Re-imagining the Making of Climate Law and Policy in Citizens’ Assemblies

Abstract In recent years, climate citizens’ assemblies – randomly selected representative citizens gathered to make policy recommendations on greenhouse gas emissions targets – have gained in popularity as a potential innovative solution to the failure of governments to design and adopt ambitious climate change laws and policies. This article appraises the process and outcomes of three climate citizens’ assemblies held at the national level – in Ireland, France and the United Kingdom – and evaluates their contributions to the making of climate law and policy. In doing so, it first looks at whether citizens’ assemblies have the ability to improve the substance of climate law and suggests that they face difficulties in providing an integrated, holistic response to the climate problem. It then explores how citizens’ assemblies have fed into subsequent legislative processes to show their positive influence and draws lessons for our understanding of the role of citizens’ assemblies in climate lawmaking.



In the 20 years since the adoption of the Kyoto Protocol, 1 the number of climate laws and policies worldwide increased by a factor of more than 20: from 72 in 1997 to more than 2,315 today. 2 Yet, at the same time, these new legal and policy frameworks have been unable to provide an adequate response to what has now become a climate emergency. As a result, the capacity of democratic regimes to respond to the climate crisis has been questioned and the model of ecological authoritarianism explored as a possible alternative. 3 Conversely, calls for more, and better, public involvement in the transformations needed to move to a climate-resilient economy have multiplied. 4 The idea that climate citizens' assembliesrandomly selected representative citizens gathered to make policy recommendations on how to meet greenhouse gas (GHG) emissions targetscould provide an innovative solution to the failure of governments to design and adopt ambitious laws and policies has gained in popularity recently.
Citizens' assemblies on climate change are founded on the idea that deliberative governance produces good, perhaps even better, decisions. 5 Their outcomes are generally praised for being realistic while also ambitious. 6 Nevertheless, considerable uncertainty persists about whether such innovations can, and should, be formally integrated into institutional systems or whether they should remain ad hoc experimental exercises. A related, emerging question is whether they have the potential to become significant sources of inspiration, and positive disruption, for the making of climate law and policy.
This article appraises the process and outcomes of the climate citizens' assemblies organized in Ireland, France and the United Kingdom (UK)as the first three national climate assemblies held worldwideto assess their contributions to climate lawmaking. My objective here is not to provide an in-depth analysis of each assembly or a detailed comparative study of the three experiments. Each assembly had its unique features and operated within a specific institutional, political and social context; a dense network of deliberative theory scholars is in the process of evaluating each case separately. 7 The aim of this article is not to dwell on important questions relating to the deliberative process, including participant selection, the place of experts and internal governance structures, but rather to speak to the broader interests of transnational legal scholars. I am interested here in the implications of the use of this new deliberative instrument for the field of climate law and policy.
It is now well established that climate change is pushing, as well as exposing, the limits of existing systems of law and governance. The scholarship has been particularly interested in how climate change has disrupted the field of litigation. 8 In this article, I look at the other side of the coinlegislationto understand how citizens' assemblies challenge traditional lawmaking processes. The three case studies it analyzes serve a triple purpose: firstly, to better understand the reasons for and implications of this burgeoning trend in transnational climate governance; secondly, to tease out initial findings about climate citizens' assemblies which are becoming models for organizing similar fora in other countries; 9 and, thirdly, to gain a better understanding of the new roles assigned to citizens in climate lawmaking processes.
The analysis examines the potential contribution of citizens' assemblies to the making of climate law and policy from two different perspectives. The first question is whether citizens' assemblies can modify the substance of climate lawthat is, whether such democratic innovations can lead to new and different outcomes for climate law by offering solutions to the difficult trade-offs and complex linkages between sectors that have to be considered by climate law and policy. The second issue is one of procedure, namely, the impact of citizens' assemblies on the process of lawmaking. The aim is to assess whether climate assemblies can influence lawmaking, and to understand how the role of citizens selected by lot is conceptualized in this process.
The article proceeds as follows. Section 2 introduces the concept of citizens' assemblies and reflects on their increased popularity as a governance tool to respond to the climate emergency while Section 3 briefly presents three climate assemblies, the first three to be held at the national level worldwide. Section 4 then investigates how citizens' assemblies approach the substance of climate law and shows that they face difficulties in providing an integrated, holistic response to the climate problem. Section 5 explores how citizens' assemblies have fed into subsequent legislative processes to highlight their positive influence despite varying understandings of their role vis-à-vis lawmaking. Section 6 draws lessons from this new trend for climate law and policy, and Section 7 concludes.

' :         
A citizens' assembly is a group of representative citizens brought together to deliberate upon and make recommendations on a particular issue. It generally qualifies as a 'minipublic' that meets over at least two weekends (or the equivalent) and includes at least 40 members. 10 Citizens are selected by lot and demographically stratified to reflect the wider population, based on a number of parameters such as age, gender, ethnic background, education level, location, and employment type. They then meet (in person or online) to learn about the topic by listening to presentations by experts and stakeholders, and discuss in small and large groups. Eventually, they engage in a series of votes to agree on workable recommendations. Citizens' assemblies are founded on the idea that collective deliberation leads to better outcomes. 11 Evidence suggests that social interactions enhance human reasoning: as participants express and discuss their individual preferences, values and interests, they reflect on them, justify their positions, evaluate their beliefs and can become more open to preference change. 12 Alongside other small-scale experiments aimed at democratizing our societiessuch as participatory budgeting, e-democracy and consensus conferencescitizens' assemblies have multiplied over the last two decades. The first citizens' assembly was established in 2004 in British Columbia (Canada) to deliberate on a new electoral system. 13 Citizens' assemblies have since been held in multiple countries, including the Netherlands, Ireland and the UK, to deliberate on a wide range of topics, which include polarizing issues such as same-sex marriage and abortion rights. 14 Deliberative assemblies are often executed at the local and regional levels; national citizens' assemblies represent only 20% of deliberative initiatives. Climate change has not traditionally been a core interest for theorists of deliberative democracy. The use of citizens' assemblies as an instrument to address the climate problem is a new phenomenon, encouraged principally by the rise of grassroots movements, including Extinction Rebellion, which see it as a potential solution to the dual crises of climate change and democratic legitimacy. 16 In recent years, general dissatisfaction with representative democracy has grown in western societies, leading to an erosion of trust in political institutions and traditional political parties. 17 At the same time, public awareness of the climate crisis has increased since the adoption of the Paris Agreement in 2015, 18 to the point that climate change is now a main source of concern for European populations. 19 While there has been a decline in citizen participation in conventional politics, as demonstrated by low election turnout, 20 an increasing number of citizens have sought to have their voices heard by taking to the streets to demand urgent climate action, joining environmental protests such as Extinction Rebellion, Fridays for Future and climate strikes. 21 Policymakers have attempted to respond to such discontent by, inter alia, giving environmental issues more prominence in their political programmes, 22 organizing surveys on citizen behaviour and preferences in relation to climate action, 23 holding public debates, 24 and consulting multi-stakeholder public institutions. 25  this has also meant experimenting with the innovative tool of citizens' assemblies, at national, regional and local levels. 26 Citizens' assemblies are presented as a response to a suboptimal lawmaking process which produces climate laws that lack ambition and are unable to act adequately upon scientific evidence. These shortcomings are generally explained by the complexity of climate change as a policy issue, shaped by powerful vested-interest groups which compete to influence decision makers. 27 Climate law and policy is designed through a multi-stakeholder process that generally favours actors exerting greater socio-political power. 28 Often, citizens' voices, experiences and opinions are not adequately considered while commercial interests prevail. 29 Governments and other mandating entities that hold such citizens' assemblies present deliberation as a positive solution, capable of generating better responses to the climate emergency. The main argument in favour of such assemblies is that citizens need to be more involved in the making of climate policies and legislation that will significantly shape their lives and futures. According to mandating entities, empowering citizens' voices in climate assemblies increases the acceptance of new decisions, enhances their legitimacy, and facilitates behavioural change. 30 An additional justification addresses the climate and democratic crises together, arguing that citizens' assemblies respond to a need for more, or a different type of, democracy and represent a new method of governance to enhance public participation. 31 This rationale is found, for instance, in the assertion by French President Emmanuel Macron that 'citizens have asked for more democracy; they do not want to be simply those that respect laws any more.  32 Élysée, 'Échanges avec les 150 Membres de la Convention Citoyenne pour le Climat', 10 Jan. 2020, available at: https://www.elysee.fr/emmanuel-macron/2020/01/10/echanges-avec-les-150-membres-de-laconvention-citoyenne-pour-le-climat (author's translation).
Besides these openly stated justifications, decision makers can be driven by more questionable motivations, including legitimizing decisions ex post or greenwashing unambitious measures. 33 Indeed, while citizens' assemblies can be deemed to offer an alternative to traditional lawmaking procedures, in practice, they can be a 'model of state-sponsored citizen empowerment': 34 they often operate in close symbiosis with policymakers who could use them in self-interested ways, for instance, to strengthen the legitimacy of their own decisions or to absolve themselves from responsibility.
The question that arises, naturally, is whether deliberative democracy is well equipped to respond to these shortcomings and design 'better' climate laws. Citizens' assemblies are generally used to respond to three types of public policy problem: (i) values driven dilemmas; (ii) complex problems that require trade-offs; and (iii) longterm issues that go beyond the short-term incentives of political cycles. 35 As a 'wicked' 36 problem, the climate problem combines these three characteristics and, thus, appears ideally placed for a deliberative exercise. Climate change is, indeed, a complex and open-ended problem that invites diverging views on the appropriate response. Citizens' assemblies could partially respond to the rising social discontents about the heavy burdens that new climate policies place on already underprivileged segments of society. They offer a forum to deliberate over difficult questions pertaining to 'how communities wish to live and how losses and benefits are distributed in a society', 37 ensure that different viewpoints are being considered and, could, eventually, lead to more equitable decisions.
Until now, empirical evidence regarding citizens' deliberation in the field of climate change has been limited. Niemeyer held a small-scale deliberative exercise on climate change in the Australian Capital Region and concluded that: [d]eliberation increases the salience of common-good issues and engenders deeper forms of cognition on complex issues in ways that produce outcomes reflective of a strongly held, if latent, desire to achieve action consistent with long-term management of and the need for urgent action on climate change. 38 However, a better understanding of the climate problem by citizens does not necessarily lead to better governmental action, and the extent to which small-scale exercises can be 33 Members of France's Climate Citizens' Convention voiced these concerns loudly, in particular because they feared that the assembly would help to justify a carbon tax initially rejected by the Yellow Vests movement. 34 Lang, n. 11 above, p. 35. 35  The Irish Citizens' Assembly was the world's first nationwide experiment in citizens' assembly deliberation on the issue of climate change. An earlier attempt in Australia to set up a climate assembly had failed as a result of political resistance. 39 As such, the Irish assembly can be considered groundbreaking. Compared with subsequent experiences, however, the experiment was significantly smaller in size and less ambitious, as it was part of a broader deliberative exercise set up by the Irish Parliament in July 2016 to discuss some of the core challenges facing the country. 40 A government-appointed chairperson and 99 randomly selected citizens were brought together to deliberate on five issues: abortion; population ageing; fixed-term parliaments; referendums; and climate change. The assembly met for 12 weekends between October 2016 and April 2018 in a hotel in Dublin, with limited time to discuss five topics of considerable importance and scope. Climate change was not included initially in the draft Oireachtas resolution that established the assembly, but was subsequently added thanks to an amendment proposed by the Green Party. It was scheduled to be considered last, and over a single weekend only. However, in response to citizens' demands, two changes were made: the topic of climate change was brought forward to be considered earlier in the process, and two weekends (in autumn 2017) instead of one were devoted to it. The assembly was asked to deliberate on the following question: 'How can the State make Ireland a leader in tackling climate change?'.
The recommendations of the assembly were published on 18 April 2018. 41 A total of 13 recommendations were adopted, including four general recommendations on climate considerations and policymaking, and sectoral measures on energy, transport and agriculture/land use. The recommendations were accepted by majority vote, each receiving a high proportion of 'yes' votesan indication of near consensus among participants. citizensassembly.ie/en/How-the-State-can-make-Ireland-a-leader-in-tackling-climate-change/Final-Report-on-how-the-State-can-make-Ireland-a-leaderin-tackling-climate-change/Climate-Change-Report-Final.pdf. 42 89% and above, except on ending peat extraction subsidies within 5 years (67%) and accepting higher taxes on carbon intensive activities (80%).

France's Convention Citoyenne pour le Climat
The French Convention Citoyenne pour le Climat (Citizens' Convention for Climate) can be considered unique in the world in terms of its scale, length, ambitious mandate and allocated funding. 43 The creation of the climate assembly was announced by President Macron in April 2019 after the Yellow Vests (Gilets Jaunes) movement protested against fuel tax hikes. It was also a response to the perceived failure of the national public consultation (Grand débat national) that followed the protests with the aim of enhancing the democratic debate. The Citizens' Convention for Climate brought together 150 citizens over seven weekends between October 2019 and June 2020, with an eighth follow-up session held in February 2021. It met in Paris, at the Conseil Économique, Social et Environnemental (CESE). The assembly was given the mandate to make proposals on how France could reduce its GHG emissions by at least 40% compared with 1990 levels by 2030 'in a spirit of social justice'. 44 President Macron promised that the outcome of the assembly would be submitted 'without filter' to a referendum, parliamentary vote or decree. 45 The assembly's deliberations resulted in a 460-page report, consisting of 149 measures, 46 which combine aspirational recommendations and legal proposals. The Ministry of the Ecological Transition estimates that 40% of the measures are of a legislative nature, while 20% fall within the scope of the budgetary law; 10% would require an executive act (such as a decree), 5% fall within the remit of international negotiations, and the rest under a variety of other modalities. 47 In addition, the assembly identified three proposals that it wished to submit to referendum: these relate to recognition of the crime of ecocide, and reforming the preamble to, and Article 1 of the French Constitution. 48

Climate Assembly UK
The UK set up a climate citizens' assembly a few months after France. The assembly was commissioned by six parliamentary select committees 49 after the adoption, in June 2019, of a legally binding target to reach net-zero emissions by 2050. 50 The 110 citizens met over six weekends between January and late May 2020, initially in a hotel in Birmingham, and virtually for the last three sessions during the COVID-19 lockdown. They were asked to deliberate on the following question: How should the UK meet its target of net-zero greenhouse gas emissions by 2050?
The assembly's work resulted in a 550-page report entitled 'The Path to Net Zero', launched on 10 September 2020. 51 The report starts by identifying general principles which the assembly deemed should drive climate action. It then distils recommendations into nine thematic categories, each outlining the considerations that government and Parliament should bear in mind when making decisions, assessing various scenarios for possible futures, and evaluating different policy options to materialize such futures. Contrary to those of the French Citizens' Convention for Climate, the conclusions of Climate Assembly UK were not written by the assembly's members: rather, they represent a synthesis of the work carried out by the assembly, with a detailed analysis of citizens' positions, as well as direct quotations from citizens and pictures taken during the sessions.

         
Citizens' assemblies on climate change are presented with an open question, which creates unique challenges compared with those of assemblies working on discrete problems such as the legalization of same-sex marriage or abortion. 52 With climate assemblies the options are not binary, and expert presentations tend to be less oppositional. In this section I investigate how citizens' assemblies have responded to two challenging characteristics of climate law: its wide scope, and its cross-sectoral nature.

Defining the Scope of Climate Law
When citizens are asked to deliberate on climate law and policy, they are in effect given a vast and perhaps over-ambitious mandate. Indeed, climate law and policy intersect with most aspects of social and economic life and, as a result, 'all fields of law … have to adjust and incorporate the new way of doing things internally'. 53 This all-embracing mandate has had two seemingly contradictory outcomes: on the one hand, deliberations have necessarily had blind spots; on the other hand, assemblies have sought to further extend their initial mandate.
The wide-ranging scope of climate action is a major challenge for citizens' assemblies. Participants are generally overwhelmed by the density of the information shared with them and the variety of issues upon which they are asked to deliberate: in both Ireland and France, the assembly members asked for an extra session to be scheduled to have more time to assimilate the new data and deliberate adequately. Even so, climate assemblies are inherently unable to deliberate on all aspects of climate law and policy. The themes chosen for deliberation vary depending on the specifics of each country, their policy priorities and the key sources of GHG emissions. For instance, agriculture featured prominently in the discussions of the Irish citizens' assembly because the sector plays a central role in the economy and represents the single largest contributor to Ireland's GHG emissions. 54 Notably, because their mandate has so far concentrated on reducing emissions, assemblies focus on climate mitigation, with little attention given to adaptation. Thus, policy sectors such as water, which can be affected by both climate impacts and mitigation policies, were overlooked in the three climate citizens' assemblies. 55 On the other hand, citizens' assemblies generally do not conceive of the climate problem stricto sensu but rather conceptualize it within the broader question of environmental protection. 56 For instance, Climate Assembly UK paid special attention to the objective of 'protecting and restoring the natural world', which it ranked fourth in terms of priorities to achieve net-zero emissions. 57 It considered the environmental and biodiversity impacts of energy sources, noting, for instance, the negative impacts of bioenergy associated with intensive land use and deforestation. 58 In France, the focus of the assembly on biodiversity resulted in an implicit extension of its initial mandate: despite initial pushback from its governance committee (in charge of its work programme), the climate assembly reinterpreted its emissions-focused mandate to include biodiversity protection, irrespective of whether it could contribute to reducing GHG emissions. 59

Addressing Trade-Offs and Cross-Sectoral Issues
As a cross-sectoral problem, the root causes of climate change are found in a variety of sectors, which include energy, transport, industry, housing, and agriculture. Climate 54 Environmental Protection Agency, 'Ireland's National Inventory Report', 2021, p. xviii, available at: https://www.epa.ie/publications/monitoring-assessment/climate-change/air-emissions/Ireland_NIR-2021_ cover.pdf; Ireland, Recommendations of the Citizens' Assembly, n. 41 above, Recommendations 11-3. 55 President Macron noted this oversight explicitly, mentioning that the country is wasting a significant amount of water, and investment in distribution networks is needed: Élysée, 'Réponse du Président de la République aux Membres de la Convention Citoyenne pour le Climat', 29 June 2020, available at: https://www.elysee.fr/emmanuel-macron/2020/06/29/le-president-emmanuel-macron-repond-aux-150citoyens-de-la-convention-citoyenne-pour-le-climat. 56 The same remark applies to local climate assemblies: e.g., the Oxford Citizens' Assembly on Climate Change included the question of biodiversity protection in its discussions: law and policymaking therefore demand high-level coordination to ensure that new measures do not carry unintended consequences. 60 In theory, citizens' assemblies could help to resolve important trade-offs in the making of climate law and policy: by elucidating citizens' preferences, they can guide policymakers when confronting difficult choices. 61 However, this proves difficult to realize in practice. Citizens' assemblies generally consider their recommendations as forming a coherent project which ought to be considered as a whole. 62 However, the reason for their holistic approach is that they recognize the need for multiple policy interventions (i.e., selecting only one recommendation from their proposed package would not respond adequately to the climate problem) and is not related to the intrinsic unity of their project. Assembly members in Ireland and France gave little attention to how their recommendations related to each other and did not reflect on whether their combination could have unintended negative consequences. Climate Assembly UK, on the other hand, tackled the question of the multi-sectoral impacts of climate measures more directly. Participants identified and sought to mitigate the potential negative impacts of climate policies: in particular, they emphasized the value of policies with 'co-benefits' that were deemed to facilitate the social acceptability of policy measures. 63 In the words of an assembly member, '[t]he pros and cons should be constantly under review. We must have a holistic view of what is good for the planet as well as humanity'. 64 For instance, the environmental impacts of various energy solutions were assessed, acknowledging the potential for cascading impacts such as the risks that changes in the energy sector posed for food production. Citizens expressed concerns about the impact of onshore wind and solar power on land use: 65 some assembly members noted that their support would 'depend on how the land is used', 66 63 Climate Assembly UK report, n. 51 above, p. 6. 64 Ibid., p. 17. 65 Ibid., p. 369 (onshore wind) and p. 384 (solar). 66 Ibid., p. 373. water'. 67 Climate Assembly UK, therefore, provided a forum to identify points of tension between conflicting policy priorities. However, the outcomes of the citizens' deliberations did not result in clear guidance for decision makers on how to resolve trade-offs.
Moreover, the three citizens' assemblies did not represent an alternative to the siloed approach generally found in climate lawmaking. Indeed, climate assembly members often divide their time between plenary sessions and thematic subgroup meetings. In the UK, three subgroups were created: on travel, housing, and food/consumption. In France, five subgroups were formed on food, housing, production and work, consumption, and transportation.
The thematic breakdown of assembly participants is justified by the fact that time constraints prevent citizens from gaining expertise in all aspects of climate policy. However, it can be problematic in the context of climate lawmaking. Indeed, it fails to recognize interdependencies between diverse policy areas such as climate, energy, water and food, with the risk that sectoral decisions result in conflicting or unintended policy outcomes. The subgroups in France had a quasi-monopoly over the recommendations under their thematic responsibility: they discussed and drafted proposals that were only minimally debated in the plenary at a fairly late stage. In addition, researchers who observed the deliberations noted that, in reality, the thematic groups had naturally divided into more specialized groups, each working on separate proposals. 68 This is problematic because deliberative exchanges, considered essential to produce quality outcomes, 69 were in reality often limited. Additionally, the proposals were discussed in silos, without members necessarily having a holistic overview of how they might fit within the broader project and how they might affect, or be affected by, the other recommendations. Notably, UK assembly members observed that they were not given 'adequate opportunity to participate in the production of recommendations by topic groups to which they were not assigned'; this, they considered, eventually made it difficult to cast an informed vote. 70 A thematic breakdown of deliberations can also marginalize the many cross-cutting issues that cannot be considered in breakout groups. Remarkably, the role of the energy sector in climate mitigation was left out of the thematic discussions of the assemblies in the UK and France. Climate Assembly UK considered the topic of energy in its plenary, and devoted significant time to discussing the advantages and disadvantages of various energy sources. 71  significant hurdles and is considered to have been one of the assembly's main shortcomings. 72 The governance committee of the Citizens' Convention for Climate identified five cross-cutting issues that did not fit the thematic breakdown: (i) energy; (ii) constitutional reform; (iii) communication, engagement, education and training; (iv) funding; and (v) protection of the natural environment and biodiversity. 73 A cross-cutting group named escouade (squad) was created after the assembly's second session in November 2019 to discuss these topics. However, assembly participants who were not members of this group criticized it heavily: it was seen as non-legitimate because, unlike the other groups the membership of which was decided by lot, participation in the escouade was based on self-selection. In addition, the escouade was seen to have a more strategic role because of the nature of the topics it covered. As a result of these disagreements, the escouade was disbanded at the end of the fourth session: three of its topicsincluding energywere transferred to the thematic groups and twoon the constitution and fundingwere discussed between sessions, through an online platform and webinars. 74 Still, the energy sector does not feature heavily in the outcomes of the Citizens' Convention for Climate because citizens were not able to gain as much knowledge and devote as much deliberation time to the topic compared with the other policy areas they considered in their expert subgroups. 75 The decision to organize deliberations in smaller groups has notable impacts on the substance of the assembly's outcomes. However, this structure also has implications for how decision makers engage with an assembly's recommendations: they can reproduce the thematic divisions into legislative drafts, as in the case of the French Climate and Resilience Law, 76 or, on the contrary, adopt a more critical attitude and query the representativeness of proposals made by a subgroup of the assembly. 77 It is therefore evident that the wide scope and cross-sectoral nature of climate change law has fundamental repercussions for the work of the assemblies. Their success relies in part on finding a structure that enables deliberations that are focused enough to offer quality outcomes, while also encouraging policy integration and coordination to ensure crosssectoral coherence.

   '      
Various parameters can be used to assess the success of citizens' assemblies. They can be considered successful if, for instance, deliberations have been of a high quality, if assembly members have become more educated about a public policy problem, or if media coverage has increased public awareness. The parameter I use for the purposes of this article is whether, and if so how, citizens' assemblies held at the national level feed into subsequent legislative processes. Traditionally, mini-publics have been used to help to inform policymakers, but not necessarily to influence policy directly. 78 Climate assemblies generally follow the same logic: they have so far not been given the mandate to make binding decisions for the polity. Instead, they are bound to their mandating entity by a 'moral' 79 rather than legal contract. Despite their purely advisory role, citizens' assemblies appear to have had a positive influence on decision-making processes. My analysis below has two objectives. The first is to offer an overview of what happened in the realm of climate lawmaking after the assemblies submitted their work and, in particular, to describe how their recommendations were taken into account. The second, and related, objective is to understand how the relationship between climate assemblies and existing legislative processes has been conceptualized.

Climate Assemblies as Mechanisms to Elicit Informed Preferences
Climate assemblies are intended to act as consultative mechanisms for eliciting informed preferences on the basis of which lawmakers can make decisions. Although taking place in distinctly different political, legal and institutional circumstances, the climate assemblies in both Ireland and the UK have helped to shape policy and legislation by clarifying citizens' preferences: in doing so, they have empowered decision makers to act and informed policy debates.

An assembly to encourage climate leadership in Ireland
The Irish Citizens' Assembly, by identifying the people's concern for climate change and their appetite for more governmental action in this area, strengthened political mobilization and spurred the drafting of a comprehensive Climate Bill.
The resolution presenting the mandate of the Irish Citizens' Assembly closely linked the deliberative exercise to the traditional institutions of representative democracy. It provided that 'the Assembly will report and make recommendations to the Houses of the Oireachtas' and set up a follow-up process by requiring the government to respond in the Houses of the Oireachtas to each recommendation of the assembly. However, it also gave the government much discretion on how to use the work of the assembly, including an implicit option to reject proposals by stipulating that the government, 'if accepting the recommendation', would need to indicate whether it plans to hold a referendum on the proposal. 80 The Chairperson of the assembly working on climate could only express the 'hope' that 'the work of and recommendations made by the Citizens' Assembly can contribute to the wider public engagement and national discourse on the issue'. 81 In practice, the legislative branch gave significant attention to the work of the assembly. The submission of the report in April 2018 was followed by the establishment of a Joint Oireachtas Committee on Climate Action to consider its recommendations and assess how it might 'inform Ireland's climate and energy plans'. 82 After a six-month examination of the assembly's findings, the Joint Oireachtas Committee produced a report in March 2019 entitled 'Climate Change: A Cross-Party Consensus for Action', 83 which contains 42 priority recommendations.
The report was presented as a 'robust response to the Report of the Citizens' Assembly'. 84 It references the assembly's recommendations, regularly mentions that 'it accepts' them, contextualizes them in terms of state of play, scientific and technological knowledge, and sets out possible means of delivery. This is not to say that there is complete symmetry between the assembly's report and that of the Committee. Indeed, some recommendations were not discussed in detail because of time constraints, such as those relating to food waste 85 and adaptation measures, 86 while other issues not covered by the assembly were added. 87 The report was unanimously endorsed by the Oireachtas in May 2019. 88 At the same time, Ireland became the second country in the world (after the UK) to declare a climate and biodiversity emergency. 89 The work of the Committee then fed into a cross-government All of Government Plan to Tackle Climate Breakdown (since renamed the Climate Action Plan), published in June 2019. 90 The Plan identifies 183 actions that cover the broad themes discussed in the citizens' assembly, including energy, agriculture and transport. It incorporates some of the assembly's recommendations, such as raising the carbon tax 91 and accelerating the transition to electric vehicles, 92 while ignoring others, such as proposals to create an independent body on climate change, 93 reduce food waste, 94 and end peat extraction. 95 The Plan was followed by the introduction of the new Climate Action and Low Carbon Development (Amendment) Bill 2020 (then revised in 2021), which commits Ireland to net-zero carbon emissions by 2050. 96 It is evident that the recommendations of the assembly were given serious consideration and formed the basis for subsequent parliamentary and governmental work on climate change. Admittedly, the extent to which the work of the citizens' assembly influenced climate legislation is difficult to assess. Indeed, the mere organization of a citizens' assembly is indicative of the mandating entity's readiness to act. A few months before submission of the report, the Taoiseach (head of government) Leo Varadkar had described Ireland as a 'laggard' in climate policy in the European Union (EU) and expressed a willingness to remedy that. 97 Moreover, it is difficult to assess whether Ireland's climate law and policy would have been any different without the citizens' assembly because the recommendations were not particularly detailed. For instance, the proposal that the state 'should take a leadership role in addressing climate change through mitigation measures, including, for example, retrofitting public buildings, having low carbon public vehicles, renewable generation on public buildings' 98 can be followed in spirit only because it lacks quantitative or qualitative targets that would provide more specific, and constraining, guidance.
Nevertheless, the assembly helped to shape climate policy and legislation by signalling a relative consensus not necessarily on specific climate measures but rather on the will of the people to see their government assume climate leadership. The Joint Oireachtas Committee on Climate Action observed that the assembly's work provided a 'new impetus in climate policy' by highlighting 'a strong desire on the part of citizens for the State to take far more ambitious actions to make Ireland a leader in tackling climate change'. 99 Similarly, the Irish government's Climate Action Plan noted that the work of the assembly revealed a 'near consensus on the need for strong and early action to reduce Ireland's greenhouse gas emissions and to make Ireland resilient to future 91 Ireland, Recommendations of the Citizens' Assembly, n. 41 above, Recommendation 3/action 8. 92 Ibid., Recommendation 9/action 72. 93 Ibid., Proposal 1. 94 Ibid., Proposal 12. 95 Ibid., Proposal 7. 96 Climate Action and Low Carbon Development (Amendment) Bill 2020, published 7 Oct. 2020; Climate Action and Low Carbon Development (Amendment) Bill 2021, published 23 Mar. 2021, available at: https://www.gov.ie/en/publication/984d2-climate-action-and-low-carbon-development-amendment-bill-2020. 97 N. Sargent, 'Taoiseach Tells EU He Is Not Proud of Ireland's Role as Europe's Climate "Laggard"', The Green News, 18 Jan. 2018, available at: https://greennews.ie/taoiseach-tells-eu-not-proud-irelandclimate-laggard-role. 98 Ireland, Recommendations of the Citizens' Assembly, n. 41 above, Recommendation 2. 99 Report of the Joint Committee on Climate Action, n. 62 above, p. 2. climate impacts'. 100 By putting forward such views, the assembly helped to inform policy debates and empowered decision makers to act more ambitiously to tackle climate change.
An assembly to implement the net-zero target in the UK In comparison with the Irish example, the climate assembly in the UK was held in a different legislative context: as the country had already adopted a legally binding net-zero target, the assembly concentrated its work on the types of measure needed to meet such target. The mandate of Climate Assembly UK was to 'advise Parliament on how people want [it] to meet the net zero target, and to suggest policies that the Government can implement to secure success'. 101 The advisory nature of the assembly's mandate gave policymakers discretion on whether, and how, to use the assembly's recommendations. Commitments, albeit vague, regarding how the assembly's recommendations would be used were made only after the assembly concluded its work and launched its report on 10 September 2020. 102 The Secretary of State for Business, Energy and Industrial Strategy (BEIS) Alok Sharma announced that the recommendations would inform the UK's new nationally determined contribution, which would cover 'all the policy areas' found in the report. 103 A few months later, the UK government announced a tenpoint plan for a green industrial revolution, which included certain policies that had been put forward by the assembly, such as the rapid prohibition of the sale of new petrol and diesel cars. 104 To date, however, the direct impacts of the assembly on government policy are difficult to identify because decision makers do not explicitly reference the assembly's recommendations to justify new policies. Instead, the alignment of planned policies with the assembly's priorities is emphasized, 105 leaving aside recommendations that can be in contradiction. 106 Contrary to the Irish example, no clear follow-up mechanism was established to ensure that the assembly's recommendations would be considered by decision makers. However, the BEIS Committee opened a standing inquiry into the findings of the assembly to 'review on a regular basis the government's engagement and interaction with the findings of the assembly and progress in implementing its proposals', 107 seeing it as an opportunity to 'scrutinis[e] the Government's climate change policy and progress on the [net-zero] target'. 108 In addition, the citizens' assembly asked that a follow-up session, not initially provided for, be organized to explain 'how Parliament and government have taken the assembly's recommendations on board' and 'what has been taken into account, what hasn't and why'. 109 The assembly's main pathway to influence has been to offer new evidence on public preferences that policymakers consider is needed to inform the development of new climate policies that are 'feasible and equitable'. 110 The assembly's recommendations have been conceived by Parliament and government as 'new data' 111 to understand public preferences and to offer a template of publicly acceptable measures to meet netzero targets. 112 The UK government described the impact of the assembly's work in the following terms: Findings from the Assembly form an important addition to the Government's evidence base on assessing the UK public's understanding, attitudes and perceptions around net zero. This evidence base is being used in several ways, including to inform development of the Net Zero Strategy, the Heat and Buildings Strategy and the Transport Decarbonisation Plan, as well as to inform the development of public engagement approaches. 113 The holding of the climate assembly helped in raising awareness on the part of policymakers about the need to engage the public on how to achieve the net-zero target, both to 'communicat[e] the challenge, and giv[e] people a say in shaping future policies'. 114 However, because the assembly is seen as one tool among multiple strategies of public engagement, 115 its contribution to the development and implementation of climate policies has so far not been distinctively visible.

Climate Assemblies as Actors in the Lawmaking Process
Contrary to Climate Assembly UK, the Citizens' Convention for Climate in France was defined from the start in terms of its public policy impacts. At the heart of the assembly was the commitment of President Macron that the conclusions of the assembly would be submitted 'without filter' ('sans filtre') to the executive, the Parliament, or the people by referendum. However, the full implications of this mandate eventually could not be realized as a result of several obstacles described below. Instead, as the assembly progressed, its role evolved from an impossible function of pre-legislator to that of unique participant in the lawmaking process.

The impossible mandate: Lawmaking 'without filter'
The quasi-legislative mandate given to the citizens' assembly in France became the defining element of the relationship between the citizens and the mandating entity, to the extent that whether the assembly's work would be implemented 'without filter' became the only way to measure the success, or failure, of the experiment. Yet, this mandate was in effect impossible to meet. Indeed, while the no-filter promise was based on the assumption that traditional lawmaking processes need to be challenged to enhance climate action, it lacked detailed consideration of its implications for the constitutionally protected work of the executive and the legislative branches.
The no-filter commitment was not defined before the Citizens' Convention for Climate started its work. The assembly's mission statement written by the Prime Minister remained vague on this important point; it merely explained that the outcomes of the assembly would take the form of a report summarizing the discussions, identifying the legislative and regulatory measures deemed necessary to meet the emissions reduction objective, and highlighting the measures that the citizens wished to see submitted to a referendum. 116 The citizens of the assembly became very attached to President Macron's promise that their work would be used 'without filter'. While the citizens' assemblies in Ireland and the UK perceived their function to be advisory, the citizens' assembly in France had a significantly more tense and more complex relationship with the government, which was viewed with suspicion and criticized for failing to respond adequately to the climate crisis.
The no-filter commitment represented the potentially transformative power given to the assembly to change the status quo, but also raised the question whether the assembly could, indeed, produce an unfiltered piece of law. Arguably, the essence of lawmaking requires a filtering process: given constitutional constraints, 'without filter' cannot be understood to mean that a text would be transformed into law without debates and amendments. More plausibly, it would mean that the executive will not 115 Ibid., paras 35-6. 116 Lettre de mission du Premier Ministre, n. 31 above. cherry pick between proposals but instead submit the whole package for legislative discussion. 117 However, this is not how 'without filter' was understood either. Instead, the pre-legal mandate of the citizens was protected only by the moral commitment of the President, who used the vagueness of his commitment to interpret it differently throughout the process. In January 2020, when he came to talk to the citizens' assembly at a questions and answers (Q&A) session, he defined the notion of 'without filter' in terms of the nature and specificity of the recommendations: The more precise, clear and detailed it is, the more 'without filter' it can be … if it is just an idea, an intention, then, in a sense, you are giving it to people who will need to set it to music and therefore there will be a filter … On the other hand, if, at the end of your work, you give legal texts … I commit that they will be given either to the Parliament or the French people as you propose them as long as they comply with our constitution. 118 He also gave himself the right to sideline certain citizens' proposals, which he used in June 2020 to reject three recommendations. 119 In the end, the assembly did not produce a piece of legislation that could be sent directly to the Parliament. Rather, it offered a complex mix of recommendations of varying prescriptiveness and force, which made the process more convoluted.
In December 2020, at another Q&A session with the citizens, the President appeared to have changed his interpretation of the no-filter promise: he argued that filtering meant ignoring a proposal, but that discussing a recommendation and putting it aside for valid reasons did not constitute filtering. 120 Such an understanding significantly reduced the no-filter commitment, given that the effectiveness of a measure, in terms of carbon reduction, often hinges on the technical details, on the material and temporal scope of the measure, and applicable exceptions. As a result, many of the assembly's recommendations have been translated into climate law or policy, 121 but with significant caveats, which, many would argue, implied that a filter had been applied to the recommendations after all. 122 Such is the case, for instance, with the government's plan to phase out domestic flights on routes where rail alternatives of under two-and-a-half hours are available, 123 which is less far-reaching than the assembly's proposed four-hour threshold. 124

Citizens as participants in the lawmaking process
The effect of the citizens' assembly on climate law in France should not be measured solely against the theoretically and practically flawed benchmark of the no-filter promise. Indeed, the work of the assembly has led to significant changes in climate legislation and policy. Legislative work on climate change after the assembly's report has explicitly been grounded on the citizens' recommendations and resulted in initial regulatory measures adopted by the Conseil de Défense Écologique (a special council of ministers) in July 2020 125 and a new climate bill adopted by Parliament in July 2021. 126 A constitutional reform was also planned but eventually abandoned. 127 The results, even if often considered disappointing compared with the assembly's recommendations, 128 are acknowledged to have increased governmental ambition. 129 Beyond actual law and policy changes, the citizens' assembly transformed the process of climate lawmaking. This process has not necessarily been one of 'co-construction of solutions', 130 as the Prime Minister had vowed, 131 nor the 'partnership' that President Macron promised. 132 On submission of the assembly's report, the executive branch took complete control and became the sole architect of the new climate law. The assembly participants did not become lawmakers but they deeply challenged the foundations of lawmaking. Notably, the assembly was given two opportunities to meet with the President to ask questions, express criticism and discuss concerns, with each meeting lasting several hoursan exercise that would not be possible before elected representatives as the President is constitutionally prohibited from debating openly with the Parliament. 133 By enabling an open debate between citizens selected by lot and elected representatives before the lawmaking stage, the citizens' assembly played a unique role in the pre-construction of the law.
In addition, the assembly also influenced the lawmaking process itself: indeed, after having submitted its report in June 2020, members of the climate assembly were invited to have regular meetings with government officials and to attend governmental consultations with external stakeholders. 134 By explaining and defending their recommendations, they extended their role in the lawmaking process significantly, outside any official mandate or structure. Their presence can be assumed to have increased the pressure put on lawmakers to take urgent climate action, but the assembly members found themselves unprepared for engaging with lobbying activities and lamented the fact that they were considered as only one voice against more powerful interest groups. 135 In sum, the place of the climate assembly has been ambiguous vis-à-vis legislative processes: on the one hand, it was given the unique capacity to pre-legislate and its recommendations were used subsequently as the basis of a new climate law; on the other hand, its relationship with political institutions was ill-defined as it rested on an impossible-to-keep presidential promise and required significant involvement of assembly members after submission of their report to take part in the actual drafting of the law.

   
This final section highlights initial lessons that can be learnt from the deliberative turn in climate law and policymaking, and remaining questions which may influence the success of making climate law through deliberation. I consider the consequences of my analysis in Sections 3 and 4 to emphasize that the nature of the climate problem raises multiple challenges for deliberation, and that the different conceptions of the role assigned to climate assemblies raise important questions about how to involve citizens to make 'better' climate laws.

Deliberating on Climate Law
Citizens' assemblies on climate change have been faced with the complexities of using law to solve the 'wicked' problem of climate change, as illustrated in Section 3 on the intersectionality of climate law. Below, I highlight three attributes specific to climate change, which need to be taken into account when designing citizens' assemblies and which entail challenges for the deliberative process: (i) its multi-scalar governance; (ii) the variety of legal and non-legal intervention tools; and (iii) the need to address the rights of future generations.
Firstly, citizens' assemblies deliberating on climate are faced with the difficulty of advising national governments on a multi-scalar issue. While the societal or institutional issues addressed by citizens' assemblies can be country-specific, the climate problem commands a 'glocal' response, 136 spanning global as well as subnational levels. As a consequence, citizens have to work within existing legal frameworksrelative to local, EU and international levelswhich they do not have the power to change in the context of the assembly. Climate assemblies have responded differently to this difficulty. In Ireland, for instance, the assembly rejected recommendations that did not fall within state competences, 137 while in France the assembly made proposals to guide the national government on how it should position itself in decision-making processes at local and supranational levels. 138 The organization of local citizens' assemblies on climate change suggests that deliberative democracy exercises held at different governance levels can be complementary, but the potential tensions or synergy between such exercises at different governance levels will require further attention.
Citizens' assemblies are also faced with the need to decide on the place given to legal interventions in the fight against climate change. The outcomes of citizens' assemblies have shown that citizens are able to break down their vision into a variety of measures that range from the very technicalincluding standard setting 139to the vastly transformationaladvocating the creation of new institutions 140 or constitutional revisions. 141 Climate assemblies generally see legislation as one tool, but not the only one, to build the new societal project that was imagined during deliberations. The recommendations of the citizens' assemblies combine aspirational objectives to steer policymaking towards a certain vision of the future with economic instruments (tax incentives, public investments) and regulations (prohibitions, obligations, sanctions, technical norms). In addition, pedagogical approaches to encourage individual behavioural change (information and education) feature prominently in the recommendations of assemblies, as they insist on the need to empower citizens instead of constraining them through legal measures. 142 A final question related to climate deliberations is whether citizens' assemblies, which do not have the electoral mindset of representative democratic institutions, can protect the young and future generations that will be most affected by climate change. Our political and legal institutions have not been conceived to govern the future. Legislators generally fail to respond to climate change: while they are expected to pursue the 'public interest', 'they make decisions on how to use other people's resources, not their own', 143 which is particularly detrimental to the young and future generations. Citizens' assemblies, at least hypothetically, offer two grounds to overcome this myopia. The first stems from the theory of deliberation, which considers that deliberative assemblies can better contribute to long-term planning because careful listening and thinking encourages us to fight our cognitive biases against the future. 144 This has been visible in the work of assembly members, who have regularly mentioned their willingness to recommend ambitious action to preserve the rights of 'future generations', 145 or at least of their children, as symbolized by the planting of a tree at the CESE for the newborn of one of the assembly's citizens in France. The second ground relates to the institutional design of citizens' assemblies, which can accommodate better representation of younger generations, including by increasing their comparative share in the participant selection, 146 by inviting organizations representing young and future generations to present their views to assembly members, 147 and by engaging with other innovative structures such as children's parliaments. 148 6.2. Making 'Better' Climate Laws?
The recommendations of citizens' assemblies help us to imagine climate law in its ideal formthat is, detached from external influences and lobbying, and putting citizens' interests, social justice and long-term perspectives at the heart of lawmaking. However, a study of climate citizens' assemblies also highlights that further reflection about what is needed to make 'better' climate law is essential. The way in which assemblies are used by decision makers reveals different assumptions regarding how to improve legislative responses to the climate emergency. A first model assumes that a better understanding of public preferences will benefit climate legislation, while a second model prefers to give citizens a more direct role in lawmaking. Both models have inherent shortcomings. The public preference model, favoured in Ireland and the UK, responds to a broader trend of public dissatisfaction with government action on climate, expressed most visibly in climate protests and litigation attempts. It is driven by the argument that 'meaningful public engagement will help build stronger public consent' for the net-zero transition. 149 Gaining citizens' support for climate policies is seen as necessary to facilitate their implementation and to increase ambition levels over time. 150 However, better understanding public preferences is by no means a magic bullet to solve climate change. Indeed, past experiences have shown that even when governments seek to clarify public preferences and concerns, they do not necessarily take them into consideration; 151 and, when they do, it is more often out of self-interest rather than to legislate for the common good of society. In addition, it is important to acknowledge that such preferences have inherent shortcomings: first, they embody citizens' choices between different options, selected by and presented to them by experts, but not automatically preferences in an absolute sense; second, they represent views of a uniquely well-informed group that might have initially been designed as a microcosm of the larger population, but the views of which evolved as a result of the process.
The second model assumes that closer citizen involvement in lawmaking results in better laws, but this hypothesis raises a fundamental question about whether citizens have the ability to legislate. Indeed, what distinguishes a citizen from a legislator, apart from electoral legitimacy, is the legislator's knowledge of the legal texts and processes. In contrast, citizens' assemblies are designed to be representative of the general public, with citizens chosen through stratified random sampling, based on common categories (including age, gender, ethnicity, education level and/or social class, place of residence). As a result, the vast majority of citizens have very limited knowledge of, inter alia, constitutional law or environmental law. Yet, the Citizens' Convention for Climate was designed on the basis that citizens are able to produce legislative text if they are presented with the necessary foundational knowledge and supported by a team of lawyers. 152 However, even with the tools to pre-legislate, significant doubts remain about whether citizens can overcome the powerful private interests at play in the legislative process, as illustrated by the discrepancy between the recommendations put forward by the French Citizens' Convention for Climate and the measures found in the Climate and Resilience Law. 153 The alternative, subject to constitutional rules, would be to organize a referendum that legitimizes the work of citizens selected by lot, a common option for citizens' assemblies. However, the pertinence of this option is limited by the broad scope and technical nature of climate legislation. 154 In sum, both models illustrate the need to think further about what the aims of the process arebeyond the general objective of responding to the climate emergencyin order to tailor the deliberative mechanism adequately.


In a 2019 article published in Transnational Environmental Law, Natasha Affolder called on environmental scholarship to focus more on its 'people', on 'who' influences transnational environmental law. 155 While she highlighted the role of judges, scholars and funders, my focus here has been to understand what it means for law when we include citizensand not only their democratically elected representativesin the lawmaking process. The climate crisis demands new ways of thinking about the law, and citizens' assemblies, founded on the idea that deliberation can lead to better outcomes, are increasingly perceived as a promising solution. The examples of three citizens' assemblies held at the national level provided an opportunity to examine how the aspirations of deliberative theory can be realized in practice.
This article has shown that citizens' assemblies cannot be considered a 'silver bullet' solution to the climate emergency because they face significant limitations in terms of how they can renew the substance and procedure of lawmaking. However, they offer new avenues to complement more traditional modes of environmental public participation, such as focus groups or public debates. They appear to be a positive force which can disrupt the status quo: so far, the risk that ultimately the recommendations of climate assemblies would be set aside and forgotten has not materialized in the three countries studied. As the legacy of the three citizens' assemblies consolidates over time and as the number of countries holding climate citizens' assemblies increases, how assemblies reshape climate lawmaking will become clearer. Importantly, how to square the complexity of climate change governance with deliberative ideals and how to conceptualize the relationship between citizens' assemblies and traditional decisionmaking processes will need further reflection. Overall, the emergence of citizens' assemblies as a new instrument in the toolbox of climate law teaches us important lessons about climate lawmaking, and offers fresh perspectives on how to implement the objectives of the Paris Agreement. 156