Sortition refers to the selection of political representatives by random drawing from the entire citizenry. This procedure, used for centuries in the republican regimes of Athens and then Rome, nearly disappeared at the beginning of the 19th century in favor of elections. Since the 1970s, researchers and institutional frameworks have been reintroducing it as a tool for participation and deliberation.
The participation bias, a blind spot of contemporary sortition
Most presentations of sortition emphasize the equality produced by chance: each citizen has the same probability of being selected. This theoretical principle faces a concrete obstacle: the acceptance rate of those drawn remains low.
According to the OECD database published in 2023, the median acceptance rates observed in reported deliberative processes are low. A significant portion of citizen assemblies do not even communicate this information.
Those who refuse to participate do not constitute a neutral sample: a study published in the European Journal of Political Research in September 2026 shows that non-participants differ in their electoral behavior, their trust in government, and their level of concern about certain issues like climate change.
The random procedure alone is therefore insufficient to guarantee representativeness. Compensation, logistical support (childcare, transport, accommodation), and the recruitment method weigh as heavily as the draw itself. The same study tested door-to-door outreach as a lever: it increases the registration of solicited individuals, but some barriers require stronger measures than a simple invitation.
This structural problem distinguishes contemporary Sortition from its Athenian ancestor, where participation was backed by a binding civic framework and a much more restricted eligible population.
Sortition and election: two logics of political legitimacy
Elections are based on a mandate: the people choose a representative who acts on their behalf. Sortition is based on a sample: a group of citizens, statistically close to the population, deliberates on a given topic. These two procedures do not produce the same type of legitimacy.
Voting confers legitimacy through explicit consent from voters. Sortition aims for legitimacy through resemblance: the randomly drawn assembly is supposed to reflect the social, economic, and demographic diversity of the civic body. Bernard Manin, in his work on representative government, showed that sortition and election coexisted in most historical republics, each fulfilling a distinct function.
Elections favor the selection of profiles deemed competent or charismatic, which mechanically creates a distance between elected officials and voters. Sortition, on the other hand, produces assemblies composed of individuals without prior mandates, which reduces this distance but raises questions about technical competence and accountability to citizens who are not drawn.
What sortition does not replace
No contemporary framework proposes to fully substitute elections with sortition. Recent citizen assemblies (climate conventions in France, deliberative assemblies in Ireland or Scotland) function as advisory or pre-legislative bodies. Their power depends on the institutional framework that hosts them.
The French Court of Auditors has indeed made recommendations to clarify the legal status and decision-making scope of these citizen participation mechanisms, highlighting the gap between the expectations raised and the non-binding nature of the opinions rendered.
Descriptive representativeness and sortition: practical conditions
Sortition produces descriptive representativeness: the assembly resembles the population on measurable criteria (age, gender, socio-professional category, territory). For this resemblance to be effective, several conditions must be met:
- A sufficiently broad and up-to-date drawing register, covering all eligible citizens, not just electoral rolls that exclude non-registered individuals
- A stratification or weighting mechanism that corrects self-selection biases, as volunteers do not represent the general population
- Real compensation covering income loss, travel expenses, and childcare, otherwise precarious categories remain underrepresented
- A follow-up and support protocol that reduces the gap between solicited individuals and those who actually agree to serve
Without these measures, sortition reproduces the participation inequalities it claims to correct. OECD data confirms that actual representativeness varies significantly from one mechanism to another depending on the resources allocated for recruitment.
Citizen assembly and deliberation: the French framework
In France, the Citizen Convention for Climate has been the most visible test of sortition applied at the national level. Its participants, drawn by lot, produced legislative proposals, some of which were adopted, modified, or rejected by the government and Parliament.
This experience highlighted a recurring problem: the link between citizen deliberation and political decision-making remains unclear. Citizen conventions do not have their own legislative power. Their recommendations pass through the filter of elected assemblies, which generates a sense of unfulfilled promise among participants.
The Senate and the Court of Auditors have recommended better regulation of these mechanisms to avoid an inflation of expectations without guaranteed results. The question of the “last mile” between citizen opinion and the voted law remains the main point of friction in the French debate on participatory democracy.
What institutional future for sortition
Several proposals are circulating: a randomly drawn upper chamber, permanent citizen evaluation commissions, deliberative panels integrated into the legislative process. None have advanced to the stage of formal proposal within the French constitutional framework.
For now, sortition remains a sporadic complement to representative democracy, not a replacement. Its credibility will depend less on the random procedure itself than on the institutions’ ability to make its results binding and to resolve the structural problem of effective participation of solicited citizens.
