Could Random Selection and Deliberative Democracy Revitalize Politics in the 21^st Century?

Having vanished for centuries, sortition now seems to be returning to the world of practical politics. [1] Recent experience in Iceland illustrates this. After the economic crisis of 2008 when the country nearly went bankrupt, the desire to change the government and the rules of politics was expressed in huge street protests. The election that was brought forward to April 2009 resulted in a coalition between the Social Democrats and the Greens, and a trial of the former prime minister on charges of negligence took place in March and April 2012. At the same time, in 2009 a Citizens Assembly consisting of 1200 people chosen by drawing lots and a few hundred qualified figures was formed on the initiative of civic associations, to identify the values that the country should be refounded upon. This experiment was repeated in November 2010, this time with governmental support, with a view to adopting a new constitution. The task of the second Citizens Assembly, following up on the results of the first, was to determine the main principles of the future fundamental law. Soon after, a “Constitutional Council” was elected by the people, consisting of twenty-five “ordinary” citizens. The 523 competing candidates were purely individual: members of parliament were ineligible, and electoral campaigning was legally reduced to a minimum in order to set this event apart from the normal habits of the widely-discredited political class. During the spring and summer of 2011, this Council worked on the draft of a new constitution. The most notable of the main innovations were a thorough reform of the balance among the different governmental powers, greater transparency in the decision-making process, a major extension of the devices of participatory and direct democracy, and more consideration of environmental issues. Draft articles were posted online as they were being written, and members of the public could comment and make suggestions via the Council’s pages on Facebook, Twitter or Flickr. The proposed constitution was submitted to Parliament in the summer of 2011 and should be put to a referendum in 2012; this will be the third referendum in the space of a few years, two others having twice prompted the Icelanders to reject (in March 2010 and again in April 2011) governmentally-agreed plans for the payment of the debt resulting from bank failures. Icelandic policies have essentially preserved the welfare state, and they have set into motion a reorientation of the economic model.

This experiment is only the cutting edge of hundreds or even thousands of others in which random selection is being used or proposed. Restricting ourselves to French examples: the Europe-Ecology-Greens (EELV) party in Metz has chosen by lot its cantonal and legislative candidates. The Foundation for Political Innovation, close to the UMP, proposes that from now on 10% of municipal councillors be chosen by lot. [2] The centrist Montaigne Institute suggests using a citizen conference to discuss the financing of the health care system. The Jean Jaurès Foundation, linked to the French Socialist Party, is reflecting on citizen juries. [3] The ecologist Hulot Foundation is calling for the creation of a third legislative chamber to be chosen by lot, [4] while the directors of ATTAC, close to the Radical Left, are talking about replacing the Senate with a chamber chosen that way. At the international level, there is growing interest in sortition in political theory.

Why did sortition disappear in modern democracies after the revolutions of the seventeenth century? Why is it coming back today? What are the possible justifications for a significant use of this procedure in contemporary democracies?

The Disappearance of Sortition in Modern Democracies

Republican and democratic practice has rarely hinged on a single procedure, and the modern single-mindedness about elections is historically rather exceptional. Bernard Manin first raised the question of why with modern revolutions sortition disappeared from the political scene. [5] His answer was based on two observations: first, the founding fathers of modern republics wanted elective aristocracies, and for this reason they rejected random selection, which Plato and Aristotle had connected with democracy. Second, the theory of consent, deeply rooted in theories of natural law, was so widespread that it seemed difficult to legitimize any political authority that was not formally approved by the citizenry.

Both of these arguments are important, but they do not explain everything. In particular, they do not help us understand why radical advocates of descriptive representation (in which the representative body resembles in its characteristics the whole population) did not argue in favour of the political use of sortition, the democratic character of which was still being pointed out by Montesquieu and Rousseau. To explain this enigma, we have to descend from the airy heights of “pure” political thinking and grub around in the ways that these ideas materialize in techniques of governing, in tools and practical devices. The idea of the representative sample is familiar to twenty-first century readers, who have been made receptive by decades of intensive use of statistics and opinion polls. However, it was not invented until the end of the nineteenth century. There could be no relationship between sortition and descriptive representation before, because the idea that a random selection statistically produces a sample that has the same characteristics as the original set was not yet scientifically available. [6]

The unavailability of the statistical concept of the representative sample (even though probability calculus was already well developed by the time of the American and French Revolutions) is the key to understanding why political sortition seemed useless in modern democracies, whose size – as almost no political writer in this period failed to point out – made it impossible to have self government similar to that of the ancient democracies. In this conceptual world, drawing lots meant arbitrarily giving power to someone. Lacking the idea of the representative sample, the proponents of descriptive representation were forced to choose other tools to advance their ideals.

Deliberative Mini-Publics

Conversely, the current significance of the comeback of sortition in a number of experiments can be explained largely by the spread of the concept of the representative sample, which through opinion polling had gained some political legitimacy. Contemporary experiments are novel in considering sortition as a way of selecting a representative (or at least diverse) sample of the population, a kind of microcosm of the city, a mini-public that can opine, assess, judge and possibly decide in the name of the community, in cases where everyone cannot take part in deliberation and where social heterogeneity hinders thinking of all individuals as interchangeable. Some philosophers and historians of science have described how a “taming of chance” arose from calculations of probabilities as these gradually came to be used as a tool in scientific, administrative and commercial activities. [7] We can extend this reasoning to politics and argue that the calculation of probabilities – or more precisely a variation of it in the idea of the representative sample – helped to bring about in the 1970s a taming of chance in the form of the mini-public.

Some classical ideals such as the equality of every citizen in a random selection and the idea that everyone can make a useful contribution to solving communal problems are revived in current experiments. However, in cities like ancient Athens or medieval and renaissance Florence, each member of the lot-drawing group took turns at being governed and governing. In this system, rapid turnover in public offices coupled with random selection allowed a kind of self government difficult to see happening at the national level in modern democracies. Deliberative democracy rests on different reasoning. [8] It is based on mini-publics that make it feasible to create a counterfactual public opinion, different from elected representatives but also from the public opinion of large numbers of people. John Adams could claim that representatives “think, feel, reason, and act” like “the people at large.” [9] For theorists of deliberative democracy, statistical similarity between “descriptive” representatives and the people is only a starting point. The mini-public, once it has deliberated, is supposed to be able to change its opinion; in fact, such a change is taken to be a sign of a high-quality deliberation. This can be seen very clearly in James Fishkin’s presentation of the reasoning behind the deliberative poll, an instrument that he invented:

<quote<Take a national random sample of the electorate and transport those people from all over the country to a single place. Immerse the sample in the issues, with carefully balanced briefing materials, with intensive discussions in small groups, and with the chance to question competing experts and politicians. At the end of several days of working through the issues face to face, poll the participants in detail. The resulting survey offers a representation of the considered judgments of the public....

The aim is to have this process differentiated from the epistemological and political reasoning of conventional polling. The latter is merely “a statistical aggregation of vague impressions formed mostly in ignorance of sharply competing arguments,” while a deliberative poll aims to tell us “what the public would think, had it a better opportunity to consider the questions at issue.” [10]

The Political Justifications

While there are many different combinations, several standard sortition and deliberative democratic devices have now been put to the test. The oldest, invented in the 1970s in both Germany and in the United States, is the citizen jury. Inspired from the trial juries, the citizen jury lets a group of people (from a dozen to a few tens) selected by lot to deliberate under optimal procedural conditions, alternating for one or two weekends with internal discussion and hearing from experts taking contradictory positions. The experts are selected (sometimes in collaboration with the jury if it has several sessions) by the hosts of the proceedings, who must be independent from the organizers. The purpose of the jury is to produce public recommendations on the public policy issue on which it has been convened. [11] Similar to citizen juries are consensus conferences, developed in Denmark at the end of the 1980s; these focus on scientific and technical issues. [12] Deliberative polls, tried out by Fishkin in the 1970s, are distinguished by their size (they can bring together several hundreds of citizens selected by lot) but also because they end not with a consensual opinion but with a poll of opinions that have been informed by high-quality debates. Conversely, citizens assemblies of the Icelandic variety – first held in 2004 in British Columbia, Canada [13] – which can be as large as deliberative polls, have the power to make decisions or at least to submit proposals to the whole population for a referendum.

What sources of legitimacy can these devices rely on? The mini-publics selected by lot clearly cannot be justified either by their numbers or by their expert knowledge. However, historical experience along with philosophical reflection shows us that they can claim several kinds of legitimacy.

More deliberative politics. The assumption underlying current devices, namely that deliberation by “ordinary” citizens conducted in good conditions can lead to reasonable results, tends to be broadly corroborated by empirical social science studies conducted by observers who are not part of the process being analysed. To the extent that the opinion of a mini-public is more “reasonable” than mass public opinion, it can go on to influence the latter if it is reported by the media. In the context of a widely-shared dissatisfaction with the political system, this can counterbalance politics-as-spectacle and the autonomy of the political class, and help make that class more responsible to citizens. The aim is to promote better communication between policy makers and citizens, as well as high-quality deliberation with citizen involvement.

Diversity of social experience. In addition, the deliberative devices that bring ordinary citizens together can have epistemological advantages over representative government and committees of wise men: good deliberation must include diverse points of view, so that the range of arguments considered will be broadened and the discussion will be more inclusive. [14] From this point of view, randomly-selected mini-publics have the advantage of being socially – and therefore epistemologically – richer than committees of experts or of political leaders; they are also socially and epistemologically richer than publics where the participants come purely from volunteers or only from already organized civil society. The input of such mini-publics is important in a world of increasing complexity.

A substitute for radical democracy. The third argument comes from a radical democratic perspective. Advocates of representative government sometimes argue that the best political system would no doubt be a true “government of, by and for the people,” but that because such a system is impossible in large mass societies, representative government is the least bad option that is available. However, it is possible to maintain that in modern democracies the least bad solution is to replace the self-governing people, or at least to back it up, with randomly-selected mini-publics, since this device affirms the fundamental democratic value of equality: each citizen has an equal chance of participating in the decision, and the diversity in the social composition of a mini-public reflects that of the people.

Impartiality. The fourth, more consensual argument for randomly-selected mini-publics, derived from wide-ranging historical experience, is their impartiality. Elected officials, experts and organized interests have a strong tendency to defend particular interests. Conversely, random selection tends to recruit non-partisan people without career interests to defend, and they are encouraged by the deliberative procedural rules to reach a judgement tending towards the public interest. This feature is particularly valuable when it comes to dealing with long-term issues such as the preservation of environmental balances and of living conditions for future generations. [15]

The Challenges Confronting Deliberative Democracy

However, random selection devices confront a number of challenges.

Deliberation and social inequalities. How can speaking be divided in an egalitarian way in a group that is socially and culturally heterogeneous, where some are more accustomed than others to public speaking, where having cultural capital tends to lower inhibitions and to reinforce self confidence, and where discussions within a given procedural device can be strongly influenced by the experts being interviewed and the professionals running that device? The procedures try to minimize the asymmetries. In order to appreciate the potentials here, you need to have seen previously silent people speak up in small group (“breakout”) sessions and return to the plenary session with a greater ability to speak. However, the equality is never perfect.

The effects of the deliberation on individuals. What are the real effects of deliberation on the individuals taking part in these mini-publics? Theorists of deliberative democracy often assume that by informed discussion citizens are enabled to persuade each other through the unconstrained force of the better argument. However, this assumption is difficult to prove, and what actually happens in deliberations largely remains a “black box.” Psychologists have even argued that small-group discussions strengthen pre-existing differences and make it more difficult to find a compromise. Although this logic based on laboratory experiments does not seem to appear in the actual practice of mini-publics, it must be admitted that the precise effects of deliberation on individuals have yet to be demonstrated, though there is a fast growing literature on this subject. [16]

The issue of responsibility. The new devices clearly raise the issue of the responsibility of the citizens who are randomly selected. In ancient Athens, those who became members of the colleges of magistrates had to account for their actions. Even though they certainly did not always respect their electoral promises, elected officials had legal responsibility when they held executive positions, and were answerable to their electors if they sought to be re-elected. To whom could and should today’s juries be accountable?

Deliberation and publicity. While deliberative polls are public, this is not true of citizen juries or consensus conferences, which take place entirely behind closed doors even if the final verdict is rendered publicly. In another context, Jon Elster has argued that public sessions encourage members to rigidify their positions and to deploy rhetorical arguments. [17] Others have countered that publicity is useful in order to stop juries from being influenced by lobbies. From a Habermassian or Kantian perspective, publicity is rather one of the strengths of discussion, in that it encourages speakers to consider the general interest or at least to try to show in what ways their arguments are compatible with it, and the pressure of publicity curtails deal-making more effectively than closed-door sessions. In any event, non-public discussions have the major disadvantage of making involvement of the general public more difficult.

Deliberation by mini-publics versus deliberation by the mass. By definition, deliberative mini-publics aim to reach a counterfactual opinion (what public opinion could be), better informed, enjoying a reasonably satisfactory setting in which to be formulated, but which therefore may well differ from the mass opinion. The way in which juries’ recommendations have been rejected in several key experiments amply demonstrates that this is not a merely speculative risk.

The issue of social transformation. How could mini-publics have real relevance to the world? Randomly-selected citizens are dug up from their previous social contexts and artificially placed together. Given that their coming together follows solely from the willingness of public authorities, it is unlikely that they can be really subversive about power structures. To impose positive changes in a world where the structural resistance of the dominant interests is enormous, instead of reasonable discussions in modest committees, would it not be more effective to call for indignation and a mass uprising against injustices and perils threatening the planet?

Changes in Democratic Representation

In spite of these challenges, the current wave of sortition experiments is evidence of an attempt to enrich democracy. Take for example the Citizens Assembly in British Columbia in 2004, which had been tasked with coming up with a proposal to reform this Canadian province’s voting system. The organizers felt that in order to avoid conflicts of interest it was best to entrust the reform of an electoral law to someone other than the political parties. After working for a year, the Citizens Assembly proposed to do away with the first-past-the-post system, which crushes minorities, and to introduce a more proportional logic. The proposed law as drafted was put forward for ratification by a referendum in May 2005. Gordon Gibson, an adviser to the province’s Prime Minister and creator of the Citizens Assembly, justified this innovation as follows:

We are ... adding new elements to both representative and direct democracy.... As things stand now, both streams of decision making are highly influenced – almost captured – by experts and special interests. The idea of deliberative democracy is essential in order to make the public interest enter the fray, as represented by panels of citizens randomly selected. The traditional representatives we elect are chosen by majoritarian consensus, for an extended period, as professionals, with unlimited jurisdiction to act in our name. The new kinds of representatives we are talking about are chosen at random, for a short period, as ordinary citizens and for specified and limited purposes. [18]

{}To appreciate these remarks, we must recall that modern democracy has never been reduced to representative government on its own, i.e. to regular elections of representatives who, in the name of their constituents, have a monopoly on decision making. Also part of the panorama are the establishment of an independent judiciary, the activity of committees of experts, social democracy implying social partners, and elements of direct democracy such as referendums and popular initiatives. But now, with social networks and the end of mass parties, a pluralisation can be seen in forms of democratic legitimacy. [19]In the future, it would be desirable for sortition to be again associated with elections, as it was in most of the democratic and republican experiments down through history. For this innovation to be truly meaningful, it must be legally institutionalized and not rest only on the political will of leaders. Of course the idea is not to suppress elections, but to enrich the democratic dynamic by bringing into it this new element on a significant scale. Iceland’s experience clearly shows the potential that this element holds for the democratization of democracy – such a contrast with Greece, which for its part followed the path of reform from above, leaving the institutional system as it was!

It would in any case be naïve to think that politics in the twenty-first century will just continue, with marginal modifications to the politics of the preceding century. Given the size of the crisis of financial capitalism and the increasingly evident impasse into which we are being led by the production model followed up to now, and in view of the massive disrepute into which institutional politics has fallen, the status quo would seem to be neither realistic nor adequate.

Previously published in French in laviedesidees.fr. Translated by John Zvesper with the support of the Institut Français.

{{}}