Just recently, Liquid Democracy enthusiasts suggested alternative delegation models different from the transitive delegations used in LiquidFeedback. These models are the so-called “preferential delegations” and also different hybrids combining the ideas of preferential and transitive delegations.
In a paper titled “Preferential delegations and the problem of negative voting weight”, Jan Behrens and Björn Swierczek demonstrate that preferential delegations as well as all hybrid solutions do not satisfy at least one of seven properties to be demanded from a Liquid Democracy system.
The mathematical proof uses these properties and resulting rules to construct a logical contradiction, which exists whenever preferential delegation is allowed. This includes any hybrid independent of its implementation. In a second step, the number of properties necessary to demonstrate the deficits of preferential delegations has been reduced to five.
Primary publication in “The Liquid Democracy Journal on electronic participation, collective moderation, and voting systems“, Issue 3:
↗ www.liquid-democracy-journal.org/issue/3/
Addendum from March 28, 2015, which reduces the number of necessary properties from seven to five:
↗ Addendum (advance publication)
“The Liquid Democracy Journal on electronic participation, collective moderation, and voting systems“
↗ www.liquid-democracy-journal.org/
“The Liquid Democracy Journal on electronic participation, collective moderation, and voting systems“ is dedicated to the idea of Liquid Democracy, which is a democratic principle that uses transitive delegations to unite the best of direct and representative democracy.
But this journal is not just limited to Liquid Democracy; it also covers those topics coming up when implementing it: electronic participation, collective moderation, and voting systems.
###
Two of the 63 figures, which make the mathematical proof understandable even for the layman Photo: IAD
Jan Behrens Photo: IAD