Dismounting the so-called Condorcet Drivers Paradox

Abstract

In a multi-agent dynamic common-policy decision-making environment, by asking questions, an agent can generate an agreement on dynamically changing (on the fly, during the decision-making process) the common policy shared by the agents, and therefore, it can change the current list of possible options from which the decision is made. Important efforts were previously undertaken to define the Logic of Preferential Choice (LPC) as a model in which the so-called Condorcet Driver Paradox (CDP) and, more general, other apparently paradoxical decisions are legal syntaxes and valid formulas. On the contrary, we prove here that CDP can be easily dismounted in many formal languages – such as English, Arithmetic, SQL, Binary Logic, and Probabilities) - languages that are not only appropriate for modeling the process of question asking, but they are also proved as productive perspectives in decision-making, much popular in the community, much simpler and much largely accepted than the Logic of Preferential Choice.