Electing the Doge of Venice: analysis of a 13th Century protocol [pdf] (2007)
rangevoting.org
rangevoting.org
Somewhere like India or the US uses elections to work out which major faction has the most power. This is a big improvement on the factions actually fighting, the process actually simulates a battle in a loose sense and something like first past the post voting (while ugly) is good enough.
A protocol like the Venetian use is obviously satisfying very different goals. As the article points out, there are interesting properties of the system that influence negotiations and consensus building. And I assume the oligarchs already know roughly how the power breaks down if there is a big fight, there would be less factions to keep track of. I really liked the article because it explores all this.
I'd also observe that (in politics, anyway) a system where nobody quite knows how it works encourages a certain level of moderation. If people can figure out how to optimise a social technology to their political benefit they will (gerrymandering, strategic lawsuits and judge selection, regulatory capture, purposefully choosing policies to get just enough votes and no more, etc, etc). If the actors have no idea how in practice they can't do tight optimisations.
We have these words, like "elections" and "democracy" and have no idea what they are. imho.
Previous discussions
I am wondering how HN allowed a repost just a few months after that. Typically you have to wait a year or more before being able to post the same link again.