Summary

It seems unlikely that green growth alone (or absolute decoupling, measures of efficiency) will be sufficient to reduce greenhouse gas (GHG) emissions in time ^[1]. In order to not spend the world’s carbon budget in the coming decade(s), it seems we will also need to change our behavior (measures of sufficiency). However, it is very difficult for political parties to advocate such measures - and to properly inform the general public - out of a fear of electoral loss. Yet the longer we wait with effective GHG emission reductions, the more impactful such measures will need to be. The solution this text proposes is an (online-first) preferendum, in which a government or parliament polls the broad population about which of a set of additional measures, compiled ideally through a deliberative process, they would prefer to see taken up (first). The platform would provide background information for each of the measures, estimates of their GHG-reduction capacity, financial cost, and social and broader ecological impact. Citizens select several measures with varying ‘ambition levels’, ideally until the targets are met - or, agree to pay the price for not doing so; translated into e.g. GHG emission rights.

Sample platform

This sample application demonstrates such a platform: https://ecorendum.be/

Some notes:

• These are dummy proposals, costs, impacts and targets. The actual numbers should be calculated by governments and administrations, universities and other experts.
• Ideally this set of possible measures gets compiled through citizens’ assemblies, such as the Convention Citoyenne pour le Climat in France; and/or another online platform in which the entire population and all universities can participate (see below why I am not/no longer focussing primarily on this initial deliberative phase).
• The site currently lists only a couple of dummy proposals. In reality there could be many more (e.g. the 149 proposals from France’s Convention Citoyenne pour le Climat), and the site’s design should be adapted accordingly. E.g. as to not overwhelm or discourage users, the number of proposals listed could be randomly limited (but weighed by impact), along with the reduction targets.

Introduction

The problem and traditional democratic solutions

It is widely known that, and understood why, it is so difficult for the party-political democratic system to effectively deal with climate change and similar problems: problems that manifest themselves on a long timescale, yet their mitigation would pose a (deemed) short-term sacrifice. No political party dares to take such action, fearing electoral loss, instead postponing the solution to future generations or at least legislatures; being tempted to bet on technological innovation (decoupling; measures of ‘efficiency’ rather than ‘sufficiency’); or free-riding. At best, green parties will attempt to describe, ‘package’ the solutions as desirable; focus on the upsides; e.g. cleaner air.

Politicians are not correctly informing the electorate on the severity of the problem and the impact of the necessary measures to deal with them; hushing citizens with promises of future innovation, downplaying the problem, or focusing on other political topics. Neither are commercial media, social media or mainstream entertainment (exceptions aside, it’s not what generates most revenue). Citizens, especially older generations who no longer receive an education and are fixed in their worldview, are thus in no way forced to acknowledge the severity of the problem. As long as there are political parties or factions which bring a message that everything will be alright, people vote for this perceived legitimate option. Again, the only way out of this seems to be either to try to ‘tempt’ voters by focusing on upsides - which becomes a less realistic story every day and typically only works for certain (‘nature-minded’) groups - or by appealing to reason and morality. It doesn’t help that ecological issues are ‘politicized’, seen as a left-wing issue (some might say this is rather arbitrary; one can certainly imagine reasons why it could have been a right-wing issue).

It becomes increasingly clear that the party-political system and their traditional solutions (betting largely on future innovation and/or trying to ease voters into accepting measures by portraying them as desirable) have not resulted in the necessary rate of change ^[1:1] and seem unlikely to in the near future ^[2].

The ironic and poignant result of this inability to act decisively now, is that the measures which will need to be taken will likely only get ever more drastic and thus less ‘politically realistic’. It seems likely that this will result in increasing polarization, with on the one hand scientists and climate and ecological activists demanding ever more extreme measures to try and stay within identified boundaries; on the other hand people who refuse to accept these ever more extreme measures, provoked by (typically conservative) politicians and social media bubbles. It seems almost unavoidable that this will result in increasing social unrest, even in the coming decade, as we lose ever more hope of still being able to avoid the worst.

Litigation

The only way out of such a game theoretical suboptimal equilibrium is to change the rules of the game. It is typically precisely the role of a government to change the rules. But what if the executive and legislative powers in themselves are trapped in a stalemate position? Then the judicial power can still come in. This is precisely what’s happening. After decades of cases against individual polluters, recently, in several countries, more and more cases ^[3] are started and won against governments (started by the ‘Urgenda’ case in The Netherlands) and large companies - giving politicians an ‘alibi’: they had no choice but to impose stringent measures, they were forced by court. Unfortunately, this is an extremely slow process, and not guaranteed of success: in Belgium, the ‘Klimaatzaak’ case lasted 6 years and even though the Belgian governments were convicted, they were not given any binding targets, and a year later, they seem to largely ignore the ruling.

Deliberative democracy as part of the solution

It is being thoroughly debated that deliberative forms of democracy, ideally with randomly selected citizens, either on an ad-hoc basis (such as la Convention Citoyenne pour le Climat (CCC) in France) or on a permanent, institutional basis (such as the futures assemblies proposed by e.g. William McAskill ^[4] and others), might be the democratic alternative which would depolarize, depoliticize these topics. Indeed, the results of the CCC, along with many other examples ^[5] have shown that such assemblies can come up with a coherent, sufficient, realistic set of proposals supported by all slices of the population.

However, it is becoming clear that the bottleneck resides not in identifying effective solutions, but in implementing them. Academics largely agree that the role of deliberative panels should be only to _propose _solutions, ideally raise public awareness and broaden the public discourse. The deliberative body does not have sufficient legitimacy for the proposed measures to be binding, even when organized by a country or region’s executive or legislative powers. The constitutions of some countries like Belgium even explicitly forbid non-elected citizens to create legislation. It is typically agreed upon therefore that the output of the deliberative process (the proposed measures) should be passed back to the legislative and/or executive powers - elected politicians, who do have the necessary legitimacy to decide on their implementation. However, it turns out that it is far from certain that measures will get implemented. For example, after the CCC, even though president Macron proposed beforehand he would implement the measures ‘without filter’, only 15 out of 149 proposals were taken up without modification, 55 after being modified or weakened, while the remaining 79 or 55% were not applied (situation 2/04/2021) ^[6].

The problem remains the same: when the proposed measures are transferred back into the party-political system, the well-known flaws described above result in them being largely ignored. It turns out that party politicians still assume that, even though an informed representation of the country’s inhabitants were in favor of the measures, the uninformed general public will not necessarily agree to blindly defer their democratic power to them (a problem well described by, among others, Christina Lafont in ‘Democracy without shortcuts’). Although there is some theoretical evidence that measures proposed by a well-organized citizens’ assembly are more easily accepted than the same measures getting proposed top-down, and there is some initial empirical evidence which backs this up ^[7], there is reason to believe that indeed, party politicians’ fear for electoral loss is not entirely ungrounded. After all, no matter how small the loss, in the party-political zero-sum game, the best strategy for all parties is to avoid the risk.

So how to get the broad public informed, supportive of the necessary measures; not just the randomly selected citizens who participated in the assembly?

Preferenda

Whether a set of possible measures was compiled through deliberative, party-political or technocratic means, this text proposes to put the measures to a vote to the entire population - as a form of direct, participatory democracy. Not as a typical referendum (e.g. “Do you accept this set of 149 measures produced by the climate assembly, yes or no?”), but as a preferendum or multiple-choice referendum. In a preferendum, the enquirer polls the preference of the enquiree for a set of options; in this case, all or a subset of the proposed ecological measures. This concept was recently proposed as part of the solution to solve the climate measure problem by Belgian historian and one of the world’s leading experts on deliberative democracy, David Van Reybrouck ^[8].

Advantages

A preferendum combines the advantages of representative, deliberative and direct/participatory democracy. With representative democracy, it shares the ‘one person, one vote’ principle. With deliberative democracy it shares the fact that citizens must inform themselves; both regarding the severity of the problem (the sort of measures needed to meet targets), as well as regarding possible (and impossible) solutions. It also avoids (party-political) polarization and encourages nuance. Finally, it has the advantage of direct democracy that citizens feel directly involved and heard in politics, and that it is difficult to question the legitimacy of the outcome.

For politicians, preferenda also offer large advantages. Inherently, the preferendum informs them of the relative support for different proposed measures. As described above, it takes two difficult tasks out of their hands – informing the populace of the severity of the challenge, and imposing the required set of strict measures – all without risking electoral loss.

Issues with Van Reybrouck’s proposal

In his original text ^[8:1], Van Reybrouck proposes to offer a subset (e.g. 20) of the entire set of measures to each participating citizen, so as to not overload people with dozens of measures (e.g. the 149 proposed by the CCC). In his proposal, citizens would rate and rank these measures: for each measure they would indicate the degree to which they ‘agree’ with them, e.g. ‘strongly disagree’, ‘somewhat disagree’, ‘neutral’, ‘somewhat agree’, ‘strongly agree’ (rating) and they would be able to select their top 4 priorities (ranking).

I fear that this way, citizens would simply select those measures which would have the least impact on their personal life - and that these are not necessarily the ones which would most effectively reduce GHG emissions. This seems to be confirmed through a recent poll performed by Kantar Public among 9000 adults in 9 Western countries ^[9] - for instance, citizens think it is very important to ‘reduce waste and increase recycling’, ‘stop deforestation’ or ‘protect endangered animal species’, which are all rather abstract, while almost nobody thinks it is important to ‘reduce meat consumption’, ‘ban fossil fuels vehicles’, ‘increase the price of products that do not respect environmental criteria’, ‘reduce travel by planes’ or ‘favor the use of public transport over cars’ - which, in general, are precisely the sorts of behavioral changes which would most effectively, surely, reduce GHG emissions, for the lowest public cost.

Additionally, there is no guarantee that the selected measures would add up to the required GHG reduction target; cheaper measures are not necessarily preferred; and there would be no penalty to choosing solutions which would have a high environmental or social cost, or a high degree of uncertainty.

I believe all of these concerns can and should thus be tackled, by a) estimating the GHG emission reductions of each measure; b) estimating the financial cost, social impact and broader ecological impact such as nitrogen pollution; and c) stimulating respondents to select sufficient measures in order to meet internationally agreed upon targets (possibly not only GHG reduction targets, but also other social and environmental targets), or clearly lay out the consequences for not doing so.

Organization

Like the organization of citizens’ assemblies ^[10], the initiative should be taken by the government or the parliament, while it should be executed by an independent group of experts, lead by a varied coordination group and scrutinized by a supervisory panel including prominent citizens, a wide range of experts (legal/political, communication, security and privacy, IT/UX, experts in deliberative democracy,...) and ideally participants of the preceding citizens’ assembly.

The extent to which the constitution or legislation with regards to referenda can be applied would differ by country and must be further researched. As a starting point, every citizen who is otherwise allowed, or obliged, to vote, should also do so in the preferendum. Obligating the participation in the preferendum (especially when voting is not obligatory) has both advantages and disadvantages which should be weighed per country, based on its history, previous experiences with referenda, etc.

Process and tool

Compiling and preparing the measures

1. A set of possible measures is selected, independent of party political dogmas, spanning the entire ideological spectrum. It is ideally constructed through a deliberative process with randomly selected citizens representative of the region’s populace, assisted by the help of experts (domain experts, legal experts, administration, stakeholders, scientists,...) and optionally politicians. Only measures are retained which a) have an effect on the main variable to influence (e.g. GHG reduction), and b) the organizing level of government has control over (e.g. with regards to GHG reductions in European member states, measures concerning sectors under ETS are less relevant - unless the participation in, or functioning of, this system by itself is questioned).
2. The set of measures is translated into concrete bills when possible (changes to legislation or decrees).
3. Measures might often have several ‘ambition levels’, with different target amounts, start dates or a different implementation.
4. For each measure and ambition level, its main impact is calculated, e.g. the estimated GHG reduction by 2030 (possibly laid out over a certain timespan); along with its wider ecological impact (e.g. on other planetary boundaries ^[11]) and social and humanitarian consequences (both locally and globally). Also, the financial cost of implementation is calculated (again possibly laid out over a certain timespan).
5. All political parties are allowed to give their brief opinion on each of the measures.
6. Each measure, although also summarized briefly, is extensively documented, linking to: the original material which it was based on during the deliberative process; any reports from think tanks or other institutions relating to the matter; separate academic papers; newspaper articles; other media such as recordings of debates, documentaries; explainer videos on YouTube; related measures; etc.

(Steps 5 and 6 can continue throughout the rest of the procedure)

Call to vote

7. All citizens who are entitled or obligated to vote in regular elections, should also be called to vote on the preferendum. Largely the same means can be used: an official invitation, frequent mention on national television, etc. Ideally the referendum is obligatory.
8. The default way to vote is through an online website, but with offline backups.
9. Citizens receive a number of days or weeks to vote, so they have time to inform themselves. On the online platform, their results are stored during this time.

The online preferendum tool

10. All these measures, along with the primary and secondary targets and total cost, are presented on an extremely user-friendly and accessible website. Visitors are shown how to use the website through both text and video; and the actual tool ‘guides’ the user through the different sections, explaining step-by-step how the site works.
11. Each measure has a ‘frequently asked questions (FAQ)’ section, answering critical questions, both in short and long form, again with plenty of links.
12. Measures are displayed in a random order. If there are too many (e.g. more than 30), only a random subset of measures would be presented to each user; and the targets would be scaled down respectively.
13. Ideally, it should be possible for all (authenticated) citizens to comment on proposals; however, strict rules apply to ensure polite, constructive dialogue. Ideally this is community-moderated, with a reputational system (cfr. e.g. Stack Overflow).
14. Top-rated comments should be included in the FAQ if possible, with the debate which they stemmed from linked.

An offline backup

15. Citizens who do not have access to the website, or lack the digital skills, can go and cast their vote in physical locations, much like digital voting booths. They receive ‘training’ on how to use the tool at the voting campus, and are assisted if necessary.
16. If digital voting booths are not possible, a physical alternative with less interactivity is possible.
17. In order to prepare for their physical vote, citizens who decide not to vote online, receive an information brochure describing the possible measures.

The voting process

18. The user can select as many proposed measures as he or she wants. If a measure has different ‘ambition levels’, one of these ‘variants’ must be specified.
19. As the user selects or deselects measures, he or she sees the total cost, primary target, secondary targets and total impact change.
20. If not selecting sufficient measures to meet the main target translates into a penalty; typically a financial cost, this is displayed (e.g. the cost of emission rights on the European emissions trading market to make up for unmet GHG reductions). Optionally, this amount is even translated into an estimated tax increase; on average, or even applied to the person’s specific situation, which he would optionally be able to specify.
21. The user is encouraged to keep selecting measures until the target is met.
22. When satisfied with their selection, the user submits their vote.
23. When their selection of measures does not meet the primary target, the possible consequences (additional to the penalty) are clearly explained, and the user must explicitly indicate to have read and understood these consequences.
24. Each citizen can only vote once.

Results

All users’ results are anonymously (yet if possible and legal, along with some metadata such as age, education level, region,...) stored in a database: the measures and ambition level users selected, or, if they did not meet the primary target, the penalty they instead agreed to pay. This allows ranking measures relatively in terms of approval rating or acceptance. It sketches the willingness of citizens to take the necessary measures for the common interest; even when many citizens ultimately claim they are not willing to take sufficient or even any measures and, after being explained the implications, instead pay the consequences.

Remarks and challenges

Acceptance

In order for citizens to accept both the exercise and the set of proposed measures, several precautions must be taken:

• Ideally, the set of measures was compiled in a deliberative fashion; ideally through a combination of a citizen’s assembly such as the CCC, and online (crowd-moderated) input. The former is widely researched and tested; the latter is less well known, certainly interesting to investigate, but outside of the scope of this text.
• It should regardless still be possible to accept input from citizens on the proposals. They should not get the feeling they are not heard or are not taken seriously.
• Any often requested measures which were not withheld (either because they are physically, financially, socially or legally impossible; or because they are outside of the domain of the organizing government), should regardless be mentioned somewhere, with a thorough explanation of why they were not included.
• It is vital that interest groups and public figures trusted by all sorts of societal groups publicly endorse the exercise, acknowledge the need for stringent measures, and the legitimacy of the proces. For this, they should be involved early on in the trajectory; already during the deliberative phase.

Choice of subject

A legitimate question is whether the public will accept the top-down choice of the subject of the preferendum - as proposed here, the selection of measures to mitigate climate change, while limiting broader ecological impact and ensuring social justice. E.g.: why not a selection of measures to prevent migrants from entering the country?

As argued by MacAskill ^[4:1] and others, a possible solution is to focus on the rights of future generations, who do not yet have a voice/vote. Second, the form of preferendum as described here is particularly useful when certain targets must be met by combining different measures; maintaining a broad overview of the impact on a broader ‘system’, in this case the ecosystem. This is particularly difficult to achieve through any other traditional, non-digital, democratic tool.

Finally, another reason why this subject lends itself well for this kind of preferendum is that, in European member states and possibly other states, the targets have been imposed by the European or another higher level of government: the target itself is not so much a political option, it is a given, and clearly, the typical party political tools on the national level have proven to be insufficient to meet them.

Of course, if this exercise proves to be a successful precedent, it could be used to achieve other kinds of (ecological) targets.

Bypassing the parliament

If, as is most often the case, the preferendum gets initiated by the government (or the president or prime minister), on a subject of their choosing, it can be seen as effectively bypassing the scrutiny of, and possibility of opposition in, the parliament.

A possible answer relates to the above remark concerning the choice of subject. If the subject is imposed, an a-priori given not subject to political debate, at least this decreases the weight of the counter-argument. Additionally, nothing obstructs the parliament from questioning the government’s decision to organize the preferendum or its high-level form. The rest of the implementation should be left to independent instances.

Online nature, inclusion, practical difficulties

The typical argument against e-democratic innovation is that it risks leaving out a large portion of the populace who are either unable or less able to participate online. This will certainly remain a valid argument for several decades to come, and systematically responding to it would be the subject of many studies in itself and is outside the scope of this text.

However, in the West and particularly in Europe where this proposal is most relevant, the argument is decreasing in validity every passing year as the percentage of people with access to (cheap) mobile devices and the internet increase, along with the number of people getting accustomed to working online. The question then becomes whether the (organizational) challenge posed by reaching and involving those still left out still outweighs the enormous potential of this sort of extremely effective new form of democracy. Is the cost of not leveraging these e-democratic tools not getting much higher than that of taking the necessary organizational measures to involve them? For example, The Netherlands have statistics of access to the internet at home and daily usage rates. Even in the age group of 65-75, 96,3% have access to the internet at home and 79,3% use the internet daily. At the age group of 75+, this drops to 79,6% and 52,4%, respectively. In total, 3% of the population does not have internet access at home. That amounts to about 400.000 adults in The Netherlands.

Since every citizen who is allowed or obligated to vote would get a written call to participate in the vote by postal mail, just like it would happen during regular elections, the argument that certain groups of the populace are difficult to reach also does not apply. Citizens are assumed to by default vote online, authenticating ideally through their id card or some other national way to securely and uniquely identify (possibly combined with more convenient ways to authenticate afterwards; such as email/password, facebook, etc). The off-line alternative is opt-in, and only for these citizens, an information brochure must be published and accommodation must be foreseen - presumably using the same infrastructure and organizational channels as used during the regular parliamentary elections.

Let us not forget that one reason why people have difficulties finding their way online, is because often, web applications are extremely badly designed - not in the least government websites, which citizens typically have to interact with. Nowadays however, developers have sufficient experience with UX (user experience) design to make the experience as easy and frictionless as possible. Much of the preparation (educating people how to use the tool) must also only be done only once and can then be re-used or at least best practices and expertise can be shared. With such numbers of users, it should be possible to provide the absolute best possible user experience.

Let us also keep in mind that the proposed solution should not be perfect, it need only be better than the current party-democratic system - which is rapidly steering our society towards the brink.

Regardless, this remains a large challenge and potential achilles’ heel, which must be thoroughly researched and possible risks mitigated.

Adoption

The representative democracy is typically skeptical of handing over her power. Although the ‘deliberative wave’ probably cannot be stopped anymore, direct democracy is yet a step further. Getting executive or parliamentary power to resort to this kind of new democratic tool would require two things.

First, the current representative-democratic powers must be convinced that it is in their benefit to resort to this new tool. It becomes ever clearer that the alternatives are rather bleak. As explained above, not taking the required measures today risks causing more and more societal polarization, and the measures to take tomorrow will get ever more extreme. The sooner measures can be taken, the easier. And in order not to suffer electoral loss by taking the required measures of sufficiency, this tool allows to ‘outsource’ them.

Even so, for various reasons beyond the scope of this text, it is unlikely that politicians will choose to resort to this experimental tool when they are not sure of broad support - when citizens are not demanding for it. The time for these tools is now. The reason why people aren’t asking for them is first of all because people generally still associate democracy with elections - not with deliberative or direct alternatives. It is the task of various actors in society to popularize these alternative democratic concepts, e.g.:

• Activist/civil society groups can plead for it, such as ‘Extinction Rebellion’ world-wide, ‘Het Burgerparlement’ in Belgium or ‘Burgerberaad voor Klimaat en Milieu’ in The Netherlands;
• New political parties can put it high on their agenda and use deliberative or participative tools themselves, such as The Five Star Movement in Italy, Agora in Belgium, or Beyond Politics in the UK;
• Academics can further research the topic; and together with communication and IT experts assist in organizing local experiments, and learn from them;
• Local governments can continue to experiment locally, using commercial facilitation from companies such as CitizenLab;
• Media can report on the issue, and public intellectuals can write opinion pieces;
• Celebrities and thought leaders can mention it;
• Artists and especially directors can include it on television, in series and films.

Activists, civil society groups and public intellectuals should, backed by all the above, lobby at elected politicians and the government, and pave the way as much as possible.

Next steps

Ideally, this concept would need a lot more research on various domains, as described below. However, unfortunately, the time we need new democratic tools to solve the question of how to implement measures of ‘sufficiency’ is now. The West is spending the world’s remaining carbon budget - and deteriorating the environment - quickly, and with every passing year, remaining hope decreases and chances for extreme polarization increase. That is why I believe we must advance on both fronts at the same time; trying to maximize effect while minimizing the risk of creating negative precedents.

Research

Several fields related to this proposal could use extensive theoretic and especially empirical research. The academic field of deliberative democracy has been booming over the past decade and the past years particularly, experiments and adoption are increasing exponentially (although set back by the Covid pandemic). But in recent years the field seems to have especially come to understand its limitations, thanks to the work of e.g. Lafont, and hence, thought leaders like Van Reybrouck started to shift their attention to possible ways to resolve the issues through direct democracy. A few possible research topics that come to mind - although far from exhaustive:

• With a more basic version of the tool, as proposed by David Van Reybrouck, would citizens systematically choose the least infringing and less effective measures? There is some research suggesting this ^[9:1].
• Do citizens accept the top-down selection of this subject, even when it is well explained?
• How many citizens will refuse to acknowledge its legitimacy, refuse to partake and will simply not vote?
• When a government follows the results of the preferendum, do citizens still ‘punish’ the majority parties?
• To what degree are countries’ constitutions adapted to these tools?
• How to best explain the required concepts to a broad audience, or how to differentiate between different audiences?
• Which ways of presenting the measures (online and offline) is most effective; how many can be shown at once;...?
• How much would the exercise cost; both the online and offline part?
• To what extent is it possible to actually estimate the cost and impact of different measures? (presumably this is a separate and already mature field of research)

Real-life experiments

Of course, any real-life experiments would count as invaluable empirical research projects - like the CCC has been in France for deliberative democracy. Identically, while practically applying these concepts, using the best knowledge at that time and only if high quality is guaranteed, any of these real-life preferenda should be carefully followed (and guided) by academics. As such, experiments do not necessarily have to be executed by governments. Possible real-life experiments initiated by non-elected bodies could include:

• A media platform, or coalition of platforms, organizing a large-scale public enquiry; e.g. in France based on the results of the CCC;
• An activist or civil society group organizing it, sponsored by e.g. foundations supporting democratic innovation.

Campaigns

As described in the ‘Adoption’ section, these sorts of democratic renewal can only get adopted if they - aside from being well backed up by research - are demanded by the broad population. They will have no legitimacy except when organized by legislative or executive powers, who will not do so without clear support from their voters. As described above, all societal actors believing in this concept can play their role in raising public awareness around this new form of democracy - one worthy of the 21st century.

Notes

---

1. https://iopscience.iop.org/article/10.1088/1748-9326/ab842a ↩︎ ↩︎

2. https://eeb.org/library/decoupling-debunked/ ↩︎

3. https://en.wikipedia.org/wiki/Climate_change_litigation ↩︎

4. https://philpapers.org/rec/JOHLIR ↩︎ ↩︎

5. https://www.appropedia.org/List_of_climate_assemblies ↩︎

6. https://reporterre.net/Convention-pour-le-climat-seules-10-des-propositions-ont-ete-reprises-par-le-gouvernement ↩︎

7. http://www.odoxa.fr/sondage/mesures-de-convention-citoyenne-seduisent-francais-a-lexception-notable-110-km-h/ ↩︎

8. https://www.nrc.nl/nieuws/2021/10/08/geef-burgers-echte-invloed-op-het-klimaatbeleid-met-het-preferendum-a4061204 ↩︎ ↩︎

9. https://kantar.turtl.co/story/public-journal-04/page/1 ↩︎ ↩︎

10. https://researchportal.vub.be/en/publications/deliberative-minipublics-has-made-it-to-mainstream-politics-a-dis ↩︎

11. https://www.stockholmresilience.org/research/planetary-boundaries.html ↩︎