Grassroots Federation: Fair Governance of Large-Scale, Decentralized, Sovereign Digital Communities
Ehud Shapiro, Nimrod Talmon
TL;DR
Grassroots Federation targets egalitarian, sovereign governance for massive digital communities by replacing large-scale voting with small, sortition-based assemblies. It develops dynamic federation graphs and a federation transition system to model growth while enforcing fairness in representation and participation, both in the static and dynamic limits. The paper proves that a restricted protocol FF achieves persistently-fair representation and eventual fairness in eventually-stable runs, providing a mathematically grounded path toward scalable, sybil-resilient digital democracy. Its framework supports upward mobility and outlines practical directions, simulations, and extensions toward global digital democracy with grassroots foundations.
Abstract
Grassroots Federation aims to address the egalitarian formation and the fair democratic governance of large-scale, decentralized, sovereign digital communities, the size of the EU, the US, existing social networks, and even humanity at large. A grassroots federation evolves via the grassroots formation of digital communities and their consensual federation. Such digital communities may form according to geography, jurisdiction, affiliations, relations, interests, causes, and more. Small communities (say up to $100$ members) govern themselves; larger communities -- no matter how large -- are governed by a similarly-small assembly elected by sortition among its members. Earlier work on Grassroots Democratic Federation explored the fair sortition of the assemblies of a federation in a static setting: Given a federation, populate its assemblies with members satisfying ex ante and ex post fairness conditions on the participation of members of a community in its assembly, and on the representation of child communities in the assembly of their parent community. In practice, we expect a grassroots democratic federation to grow and evolve dynamically and in all directions -- bottom-up, top-down, and middle-out. To address that, we formally specify this dynamic setting and adapt the static fairness conditions to it: The ex post condition on the fair representation of a child community becomes a condition that must always hold; the ex ante conditions in expectation on the fair participation of an individual and on the fair representation of a child community become conditions satisfied in actuality in the limit, provided the federation structure eventually stabilizes. We then present a protocol that satisfies these fairness conditions.