Towards Global Cooperation: The Case for a Deliberative Global Citizens' Assembly

In an important article published in this journal, Dryzek, Bachtiger and Milewicz (2011) champion the convocation of a Deliberative Global Citizens’ Assembly (DGCA). In this article, I aim to further strengthen the case for a DGCA by addressing: (i) why a DGCA is likely to take a long-term perspective in the global interest and (ii) why it is so vital that a global institution should do so. I start by analyzing the nature of the issues requiring global policy. These issues, I will argue, are typically global cooperation problems. Cooperation problems pose two major challenges. The first is to prevent freeriding – i.e. serving one’s immediate interests at the expense of the global interest. The second is to align on an efficient global policy. In both respects, I will argue, a DGCA is a good candidate to yield desirable results (and is likely to do better than current supranational institutions).

• Citizens' deliberation is well-suited to solve free rider and coordination problems. • By acquiring legitimacy, a deliberative global citizens' assembly could pressure sovereign nations to follow its recommendations.

The case for a deliberative global citizens' assembly
In an important article published in this journal, Dryzek et al. (2011) argue that in order to address the democratic deficit in global governancewhich is widely recognized as they point outwe should consider a 'deliberative global citizens' assembly' (DGCA). In their article they target proposals advancing alternatives such as a 'popularly elected global assembly' (PEGA) (e.g. by Falk and Strauss, 2001).
Dryzek and colleagues offer a number of reasons why a DGCA is preferable to a PEGA.
A DGCA, they argue, would be easier to accept for nondemocratic stateswho would not have to hold elections to provide delegates for the global parliamentand states standing on their sovereigntysince it is less likely to be perceived as a direct challenger of their national parliaments (Dryzek et al, 2011). Moreover, it doesn't pose a problem of 'expressive voting' in which voters would use elections for the global parliament erroneously to pass judgment on their national government. Finally, it could serve as a 'focal point', making international politics in general more deliberative.
While these are all valid reasons, the most important one they offerin my opinionis that: a global institution needs to be able to take a long-term perspective and consider the global good (Dryzek et al, 2011). A PEGA, they point out, is not likely to take such a perspective because elected and mandated representatives of states are likely to focus on the short-termthe next electionsand view a global assembly as a venue to further the interests of the particular group they represent. A DGCA, to the contrary, is likely to take a more long-term perspective. The participants, as they point out, have no strong connection with their state's government. Moreover, empirical evidence suggests that citizens' deliberation often leads to proposals taking a long-term perspective (Dryzek et al, 2011).
I fully agree with this analysis. In this paper, I aim to further strengthen the case for a DGCA by addressing: (1) why a DGCA is likely to take a long-term perspective in the global interest; and (2) why it is so vital that a global institution should do so. I start by analyzing the nature of the issues requiring global policy. These issues, I will argue, are typically global cooperation problems. Cooperation problems pose two major challenges. The first is to prevent freeriding, that is, serving one's immediate interests at the expense of the global interest. The second is to align on an efficient global policy. In both respects, I will argue, a DGCA is a good candidate to yield desirable results (and is likely to do better than current supranational institutions). As signposted earlier (1), in addition to pointing at the empirical evidence in support of this claim, I will offer an explanation as to why I believe this is the case.
In the following section, I describe the global cooperation problems we are facing and the current failure to solve these problems in an efficient manner. In the third section, I argue that the current institutional design of the UN (and of other supranational institutions), which offer a platform for 'inter-national' negotiation, is the culprit. In the fifth section, I argue that the integration of a DGCA in the UN could help to address these shortcomings. I support this claim by pointing out evidence from social experiments with deliberative citizens' assemblies and offer an analysis of why such deliberative citizens' assemblies are better suited to deal with (global) cooperation problems. In conclusion, I briefly address the realist concern that such an assembly would wield no power over sovereign nations. I argue that its proposalseven if advisory in naturecould gain the necessary proverbial 'teeth' by acquiring legitimacy.

A vicious circle of global cooperation problems
Among other problems, climate change, poverty, mass migration, overpopulation and conflict present us with a vicious circle of global problems that mutually reinforce each other. These problems are global problems not only because they affect (directly or indirectly) all of humanity, but also because they can only be dealt with efficiently at the global level. 1 Briefly and grossly oversimplifying, climate change threatens many people with destitution due to failing crop yield and the rising sea level. This in turn would force those climate victims to migrate. According to the climate scientist Norman Myers (2002), up to 200 million people could be forced to migrate if the average global temperature would rise to 2 degrees Celsius above pre-industrial temperatures. In a globalized world, however, people do not only migrate to escape famine, they also migrate to wealthier countries to escape relative poverty. Mass migration in turn keeps the poorer regions from which many of the young and entrepreneurial people migrate poor and creates tension (conflict) in the host countries (Collier, 2013). The arrow between migration and conflict points in both directions since conflict, of course, leads to mass migration. Today almost 70 million of so-called refugees have fled their home because of armed conflicts. 2 Conflict also leads to poverty and poverty to conflict. This is called the 'conflict trap'. Armed conflicts destroy local economies and (mainly young and male) people in poor regions will more easily take up arms (not in the least because they are desperate and see it as a means of survival). Poverty also leads to overpopulation. As countries and regions within countries get wealthier they typically witness a drop in fertility (especially if girls have access to education). Overpopulation in turn keeps groups in poverty and finally could exacerbate climate change in the future. While at presentit is important to emphasizethe bulk of greenhouse gas emission comes from rich, industrial nations, a larger world population could lead to increased fossil fuel burning, cattle raising and deforestation.
The way out of the vicious circle requires us to eradicate poverty and prevent escalating climate change which are the supply sources of the vicious circle described above. If we control climate change and eradicate poverty, we prevent mass migration and the conflict that follows in its wake as well as unsustainable population growth which threatens to burden our planet even more. Eradicating poverty and dealing with climate change present us with a cooperation problem at the global level (as does dealing with mass migration). Cooperation problems arise when a commonly beneficial outcome requires us to coordinate on a successful solutionwhich I refer to as a 'coordination problem'or to prevent freeriding by some of the actors involvedwhich I refer to as a 'free rider problem'. Freeriding occurs when parties profit from the effort of others without contributing anything themselves, thereby preventing the group beneficial outcome of cooperation to materialize. Note that cooperation problems do not fall neatly in one or the other kind. In fact, most cooperation problems contain both coordination and free rider problems.
With regard to poverty, the economist Jeffrey Sachs (2005) has argued that the poorest countries are caught in a 'poverty trap'. Due to a lack of resources, infrastructure, technology and human capital, they cannot enrich themselves in the current global economy. Sachs estimates (writing in 2005) that it would cost around 175 billion dollars a year to eliminate extreme poverty over time. According to Sachs, given such a contribution, the poverty trap would largely disappear over time and the poorest countries could end up in a positive spiral and be able to develop on their own. Provided of course that the political leaders of the developing countries implement sensible policies in the interest of the citizens. Ironically, donations from developed to developing countries total much more than the required 175 billion dollars. In 2017 official donations from developed countries to developing countries totaled around 147 billion dollars. Add to that charitable donations (by individuals, foundations and corporations) which totaled 410 billion dollars in 2017 and of which a sizable portion goes to developing countries and it should be clear thatif Sachs' estimates are in the right ball parkthere are actually enough funds being donated to eradicate extreme poverty. 3 While eradicating poverty is a complex and multifaceted problem, one major issue preventing us from doing so is that foreign aid lacks coordination. What 'USAID' does in terms of development is independent of what European Mass migration Overpopulation countries do and certainly to what China does. In addition to a total lack of vision, this fragmentation also leads to very high administrative costs. It makes aid depressingly inefficient. The need for international policy coordination has not escaped the attention of the donor and developing countries. In 2005, an agreement was signed in Paris in which a hundred ministers and representatives of aid organizations promised to make efforts to increase the effectiveness of development. Their main resolve was to strive for the harmonization of policies and a result-oriented approach. 4 So far, however, they have not succeeded. Too often aid does not result in economic development (Moyo, 2009;Rajan and Subramanian, 2008). An important reason for this lack of coordination leading to ineffective aid, I will argue in the next section, is that the context of international politics stands in the way of coordinating global policy efficiently. This difficulty to coordinate on a supranational level is also evident from the migration crisis in Europe, in which EU countries prove unable to agree on how to deal with and absorb the migration stream. So, dealing with poverty (and migration) presents us mainly with the challenge to coordinate the actions of many different parties to achieve a common goal (since the current donations should suffice to eradicate poverty given a coordinated and efficient policy) but also harbor a free rider problem (nations and other potential donors are still incentivized in the short-term to restrict donations and keep the funds for themselves or close their borders in the case of migration, thereby freeriding on the effort of other nations to deal with these issues).
Climate change, on the other hand, presents us with a free rider problem. While reducing the emission of greenhouse gasses is in the long-term interest of all nations, they do nevertheless profit in the short-term by not bearing any costs in reducing emission (such as reducing industrial activity that emits CO2, investing in technology that harvests climatefriendly energy or in the research on new ways to harvest climate-friendly energy). To deal with climate change we must therefore devise global policy measures in the long-term interests of all of the stakeholders of this planet and prevent freeriding by actors heeding their short-term economic interests (e.g. ensuring that all or at least the great majority of nations agree to contribute to reducing greenhouse gas emission). Free rider problems, however, typically also contain coordination problems. In addition to preventing freeriding, the actions of all parties involved must be coordinated to achieve forceful global action in the light of the climate crisis.
Today freeriding in the context of the climate crisis is rampant. Since 1995, 24 climate summits have been organized and the result can only be described as disappointing.
The last summit in Katowice ended on a disillusion note: we would no longer strive for an increase of 'only' 1.5 degrees above pre-industrial world temperature, but for an increase of 2 degrees. The reason is simply that many nations do not want to bear the necessary costs to deal with the threat effectively. Again, the culprit -I will argue in the next sectionis the institutional design in which international politics operate.

Diagnosis: the failure of inter-national negotiation
In order to solve these global cooperation problems, we need a global institution that orchestrates global cooperation in the long-term global interest. It must outline the way nations are to coordinate their actions and prevent freeriding from nations. Thanks to the global institutions we have todaythe United Nations in the first place, but also global institutions such as the World Bank, the International Monetary Fund (IMF) and the World Trade Organization (WTO)many initiatives for global cooperation have been undertaken (e.g. summits to deal with climate change and to eradicate poverty). The UN is an institution with much merit, not in the least because of its important contribution to making our era the most peaceful era in human history. Nevertheless, it often fails to deal with global cooperation problems effectively.
The root cause of these failures, I believe, lies with the current institutional design of the UN (and other supranational institutions such as the NATO and the EU). They offer a platform for 'inter-national' negotiation by representatives of nations (or other political units). Global conferences such as climate COPs (conferences of parties) bring together delegates from nations who negotiate with each other on the course of action to take. This, I will argue, prevents an efficient solution to these global cooperation problems. It creates incentives and a dynamic which predictably lead to outcomes which are not in the global long-term interest. On the one hand it creates an incentive to free ride (it falls short of solving free rider problems), on the other hand it often stands in the way of reaching an agreement on how to tackle the issues at stake (it falls short of solving coordination problems).

An incentive to free ride
The representatives negotiating with each other on global policy, are typically (but not exclusively) elected politicians or diplomats mandated by elected politicians. The electoral system in force in democratic countrieswhich make up the majority of influential countries in those meetings (but by no means all influential countries, consider China and Russia)holds those representatives accountable to the population of the nation they represent. If public opinion turns against them, they stand to lose the upcoming elections. This, of course, is a virtue of democratic governance: it prevents kleptocratic regimes. However, when it comes to international politics and dealing with global free rider problems, this virtue often turns into a vice. In such a context, politicians are incentivized to negotiate in the short-term interests of the group they represent, not in their long-term interest let alone in the global long-term interest. Agreeing to introduce taxes on CO2 emission or impose other measures of austerity, for instance, can turn public opinion against a politician. This problem is exacerbated by the competitive context of party-based democracy in which other parties can be expected to rally up protest against such 'unpopular' policies in an attempt to gain market share on the electoral market. Leaders of non-democratic nations, however, are equally prone to devise policy in the short-term interests of their group. While they do not face the uncertainty of re-election and electoral competition, they do risk being overthrown. This often leads them to focus on the short term, as Dryzek et al. (2011) rightly point out. This averseness to inforce policies which are costly in the short-term (but beneficial in the long-term) is evident from the way most nations have been handling the climate issue. Often (but not always) they will pay lip service to the importance of dealing with climate change and insuring a prosperous and safe future for all, but when push comes to shove these intentions and promises are (in many cases) not turned into actual greenhouse gas emission cutting efforts for fear of losing public support. The latest UN climate change conference or 'Conference of the Parties (COP) 24'at the time writing this articleheld in Katowice showed once again that despite important efforts by some of the delegates to tackle the issue head on, international disagreement prevents this. They did agree on a rulebook to implement the 2015 Paris agreement, but once again postponed pledging to implement the necessary reforms to meet the targets set in Paris. 5 Predictably, when nations pursue short-term (economic) group interests, we get what Hardin (1968) famously called a 'tragedy of the commons'. When faced with a common good (such as the climate), self-interested actors (seeking to maximize their personal benefit) will deplete that common good to the detriment of all. Suppose there is a pasture that can sustain a hundred sheep. Ten shepherds use the pasture and each shepherd has ten sheep. Adding more sheep would cause to pasture to be overgrazed and ultimately to turn into a wasteland. Nevertheless, each shepherd will still benefit from adding another sheep. This generates a substantial increase in income (10% more wool) and only a relatively small cost (1% less grass for his sheep). Shepherds aiming to maximize their individual payoff will therefore increase their livestock beyond the capacity of the pasture leading to the demise of this common good. The problem is that the cost (of adding another animal) is shared by the entire group, while the benefit (of adding another animal) is reaped by each individual shepherd. 6 In order to prevent such a tragedy of the commons, two conditions must obtain. First, (most of) the actors must realize the predicament they are in. They must be aware of the importance of preserving the common good and be willing to preserve it. Second, these actors must know that (most of) the other actors involved are aware of this and are also willing to undertake the necessary measures to preserve the common good. Otherwise, their actions to preserve the common good would be (perceived as) futile and they will no longer want to bear the costs to preserve the common good. What needs to happen, in other words, is that most involved actors take a long-term perspective and cooperate (refrain from freeriding) in order to preserve the common good and this will only happen if they can count on the others to cooperate as well. This mutual cooperation in the long-term interest of all is impeded by the current institutional context in which international politics operate. The tragedy of the commons that's unfolding before our eyes is not so much caused by the callous, short-sighted or immoral nature of the people in charge butas pointed out aboveby the incentives of those representatives created by the political context in which they operate (Vlerick, 2019).
The problem is further exacerbated by the fact that the interests of nations are not entirely aligned. With regard to the climate issue, the short-term interests of industrial countries depending on fossil fuel for their industrial activities in general and of fossil fuel producing countries in particular are manifestly not aligned with the interests of certain small island groups (such as the Solomon Islands) with little to gain from fossil fuel usage (since there's little industrial activity) and much to lose (since they face the immediate threat of disappearing below the surface of the sea in the next few decennia because of rising sea levels caused by climate change). Nations that have more to lose than other nations with the implementation of certain policies, often oppose themselves to these policies (even if adopting these policies would be in their long-term interest) because they do not want to be at 'the losing end' of the negotiation.

Obstacles to coordinate on a global policy
The destructive dynamic of international negotiation does not only create freeriding (refusing to bear the necessary short-term costs in the global long-term interests), it also hinders agreement on a global policy. Take the ongoing negotiation between developing and developed countries on how to fund the necessary measures to tackle climate change. At the 2009 Copenhagen COP, developing countries generally insisted on public funding, while developed countries pushed for market mechanisms (CO2 taxes) and private funding. Moreover, they disagree on the issue of intellectual property rights. Should technological innovations for harvesting green energy or otherwise reducing CO2 emission typically developed in industrialized countries be made freely available to developing countries or should intellectual property rights be maintained (Dimitrov, 2010)?
When it comes to the difficulty to coordinate global efforts to deal with global problems, the current context of inter-national negotiation is not only problematic because the short-term interests of nations diverge, but alsoperhaps even more sobecause of the incentives of the representatives involved. Representatives go to these negotiations with a lot of baggage. Often they have publicly defended the way they believe these problems should be tackled: 'we'll do things this way'. Even if they haven't, they rarely enter the negotiation with an open perspective. They are often linked to political parties, by being members of a political party themselves or (more commonly) by being mandated by elected members of parties carrying out executive functions. Those political parties typically have preestablished views on how to deal with the issues at stake. Different parties are often divided along ideological lines and consequently propose substantially different ways of tackling similar problems. Neo-liberal parties will have different views on how to allocate the costs for curbing greenhouse gas emission than socialist parties. The former may be wary of imposing too many restrictions on companies while the latter will mainly worry about the consequences of policies (such as increasing taxes on gasoline) that might affect citizens with modest means. Unsurprisingly, representatives of different parties, taking different ideological lines, have a hard time agreeing on policies. This, of course, does not mean that agreement cannot be reached between different party representatives, merely that ties with parties can impede agreement between representatives.
Even more so, in the context of competitive party democracy, politicians from different parties actually have an incentive to disagree with each other. They compete for votes trying to convince voters that their program is different and better than what other parties have on offer. This is exacerbated by heavily mediatized public debates where these politicians affront each other and attempt to 'win' the debate. This does not only complicate efficient policy making on a national level, but also on a global level since these ideological divides run across countries and are no less present on the international scene.
If anything, agreement is further complicated in an international context because of ideological differences between nations. Representatives from the more liberal United States could easily clash on policy with those of the more socially inclined Scandinavian countries and with those of more authoritarian countries such as China. International politics often leads to discord because all parties want to impose their own 'social contract'their own way of organizing the cooperative effort to deal with the issues at hand (see Vlerick, 2019, forthcoming). As Robert Putnam (1988) rightly points out in his 'two-level games' model, international negotiation simultaneously consists of two negotiations: the first is negotiation at the domestic level and the second is negotiation at the international level. Representatives of nations must first negotiate with other stakeholders (e.g. pressure groups) and government officials of their nation to determine the national interests. These preceding (domestic) negotiations determine which outcome(s) of the international negotiation will be acceptable to negotiating representatives. International agreement can only ensue if there is an overlap in acceptable outcomes for all representatives. Such an overlap does not always exist and if it exists it often concerns policy proposals that end up being inefficient because too many compromises have been made to make them acceptable for all. This difficulty to coordinate on an efficient global policy does not only affect dealing with climate change but also with poverty. While tremendous progress has been made in eradicating poverty in the last decennia. 7 There are still some 736 million people living in extreme poverty today 8 and many more in relative poverty. The main obstacle to efficiently eradicating poverty is notas pointed out above a lack of altruism displayed by affluent nations and people, butamong other things such as the structure of the global political economya lack of coordination.

A solution: citizens' deliberation at the global level
In order to address the pressing global cooperation problems, we need efficient global policy in the long-term interest of all stakeholders. Currently this is impeded by the context of international politics shaped by our global and supra-national institutions which provide a platform for international negotiation by representatives of nations. In order to overcome the free rider problems and lack of coordination that plague current international politics, we need to address the current institutional design. The proposal advanced by Dryzek et al. (2011) to erect a DGCA is particularly promising in this light. It provides -I will arguethe key to solving both kinds of cooperation problems (free rider and coordination problems) that currently stand in the way of successful global cooperation.

Solving free rider problems
Dryzek et al. (2011, p. 38) state that: 'While there are no guarantees, we would expect a DGCA to adhere more closely [in comparison to a PEGA] to long-term considerations and bring global public goods to the fore'. In other words, they expect deliberating citizens to put forward policy that is less likely to free-ride on the common good (and produce a tragedy of the commons) than when policy is left to a parliament of elected politicians. This point has been repeatedly stressed by scholars writing on deliberative forms of democracy. They emphasize the importance of deliberation in this. According to O'Flynn (2010, p. 301) 'deliberative democracy is well placed to deliver the public interest'. The reason he offers in support of this claim is that deliberation 'obliges us to take a broader or more encompassing view of important decisions of law or policy than merely consulting our own special interest in them' (O'Flynn, 2010, p. 307). In a similar vein, Young (2002, p. 26) claims that deliberation 'promotes cooperation' and 'solves collective problems'. 9 This is not mere conjecture. As Dryzek and colleagues point out (2011, p. 38): 'many empirical studies on citizens' deliberation show that as soon as participants start making an effort in seriously listening and take a respectful attitude toward other positions and arguments, their own thinking is increasingly enlarged in both time and space'. Meaning that they take the interests of a larger group of people into consideration and consider the long(er) term.
This was clearly the case in the first multinational and largescale initiative of citizens' deliberation on the climate issue. The project was called 'World Wide Views on Climate and Energy' and took place a few months before the UN climate summit in Paris in 2015. 10 Groups of citizens spread over the globe were informed about climate change (causes, potential solutions, etc.) through videos and deliberated on how to deal with the climate issue. The measures proposed by these deliberating groups of citizens generally went well beyond the guidelines proposed by the UN summit a few months later. Remarkably, in contrast to (many of) the representatives in the UN summit in Paris, the vast majority of participants put their group interests aside. No less than 79 per cent of the participating citizens felt that their country should take strong measures to tackle climate change even if other countries did not. Given these promising findings from social experiments with citizens' deliberation and the fact that our current democratic systems are poorly equipped to deal with a complex global problem such as climate change, we should as Dryzek and Niemeyer (2019) forcefully advocate in their recent paperconsider the implementation of such deliberative 'mini-publics' in existing global governing bodies. The question remains why citizens' deliberation does not exhibit the same shortcomings as negotiation between representatives of nations. After all, citizens also have a self-interest and belong to certain nations. Why don't they pursue their own interests and that of the particular group to which they belong (or at least not to same extent as representatives do)? How come they seem more prone to take a long-term perspective in the global interests? Three important factors, I believe, contribute to this.
First of all, citizens are not subject to the same kind of damaging incentives as professional politicians. They do not need to please their home electorate in the short-term in order to be re-elected. Second, the dynamics of deliberation are fundamentally different from the dynamics of negotiation. Whereas the latter take the form of what Steiner (2012, p. 4) calls 'strategic bargaining' in which the actors typically have fixed preferences 'and maneuver to arrive at an outcome that is as close as possible to their preferences', the former (ideally) takes the form of an open-minded exchange of ideas, views and arguments. It leads to mutual justification (Mansbridge et al, 2010) rather than deal making. In such a context, participants are forced to reflect upon the positions they advocate and justify them to their fellow participants (with different interests and backgrounds). This makes it harder to maintain and defend positions that are self-interested or only in the interest of the particular group one belongs too. It also makes it harder to defend positions that are beneficial in the short-term but detrimental in the long-term. This is not mere conjecture, as Dryzek et al., (2011) point out, analysis of the dynamics of citizens' deliberations show that such deliberation often produces a shift toward considerations aimed at preserving public goods and considerations taking into account the interests of others and society as a whole.
Finally, there is the diversity of the group of deliberating citizens (ensured by random sampling). This strengthens the virtues of deliberation, often leading people to adopt more long-term views in the interest of a wider circle of stakeholders. On the one hand, it exposes the participants to the views and interests of people in very different positions (be it because of socio-economic gaps, cultural background, difference in gender, age, etc.). On the other hand, since participants need to justify their position to others with diverging interests, they have to take the interests of these people into consideration if they want to be successful. Proposals that go radically against the interests of others within the assembly do not stand a chance.
Interestingly, in this regard, deliberation appears to be a very effective means to bridge ideological divides in highly divided societies. According to Gutmann and Thompson (1998) deliberation generally leads to a better understanding of the views of others and ultimately to more tolerance for those views. A number of empirical studies support this claim. Under the right conditions, deliberation has been shown to bridge ethnic, linguistic and other divides (Dryzek, 2005;O'Flynn, 2017;Steiner, 2012;Ugarriza and Caluwaerts, 2014). Deliberation, it seems, has the power to open our perspectives even to the viewpoints of people we originally met with suspicion or even hostility.

Solving coordination problems
In order to deal with global cooperation problems, we must not only solve the inherent free rider problems they present us with, but also the inherent coordination problems. In this regard too, a DGCA seems promising. In any case, more promising than the current international political context plagued by a difficulty to agree on particular global policies. Empirical data suggest that citizen's deliberation often yields a high degree of consensus around certain solutions. According to Dryzek (2009Dryzek ( , p. 1390) 'a large public policy literature points to the effectiveness of deliberation on the part of those concerned with a common problem in generating solutions that are both effective and mutually acceptable' (my italics). In a similar vein, Steiner (2012) concludes from his survey of empirical studies on deliberation that under favorable conditions (such as a relatively even distribution of initial positions among the deliberators) deliberation helps to reach consensus decisions.
Incidentally, this was also the case with the World Wide Views on Climate and Energy. 11 Not only did the deliberations generally lead to proposals that went well beyond those advanced by the representatives a few months later in the climate summit in Paris (see above), the degree of consensus achieved was also considerably higher than among the politicians and diplomats. This, of course, doesn't mean that a full consensus is typically achieved in deliberative citizens' assemblies, let alone strived for. The latter can be undesirable as some scholars point out (e.g. Hansen, 2004;Steiner, 2012) since it could create pressure on dissenting minorities. It merely means that proposals emanating from citizens' deliberation often seem to have a relatively large support base in the group of deliberating citizens. 12 Again, we have to ask ourselves why? Why do deliberating citizens seem to have an easier time agreeing on policy than negotiating representatives?
A number of important factors, I believe, contribute to this. First, in contrast to elected politicians and diplomats, randomly selected citizens are free from political allegiances or ties. They do not take directives from political parties who have an incentive to disagree with the views and ideological lines of other political parties. They are free to make up their own minds after having been well-informed by experts.
Second, randomly selected citizens are not only free from political ties but also from public opinions. Public commitment to a point of view (taking a public stance in certain matters as professional politicians are required to do) makes people more resistant to moderating their views in the light of subsequent argument. As MacCoun (2006) points out, this is a basic finding in social psychology. Moreover, politicians are often criticized for not sticking to their opinions (Steiner, 2012). This makes them even less likely to be swayed by good arguments and agree with other politicians. To avoid this issue in citizens' deliberation, Chambers (2005) argues that such deliberations profit from taking place behind closed doors. This protects the deliberators from 'the harmful effects of the glare of publicity'. Deliberating in private makes it easier for participants to be swayed by good arguments and coordinate on promising policy proposals. Chambers does, however, point out that citizens' deliberation should be transparent to the public at large in order for their proposals to carry legitimacy and looks for ways to reconcile both (e.g. by introducing transparency and publicity later or elsewhere in the process, as is also suggested by Goodin (2005) and Dryzek (2009)).
Third, citizens' deliberation is more likely to yield agreement on policy proposals than negotiation between representatives of nations because they engage in deliberation rather than negotiation. While the inherent strategic bargaining in negotiation (Steiner, 2012) can of course (and sometimes does) lead to an agreement, the context of deliberation is more geared towards reaching agreement than the context of negotiation. Whereas deliberation consists of justifying one's points of view and evaluating the justifications of others, negotiation is more a matter of attempting to have one's own (fixed) preferences reflected as much as possible in the outcome (i.e. in the policy). The latter often prevents an agreement (which is clearly the case in the negotiations on the climate issue, as pointed out above). Parties that feel dupedthat is, they believe are unable to have their fixed interests reflected in the policywill often block the process of coming to a global policy by sticking to their position. While this predicament could of course also befall deliberating citizensa sizable minority of participants could be strongly opposed to what the majority proposesand the deliberation would therefore fall short of reaching its goal, namely to put forward a proposal that is at least acceptable to all participantssocial experiments with citizens' deliberation indicate that this is not commonplace (Dryzek, 2009).
A fourth aspect that contributes to that is the cross-cultural similarity in fairness norms. Since ordinary citizens are typically not (publicly) committed to certain points of view and political ideologies, they tend to evaluate the arguments of others in terms of fairness. There is good evidence that people in all cultures have a similar sense of fairness (and accord much importance to it) (Vlerick, 2017;Vlerick, 2019). In all cultures fairness is associated with the equal distribution of resources and/or the distribution of rewards in function of one's contribution (Tomasello, 2005). Binmore (1994,1998,2005) argues that these fairness intuitions are innate. Humans possess what Binmore calls 'a deep structure of fairness' (an innate sense of fairness) with which they evaluate the behavior and opinions of others. The innate origin of fairness norms is supported by studies on young children showing that they harbor a number of intuitions about distributive fairness they couldn't have absorbed through cultural learning. Young children share spoils equally with peers after having cooperated to obtain the goods even if they could easily monopolize the goods (Warneken et al, 2011), understand and defend the entitlement of others (Schmidt et al., 2013), understand fair as equal (Wittig et al., 2013), and give less to free riders than to cooperators (Melis et al., 2013).
Since this sense of fairness is universally shared, it provides a common ground for agreeing on policy proposals. Fair proposals can and often will be recognized as such by all participants regardless of their cultural background or political preferences. In short, in the absence of strong ideological barriers dividing participants (as is often the case with politicians and diplomats), deliberating citizens evaluate proposals with a shared set of (intuitive) fairness norms and are likely to coordinate on proposals that are detected as fair.
An existing proposal Dryzek et al. (2011) aren't the only one's championing the convocation of a DGCA. Recently, Vergne, Schecter, Van Reybrouck, Bouley de Santiago, Walker, and Bouricius have worked out in quite some detail how citizens' deliberation could be organized at the global scale. In their proposal, first national or regional citizens' assemblies are held. Participants are randomly selected by sortition, ensuring that the group of deliberating citizens is representative of the population of the nation or the region they are selected from in terms of age, gender, ethnicity, cultural background, socioeconomic status, etc. These regional assemblies are composed of 50 to 1,500 people, depending on the size of the region or the country.
Before the start of these deliberations, the agenda should be set. Given the complexity of the global issues at stake, I think it is advisable that every iteration leading up to the DGCA would be devoted to a single global cooperation problem. The important but limited task of setting the agenda (determining the global issue that will be discussed) should be integrated in the process. In their proposal, Vergne and colleagues suggest organizing an 'agenda council'. This council would be composed of around 200 members drafted by lot from the world citizens who have been selected to participate in the regional councils.
Once the agenda has been set, the various assemblies are informed by experts on the (global) issue at stake. These experts, it is important to note, should be selected by the members of the assemblies themselves to prevent a biased selection of experts. After being thoroughly informed, various interest groups such as lobbyists, NGOs, industry and nations have the opportunity to propose and justify their researched solutions. Finally, the assemblies start deliberating with one another on possible solutions. After the national and regional citizens' assemblies took place, the global citizens' assemblycomposed of 800 citizens which are randomly selected from the national and regional assembliesis organized. As is the case in the national assemblies, citizens deliberate in groups of six to ten. Subsequently, the proposals emanating from these round tables of deliberating citizens are presented to the entire group with the aim of reaching a high degree of consensus around the most popular proposals. By first organizing national assemblies from which the participants to the global assembly are (randomly) selected, the deliberation indirectly includes many more citizens than can realistically be brought together in a single assembly. This benefits the global citizens' assembly on an epistemic level: many more people are involved in coming up with creative proposals to deal with the issues at hand. It also benefits the legitimacy of that assembly, since it involves the indirect input of a much larger representative sample than those 800 citizens who will ultimately participate to the global citizens' assembly.
Finally, between 100 and 300 citizens are randomly selected from the global citizens' assembly to participate in an oversight council in order to discuss how the whole process could be improved and how to implement these improvements in the next iteration. That way it becomes a self-learning process that improves over time. The cost of the entire process has been estimated at 4 million euro (per iteration). This, as they point out, should not be a major problem since the yearly budget of the UN is substantial enough.
Importantly, Vergne and colleagues propose to integrate such an assembly in the UN. They do not propose to replace any existing institutions with a DGCA. Rather, they argue, a DGCA should be 'bolted on' to existing institutions. I follow them wholeheartedly in this. It is, I believe, neither realistic nor desirable to replace existing institutions in a revolutionary way by a DGCA. The aim is to strengthen the global institutional level in the light of pressing global problems. Dismantling important existing global institutional structures would do the opposite.
Many issues with which UN institutions are concerned would fall beyond the scope of a DGCA. Since the 'raison d'être' of a DGCA is to improve policy in those areas of global governance that are faced with (global) cooperation problems, the DGCA should only be concerned with such problems. The policy proposals emanating from the DGCA would then be carefully communicated (and justified) to the members of other national and international organizations concerned with these problems (e.g. climate COPs), world governments and the population at large. This, as I will address below, is crucially important for the policy proposals to acquire legitimacy and carry the necessary weight.
While digging any deeper into the details of this proposal is beyond the scope of this paper, I hope that including it makes the case for a DGCA more tangible (and highlights its feasibility).

Conclusion
The convocation of a DGCA, I have argued, would be a promising course of action to reinforce global governance in the face of global cooperation problems, such as climate change and poverty. Those problems require solving inherent free rider and cooperation problems and a DGCA is well-suited to solve both. However, as any political realist will undoubtedly immediately point out, heads of state may very well ignore the proposals of such a DGCA. Given the sovereignty of states these proposals would not be binding to nations. How then is a DGCA going to prevent nations from 'defecting', pursuing their own short-term self-interest at the detriment of the long-term global interests? What could incite sovereign nations to follow the recommendations of such a global institution?
I believe that the extent to which nations are prone to follow such recommendations, depends crucially on the perception of the legitimacy of the global institution from which they emanate as well as the perception of the legitimacy of the recommendations themselves. When the process through which these proposals come about is widely perceived as a legitimate process, leaders of nations (and representatives of nations seated in international organizations) can be expected to be pressured to follow these proposals. This in turn requires the process to be transparent to the public at large and that process to be perceived as fair. Bridging the gap between the 'deliberating microcosm' and the population at large is a crucial, challenging but not insurmountable requirement for well-functioning citizens' deliberation, as Fishkin (2009) points out. In order to be legitimate and to be perceived as such, a DGCA must strive for (and succeed to a certain extent in realizing) the Habermasian ideal of equal and unconstrained participation of all of its participants (which make up a sample that is representative for the world population) (Habermas, 1996).
In democratic nations, the electorate can readily pressure their leaders to follow directives they consider to be legitimate and in their long-term interests. In non-democratic nations, which still govern over some 3 billion people, 13 there is no or little electoral pressure. However, the population can and does still exert pressure on the people in power (ever more so in fact in the digital era where citizens have the means of coordinating political action through social media and platforms), since they may have the power to overthrow governments or at least threaten the regime with instability (Dryzek et al, 2011). Whether people in power will be pressured by their citizens into following the global directives emanating from a DGCA, however, also depends on the perceived legitimacy of the directives themselves.
The influence of a DGCA will, therefore, crucially depend on the transparent communication of its modus operandi (legitimacy of the process) and of its proposals as well as the reasons for those proposals. Justification is not only key within the deliberative assembly, but also in communicating the proposals to the population at large. Equally important to their legitimacy, of course, is the quality of the proposed solutions to the issues at stake. In this context too, the empirical evidence on citizens' deliberation stems hopeful. Given enough diversity in the sample of deliberating citizens which would be the case in a representative sample of the world populationcitizens' deliberation can be expected to lead to well thought out proposals (Fishkin, 2009;Landemore, 2013;Mercier and Landemore, 2012). In fact, as Hong and Page (2004) forcefully claim: diversity trumps ability. A diverse sample of moderately competent individuals will often outperform a group of experts. Cognitive diversity, Page concludes (2007), contributes more to the competence of the group than individual ability.
For a DGCA to acquire the necessary legitimacy to incite nations to follow its directives would take time. Even in the best of scenarios, this can be expected to take a considerable number of iterations. That it should eventually do so, is of course not guaranteed. But, provided that the proper communication channels are set up between the deliberating microcosm, the world leaders and the population at large, a DGCA could -I firmly believegrow into a global institution with substantial influence on national policies. Moreover, as Dryzek et al. (2011) point out (and as I have mentioned in the introduction) proposals of a DGCA could be met with less resistance by nations standing on their sovereignty and non-democratic nations than those of alternatives such as a PEGA. All of these considerations make it the best candidate to my knowledge to deal with the pressing global cooperation problems of our time.