By proposing to take seriously the idea of a legislative assembly drawn by lot, the work edited by John Gastil and Erik Olin Wright invites us to rethink some of the fundamental concepts of constitutional analysis.
Drawing lots for members of a legislative assembly: incongruous just a few decades ago, this idea is nothing but audacious today. In the meantime, many institutions (Parliaments, Governments, communities) have used “mini-publics”, that is, provisional bodies composed by drawing lots in order to formulate proposals or give an opinion on a specific topic. One of the most successful experiments in this area was the work of the Irish Parliament, which successively convened two assemblies, composed partly (for the first) or entirely (for the second) of citizens drawn by lot. Some of their proposals were submitted to a referendum, leading to the legalization of marriage for same-sex couples in 2015 and that of abortion in 2018.
Based on these experiences, John Gastil, professor of communication and political science, and Erik Olin Wright, sociologist, propose to imagine what a randomly selected parliamentary assembly could be like. The collective work that they coordinated, Legislature by Lot. Transformative Designs for Deliberative Governance opens with this proposition, to which twenty or so of the best specialists in deliberative democracy respond.
A bicameral parliament, combining election and drawing of lots
The initial proposal by John Gastil and Erik Olin Wright is relatively simple: establish a bicameral parliament composed of an elected chamber and a randomly selected chamber. These two chambers would have equal powers and would each assume all the missions of parliamentary assemblies, from voting on laws to monitoring government action.
At the heart of their proposal is an analysis of the inadequacies and flaws of political elections. Two criticisms in particular are directed against them. On the one hand, the idea that elected parliamentary assemblies no longer deliberate (p. 7): the people elected to them have made campaign commitments, carry a political project and are subject to group discipline, which prevents them from changing their minds and being convinced by the arguments of other parliamentarians. As a result, “the force of the best argument” no longer carries weight and decisions are taken primarily on a partisan basis and not within the framework of a deliberative process. On the other hand, elections fail to create a Parliament that reflects society. American senators, for example, have a median wealth of $3.1 million and are 79% men, as Tom Malleson notes (p. 172).
Conversely, a randomly selected assembly would fully reflect society, of which it would constitute a representative sample and would be able to deliberate more easily, that is to say, attempt to develop, through discussion and debate, proposals of general interest. In Gastil and Wright’s proposal, the randomly selected individuals would be appointed for two or five years and would benefit from guarantees and resources (in particular an allowance) identical to those of elected parliamentarians. An internal revocation mechanism within the chamber would make it possible to exclude, on an ad hoc basis, individuals who would harm its work or damage its image. The organization of the randomly selected chamber would be largely modeled on that of an elected assembly. Its members would be divided into committees, which would allow for relative specialization. In conducting their work, they would be supported by a dedicated administration and would benefit from the possibility of hearing experts before deliberating on a specific topic.
Unlike the authors of other contributions (in particular Hennig, p. 301, and Bouricius, p. 313), John Gastil and Erik Olin Wright propose to maintain an elected chamber, with the triple aim of encouraging the development of political parties, of retaining a capacity to arbitrate conflicts which divide society, not all of which can be resolved by consensus, and of bringing out political leaders.
A proposal that is a shock to certain constitutional concepts
While this proposal is not new, the format of Legislature by Lot has the merit of drawing attention to its institutional implications and, implicitly, to the role that our political systems give to elections, which goes well beyond a simple procedure for designating people to a political office.
The proposal for a randomly selected assembly first raises the question of political representation. One of the arguments used in favour of random selection is that it allows for greater representation, for example in terms of gender, age, socio-professional categories or minorities (p. 17). However, this quality of random selection is only valid if participation is compulsory for those selected. This therefore raises the question of constraint, similar to what exists for assize juries. However, the duration is not comparable: a few weeks on one side and several years on the other. Furthermore, over such a duration, the risk of professionalisation of those selected is not insignificant, which would reduce the expected benefits of random selection. If participation remained optional, self-selection biases would conversely be to be feared.
Another important issue is that of responsibility. Political elections as we know them are generally seen as a dual mechanism of designation and sanction (by being able to oppose the renewal of the mandate of an outgoing elected official). This latter characteristic no longer exists with the drawing of lots since citizens have no possibility of acting on the people who are designated. Jane Mansbridge, whose contribution deals with this subject, recognizes this difficulty: ” I would oppose any legislative system based entirely on lot. Too much of value to democracy would be lost. » (“I would oppose any legislative organization based entirely on the drawing of lots. It would be too much to lose of what makes democracy valuable”) (p. 193). However, equating election with accountability can be at least partly contested. In the case of second presidential terms in the United States or in France, voters have no means of sanctioning the accountability of the incumbent president, who cannot run again. On the other hand, political accountability can also be given a broader conception, resulting in the surveillance of civil society and the obligation to regularly report. This is what Jane Mansbridge calls the “ deliberative accountability » (“deliberative accountability”) (p. 194). In this sense, a randomly selected assembly would also be responsible since it operates under public scrutiny.
Finally, the question arises of the relations between a randomly selected assembly, on the one hand, and the other constitutional public powers, and in particular, the other chamber and the executive power, on the other hand. Indeed, within political institutions, two forms of legitimacy would coexist, one based on election and the other on drawing lots. However, the contemporary logic of representative democracies is based on electoral promises and commitments, intended to be fulfilled once the election has been won, the vote being deemed to be both the designation of a person and adherence to a program. Conversely, the interest of a randomly selected assembly is to allow informed deliberation simulating ” what the people would think about the issue under good conditions for thinking about it ” (“what the people would think about this subject if they were put in the right conditions to think about it”) (Fishkin, p. 98). These are two different logics, likely to produce opposite results and to generate a conflict of legitimacy between the two chambers, because ” elections and exit embody competing and mutually undermining conceptions of representation » (“elections and drawing lots embody competing and mutually undermining conceptions of representation”) (Vandamme, Jacquet, Niessen, Pitseys and Reuchamps, p. 130).
Many other questions are also considered, such as the competence of citizens drawn by lot to legislate or the link between the randomly selected assembly, social movements and participatory democracy.
A realistic utopia?
Finally, there remains an essential question, that of the realism of such a proposal. Indeed, if drawing lots is increasingly studied as a method of selecting governments, it remains little known to the general public and little supported by political leaders.
Chapter 6 (Vandamme, Jacquet, Niessen, Pitseys and Reuchamps) provides interesting data on this subject. It shows that 77% of the Belgian parliamentarians questioned are hostile to the prospect of a second randomly selected assembly (with a notable exception for the Green deputies, who are overwhelmingly in favour), while Belgian citizens are more neutral, since only 40% are opposed to it. There is therefore neither social consensus nor political demand, at present, for establishing such an assembly, even in Ireland, which has nevertheless had successful recent experiments (Arnold, Farrell, Suiter, p. 120).
Moreover, studies are lacking, for want of an assembly to study. Indeed, a parliamentary assembly drawn by lot did not exist anywhere, not even in Athens of the IVe century, whose political system was rather based on a multitude of bodies drawn by lot, but none of which had the full powers of a parliamentary assembly. While many deliberative mini-publics (consensus conferences, deliberative polls, etc.) have been studied in recent years, their characteristics are profoundly different since, in general, they only lasted a few days, they focused on a single subject and they had no decision-making power. Consequently, questions such as professionalization, lobbying and even corruption did not arise. There is therefore a ” scale-transformation problem” (Malleson, p. 174) between a mini-public and a randomly selected assembly, which is currently insurmountable, and which equates the creation of a randomly selected assembly with a ” leap of faith » (“act of faith”) (Bouricius, p. 327), even leading some to fear that a bad experience would permanently discredit the drawing of lots as a political instrument (Burks and Kies, p. 260 et seq.).
This is why many contributors promote a gradual approach, which Gastil and Wright subscribe to in their final response. The most frequently proposed path would be to multiply the mini-publics drawn at random on defined themes (Fishkin, p. 100). This is a trend that can currently be observed in certain European countries and that France is about to experiment with at the national level with the citizens’ convention on ecological transition, which is due to meet in the second half of the year.