PARTICIPATORY DEMOCRACY, cooperative democracy, deliberative democracy, permanent democracy, semi-direct or civic-tech democracy, 2.0 or 3.0 democracy, random selection democracy—there are certainly plenty of proposals on how to rebuild our political systems. Many researchers, politicians, and ordinary citizens participating in public debate have given thought to the question of democracy, particularly representative democracy. “One person, one vote”: the very expression seems dated, and low voter turnouts tell us how deeply rooted is the popular distrust today. Has the last word really been said? Is it too late in the day? No, I don’t think so. There is much more to be said—for the simple reason that the history of democracy is just beginning. What are our few decades of apprenticeship in electoral democracy, when weighed against all the past, and especially future, centuries of human history?
Not everything has been said, and it is important to begin by reviewing the existing proposals for participatory democracy, as well as the successful and many unsuccessful previous experiments. This will help us to construct a truly innovative model for twenty-first-century democracy, one that draws on the lessons of the past so that we can better prepare for the future.
I know only too well that summaries in a few paragraphs, sometimes in a few lines, risk not doing full justice to the ideas developed by various authors. But I will take the risk anyway. I cannot do without the support of their thinking, but nor can I allow it to have an ossifying effect. For the foundations of democracy are both collective and dynamic.
With a boldness enhanced by the sense of risk, my own proposals will follow next. In Chapter 10, I propose a completely new basis on which to fund the political process—and I hope you will give me a fair hearing. If I place the funding of democracy at the center, it is because the current system has been corroded by money; democracy will fulfill its promise of equality—at the ballot box but also in the representation of people’s preferences—only if the funding problem is solved at the outset. This is not a sufficient condition, but a necessary and urgent condition. Of course, to speak of fiscal matters and the public funding of political democracy is not very “sexy,” and the kind of activists who have fought historically for democracy—the men and women who hand out leaflets each week at street markets or metro station entrances and who attend branch meetings several times a month—might accuse me of staying aloof from the reality of political representation. What is the point of focusing on the relative weight of private and public money in election campaigns, when what matters on the ground is to convince voters one by one? Although the commitment of such activists is sincere and important, and although they devote tremendous energy to it at the expense of family life and leisure time, the present workings of the democratic process mostly end up nullifying its effects. What counts in politics is money: our votes have a “price,” and that is why more and more politicians respond to the preferences only of the richest groups in society (those who contribute financially to their election campaigns) and not at all to the preferences of their own party activists, who give up so much of their time.1 This is the reality that we must address, and in doing so we should not be scared of the figures involved. Democracy has a cost, and if we no longer want the rich to “bear” this cost—and to benefit from everything they get in return—then, well, we have to think in terms of its being borne by the public purse.
To be sure, the democratic revolution that I propose in this book cannot end with the question of funding. It is equally urgent to improve how everyone is represented. Here it seems to me necessary—as I discuss in Chapter 11—to reshape the functioning of political democracy in accordance with the model of social democracy. In particular, the reconquest of electoral democracy—against the extremes and against populism—cannot happen without a profound reform of our parliamentary system. And my proposal, in short, is to introduce an element of social representation into a dual and sovereign parliamentary assembly. To put it simply, this would guarantee the presence of the popular classes (employees, blue-collar workers, those without secure employment) in the National Assembly, Parliament, or Congress, from the benches of which they are presently missing. We shall see that their absence is not without consequences: it directly affects the legislation that is adopted or rejected in France, as well as in the United Kingdom and the United States. But before we go into details, we must run through the various proposals for democratic renewal that already feature in public debate.
Oh yes, if you skim this page again and see there is no mention of “the internet,” well, there is really no reason to be surprised. True, it is not there. But if what you are asking for is Web 2.0, I am more than happy to agree: I never tire of using the social media, sometimes, I must admit, to the detriment of efficient, optimal use of my own brain time. But the internet is a tool. It is not the solution. Those with the power are only too happy to let people amuse themselves online; it allows them to express their preferences on paper (or rather, on the screen), giving them the illusion of participation at the very moment when the die is cast.
I am not unaware of the power of this tool. In fact, I showed in Chapter 8 the growing sums of money that parties and candidates spend on it during election campaigns. Nor am I unaware that the internet has partly altered the parameters of the electoral battle—negatively, in the spinoffs from the misappropriation of private data on the social media, but sometimes also positively, by permitting the online mobilization of young people, who, put off by traditional forms of activism, feel drawn to the idea of doing politics differently. The internet makes it possible to mobilize underrepresented constituencies; Alexandria Ocasio-Cortez tailored her 2018 tech-savvy campaign to Latino and younger voters, with a strong presence on social media. The internet also makes it possible to mobilize small donors, sometimes by directly winning their support on the social media; that is partly how Bernie Sanders was able to do as well as Hillary Clinton financially during the Democratic primaries of 2016. Would the Five Stars Movement in Italy have had such success just twenty years ago, before the internet became a ubiquitous tool of communication? And was not Jean-Luc Mélenchon’s unexpected showing in the French presidential election of 2017 partly linked to his mastery of the social media? A candidate rather “badly treated” by the traditional media—who gave as good as he got—was able to reach out to citizens more easily through the social media.
But make no mistake, this new technology in no way solves the central problem of campaign finances. I have been told by some politicians that it is no longer so essential to regulate the funding of election campaigns, because in the internet age it is possible to campaign with a few cents and lines of code—in other words, that campaigns awash with money no longer have any advantage over campaigns starved of private funds. But that is false—as I showed in Chapter 8. Sure, advertising online costs less than on TV, but in the internet age the cost of campaigning on the internet can prove very high indeed.
The battle for democratic equality will not be won through some online app that brings politicians virtually closer to their constituents. Such gadgets can sometimes be fun, but all they offer is an illusory participation. The reality of participation will be acted out in a remodeled Parliament; and equality will stand a chance only once the sluice of private funding has been closed.
But let us begin at the beginning, with the proposals already made to breathe new meaning into the idea of democracy. And before we set out on a little world tour, allow me to linger for a while in the French Hexagon. Here, the historian Pierre Rosanvallon is undoubtedly the author who has thought most about the history and future of democratic government. As he puts it, in terms that summarize what we have seen up to this point: “Our regimes are democratic, but we are not governed democratically.”2 Of course, Rosanvallon is referring not to the funding of the democratic process but to the representation deficit that he highlights in his critique of the French presidential system; he evidently prefers a representative parliamentarianism to the personalized presidential model. But the most fascinating aspect of his study of changes in contemporary democracies is his analysis of the “desocialization” of politics.3 For Rosanvallon, the party system no longer has any representative function; whereas political parties for a long time animated public debate and served to express social identities, they seem today to have been reduced to the residual function of selecting candidates4—a development largely due to their professionalization, which entails that social groups have long ceased to identify with them. As I will show in Chapter 11, this is not so surprising if we think that political parties no longer mirror the reality of the social groups in contemporary society. Only the UK Labour Party has for a long time had workers in its representative echelons, although this characteristic was already being eroded in the mid-1980s.
I cannot but agree with Rosanvallon’s conclusion, although I would add, as an alternative explanation, that the identification deficit has its roots in a failure of representation. How could I—as a group or simply as an individual—identify with a political party on the left or right that, once in power, responds only to the political preferences of the lucky few? How could I see as mine a party that devotes more time and energy to convincing the rich to fund its candidates than to mobilizing what has survived of its members’ energies?
Rosanvallon speaks of society as having been “forgotten.” And he proposes that it should be rediscovered through a process of “continuous” or “permanent” democracy, in which citizens are called upon to express themselves at all times, not only at elections. He suggests, for example, that democratic debate should give greater consideration to citizens’ petitions.5 I, too, think that citizens should be able to express themselves more regularly than once in four or five years, and this is why I propose—in order to give new dynamism to the expression of political preferences—that they should have the option each year to fund the political groups to which they feel closest. But I also think that, for such a society to be rebuilt, allowance should be made for genuine representation within its legislative bodies—in other words, that society should be brought into the Parliament, so that it is allowed to express itself directly in the political sphere. As we shall see in a moment, although citizens’ petitions are a good idea, they usually fail because a flood of private money and corporate lobbying determines the ultimate fate of this expression of the people’s will.
The legal theorist Dominique Rousseau also champions the idea of “continuous democracy,” albeit in different forms from those advocated by Pierre Rosanvallon.6 In particular, he has more to say about the incorporation of a social element into the legislative process, arguing for the creation of a Social Assembly.7 This would be a deliberative body, expressing the will of civil society, alongside the Congress and the Senate (in the French context, such an assembly would take the place of the existing Economic, Social, and Environmental Council). Its members would be elected in such a way as to take into account “the productive forces in economic and social life, the broad sectors of activity, … and the forms in which those forces and activities are organized.”8 The novelty of this idea is that, by making it a deliberative (not simply consultative) body, Rousseau gives this third chamber the character of a “genuine” assembly. It is a pity, however, that he does not follow his argument through to the end; in particular, he does not discuss in detail how pieces of legislation would proceed through the different chambers, and he does not consider the question of which chamber would have primacy. But we cannot really judge the importance of such a Social Assembly unless it has been decided which of the chambers will have the final say in the event of disagreement. We can only assume that Rousseau would uphold the existing primacy of the National Assembly in the French context, which would limit the scope of the reform.
Rousseau is right to think that social forces should be given deliberative powers, but in my view he should allow himself to be more “radical.” For his idea to take on its full meaning, it would be preferable to bring the representation of the social directly into the existing Parliament or Congress—which is what I propose with a mixed assembly. Otherwise, the kind of Social Assembly advocated by Rousseau might end up as an illusory form and lead to disappointments.
Some readers may think they can hear strains of a Venezuelan tune in this proposal for political representation of the social. But let me make it clear that, as I see it, the failure of Venezuela’s constituent assembly elected in 2017 is not the failure of occupational representation as such; rather, it is a further symptom of the gradual breakdown of the Maduro regime and the “Bolivarian” experiment. We should not extrapolate from it to condemn the idea of better minority representation, especially as the precise form that I propose (and here, as in so many other questions, the devil is in the detail) does not have much in common with the one tried out in the Caribbean oil-producing country.
What happened exactly in Venezuela? After Nicolas Maduro proceeded with the election of a Constituent Assembly in July 20179—an event boycotted by the opposition—it took over (in August) most of the legislative powers held by the National Assembly,10 effectively giving the president’s party full authority to rule the country. Since then, Venezuela has been sliding a little more each day into dictatorship.
But why mention this development in a book on democratic renewal? The reason is that—if we leave aside a political context that, today in 2018, on the eve of early presidential elections again boycotted by the opposition, can only serve as an example to avoid—the way in which the Constituent Assembly was elected is full of interest. One part (364 members elected by universal suffrage) represented territorial constituencies where each vote was based, as usual, on a residential qualification, while the other part (181 members) represented social groups and minorities: 173 from sectoral categories (farmers, students, pensioners, workers, handicapped people, company directors, and so on) and eight from indigenous communities. Each elector voted twice: once in her local constituency, and once by virtue of her social group attachment. It was a way of combining political and social representation at the heart of one and the same assembly.
Such dual representation—which in a way is what Dominique Rousseau proposes, as well as being one reading of what I propose in the final chapter of this book—could add up to a veritable democratic revolution. In a different context from today’s Venezuela, it could permit the rediscovery of a “forgotten society,” though with an important change in the way people vote. For although I agree that everyone should vote twice—once locally by universal suffrage and once by proportional representation for lists reflecting the reality of social groups—I do not think that there should be two different electoral colleges. In other words, I do not think that each individual should vote by virtue of his or her social group attachment; separate electoral colleges of this kind are a democratic aberration and have never really been established anywhere, except in extremely tense historical contexts (such as Bosnia, with its separate Croat, Serb, and Muslim electorates, or colonial India, with its Hindu and Muslim electorates) or for the representation of highly specific minorities such as the New Zealand Maoris. What is necessary is that all citizens should vote for lists under a system of proportional representation (PR) that reflects the reality of social groups, including 50 percent members of popular socio-occupational categories (employees, blue-collar workers, and the whole group of workers without secure employment). There is a similarity here to the “gender parity” lists for regional elections in France or to the Tunisian gender parity law,11 except that the active intervention would be in favor of the working classes rather than women.
In the case of Venezuela, the other major limitation—to say the least—of the electoral system proposed for the 2017 Constituent Assembly was that political parties were barred from running candidates for the “occupational” seats. Clearly this was a way for the regime to exclude the opposition and to ensure that a large number of seats were occupied by members of socio-occupational groups close to the ruling party (since only such groups could run candidates). The same kind of manipulation can be found in many of the puppet assemblies featuring in the constitutions of “real-world socialism,” including that of China today, where far-fetched types of social representation are invented to circumvent universal suffrage and to keep the existing parties in power. However, these shams should not stop us from going further: the idea of better social representation is certainly not a bad thing in itself, as long as it is based on transparency, universal suffrage, and the right to stand as a candidate.
In practice, how should citizens be able to express themselves outside election periods? It is a question asked by numerous researchers as well as governments, and specific measures have already been introduced in a number of countries. Italy, for instance, has gone a long way in the practice of permanent democracy and—as we saw with the elimination of public funding for political parties—has had frequent recourse to referenda on the basis of citizens’ initiatives. People’s votes of this kind are also commonly held in Switzerland, as we saw recently in the (fortunately unsuccessful) attempt to abolish national broadcasting license fees.
What are referenda on the basis of citizens’ initiatives? Although their forms vary from country to country, they may be generally defined as referenda organized on the initiative of a section of the electoral body, where everything depends on the size of that section.
In Italy, the people’s referendum was made part of the Constitution of 1948.12 It is what is known as an “abrogative referendum”: it can be held only to repeal a law or part of a law.13 The proposed change must come from at least 500,000 electors, or roughly 1 percent of the adult Italian population.14 For the result to count, a majority of the electorate must take part in the vote—a requirement that often leads to the rejection of a proposal, even if it obtains a majority of the votes cast.
In Switzerland, the referendum based on a citizens’ initiative—which is called a votation in French—can be traced back to the establishment of “optional referenda” in 1874 and the introduction of people’s initiatives in 1891. These may take several forms. If, within the hundred days following the adoption of a law in Parliament, 50,000 citizens sign a text requesting a vote by the whole electorate, the law can come into force only if the voters approve it (“optional referendum”).15 Or, any elector who collects 100,000 signatures may propose a change to the Constitution (citizens’ initiative referendum).16 In Switzerland, these two practices are very popular—the appropriate word here: between 1848 and 2010, 167 optional referenda and 158 people’s initiative referenda were held. (A new “general people’s initiative” introduced in 2003 allowed 100,000 citizens to request the adoption or revision of a federal law—not just a change in the Constitution.17 However, this possibility was eliminated in 2009 without ever having been used.)
Switzerland is often cited as a model in France, where the popular initiative referendum has been the object of debate for many decades.18 However, the popular initiative referendum has existed de facto in France for a number of years. You didn’t know? That’s not surprising, since the relevant measure, introduced during the constitutional revision of 2008 but only in force since January 1, 2015, is such that it was taken up for the very first time in 2019.19 For, instead of “simply” providing for a referendum to be held if enough citizens mobilize on the question at issue, it requires that at least one-fifth of members of Parliament should first table what is called a “referendum law.” Hence the name “shared initiative referendum”—not “popular referendum.” Nor is there anything “popular” in the fraction of the electorate that has to file the request: no fewer than 10 percent, or 4.7 million citizens.20 In other words, another fine but unworkable promise.
This is why the idea of a genuine popular initiative referendum comes up again at each presidential election in France. In 2017, it was part of the program of five candidates, including Jean-Luc Mélenchon—who did not, however, specify the number of signatories that would be required—and Marine Le Pen—who supported the idea of a popular initiative referendum with at least 500,000 electors behind it. Mélenchon went further and suggested that citizens should be able to call referenda themselves, adding that they might even be able to propose legislation. His institutional program also had at its core the principle of a “recall referendum based on a citizens’ initiative,” which would allow elected representatives to be removed before the end of their term.
Although the idea of a “recall referendum” may seem quite radical—or even “populist,” to judge by the words often used to defend it (“Out with them all!” for example)—everything depends on how it would actually work. The possibility of “recall” has existed for decades in parts of the United States,21 most notably in California, and it is interesting to look at the reasons that led to its introduction there in 1911. There is talk today of a new “gilded age,” but, as we saw earlier, the term was first applied to the explosion of inequalities in the United States around the turn of the twentieth century, which was when recall provisions were introduced to fight endemic corruption and the growing economic and political dominance of a handful of large corporations. The economic giant of the time was not called Google or Facebook, but rather the Southern Pacific Railroad.22
Recall petitions were periodically organized throughout the twentieth century, but until 2003 all of them failed for lack of signatures.23 The number required in California for a vote to be held on a petition is equivalent to 12 percent of the voter turnout at the preceding election, somewhere around 1 million in total. Yet in 2003, the recall procedure initiated against Governor Gray Davis, just eleven months after his election, led to his replacement by the political newcomer Arnold Schwarzenegger.24
If recall procedures are still very rare—at least in the United States, though less so in Latin America—we may wonder how much they and other popular electoral initiatives are illusory and how much they involve real direct democracy.25 Or, to put it in another way, how much do such instances of participatory democracy really help to solve the problem of the representation deficit? Do they not anyway create other problems, particularly with regard to the legitimacy of the vote?
Take the example of the abrogative referendum. To what extent is it legitimate that a vote by a small number of individuals can lead to the repeal of laws passed by the elected representatives of the majority of citizens? Of course, the answer to this question largely depends on the percentage of citizens required for the referendum to take place: 1 percent as in Italy or 0.7 percent as in Switzerland is very different from the 10 percent threshold often discussed. The higher the threshold, the more the referendum may be considered legitimate; but too high a threshold may mean that the whole procedure is unworkable (as we can see from the French case). You may say that a sufficient number of signatures does not necessarily result in abrogation: the proposal still has to secure a majority at the ballot box, where all citizens are called upon to express their will. But the problem is that all too many citizens do not bother to turn out, not being much concerned about the issues on which they are asked to vote. In Switzerland, for example, although people’s votes are a common occurrence, the numbers of those participating in them are usually very low. From this point of view, the Italian option of requiring a 50 percent turnout for a referendum victory is extremely interesting: direct democracy then means that a majority of citizens are involved in saying “yea” or “nay.”
Is this enough to legitimize abrogative referenda or the holding of votes based on citizens’ initiatives? Most opponents of these tools of direct democracy warn against populist aberrations and argue that they threaten to lead to greater political polarization. In this connection, the Swiss popular initiative “against the construction of minarets,” resulting in a referendum held in November 2009, gives us food for thought.26 The two key questions here are the legitimacy of a majority “yes” vote that contravened such a basic right as religious freedom, which one might have thought would be guaranteed under the constitution; and the character of what citizens intended, and what information they had at their disposal, at the time when the vote was held. Did the Swiss really express themselves for or against minaret construction, or did the debate center more on the issue of immigration? As to the information available to citizens at the time of the vote, this is by no means a problem only for popular initiative referenda; we need only think of the Brexit referendum, when no one was clear about what would happen if the “no” vote won (nor is anyone very clear about it today, three years later). Important political debates usually call for deep deliberation and involve multiple formulations and amendments that cannot be reduced to a simple “yes” or “no.”
In 2016 in California, voters in the presidential election also took part in a number of popular initiative referenda; the media focused mainly on the marijuana issue, but there were also several highly technical questions that were surprising to find in a people’s vote. Take, for example, Proposition 52: “Continued Hospital Fee Revenue Dedicated to Medi-Cal Unless Voters Approve Changes.”27 Is it really credible that voters—not the legislature—are best able to decide on the allocation of hospital revenue and on ways to fund health insurance for the poor?
We need to ask, then, whether advocates of popular initiative referenda should not address the question of expertise. Without siding with partisans of technocracy, I think it is necessary to ask whether citizens have the capacity to vote on certain technical matters. Let us be clear: it is not at all a question of their capacity “in the absolute,” but rather of their capacity as citizens at a particular point in their lives, with all the constraints such as limited time to absorb relevant information. Evidently, each citizen can acquire this capacity—for example, by becoming a parliamentarian. If parliamentary representation was introduced in our modern democracies, it was precisely because technical expertise is necessary in various fields, and because it takes time to consider a draft law, to digest its complex implications, and to propose improvements.
But let us suppose for a moment that there is a magic formula for practicing direct democracy, through people’s votes on issues about which citizens have the expertise to decide in an informed manner. Would that be an appropriate and sufficient response to the representation deficit? The answer has to be no (I almost feel like adding: unfortunately). Let me try to explain why.
Imagine that a popular initiative has obtained enough signatures for a referendum to be held. What happens next? The referendum campaign itself—which can cost a lot of money. But when a campaign costs a lot, in comparison with the statistical average, victory usually goes to those who have spent the most (remember the results in Chapter 8). What does that mean? Well, if the opposition is well organized, ready to spend what it takes and to mobilize the usual kind of lobbies, the answer is that, sad to say, nothing is likely to change in the end. In Switzerland, between 1848 and 2010, 44.3 percent of “optional referenda” and 90.5 percent of popular initiative referenda had negative outcomes.28 To take just one (striking) example, a referendum proposal in 1984 “against the abuse of banking secrecy and the power of the banks” was rejected by 73 percent of voters; the amount spent by the UBS bank alone on press ads in favor of a “no” vote was ten times more than the total funds available to the committee organizing the initiative (800,000 Swiss francs).29
This brings us back to where we started: to the need for the regulation of election spending. Without that, popular initiative referenda do little more than give “the people” the illusion that it can express itself freely between two elections,30 when in reality the dice are loaded from the start. The same holds true for the “recall” procedure. I mentioned above the vote in 2003 in California that led to the removal of Governor Gray Davis, just eleven months after his reelection. Was that a “people’s vote?” On the contrary. The campaign was not launched by an ordinary man on the street, a California citizen rightly dissatisfied with the governor’s record. No, the person behind it was Darrell Issa, a superrich businessman and Republican politician, who spent no less than $2 million to collect the required hundreds and thousands of signatures.31 A multimillionaire with two motives and no particular concern for the public interest. On the one hand, Darrell Issa never concealed that he would have liked to replace Gray Davis—a pity that in the end he had to withdraw in favor of the Terminator. On the other hand, he saw the point of the exercise as being to swing California over to the Republicans, but although his campaign, “Rescue California, Recall Gary Davis,” centered on the $38 billion hole in the California budget, it has to be said that the deficit kept growing and growing under Schwarzenegger’s governorship. He came first among 135 candidates in the election to replace Davis, paying by far the most ($13.4 million) to see off his rivals.
Perhaps you will say that all this is well and good but only anecdotal. However, the causal impact of election spending on the success of popular initiative referenda has been systematically evaluated for California in the period from 1976 to 2004. John M. de Figueiredo and his coauthors have shown that spending in favor of a citizens’ initiative strongly increases the probability that it will be adopted.32 The weight of money again, as always.
So, you can see what I am getting at. My point is not that the citizens’ petition is a bad thing in itself; it is a tool of direct democracy that can be usefully discussed. But it would be wrong to see it as the solution to the representation deficit. As long as the problem of campaign funding has not been solved, citizens’ initiatives will be captured initiatives. Sure, people can go to the polls more often, but the die is already cast when it comes to any real regard for their preferences. And their lack of power is compounded by the hypocrisy of suggesting that they can at last make meaningful decisions. This complete blindness to the issue of funding is, in my view, the big mistake made by numerous theorists of direct democracy (or “permanent” or “continuous” democracy, as it is sometimes called). But it must be settled first before real progress can be made in linking up with the people and restoring a genuinely representative dimension to our democracy.
Equally mistaken are all those who focus only on voting systems. Again, it is not that the question is unimportant—it would be good, for example, to have a strong element of proportionality in parliamentary elections in France or the United Kingdom. Yes, it might also be worth thinking about alternatives to proportional representation and first past the post. The “Borda” method, whereby voters rank all options or candidates in order of preference instead of having to choose just one, might help to raise political participation and to reduce the polarization of debate. An alternative vote system is already in use in Ireland for the presidential election, and in Australia for legislative elections.33 But again, the debate on voting methods should not make us forget that, unless the question of election funding is solved, the finest system in the world will not redeem the capture of democracy and the fact that elected representatives cater overwhelmingly to the rich.
There is another option, however, which is regularly evoked by researchers, politicians, and the new democracy activists, and which has given rise to many experiments in recent years: random selection. If the aim is to end the capture of the electoral process and to solve the representation deficit, then why not put our trust in chance? Why not replace our current legislative assemblies with parliaments consisting of randomly chosen citizens? With a sufficient number of members, chance would guarantee representativeness—and if there were no longer elections, all the negative features associated with them would disappear.
To understand the beauty of such a system, it is not necessary to go back to the often-cited example of ancient Greece. Present-day Iceland will do nicely, with its participatory experiment (between 2010 and 2013) in the rewriting of its constitution.34 First, the government brought together a representative panel consisting of 950 randomly chosen citizens (the National Forum), which met on November 6, 2010, and formally stated the principles that should underlie the new constitution. Then a twenty-five-member constituent assembly was elected from civil society. Many political scientists have studied this experiment, beginning with Hélène Landemore, who sees it as a success despite the fact that in the end the new constitutional text was not adopted. In her view, the multiple versions of a constitution drafted by the twenty-five representatives of civil society were “better, smarter, and more ‘liberal’ ” than the one produced around the same time by a group of seven government experts. In particular, they were much more open to the rights of religions and laid greater stress on the importance of having a genuinely democratic society.
Beyond this Icelandic example, Landemore has developed a number of arguments in favor of participatory democracy and random selection, her key idea being that of a “collective intelligence,” or “democratic reason.”35 The intelligence of a group of individuals, she argues, will always be greater than the combined intelligence of all the individuals in the group, since inclusive decision-making permits the emergence of “cognitive diversity” (for example, different viewpoints on the world or different perspectives), and such diversity is itself a key element in the emergence of collective intelligence. In other words, what is important in determining the quality of decision-making is not so much the intelligence—or IQ—of each member of a group, but rather their diversity. It is this diversity that ultimately determines the IQ of the group. And even if, taken separately, individuals are not competent in such and such a field, cognitive diversity enables the group to be competent by virtue of discussion and the different perspectives that each member brings to it.
This concept of “collective intelligence” has given rise to numerous empirical experiments; the researcher James Fishkin, in particular, has tested out “collaborative” or “deliberative” polling in various places around the world. These often fascinating experiments, from Australia through the United Kingdom and Italy to China and Thailand, which are described in his book When the People Speak, point to a number of lessons for the renewal of our democracies. Among the most noteworthy of these is that deliberation by a group of citizens reflecting the country’s diversity leads to changes in attitudes, as the citizens become better informed, more concerned, and ultimately expressive of more moderate opinions than those they held at the outset.36 When we see today the extreme polarization of political life in the United States, where it sometimes seems to have become impossible to enact a law, one cannot but be tempted by this idea of deliberative democracy.37
For my part, I think that deliberation—above all, the giving of high-quality, independent information to the largest number of people—must be at the heart of discussions about solving the current crisis of representation. “Diversity” is also a key concept: no one can be competent ex ante in all matters that come up for deliberation in a parliamentary term; some will be knowledgeable about economics, others about environmental questions, but no one can be expected to have expertise across the full spectrum of issues. The combination of deliberation with a diversity of representation—and therefore of points of view—helps to ensure a higher quality of democratic decisions. However, I think we can do better than random selection, and this is why in Chapter 11 I argue for a mixed assembly with a guaranteed degree of social parity, but which maintains the principle of election.
Many arguments have been put forward against random selection. I shall not review them all here, especially as I think some of them are not altogether convincing. But, of course, what comes to everyone’s mind is the question of competence. To borrow a formulation from Myriam Revault d’Allonnes, people do not want to “appoint just any citizen without a particular qualification”;38 election, by contrast, allows citizens to be chosen on the basis of competence. As we have seen, however, advocates of random selection show that competence emerges “collectively,” so a first answer would be to say that the question is not the competence of each citizen chosen by random selection. More important, we must not distort the thinking of those who support randomness: what they propose is not usually to select citizens at random from the electoral register, but only citizens who wish their names to be fed into the machines that do the random selection. This implies that only those who feel capable of exercising the burden of authority can be chosen at random. Bernard Manin eloquently points this out in the case of Athenian democracy.39
Furthermore—and Manin also clearly explains this in detail—random selection does not entail absence of responsibility. In ancient Athens, it was current practice to initiate proceedings “on grounds of illegality”: that is, any citizen could proceed on such grounds against a proposed law or decree submitted to the citizens’ assembly, or even against a law or decree that had already been adopted. If the courts then ruled in favor of the complaint, not only was the decision of the assembly quashed, but its initiator was penalized with a fine. By all accounts, this encouraged legislators—though randomly selected—to be attentive in discharging their functions, especially as such proceedings were a common occurrence.40
The most telling argument against random selection seems to me the following. Election together with social parity offers a superior solution, because it allows us to build on our collective capacity to listen to different candidates, to watch them debate with one another, and to choose the ones who seem most suited to represent us and play a useful part in collective deliberations and parliamentary decision-making. To choose representatives is not only, or even mainly, to choose the most competent individuals; it is to choose those most suited to take part in complex debates on a wide range of issues, many largely unforeseeable at the time when they were elected. It is this capacity to listen and debate that parliamentary candidates have to display in an election campaign, and the replacement of all that with a game of dice would seem to bespeak a kind of democratic nihilism. Random selection foregrounds the capacity of any human group to debate and deliberate—but, paradoxically, it also deprives virtually all citizens of their right to discuss and decide who should eventually be entrusted with voting on laws. Random selection may be used for specific decisions; one thinks of people’s juries for certain courts or of “consensus conferences” required to reach an agreement on urban planning projects. But to make it the principal means of constituting the assembly that votes on legislation would be to surrender our collective capacity to delegate that highest of all rights.
To sum up, the debate on random selection highlights a real problem concerning the nonrepresentation of whole social groups in classical parliamentary elections. But it does not provide the best solution to the problem. A preferable way forward would be to maintain the elective principle while introducing a minimum of representation for different social groups, as I shall argue in detail in Chapter 11 with my proposal for a mixed assembly with social parity.
Fortunately, I am by no means the first to underline the importance of stricter regulation of election spending and contributions. But it has to be said that this question, rightly central in American debates, hardly features at all in political discussion in France or Germany, although, as we have seen, private money plays a not insignificant role there.41 In previous chapters, I have looked at the writings of legal theorists such as Timothy Kuhner, Robert Post, and Richard Hasen who have dissected and excoriated recent decisions of the US Supreme Court; they mostly also put forward clear proposals on how to reverse the Supreme Court decision that changed everything in 2010, the “Citizens United” ruling. In the same vein, political scientists such as Benjamin Page, Martin Gilens, and Larry Bartels do not stop at highlighting the uneven attention paid to the preferences of American citizens; they also propose ways of ending it and support the idea of better regulation of election finances. Each in his way is fighting for a ceiling on both private contributions and campaign spending, and above all they advocate the elimination of super PACs. (If I have one criticism of their efforts, it is that they focus entirely on the American case, whereas the problems they raise exist, in different forms, in many other democracies.) In the book they published in 2017, Page and Gilens actually emphasized the need for public funding of elections, in order to break up the weight of private money.42
To keep the attention of all my readers, albeit at the risk of not doing entire justice to all these authors, I shall pause here on just one of these proposals that strikes me as particularly interesting and echoes the ideas I develop in Chapter 10: that is, the proposal for democratic vouchers.
How can citizens be made equal in relation to the funding of political democracy? One solution might be to give everyone a democracy check or voucher at election time, with a face value of, say, ten, fifty, or one hundred euros, which they can use to fund the campaign of their chosen candidate. The man behind this original proposal in the United States is Lawrence Lessig, a legal specialist in intellectual property rights, who in recent years has become one of the main supporters of the fight against lobbying and corruption in American political life.43 He also put himself forward in 2015 as a candidate in the Democratic presidential primaries, but unfortunately had to withdraw because he did not garner sufficient voting intentions (and therefore sufficient prospects of funding). This strategy for winning power was part of his campaign to end political corruption: that is, elect a president, or more generally politicians, whose sole aim is to restore the proper functioning of representative democracy in the United States.44 They would be what he calls referendum politicians: men and women whose task is “simply” to reform the funding of political democracy and to (re)establish political equality—and who resign once this is achieved.45
But let us return to the democracy voucher. Lessig’s idea is to give each citizen a democracy voucher worth $50 that she can use to fund the candidate or candidates (Senate and House of Representatives46) of her choice. During the nine months preceding the election, candidates would be able to collect these vouchers—from citizens whose support they had won—in order to fund their campaign. Would all candidates be able to benefit from them? In Lessig’s model, they would first have to show that they had sufficient public support, for example by collecting a certain number of $5 contributions.47 More important, they would have to undertake not to accept private donations higher than $100.
Some readers might wish to go further, in accordance with the presidential fund model described in Chapter 5, so that candidates opting for the public system would be banned from also accepting any private money. But my preference would be for a third solution that did not leave it up to candidates to choose between public and private funding—otherwise, there would always be a danger that public money was eventually ground beneath the weight of private contributions. If private funding was simply banned, the benefits of democracy vouchers would be open to all candidates. Lessig does not go this far, because his development of solutions is in a way constricted by the rulings of the Supreme Court. It so happens that I am writing this in France, where money has fortunately not yet been defined as a form of “speech”; nor has it in most other European countries, even if we have to be wary of the decisions of the European Court of Human Rights. Even in the United States, it is generally agreed today that the recent decisions of the Supreme Court are much more “political” than “juridical.” And so we may hope that, if things work out well under a future Democratic president, the Court will eventually decide that it may not have been justified to define corporations as individuals or money as speech.
Am I being too optimistic?
I do not think so. One piece of evidence is that such democracy vouchers have already been introduced in the United States. Since 2017, each citizen in Seattle who is on the electoral register automatically receives through the mail four $25 democracy vouchers (or a total of $100), which she can then give to her chosen candidates in municipal elections (for the posts of mayor, city councillor, and attorney general).48 To participate in this system of public funding, candidates must first have obtained at least 150 contributions (each with a minimum value of $10 and a maximum of $250) for the post of attorney general, 400 contributions for at-large members of the city council, and 150 contributions for city council districts. The point of this is to guard against a proliferation of “non credible” candidates.
Candidates who wish to benefit from these democracy vouchers must agree to respect the strict spending limits and to receive no contributions above $250 (excluding the value of the vouchers).49 Interestingly, candidates who choose not to take part in the democracy voucher program are also subject to a strict limit ($500) on the donations they receive from a given individual. And yet, Seattle is in the United States!
In case you are wondering, Seattle finances these democracy vouchers through a property tax that should bring in $3 million a year. A little economic equality on top of political equality! On average, the program costs property-owners $11.50 a year.
The other good news is that these democracy vouchers work. Six candidates have benefited from them since they were introduced in 2017, and nearly 46,000 vouchers have been used with a total value of $1.1 million in public funding.50 Moreover, candidates who choose public funding do not seem to have suffered at the polls—on the contrary. The city attorney candidate Peter Holmes, for example, who opted for democracy vouchers (and received 5,885, with a total value of $147,000), was reelected with a large majority (73 percent of the votes) against his rival Scott Lindsay, who stayed out of the program.51 And Teresa Mosqueda, who ran for the city council and signed up for the democracy vouchers, also emerged victorious; her rival in the home stretch, Jon Grant, also chose such funding, unlike most of the other candidates in the primaries.52
We should note that, before this program was launched, the campaign donations for Seattle city elections were very concentrated: in 2013, a mere fifth of donors contributed more than $500, but their donations represented more than 55 percent of the total funds raised for the election.53 The introduction of democracy vouchers, however, made it possible to diversify the donor profile. A study by the Win / Win and Every Voice organizations compared the structure of donations for candidates using the vouchers program in 2017 with that of donations for candidates in the mayoral election (when the program was not yet in place).54 And what they found was that there were more contributions from citizens with modest incomes (less than $50,000 a year) in the former than the latter case (14 percent against 9 percent of the total number), and symmetrically fewer contributions from very well-off citizens (earning more than $150,000 a year) for candidates who opted for democratic vouchers than for those who did not use public funding (13 percent against 27 percent). The democratic voucher program also made it possible to mobilize many more young voters and donors—which is a good thing, since the younger generation are on average less politically committed than their elders. It would seem that innovative forms of public funding might make a difference in this respect.
Of course, the implementation of the program did not proceed altogether calmly, and conservative groups—as one might have expected—took legal action against it in the name of “free speech”! It is always surprising for an outside observer to witness the degree to which this concept, so essential to the functioning of our democracies, is distorted in the United States. After all, it seems difficult to put up a rational argument that public funding, which enables the poor to express their political preferences, actually limits the free speech of the least advantaged sections of society. Fortunately, in November 2017, a Superior Court judge in King County ruled that the democracy voucher program was legal.55
So, why should this path not be pursued? I certainly think it is a possibility: there is nothing to lose, and it would anyway be better than the present situation.56 But I also think—as we shall see in the next chapter—that the annual funding of political groups out of tax revenue is preferable to this kind of democracy voucher system, which has the defect of focusing only on elections and election candidates. In my view, parties have a role to play in the period between elections, by expressing the preferences of their supporters, by reflecting about the future, and by drawing up electoral programs and platforms; a system that focuses exclusively on candidates threatens to polarize public debate to an even greater degree. The system introduced in Seattle is also extremely complex and forces candidates to collect the vouchers one by one in paper form; at least an electronic version could have been devised in the twenty-first century.
Under the reform that I advocate, each citizen would state on her annual tax return the name of the political movement to which her seven euros of public funding should be allocated. The funding would be equal, fluid, and transparent.