CONDORCET

“CONDORCET” stands for « CONsolidating DemOcRacy: Choosing an Electoral method transparently. » This PEPR Maths-ViVES project gathers a consortium of mathematicians, economists, computer scientists, and political scientists, who aim to compare the properties of different voting systems and enable the public and decision-makers to take advantage of this new research.

PEPR MATHS-VivES CONDORCET program

While the name “CONDORCET” indeed refers to Marie Jean Antoine Nicolas de Caritat, marquis de Condorcet, a mathematician and philosopher of the Enlightenment who was famous for his analysis of voting systems, the acronym here stands for the project » CONsolidating DemOcRacy: Choosing an Electoral method transparently. »

The CONDORCET project aims to compare the properties of different voting systems and enable the public and decision-makers to take advantage of this new research.

CONDORCET is a project of the PEPR Mathematics in Interaction (Maths-VivES, France 2030). It runs from October 2025 to June 2029. The project’s principal investigators are the mathematician Jean-Baptiste Aubin and the economist Antoinette Baujard. The CONDORCET steering committee includes Jean-Baptiste Aubin (INSA, ICJ), Antoinette Baujard (Université Jean Monnet Saint-Etienne, Gate Lyon Saint-Etienne), Antoine Rolland (Université Lyon 2, ERIC) and Irène Gannaz (INP Grenoble, G-SCOP). The coordinating partners are INSA Lyon, GATE at CNRS, ERIC at Université Lyon 2, and INP Grenoble. The CONDORCET strategic committee includes personalities from laboratories of ideas on democracy, artists and human sciences academics. The CONDORCET consortium includes over twenty mathematicians, economists, computer scientists, and political scientists from various centers in France, well recognized for their academic work and their experience in scientific outreach.

Four working packages to compare voting rules in a meaningful way

Social choice theory can compare voting procedures based on given assumptions about individual preferences, whether these are based on rankings of all candidates (as in the case of single-winner plurality, two-round, or Borda voting), or on the evaluation of each candidate (as in the case of approval voting, evaluation voting, or majority judgment voting). However, there is currently no unified framework for comparing these families of procedures. The project is designed to break this deadlock in four complementary working packages.

WP1. The project aims to explore two mathematical lines of inquiry: applying depth functions (and optimal transport) to characterize voting rules, and broadening the scope of axiomatics to encompass assumptions about preferences.

WP2. Given that comparisons between voting rules can only be useful to decision-makers if the results are interpreted in contexts which are familiar to them, the project aims to carry out simulations based on different preference distribution models, notably calibrated with real electoral databases.

WP3. Comparing voting systems also requires testing, through controlled laboratory experiments or surveys using representative samples, whether theoretical properties remain valid with real actual electoral behavior. The project also aims to study the complementary properties of voting systems, such as their explicability, their propensity for polarization or consensus, or the various determinants of voter satisfaction.

WP4. We hope to enable decision-makers, citizens, community stakeholders and elected officials to select a voting system with full awareness, and without being influenced by our own preferences. Beyond academic publications, the program therefore values dialogue with civil society partners and public decision-makers, and it places great emphasis on scientific mediation, planning to organize public conferences, comic books, documentary films, school visits, continuing education and an interactive website.