SOPhiA 2018

Salzburgiense Concilium Omnibus Philosophis Analyticis

SOPhiA ToolsDE-pageEN-page

Programme - Talk

Condorcet's Jury Theorem: General Will and Epistemic Democracy
(Political Philosophy & Philosophy of Law, English)

In my talk I will discuss what is Condorcet's Jury Theorem (CJT) and in what way is it related to epistemic democracy and Rousseau's idea of general will. CJT says this: If we assume that every voter has the same level of competence (i.e. a probability of making a correct decision when choosing between a pair of alternatives) which is greater than 0.5, then, if the voters choose independently of each other, their collective competence (i.e. a probability that the group deciding on the basis of majority rule will make a correct decision) will be greater than the competence of any individual voter.

There are many assumptions in the background of such formulation, and I will summarize them in the following way: (1) Correctness - among the alternatives there is exactly one which is correct; (2) Independence - voters decide independent of one another; (3) Homogeneity - competence of all voters is the same; (4) Minimal competence - competence of every voter is larger than 0.5; (5) Majority rule - the winner is the alternative which scores more than a half of the votes; (6) Binarity - the choice is made between two alternatives.

In the second part I will look into every assumption in more details. I will examine the connection between assumption of correctness and independent standard in epistemic democracy proposed by Cohen (1986), and connection between assumption of independence and Rousseau's notions about negative influence of factions on voting outcomes. In the end I will look at the assumption of binarity. CJT assumes that we choose between two alternatives, while in democratic elections we are usually faced with multiple options. Condorcet himself proposed that the majority voting method should be replaced with one that requires of voters to rank all of the options according to their preferences, and then declares the option which is preferred over any other by a majority of voters as a winner. The second solution is proposed by List and Goodin (2001) who argue that CJT could be applied to the situations with multiple options, as long as minimal average competence of the voters is larger than 1/k, where k stands for number of options, and majority rule is replaced with plurality rule. I will try to show that their solution implicitly leads to rejection of assumption of correctness and is therefore unsuitable for epistemic democrats.

Chair: Jan Buran
Time: 14:00-14:30, 14 September 2018 (Friday)
Location: SR 1.007

Miljan Vasić 
(University of Belgrade, Serbia)

Miljan Vasić

MA Student

Faculty of Philosophy University of Belgrade

MA Thesis: Democracy and "One Man - One Vote" Rule

Testability and Meaning deco