PmWiki

Governance

Project Persephone is, among other things, a collection of experiments in governance. Exovivaria, both real and virtual, would host societies requiring some form(s) of government. As well, the organization itself will need to be governed. A uniform framework should be possible.

The current proposal attempts to strike a balance between deliberative opinion polling¹ as envisioned by James Fishkin, and prediction markets governance as envisioned by Robin Hanson. This system would default to the prediction market decision -- one of a class of Market Mechanisms. Nevertheless, it would be possible to prevent that otherwise-automatic policy implementation through a series of policy-recall petitions and deliberations. Theoretically, this procedure can converge on pure majority rule, but only after much of the membership has been required to participate in formal deliberations on the issue. By design, such labored outcomes are highly unlikely. The idea is largely inspired not just by Fishkin and Hanson, but also by Michael Abramowicz's view of predictocracy.², and Mark Klein's Deliberatorium online argument-mapping mechanism.

From e-mail to Hélène Landemore:³

I think Hanson has said that futarchy⁴ wouldn't really work unless the futarchic policy decision were automatic. Obviously, that's not very democratic by any reasonable definition.

But maybe you could make it automatic unless challenged by a majority in plebiscite, with some reasonable number of petition signers (say, 1% of the electorate) required for the challenge. At which point it would be subject to deliberative polling. If it passed the deliberative poll, it would be scheduled for implementation but again subject to halting by petition-triggered plebiscite (perhaps requiring a higher fraction of signatures this time, like 5%). At which point you deliberatively poll again, randomly selecting only among those who hadn't been through any earlier deliberative poll on the issue. If you keep repeating, theoretically, this can get to 100% voter representation eventually. But at that point, everyone who might vote against it has already been represented in deliberative polling -- why would a majority vote against it in a referendum if they'd voted for it in deliberation? (Unless for some very good reason, like discovering fraud in the preparation of deliberative poll materials.) And if the petition signature threshold requirements keep going higher, the repeated-stalling process probably halts long before you get anywhere near 100%.

This obviously doesn't work for decisions with certain time limits. But as you point out, decision-procedures might be chosen depending on how they rank in terms of urgency, political salience and so on. More likely, a bigger problem would be that the "futarchs" might conflate their probabilities of political success for an unpopular decision with their odds for success if the decision were made in spite of its unpopularity. I.e., the sentiment "that'll never work because people won't allow it" could muddy estimates of probability for "that'll work if people did allow it." It might be worth some experiments to find out whether my conjectured "subjective probability dissonance" is real or not. (Assuming my idea is even worth thinking about for the years required to fund such research.)

One reasonable objection to any democratic procedure that incorporates Deliberative Opinion Polling (DP) is that DP virtually requires "drafting" members to involve them in deliberation. An alternative is to offer some sort of proxy vote transfer: Whenever someone is randomly selected but doesn't want to participate, can't participate, or feels that someone else's contributions would be more appropriate to the issue at hand, they can hand their vote off to another member. Software such as LiquidFeedback should facilitate this process. The final decision on who joins the deliberations could be made using Evaluative Proportional Representation.

¹ Fishkin, James S.: Democracy and Deliberation: New Directions for Democratic Reform, ISBN 0-300-05161-1, Yale University Press, 1991 ⇑

² Predictocracy: Market Mechanisms for Public and Private Decision Making, Michael Abramowicz, Yale University Press, 2008, ISBN 0300115997 ⇑

³ A Yale political science professor whose theme is "democratic reason"; see e.g., "Hélène Landemore on why we're all in this together", 6 June 2012, Princeton University Press blog ⇑

⁴ Hanson, Robin (August 2000). "Futarchy: Vote values, but bet beliefs" ⇑

This page may have a more recent version on pmwiki.org: PmWiki:Governance, and a talk page: PmWiki:Governance-Talk.