists from 17 December 2014: http://assemblee-nationale.fr/14/rap-info/i2473.asp#P794_78395

[33] Harris interactive is a marketing research and opinion poll company. As part of CCC, it was responsible for generating telephone numbers and recruiting members in accordance with the objectives set by the Governance Committee and under the control of the Guarantors.

[34] Technically, a computer was asked to randomly select, under the control of a bailiff, 300,000 telephone numbers from the exhaustive root of numbers allocated by the French Electronic Communications, Postal and Print media distribution Regulatory Authority (85% mobile telephone numbers and 15% landline numbers).

[35] The origin of this figure is unclear. Several formats were initially envisaged, often with a smaller number of citizens. A leak in the press shortly before the President’s announcements on 25 April 2019 indicated that the number of 300 people had initially been retained. In practice, 150 is a minimum for representativeness (difficult to combine 5 or 6 criteria of representativeness on a narrower sample) and a maximum for deliberation (difficult to have a wider public deliberate openly and fairly).

[36] Starting at 16 years old. It was indeed decided to include minors aged 16–17 in the panel, considering that the long-term challenges of climate change closely concern the younger generations.

[37] Presentation of CCC members: https://www.conventioncitoyennepourleclimat.fr/comment-sont-ils- selectionnes-2/

[38] To ensure the representation and regular participation of overseas citizens, a call for volunteers and a draw were organized on the basis of the Overseas Agency for Mobility (LADOM) database. This file lists overseas citizens who are temporarily students or apprentices in metropolitan France.

[39] To enable the participation of citizens who have lived in very precarious situations, the Governance Committee called upon the mediation of the charitable association Les Petits Frères des Pauvres.

[40] After failing to recruit farmers using the general methodology, a separate draw was made on the France Telecom business directory and two farmers were recruited.

[41] This study, carried out by Adrien Fabre (who has followed all of the CCC’s work) also showed that social and associative action was much more important for CCC members (40% place it at 9 or 10) than it was for the rest of the population (19% at 9 or 10); https://www.parisschoolofeconomics.eu/docs/fabre-adrien/ccc_externe.pdf or http://adrien- fabre.com/Documents/Les%20Français%20^et%20la%20CCC.pdf

[42]  Compared to the current French population, a sample of 150 people structured as the CCC was has a margin of error of 8 percentage points.

[43] It is a document distributed to citizens, composed of data and analyses reflecting international scientific consensus, as well as a simplified presentation of various public policy tools. See: https://www.conventioncitoyennepourleclimat.fr/wp-content/uploads/2019/10/03102019-convcit-socledoc-web.pdf

[44] Eating, Travelling, Housing, Consuming and Producing & Working.

[45] Membership of a particular working group quickly became a form of identity for every citizen. Many of them spontaneously presented themselves as « a member of the Housing group » or « a member of the Consuming group »…

[46] The CCC members adopted four proposals relating to the Constitution.

[47] This is a point that is still being debated in the Governance Committee. For some members, a specific working group should have been devoted to these issues.

[48]  The term is by Gilles-Laurent Rayssac in his previously cited article. It loosely translates to uninteresting, unelaborated conservation consisting in trivialities and rumor spreading.

[49] The complete list of fact-checker volunteers: https://www.conventioncitoyennepourleclimat.fr/facts-checkers/

[50] Some thirty requests were addressed to the fact-checkers during the first sessions of the Convention. Some of these requests went beyond the strict register for verification. For example, « how deadly are greenhouse gases? », a naive question in its form, but extremely demanding in its answer… Other questions called for a personal judgement or opinion from the fact-checkers, which they systematically refused. They were also able to reply that there were no answers to certain questions.

[51] The full list of speakers: https://www.conventioncitoyennepourleclimat.fr/intervenants/

We are not entering here into the debate on the nature of expertise. The notion of expertise carries with it the idea of knowledge detached from beliefs and interests. From this point of view, the expert would be a scholar guided by the sole concern for truth. Strictly speaking, representatives of companies, NGOs, trade unions, etc. would therefore find it difficult to claim this status. The reality is often different. In a certain number of fields (particularly technical or techno-scientific), it happens that the greatest expertise on a subject is embedded in structures linked to convictions or interests. And conversely, it also happens that the academic quality of a speaker does not absolve him or her from normative commitments, particularly in the social sciences. The portrait of the expert as a disinterested scholar is therefore in practice very insufficient to delimit the field of resources useful for informed deliberation.

[52] Full programs of the sessions: https://www.conventioncitoyennepourleclimat.fr/les-travaux-de-la-convention-3/ Cross-examination hearings involved two or three different points of view on an issue or topic. Their purpose was not to organize a « match » or a « duel » before the members’ eyes based on disputes (at best) or what is regularly practiced on television sets (at worst), but to offer citizens contrasting, even antagonistic, visions and arguments. The cross-examining nature of these hearings owes more to the adversarial nature as understood in the case law of fair trials in European courts than to the idea of an agonistic confrontation. The aim was not to indicate to the CCC members the options from which they had to choose, but rather to present the complexity of the subject so that they could formulate their own options.

[53] Summaries produced by Open Sources Politics: https://contribuez.conventioncitoyennepourleclimat.fr/pages/synthese-phase1

[54]  The full list of support group members: https://www.conventioncitoyennepourleclimat.fr/groupe-appui/

[55]  The complete list of the members of the legislative committee: https://www.conventioncitoyennepourleclimat.fr/groupe-appui/

[56]  However, they have declared themselves in favor of a carbon tax at the frontiers of the European Union. See p. 135 of the final report.

[57] See supra, footnote 47

[58] This idea was inspired by the practice of certain courts, in particular that of the European Court of Human Rights, which annexes to its judgments the dissenting opinion of the judges who voted against.  The dissenting opinions allow to express a negative opinion, but also a disagreement on a particular point of the proposal, without this necessarily leading to a negative vote on the whole proposal

[59] However, the CCC researchers noted that during the last working session, 42% of the 52 respondents to the questionnaires felt that more systematic use should have been made of voting in cases of disagreement in order to avoid « strong heads having the last word » on the silent majority.

[60] See CCC final report: https://propositions.conventioncitoyennepourleclimat.fr/pdf/ccc-rapport-final.pdf

[61] Amendments aimed at deleting a proposal in its entirety or adding a modality were considered inadmissible. Before they were put to a vote in the plenum, a conciliation procedure was proposed to the holders of competing amendments.

[62] https://www.conventioncitoyennepourleclimat.fr/2020/04/09/la-contribution-de-la-convention-citoyenne-pour-le- climat-au-plan-de-sortie-de-crise/

[63] Due to exceptional circumstances related to the lockdown, only 110 to 120 members were able to participate in these votes. To reinforce the legitimacy of such sensitive decisions, the Governance Committee had therefore proposed a majority vote of the registered members, i.e. more than 76 votes. For a decision to be adopted, it therefore had to pass the two-thirds mark of those present (65%). The vote on advertising received 71 votes in favor and 44 against, bringing the « yes » vote to 61.7%, a strong majority, but not enough to carry the decision. This result generated some emotion within the CCC ranks.

[64] Some of the initiatives that had been announced or taken by mid-November 2020 : prohibition of heated terraces, obligation to replace oil-fired boilers at the end of their useful life with less polluting heating systems, a flyer asking prefects to oppose any commercial development project that could spread soil artificialization, inclusion of the reinforcement of the automobile malus on the most polluting vehicles in the 2021 finance bill, increased resources for public transport, trains, bicycles, energy-efficient renovation of housing and public buildings, reinforcement of incentive pricing for waste management, introduction of a bonus for electric vehicles under car insurance contracts, raising the level of taxation for leisure aviation, introduction of a car penalty for thermal vehicles weighing more than 1,800 kg, provisions relating to ecocide…

[65] This is the opinion of lawyer Eric Buge (interview with the authors.) It is also a point put forward polemically by lawyer Arnaud Gossement, who feared in particular that the CCC’s practice would call into question environmental law in terms of public consultations. See Arnaud Gossement, « La Convention citoyenne pour le climat : innovation démocratique ou régression juridique », AOC, 3 February 2020, https://aoc.media/opinion/2020/02/02/la-convention-citoyenne-pour-le-climat-innovation-democratique-ou-regression- legal/?fbclid=IwAR1OqggRZ4foo3Bq5Q99KsOdY7Uzx6y6O9bzCUB4SycFhO1GN2z127ybN4k

[66] See Denis Baranger’s analysis of the reform; http://blog.juspoliticum.com/2020/09/05/democratie-participative- linopportune-reform-by-denis-baranger/

[67] The Cour de Cassation is the highest court in the French judiciary, it has jurisdiction over all the judicial system’s civil and criminal matters, and is the supreme court of appeals in these cases. It also has jurisdiction to review and to certify questions of the law. The Conseil d’Etat, or State Council acts both as legal advisor to the executive branch and as the supreme court for administrative justice. The Cour des comptes is the supreme body for auditing the use of public funds. Independent from the Government and Parliament, it has financial jurisdiction and is in charge of auditing, issuing rulings and certifying the State and Social Security accounts, as well as contributing to the evaluation of public policies. The Constitutional Council ensures respect of the Constitution, is vested with various powers, including in particular the review of the constitutionality of legislation.

[68] Some citizens wished to give more information, especially to journalists.

[69] Statement by Dominique Reynié in the newspaper Les Echos, «  L’aura perdue du Parlement  », Les Echos, 7 October 2020.