1. Introduction

Appearing as a democratic solution to the crisis of representative government, deliberative devices have multiplied at all scales from local to international (OECD 2020). However, these public action devices have been, with few exceptions (Courant 2021), ad hoc and temporary experiments. This multiplication illustrates both the diffusion of the deliberative imperative but also the difficulty of finding a perennial place for deliberation, especially within national representative systems (for the French case see Blondiaux and Fourniau 2011). This punctual dimension is often criticised by scholars and practitioners who advocate for a permanent institutionalisation of sortition that would complement (Sintomer 2019) the work of elected representatives or even replace them (Landemore 2020). The permanent integration of citizen deliberation would however, run the risk of suppressing the radical and subversive dimension of sortition (Bonin 2018) and deliberation. Moreover, the form of the device and its institutional place in the decision are crucial to prevent it from becoming a mere tool of political communication (Gourgues 2013; Gourgues, Mazeau & Nonjon 2021).

The recent institutionalisation of deliberative mini-publics (DMPs) in the French Economic Social and Environmental Council (ESEC1) illustrate this ‘second wave’ of deliberative experimentation, hybridising representative and deliberative principles (Courant and Sintomer 2019) as well as its ambiguities. Here, the form of institutionalisation is particularly interesting, since it involves citizens drawn by lot deliberating alongside interest representatives. While many studies have focused on ordinary citizens on the one hand and politicians on the other, one could argue that a third type of political actor has remained in the blind spot of deliberativists: the numerous individuals engaged in associations, trade unions or employers’ federations, i.e. interest groups. Initially groups were identified by deliberativists, foremost being Jon Elster and Jürgen Habermas, as belonging to the market space (Elster 1994) and driven by strategic action (Habermas 1981) oriented towards negotiation and the search for gains rather than the general interest (Mansbridge 1992). The implementation of increasingly institutionalised deliberative mini-publics (Böker and Elstub 2015) and the development of systemic approaches to deliberation (Mansbridge 2012) have led to a trade-off with these initial ideals. Deliberativists, previously critical of interest groups, are thus more favourable to their integration into deliberative arenas. However, little empirical work has focused on the interactions between interest representatives and citizens within deliberative spaces (Hendriks 2002; Rui 2016) and their impact on deliberation.

Can randomly selected citizens deliberate on an equal footing with interest group representatives? How do the latter envision citizen participation? And, more importantly, can an institution representing interest groups integrate citizen participation in its works? This article attempts to answer these questions by analysing the deliberative mini-public ‘New Generations’ set up by the ESEC in 2020. The third assembly in the French Constitution, gathering representatives from organised civil society, the ESEC has since 2019 experimented with several methods of direct citizen deliberation, mostly in the form of deliberative mini-publics.2 These experiments were significant in two regards. First, they led to the adoption, in January 2021, of an organic law3 reforming the ESEC, integrating, for the first time in France, DMPs in the ‘toolbox’ of a national and constitutional institution. Second, far from the supposed opposition between citizens and interest groups, this case offers a perfect opportunity to take a close look at the interactions between randomly selected citizens and interest representatives.

Indeed, the ‘New Generations’ experiment is particular as it consists of a hybrid apparatus: randomly selected citizens and members of the Council deliberated together on the same subject. While it cannot be considered ‘mixed’, since, as we will detail, the two groups had separated deliberative arenas, there were multiple connections between the two publics. This is rather unusual: for example, out of the 105 DMPs of the POLITICIZE dataset, only 10 can be categorised as ‘hybrid’, including both citizens and elected politicians or other stakeholders (Paulis et al. 2020:8).

A close examination of ‘New Generations’ shows the limits of integrating interest representatives alongside citizens within a DMP and the difficulties of institutionalising citizen deliberation within the ESEC. Articulating a longitudinal macro perspective (on the institutional and legislative transformations of the Council) with a sociological microanalysis (of the ‘New Generations’ DMP) allows us to underline that in the current implementation of citizens’ participation in the ESEC, the institution seems to be ‘flying without instruments’. Indeed, with few references to legal frameworks or scientific guidelines, the Council relies mainly on private consultants to implement deliberative practices. Doing so raises a certain number of problems unlikely to be resolved by the adopted reform.

To demonstrate this, our study relies upon a combination of methods. Concerning the reform of the ESEC itself, we analysed the three different legislative proposals put forward between 2018 and 2021. Furthermore, five semi-structured interviews with high-ranking councillors as well as the analysis of internal documents allowed us to gain insight on the discussions within the Council. Regarding ‘New Generations’, we first observed the activities of this DMP from January to July 2020 (around 11 days of in-person and online meetings). Second, we distributed questionnaires at five different points in the process (January, February, March, April and June). Overall, 149 questionnaires from the 28 citizens were collected, as well as 41 from various councillors who took part in the DMP. Third, we arranged semi-structured interviews throughout the process and at its conclusion, with citizens (5), councillors (8) and consultants (2). All interviews were conducted in Paris, between June and July 2020. Fourth, we also conducted a content analysis of around fifteen documents produced by the DMP (drafts, charts, reports, etc.). Those are written materials we deemed significant as they express crucial route points in the deliberative process (work framing, synthetisation of the debate, additional information, amendments).

The rest of this article is structured as follows. In section 2, after a brief presentation of the ESEC itself, we retrace the various steps towards the implementation of deliberative experiments under the aegis of the Council. Afterwards our analysis focuses on the ‘New Generations’ DMP, and especially on the effects of a hybrid process, highlighting its difficulties and limits (sections 3.1 and 3.2). In section 3.3, we assess the perceptions of citizens’ involvement in the ESEC and offer a typology of the councillors’ positions regarding the subject. Finally, in section 4, we underline three main problems in the overall process: the lack of standards of deliberation in the ESEC that makes the dialogue between interest representatives and citizens difficult, the important power of its executive board, and the seemingly instrumental nature of the reform.

2. Rise of the ESEC: Between Autonomous Initiatives and Contextual Responses

As mentioned, the ESEC is a consultative chamber, consisting of representatives of interest groups. The councillors are either chosen by a recognized organisation (eg. trade unions), or nominated by the government. It was created as a place for social dialogue. The powers of the Council are marginal: it mostly produces reports on various public issues, with recommendations that sometimes find their way into the legislation. It is thus worth understanding how the ESEC became one of the main institutions in charge of citizens’ deliberation at the national level. While such a process is, of course, multifactorial, three key elements are worth taking into account.

The first and broader element relates to large trends in Western politics in the last fifty years. The rise of electoral abstention, the abysmal levels of confidence of citizens in their politicians and the diminishing legitimacy of intermediate bodies (political parties, trade unions, etc.) all point to a crisis of representative institutions (Cautrès and Muxel 2011; Haute and Tiberj 2022). Coupled with the pressures that emerged from civil society for more participatory and deliberative avenues, these led to a necessity for political actors to transform existing models. However, since elected officials are not so keen on giving up their power, they either turn to temporary arrangements or consultative mechanisms. The ESEC offered the advantage of being a national, existing institution, with a certain symbolic power, but that remains advisory.

A second key driver, more specific to France, is the sequence of events inaugurated by the election of Emmanuel Macron in May 2017. As the first President without official ties to the historical governing parties of the Fifth Republic, and branding himself a ‘reformer’, Macron promised a constitutional overhaul, touching amongst other things at the ESEC. The initial plan seemed to involve transforming the Council into a ‘Chamber of civil society’. Its powers would have been to advise the government on the long-term effects of policy decisions, after consulting the public, including through randomly selected citizens. Blocked by the right-wing controlled Senate in the summer of 2018, this proposal fell into legislative limbo.

It took the Yellow Vest movement, in November 2018, to ignite a new spark. While the initial demands were more economical, the question of democracy soon emerged as a key component, notably through the request for a right to citizens’ initiative (the Referendum d’initiative citoyenne (RIC)) (Chamorel 2019). To stem the tide of protest, President Macron organised a ‘Great National Debate’ aimed at understanding citizens’ grievances. Following substantive dissension, the National Commission of Public Debate (NCPD), originally charged with organising consultations, stepped aside after being excluded from the proceedings.4 The ESEC, which was slowly repositioning itself as an actor of deliberative initiatives, seized this opportunity. Between October 2019 and June 2020, it notably hosted and co-organised the Citizen Convention on Climate (CCC). Through this experiment, the ESEC set itself apart in the application of deliberative democracy at the national level (Giraudet et al. 2022) even though, on the legislative front, no clear legal framework was devised.

Finally, the third element to take into account is the internal dynamics within the ESEC itself, and especially the role of its Bureau, its collegial executive body.5 In 2015, a new five-year mandate began and the new Bureau promised to look into ways to include more direct citizen participation into the Council. Following Macron’s election, an important internal campaign to persuade the sceptics of the necessity of a reform and of including more citizens. Internal documents provided to us by councillors reveal that by March 2018, a working group was in place to lay out the main points of the reform.6 At this time, the idea seemed to have been to include two to three randomly selected citizens to each Section.7 It is important to note that these initiatives have been done without any explicit request from the government: it is under its regulatory discretionary powers that the Bureau set up this working group, and eventually the DMPs. In December 2018, according to internal documents, the plan was to add six ‘associated citizens’ to specific working committees of the Council.

Within the context of the Yellow Vest crisis and the aforementioned Great National Debate, the ESEC decided to put in place a process that was both consultative and built on a hybrid model (in which randomly selected citizens and civil society representatives were to work together throughout the process). Sensing the importance of the Yellow Vests crisis, and in conjunction with its own reflections on citizen participation, the Bureau created in December 2018 a Temporary Commission (TC) entitled ‘Fractures and Transitions’. However, instead of having only the usual councillors, the ESEC also established a Citizen Group (CG) of 28 randomly selected citizens, piloted by a private firm specialised in deliberative processes. Both groups were independent of each other, each writing its own report, but there were bridges: five councillors attended the citizens’ sessions, and vice versa. In its final report, the Temporary Commission indicated the ideas and recommendations that emanated from the Citizen Group’s discussions and annexed the citizens’ own report at the end of the document (Badré and Gillier 2019). In the absence of any legal framework or internal guidelines regarding the implementation of citizen deliberation in the ESEC, no regulation to supervise the process was put in place. The Bureau thus adopted an experimental approach and evaluated the deliberation through an independent assessment (Courant 2019; Landemore and Sintomer 2019). This model proved crucial for the ways the ESEC envisioned citizen participation in its future proceedings.

This section highlighted the key factors that led to the ESEC an important place in deliberative experiments in France. Within the institution, the Council’s Bureau played a proactive role, setting up working groups and DMPs to experiment with ways to include randomly selected citizens in its proceedings. The larger political context also mattered: the election of Macron, but more especially the Yellow Vests and the Citizen Convention on the Climate opened an important window of opportunity. During this process, very few references were made either to legal norms, existing practices within other deliberative institutions (such as the NCPD) or scientific recommendations. We now propose to investigate this by examining the ‘New Generations’ Temporary Commission, which ran from January 2020 to July 2020, and during which citizens and councillors deliberated on the question of intergenerational solidarity. Uncommon due to its hybrid nature, this deliberative mini-public raised several questions and challenges for the future of citizen participation within the ESEC. Furthermore, as it drew from a previous experience, ‘Fractures and Transitions’, it allegedly had room for a more accomplished procedure design. Thus, most limits we identify cannot be attributed to the unprecedented nature of the process alone.

3. Assessing Citizen Participation in the ESEC: the ‘New Generations’ Case

During the spring of 2020, the ESEC launched a Temporary Commission (TC), named ‘New Generations’ and adjoined to it a Citizen Group (CG) of 28 citizens. The goal was to address the issue of social mobility as well as intergenerational solidarity.8 Considering the subject, the Council voluntarily overrepresented 18- to 30-year-olds in the group.9 Since the ESEC had little experience in deliberative processes with citizens, it recruited two private firms specialised in public engagement to oversee the CG, with two to three facilitators present at all times but only facilitating discussions for the whole group; they were never present at the tables with smaller groups. The Citizen Group was supposed to meet for 12 days over four weekends from January to April. However, the COVID-19 pandemic disrupted the program. In the end, the citizens met only for seven days, and the conclusion of the process was mainly done online.10 The final report of the TC was adopted in July (Dubrac and Teskouk 2020). The Citizens Group’s report was inserted within the text and differentiated by its colour and shading. As mentioned before, the approach being rather experimental, it did not rely on a legally established framework. The normative framing itself was a work in progress: actors were moved by different aspirations for the CG (sometimes contradictory ones).

The intention behind ‘New Generations’ was to integrate citizens and councillors in order for each group to contribute to the deliberations and reflections of the other. This integration was done collectively through joint hearings and working sessions, with agreement on a common plan, but also individually by the participation of citizens in the Temporary Commission, or conversely the presence of councillors during the citizens’ debates (see Figure 1 for a sumarisation of the process). Yet, despite the stated aim of reaching a co-produced document, citizens and councillors eventually handed in distinct reports.

Figure 1
Figure 1
Figure 1

Sum up of the 2020 ‘New Generations’ Citizen Group process.

3.1. Determining the Effects of Mixed-Membership

In this section, we weigh the impact of both the Commission on the Citizen Group and vice versa. While both groups were glad to collaborate, we found out that beneath the surface, almost no significant cooperation occurred. We highlight the imbalance between the TC and the GC, with the citizens having little involvement or role in the Commission’s work. The lack of clear guidelines regarding the role of councillors in the CG, as well as the different value given to the citizens’ input, seem to be crucial issues in this regard. The shift to online sessions emphasised this tendency, allowing the TC to gain independence and keep pace with its own agenda, which forced the CG to follow.

3.1.1. Impact of councillors on the Citizen Group

The choice seems to have been made for ‘New Generations’ to keep the councillors in an attitude of strict neutrality. Thus, even if they had the opportunity to take part in roundtable discussions (or in groups during the online sessions), the councillors were incited to stick to a position of observers rather than participants.11 That being said, there were no written or oral instructions that would provide a framework stating which posture councillors should hold towards citizens.

I think in the ESEC we didn’t give a thought to the specificities of working with citizens. We didn’t have a shared reflection, no introspection nor questioning about ‘how is it like to work with people who are not councillors?’ […] We didn’t get a briefing.

Interview with councillor A, man in his sixties

This illustrates a lack of guidelines on the role of ESEC members in the citizens’ deliberations, which has been deplored several times by councillors. However, it allowed a certain diversity of attitudes from councillors within the CG. While most did not seek to influence the citizens’ discussions, they sometimes acted ‘as citizens’ and participated as such in the discussion. Conversely, others remained rather in the background, hardly ever intervening. Some also tended to frame and reframe the discussions (in particular to refocus citizens on the subject of ‘New Generations’) or to provide their expertise on certain issues. In fact, we could not help but notice the vagueness that reigned around the role of councillors, including for citizens.

I remember during the first working session […] they [the councillors] really came as observers and they didn’t intervene at all in the discussion, they took a lot of notes, and then they gave back. And then, I felt that they were more observers and then another time, for that matter, they were much more involved in the discussion.

Interview with citizen A, woman in her thirties

Overall, even with their high social and cultural capital, councillors did not seem to influence or frame the debates and propositions of the Citizen Group. Even if the possibility of instrumentalisation was thus open we found very little willingness on the part of the Commission members to influence the CG during the process. A majority of councillors argued that the main purpose of their presence in the Citizen Group was to listen to citizens, to sort through their ideas and to eventually incorporate the best ones. There was a certain tendency to look down on the work of the Citizen Group: only certain citizens’ ideas were deemed good enough to be incorporated into the TC report.

[Q.] How do you see your role within the Citizen’s Group?

That it brings me things. I listened a lot. […] But what I really expected was for them to bring us new ideas.

Interview with councillor B, womanin her sixties

This attitude, expressed by several councillors during the interviews, is also reflected in the recommendations of the final report where the councillors’ plan was chosen as a writing guideline for both groups. If on the surface, the two reports share similar themes, a close reading reveals several divergences that are not reflected in the way the text was organised. For example, there are major differences between the measures proposed by citizens and those of the councillors. Issues of particular interest to the TC, such as training and professional mobility and old age, were not the subject of any proposals from the citizens’ side (Dubrac and Teskouk 2020:59–68,82–84). On the other hand, the Commission hardly dealt with several themes dear to the citizens: education or the issue of national, territorial and social cohesion (Dubrac and Teskouk 2020:50–54,135–137). While the writers of the TC report and the animation team of the CG were careful to make sure that there were no overt contradictions between the two texts, the analysis shows that the two reports were superposed, not in dialogue with each other.

3.1.2. Impact of the Citizens’ Group on the Temporary Commission

If the impact of the TC on the citizens was rather slim, the opposite is also true. Differences in the forms of discussions adopted by councillors and citizens as well as the limited involvement of the citizens in the TC meant that the mix was difficult. Indeed, while some of the councillors might take up citizens’ ideas, during the interviews few could point to concrete examples of the Commission integrating citizens’ propositions.

Observation of the TC and CG revealed significant differences in the way the discussions took place. There, it is useful to distinguish between debate, dialogue and deliberation.12 We argue that councillors were more prone to debate while citizens tended to deliberate. Councillors were more likely to try to persuade or to defend their position.

We try to stand by our opinion in the ESEC group [Group of Companies]. It’s difficult when some organisations are very harsh. I see that some trade unions are very harsh in their stance. They won’t change it.

Interview with councillor B, woman in her sixties

Here we see that the relationships between ESEC groups being rather conflictual (at least on some controversial issues such as taxation or employment). Interactions during the TC tend to emphasise debate over deliberation. This is confirmed by the observation: councillors were prone to advocate the same positions without taking into account the objections of their peers and to reiterate them, even if the subject had changed.

Discussions often seemed like a win/lose situation. Furthermore, the tight structure of the Commission favoured fast and majoritarian decisions. The fact that most councillors act as representatives or advocates of their organisation’s position amplifies this pattern.

[Regarding the definition of ‘social mobility’:]

Councillor C (woman in her fifties): I disagree! Individualism shouldn’t appear here, I don’t agree […] individualism doesn’t match with our discussions nor with the definitions we took on. If you disagree with the removal of individualism [from the text], I propose we vote.

Observation, TC session, online, 28 May 2020

Here the request for a vote stops the discussion, by the threat of being outvoted. While there was a will to attain a consensus within the TC, this seemed to occur both through self-censorship by some of its members (from a strategic perspective) and in the drafting by the two co-rapporteurs (responsible for adjusting the text following amendments by the councillors).

Sometimes we’re compromising. Sometimes it produces very heavy texts and heavy recommendations. But still we managed to pass recommendations and suggestions on to the government.

Interview with councillor B, woman in her sixties

The discussion in the Citizen’s Group was closer to the typical ideal of deliberation: citizens were more inclined to listen to each other, to learn and to weigh the pros and cons.

[During a writing session:]

Citizen B (man in his thirties): could we write down the number of people that are in favour or against basic income without conditions?

Moderator (man in his thirties): For now, we will look at it again and we will decide on which points you have vetoes. The reader needs to see where there is consensus or not.

Observation, CG session, 6 June 2020

This is congruent with previous findings. As mentioned, interest representatives are less likely to change their minds, while citizens, since they are not bound by organisational ties, can evolve more freely between positions (Hendriks, Dryzek & Hunold 2007). The councillors themselves also underlined the qualities of citizen deliberation especially the positive reception of dissensus. This confrontation with different practices of discussion can encourage councillors to amend their practices and to draw inspiration from citizens in order to further improve the quality of the ESEC’s work.

We appreciated [the functioning of the citizen group] and we said to ourselves that […]. it ought to come to shape another functioning of the sections within the ESEC.

Interview with Councillor F, man in his fifties

Apart from this illustration of deliberative practices, direct citizen participation in the Commission’s working sessions contributed little to the interaction between the two groups. The idea was that during all the Commission’s sessions, up to six (voluntary and rotating) citizens could be present so that more citizens could witness the TC’s inner workings. However, the very low investment of the CG members of the Commission and the holding of the sessions in Paris tended to minimise citizen participation. In general, only three to four citizens (including two recurring ones) out of 28 were present, and they spoke very little. When they did speak, they did not carry the ideas of the Citizen Group but often their own individual voice as there was no notion of a mandate or of incarnating a collective position (this was already the case in the 2019 experiment, see Courant 2019). A citizen who had attended a few of the TC sessions spoke of the discomfort he felt when he had to intervene:

We were invited to come and see how it was done, how it was discussed inside, what was going on and… And not to come, not to intervene in their opinion… […] I didn’t have the feeling that, since it was in their space, that we were supposed to come and give our opinion.

Interview with citizen B, man in his thirties

In addition, the attitude of some councillors may have reflected impatience or annoyance during the sessions, when citizens were (in their view) taking up too much space.

Councillor: Could you [citizen] be extremely short because we are in a TC meeting.

Citizen: [speaking]

Councillor: […] Could you be quicker, because I remind you that this is the work of the Temporary Commission.

Observation, TC session, online, 2 April 2020

In line with the literature (Courant 2021:13), it seems that less than six citizens in comparison with more than 20 councillors did not constitute a critical mass. However, a stronger citizens’ presence could have several virtues. It could make it possible to strengthen the links between councillors and citizens, encourage citizens to take an interest in or even draw inspiration from the work of the TC and, conversely, it could encourage councillors to take on board any disagreements of citizens and to identify issues common to both groups. As with the presence of the councillors in the CG, it is a question of clarifying the role of citizens in the Temporary Commission. If they are contained in a spectator role, their impact in the proceedings of the ESEC will remain minimal.

To sum up, the ‘New Generations’ experiment produced a report combining the citizens’ recommendations and those of the ESEC. Although the Citizens’ Group provided some input to the Temporary Committee, the mechanism chosen did not allow for a real co-construction.13 The councillors present at the meetings of the CG thus declared that they were here to listen and select the ‘good ideas’ coming from citizens. This attitude echoes the ‘cherry-picking’ of proposals coming from participatory processes of which Joan Font and her colleagues gave a systematic analysis, emphasising that both contextual and proposal-related factors could influence the selection. In particular the ‘proposal’s cost, the extent to which it challenges existing policy and the degree of support it has’ within the community significantly determine how or if citizens’ policy proposals will be accepted (Font et al. 2018). Similar criteria apply to this case. Some councillors highlighted the need for citizens to focus on inexpensive propositions.14 Others underscored the need to think about the degree of public support for certain proposals,15 and eventually discarding (implicitly or explicitly) some ideas during the Commission meetings. All of this shows that, despite the purported intent of GC and TC ‘feeding off each other’,16 citizens were not able to bear as much as councillors on the framing and final content of the report. Therefore an asymmetrical relationship remains, at odds with the ESEC’s goal of not wanting to appear ‘lecturing’ citizens but rather ‘dialoguing’ with them and ‘working together’.17 Indeed, the power dynamics involved by the unequal footing of the TC and CG shows that it heavily impacted the ‘co-constructive’ nature of the experiment (as citizens were tightly bound to the TC’s agenda).

3.2. ESEC councillors and citizen participation

As mentioned in the introduction, scholars have tried to gauge elected officials’ perception of the deliberative imperative, but few have studied at interest representatives’ views. Within the ESEC, one could argue that the acculturation of councillors to citizen participation is particularly crucial to the success of the institution’s reinvention. According to almost all of the 11 ESEC members interviewed, the recent DMPs, in jonction with the CCC, convinced a majority of councillors and groups of the importance of citizen participation for the future of the Council.

There is some expertise, well expertise … an experience translated through a diversity of origins […], of ages, of occupations, that, in my opinion, undeniably brings something, a freshness, if we compare with the Temporary Commission working on its own.

Interview with councillor A, man in his sixties

It was a nice experience because it was a mix of genres. […] When you’re with different people, with different points of view, with different stories, it always brings value.

Interview with councillor B, woman in her sixties

Indeed, whether in terms of publicity for the institution and its production, changes in institutional routines or the ability to move party lines, councillors praised the contribution of citizens to the process. However, not everyone shared this enthusiasm for the integration of citizens into the ESEC, there was also resistance, doubts and questions. We thus outline here the three main arguments relating to citizen participation raised by councillors (in a similar fashion to Jacquet et al. 2015). While the latter might raise all three, they were often keen on stressing one of these positions.

1) Principled arguments affirm that citizen participation in the Council is good in itself. Around four councillors saw the integration of citizens into the ESEC as necessary to give citizens a greater input. They were very open (at least on the surface) to citizens’ ideas and advanced a performative discourse on the Council’s capacity to implement this participatory innovation.

The two rapporteurs, like me, were already convinced of the value of citizen participation, as were a number of members of the committee, who were present at all times during these times of exchange between the two.

Interview with councillor F, man in his sixties

2) Pragmatic arguments underlined the citizen’s potential in shifting positions between the various groups that structure the institution. Three councillors relied on the findings and recommendations of the CG to defend their own organisation’s position in the face of the opposition. In their opinion, citizens would allow them to produce less consensual reports but with a greater political impact.

The citizens, they are there, they say their thing. […] They don’t have the track record, they don’t have the codes whatsoever, and that allows me, for example, to be more … to be able to insist on things that I wouldn’t necessarily have insisted on. For example, the ISF [Impot sur la fortune, wealth tax], we had a big debate on the taxation of wealth… […]. Citizens say it’s not normal to have rich people, and those with large estates have to pay taxes. That is good. We can rely on it. I can lean on it.

Interview with councillor D, man in his forties

3) Finally, sceptical arguments raised concerns about the citizens’ competencies and abilities. At least two councillors expressed strong doubts about the integration of citizens in the ESEC. Their criticisms related to citizens’ lack of knowledge (technical and political), their ignorance regarding the ESEC, and to the individual (rather than collective) scope of their discourse. Other councillors were dubious of the benefits of citizens’ integration and feared that citizens could be instrumentalised (by other councillors, by outside political forces, etc.).

What intrigues me about this experience is the total lack of knowledge, in relation to young people, of their rights. But on all subjects, housing, taxes… Most of them are students, they don’t even know what a decree is, what law is.

Interview with councillor C, woman in her fifties

In fact, we end up with reflections that are individual reflections, even if the collective, at the end of the day, manages to produce something.

Interview with a councillor E, man in his sixties

These distinctions are, of course, ideal types and councillors often resorted to different types of arguments within the same discussion. The line between principled and pragmatic arguments was not always clear. We did not witness a particular affinity between the councillors’ background and their position. For example, members of labour unions could either mobilise pragmatic or sceptical arguments. Similarly, councillors representing NGOs might adopt a positive attitude towards participation, or remain pragmatic about it.

Nonetheless, it is clear from our interviews that a portion of councillors – and their groups – are still dubious of the benefits of an increased citizen participation. This typology illustrates the mixed reception of citizen participation within the institution. It helps us understand the challenges brought by the integration of randomly selected citizens into the decision-making process of the ESEC.

4. Implementing Citizens’ Participation: Flying Without Instruments?

In January 2021, an ESEC legislative reform was adopted. The Council officially now has the possibility to include randomly selected citizens in its own consultative processes (Journal Officiel 2021: art. 4).18 However, the advice of the Constitutional Council on the law underlines several limits to the association of citizens to the ESEC proceedings. The Council advances that public consultations (including through mini-publics) remain a means to the end of producing a report (Conseil Constitutionnel 2021:4). It also argues that randomly selected citizens can only be associated punctually to the proceedings, and in lower proportion to the number of councillors (Conseil Constitutionnel 2021:6). Those decisions set the citizen deliberation in the consultative domain of the ESEC, placing the future citizen juries under the control of a Temporary Commission, thereby maintaining citizen participation in a secondary role within the institution. Following this reform, the ESEC is supposed to become the institution at the crossroads of consultative and deliberative initiatives in France. The introduction of sortition, deliberative mini-publics and the enhanced powers of consultations are there to complement the Council’s role as the ‘Chamber of organised civil society’. While other options such as the replacement of the ESEC by a ‘Citizens’ Chamber’ or the creation ex nihilo of another deliberative body were mentioned at various points (both by elected politicians and sortition advocates), it is in the end a modest reform of the Council that was adopted. The absence of any pre-existing formal regulation of deliberation made it even harder for the ESEC to delineate a clear path for citizen participation. This raises three main issues: the lack of a standard of deliberation in the ESEC, the important power of the Bureau, and the instrumental nature of the reform.

First of all, at the macro-level, the marginalisation of the National Commission of Public Debate by the executive and the proactive role of the ESEC allowed the latter to position itself as the main interlocutor on the subject of citizen deliberation. As a result, it is an institution with little to no experience in deliberative process and methods that ends up in charge of citizens’ participation in France. While it is in the process of devising internal guidelines, following the aforementioned experiments, it has, so far, been flying without instruments.

Overall, regarding the Deliberative Mini-Publics it organised, the ESEC had little knowledge of the national, European or international guidelines of deliberative and participatory democracy (Gastil and Levine 2005). This situation meant that the Council had to rely heavily on private consultants. With the emergence of a ‘public engagement industry’ (Lee 2014:201; Nonjon and Mazeaud 2018) in the last decades, this is not surprising. While both the Council and the consultants praised their collaboration, it is clear that the latter had the upper hand in how the citizens’ deliberations were conducted. However, this seemed to lead to a strong segmentation of the process, with the CG being managed by the consultants (supervised by the ESEC), and the TC functioning almost ‘as usual’. We believe this division of labour is one of the reasons the aforementioned objective of ‘co-construction’ was not achieved.

Second, we argue that the current reform is a result of a powerful Bureau, and that it will reinforce its power. The executive of the ESEC played a crucial role in first convincing the various groups of the necessity of a reform, and second in experimenting with different formats (online consultations, hybrid and mixed DMPs). In both the political and media arena, representatives of the Bureau were at the front lines to defend the reform and the renewed institution.19 This performative role of the Bureau was especially clear during the parliamentary debates leading up to the adoption of the reform. Amongst the arguments advanced by the deputies in favour, the fact that the ESEC had already set up DMPs stood out. As the National Assembly rapporteur, Erwan Balanant (MoDem, centrist) expressed:

I would remind you that the ESEC already uses citizen participation, as well as sortition. Whether or not we enshrine it in law will not change this. On the other hand, we are setting the rules and securing the conditions for its organisation (Journal Officiel 2020:9742, emphasis ours).

The goal of the bill was to delineate the powers of the Council and offer guarantees on the procedure. By experimenting with citizen deliberation in the perspective of a legislative reform, the ESEC gave an argument in favour of its enshrinement into law.

However, one can note that the lack of precision in the final legislative text is symptomatic of the discretionary power of the Bureau. Apart from the vague reference to the garants (guarantors) that need to be involved when sortition is used, the Bureau has a lot of latitude in deciding when, how and why associate citizens to its proceedings. This, coupled with a certain obligation of result, means that there is a strong probability that future Citizen Groups will lack autonomy and will leave little room for dissensus.

Finally, this raises the question of the instrumentalisation of citizen participation within the ESEC. While some sceptical councillors argued that citizens could easily be manipulated (by other councillors, by politicians, etc.), few reflect critically on their own institution’s use of deliberative experiments to justify its survival. Indeed, the confrontation of the idiosyncratic form of ESEC’s interest groups with citizen participation shows how ambiguous the organisation’s reception of the deliberative imperative is. Despite a strong emphasis in the institutional communication of the Council on its new participatory dimension, the impact of either open consultations or DMPs on the actual report is less clear. While some councillors and groups do refer to the citizens’ deliberations and ideas, the link is often frail. Rather, our analysis reveals two quite distinct spaces and a strongly segmented process, with the Citizen Group in a clear subordinate position (both symbolically and politically). Behind the democratic varnish of official communication, the old structures of entrenched corporatism are still there. Not that there is anything wrong with the institutionalised representation of interest groups: in this case, the issue is rather that one cannot claim the mantle of ‘major democratic innovation’ (CESE 2020) if the results are so small.

5. Conclusion

In conclusion, the 2021 reform of the ESEC has specified new processes of public consultation within the Council, including the direct deliberation of randomly selected citizens alongside interest representatives. The experiments that preceded the reform highlighted the challenges of such a hybrid process. These difficulties are mainly due to the lack of acculturation to citizen deliberation of councillors as well as the absence of clear guidelines defining the role of all participants and more largely the parameters of the deliberation. These shortcomings will not be resolved by the reform. While the ESEC Bureau has advocated for the implementation of sortition within it, citizens are relegated to the margins of the ESEC and the management of the deliberation is subcontracted. This example of institutionalisation illustrates the difficulty of empowering citizens at the national level, especially within an institution that fights for its survival and its recognition by the public authorities. The confinement of citizen participation to those specific arenas in the decision making process makes deliberative experiments even more sensitive and dependent on political stakes faced by the institutions in charge of it.

Moreover the hybridization of interest representation and citizen democracy presents many challenges in itself. The specificity of interest groups as main interlocutors particularly shows in the tension they express between a representative mandate they are bound to and professional and specialised contributions they bring to the table. This tension has at least two effects on the mode of interaction groups have with randomly selected citizens. First, the role of councillors remains unclear as they do not usually confront citizens in their activities (by contrast with political party members or elected representatives) or at least not in their role as councillors, which is rather directed towards policymakers. Second, the political authority they have in the deliberation process appears to rely mostly on procedural (and sometimes arbitrary) criteria and not substantial content of their contribution. The way councillors as well as the ESEC Bureau took advantage of citizen participation has proven to be essentially oriented towards the advocating of the institution’s and the groups’ interests. Therefore, it also questions the sheer possibility of implementing deliberative practices in an existing institution without drastically altering its functioning. Within the current political context, citizen deliberation is on its way to become an ad hoc resource, used by interest groups and institutions to defend their causes in the public sphere, but not leading to stronger implication of citizens in the decision-making process. Quite a shortcoming from the promised new part of a deliberative system.


  1. In French, Conseil économique, social et environnemental (CESE).
  2. The direct input of citizens has also taken the form of multiple consultations by the ESEC. We do not cover these dimensions in this article.
  3. In France, an ‘organic law’ is a special type of legislation aimed at specifying the general articles of the Constitution.
  4. The NCPD is an independent administrative authority whose objective is to assess the public utility of a project by organising public deliberations (local democratic devices or more broadly participatory processes).
  5. For a complementary perspective, going from 2008 to 2019, see Aubert, 2019, chap. 3.
  6. CESE, Document de travail du groupe n°1, 24/04/2018.
  7. Sections are thematic and permanent working committees of councillors from various Groups. For example, the Section for Agriculture, Fishing and Alimentation has 28 members from all 18 groups.
  8. A broad theme that was apparently not understood by all citizens. In our February 23rd survey, the question: ‘the subject of our report is clear’ scored 6.4 out of 10.
  9. As is increasingly common, the 28 citizens were chosen by stratified random selection. Other criteria included gender parity, geographical origin (especially a balance between cities and countryside) and a diversity of socioeconomic status.
  10. The Citizen’s Group met in person three times: January 31 to February 2; February 21 to 23 and March 13. Then, meetings moved online: with three hour-long meetings on April 30 and June 22 and full day sessions on May 28, June 4–6 and June 16.
  11. In the February 21st survey, the day with the most councillors and citizens present in the same room, the top three answers to the question ‘What proposition describes the role of the ESEC members’, the top three answers were ‘dialogue with the group’ (7), ‘learn’ (6) and ‘listen’ (5).
  12. Schematically, ‘debate’ is aimed at convincing the majority to rally to a position, in a win/lose logic. ‘Dialogue’ aims at exchanging views, without taking decisions. ‘Deliberation’ is at the crossroads of this two logic, moving away from the zero-sum game of debate, while insisting on decision-making (OECD 2020:11).
  13. Which was one of the explicit goals, as ESEC President Patrick Bernasconi expressed in his introductory remarks: ‘we will co-construct this report together’ (observation, CG, 31/01/20).
  14. During the CG first meeting, one of the two rapporteurs argued that what was needed was ‘realistic proposals that don’t cost too much’ (observation, CG, 31/01/20). A similar statement was made during the second session as well (observation, CG, 22/02/20).
  15. For instance, the impact of longevity on the public accounts equilibrium was mentioned in the TC session as a potential argument for austerity measures but has quickly been disposed of for being ‘too controversial’ (observation, TC, 18/06/2020).
  16. Observation, CG, 31/01/2021, President of the TC introducing the mandate to citizens during the first weekend session.
  17. Ibid.
  18. Other significant changes include a reduction of the number of councillors; the facilitation of the right to petition; a better connection between local organisations and Regional ESEC; and an enhanced importance of the Council reports.
  19. See for instance the intervention of a Bureau member in Le Temps du débat on France Culture (Nadjar 2021),

Ethics and Consent

This research was conducted in accordance with the Declaration of Helsinki. The participants of the ‘New Generations’ DMP were informed of our presence at numerous stages of the process. Informed consent from all 15 interviewees was obtained beforehand.


This paper draws from our work as scientific evaluators for the ESEC between January and July 2020. As part of a research team consisting of ourselves, Yves Sintomer and Hélène Landemore, we produced a (confidential) report on the ‘New Generations’ Temporary Commission. We would like to acknowledge our colleagues’ contribution and help in this process. All remaining errors are ours.

Competing Interests

The data analysed in this article was gathered as part of our work as scientific evaluators for the ESEC between January and July 2020. Hugo Bonin received financial retribution for coordinating the research team. Simon Baeckelandt and Zélie Wüthrich did not. All three authors worked as scientific evaluators for the ESEC from January to October 2021, and from December 2021 to March 2022 for Hugo Bonin.

Author Contributions

We contributed equally to the research and redaction; the order of our names was chosen randomly.


1 Aubert, R. (2019). La conversion participative. Réformer le CESE, une institution marginalisée. (Master’s thesis in political science, Paris: Université́ Paris-1 Panthéon-Sorbonne).

2 Badré, M., & Dominique, G. (2019). Fractures et Transitions : Réconcilier La France. Paris: Conseil économique, social et environnemental. Avis.

3 Böker, M., & Stephen, E. (2015). The Possibility of Critical Mini-Publics: Realpolitik and Normative Cycles in Democratic Theory. Representation, 51, 125–44. DOI:  http://doi.org/10.1080/00344893.2015.1026205

4 Bonin, H. (2018). La Démocratie Hasardeuse: Élections Ou Tirage Au Sort. (2nd ed.). Montréal: XYZ éditeur.

5 Blondiaux, L., & Fourniau, J. M. (2011). Un bilan des recherches sur la participation du public en démocratie : beaucoup de bruit pour rien?. Participations, 1(1), 8–28. DOI:  http://doi.org/10.3917/parti.001.0008

6 Cautrès, B., & Muxel, A. (2011). The New Voter in Western Europe: France and Beyond. New York: Palgrave Macmillan. DOI:  http://doi.org/10.1057/9780230119802

7 CESE. (2020). La réforme du CESE adoptée au Parlement : quelles évolutions pour le Conseil ?. CESE March 15, 2021 Retrieved from https://www.lecese.fr/content/la-reforme-du-cese-est-adoptee-au-parlement-quelles-evolutions-pour-le-conseil.

8 Chamorel, P. (2019). Macron Versus the Yellow Vests. Journal of Democracy. 30(4), 48–62. DOI:  http://doi.org/10.1353/jod.2019.0068

9 Conseil Constitutionnel. (2021). Décision N° 2020–812 DC Du 14 Janvier 2021: Loi Organique Relative Au Conseil Économique, Social et Environnemental. Paris: Conseil Constitutionnel.

10 Courant, D. (2021). Citizens’ Assemblies for Referendums and Constitutional Reforms: Is There an “Irish Model” for Deliberative Democracy?. Frontiers in Political Science, 2. May 2, 2022. Retrieved from https://www.frontiersin.org/article/10.3389/fpos.2020.591983. DOI:  http://doi.org/10.3389/fpos.2020.591983

11 Courant, D., & Sintomer, Y. (2019). Le tirage au sort au XXIe siècle. Actualité de l’expérimentation démocratique. Participations, 23(1), 5–32. DOI:  http://doi.org/10.3917/parti.023.0005

12 Dubrac, D., & Teskouk, D. (2020). Générations Nouvelles : Construire Les Solidarités de Demain. Paris: Conseil économique, social et environnemental. Avis.

13 Elster, J. (1994). Argumenter et négocier dans deux Assemblées constituantes. Revue française de science politique, 44(2), 187–256. DOI:  http://doi.org/10.3406/rfsp.1994.394826

14 Font, J., Smith, G., Galais, C., & Alarcon, P. (2018). Cherry-Picking Participation: Explaining the Fate of Proposals from Participatory Processes. European Journal of Political Research, 57(3), 615–36. DOI:  http://doi.org/10.1111/1475-6765.12248

15 Gastil, J., & Levine, P. (Eds). (2005). The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the Twenty-first Century. San Francisco: Jossey-Bass.

16 Giraudet, L.-G., Apouey, B., Arab, H., Baeckelandt, S., Begout, P., Berghmans, N., Blanc, N., Boulin, J.-Y., Buge, E., Courant, D., Dahan, A., Fabre, A., Fourniau, J.-M., Gaborit, M., Granchamp, L., Guillemot, H., Jeanpierre, L., Landemore, H., Laslier, J.-F., … Tournus, S. (2022). “Co-construction” in Deliberative Democracy: Lessons from the French Citizens’ Convention for Climate. Working Papers Deliberating on Climate Action: Insights from the French Citizens’ Convention for Climate. HAL March 3, 2021. Retrieved from https://ideas.repec.org/p/hal/wpaper/hal-03119539.html. DOI:  http://doi.org/10.1057/s41599-022-01212-6

17 Gourgues, G. (2013). Les Politiques de Démocratie Participative. Grenoble: Presses universitaires de Grenoble.

18 Gourgues, G., Mazeaud, A., & Nonjon, M. (2021). From the Participatory Turn of Administrations to the Bureaucratisation of Participatory Democracy: Study Based on the French Case. International Review of Administrative Sciences. DOI:  http://doi.org/10.1177/00208523211003122

19 Habermas, J. (1981). Theory of Communicative Action: Reason and the Rationalization of Society (Volume 1). Boston: Beacon Press.

20 Haute, T., & Tiberj, V. (Eds). (2022). Extinction de Vote ?. Paris: PUF.

21 Hendriks, C. (2002). Institutions of Deliberative Democratic Processes and Interest Groups: Roles, Tensions and Incentives. Australian Journal of Public Administration, 61(1), 64–75. DOI:  http://doi.org/10.1111/1467-8500.00259

22 Hendriks, C. M., Dryzek, J. S., & Hunold, C. (2007). Turning up the Heat: Partisanship in Deliberative Innovation. Political Studies, 55(2), 362–83. DOI:  http://doi.org/10.1111/j.1467-9248.2007.00667.x

23 Jacquet, V., Schiffino, N., Reuchamps, M., & Latinis, D. (2015). Union sacrée ou union forcée ? Les parlementaires belges face à l’impératif délibératif. Participations, 13(3), 171–203. DOI:  http://doi.org/10.3917/parti.013.0171

24 Journal Officiel. (2020). Compte Rendu Intégral Des Débats. Assemblée Nationale.

25 Journal Officiel. (2021). Loi Organique N° 2021–27 Du 15 Janvier 2021 Relative Au Conseil Économique, Social et Environnemental.

26 Landemore, H. (2020). Open Democracy: Reinventing Popular Rule for the Twenty-First Century. Princeton: Princeton University Press. DOI:  http://doi.org/10.1515/9780691208725

27 Lee, C. W. (2014). Do-It-Yourself Democracy: The Rise of the Public Engagement Industry. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/acprof:oso/9780199987269.001.0001

28 Mansbridge, J. (1992). A Deliberative Theory of Interest Representation. In Mark P. Petracca (Ed.), The Politics of Interests: Interest Groups Transformed (pp. 32–57). Boulder, Colorado: Westview Press. DOI:  http://doi.org/10.4324/9780429495984-2

29 Mansbridge, J. (2012). A Systemic Approach to Deliberative Democracy. In Jane Mansbridge & John Parkinson (Eds), Deliberative Systems: Deliberative Democracy at the Large Scale Theories of Institutional Design (pp. 1–26). Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9781139178914.002

30 Nadjar, V. (Director) (2021, January 16). Le Collectif Citoyen Sur La Vaccination Est-Il Autre Chose Qu’un Gadget ? In Le temps du débat. France Culture. Retrieved from https://www.franceculture.fr/emissions/le-temps-du-debat/le-collectif-citoyen-sur-la-vaccination-est-il-un-gadget.

31 Nonjon, M., & Mazeaud, A. (2018). Le Marché de La Démocratie Participative. Éditions le Croquant. Vulaines sur Seine.

32 OECD. (2020). Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave. Paris: OECD Publishing. DOI:  http://doi.org/10.1787/339306da-en

33 Paulis, E., Pilet, J.-B., Panel, S., Vittori, D., & Close, C. (2020). The POLITICIZE Dataset: An Inventory of Deliberative Mini-Publics (DMPs) in Europe. European Political Science. DOI:  http://doi.org/10.1057/s41304-020-00284-9

34 Rui, S. (2016). La société civile organisée et l’impératif participatif. Ambivalences et concurrence. Histoire, économie, société, 35e(1), 58–74. DOI:  http://doi.org/10.3917/hes.161.0058