[EM] proxified electoral college (was: PR vs Single-Winner Reform)

[James Green-Armytage]

This is James Green-Armytage replying to Forest Simmons


>In other words, every candidate that runs is elected to the Election
>Completion Convention, and members of this convention (rather than some
>semi-democratic Electoral College) decide the winner of the original
>contest by use of Condorcet, Approval, IRV, Plurality, or some other
>method.

	Probably the official method should be majority rule. That is, successive
votes until one candidate receives a majority. Or even plurality would
work fine. The real decision-making process would take place in a
majority-forming collective bargaining process. It might make sense to use
a Condorcet procedure as an unofficial part of this process, in order to
clarify things. But when we can fit all the voters in the same room, we
can expect a majority to form, and there is no need to use a binding
rank-ballot system. The drawbacks of using a binding ranked vote methods
are the drawbacks which we are already familiar with, in Condorcet's case,
the burying strategy, and in the other methods, a whole lot of other
stuff.  
>
>But whatever the voting method used within the Completion Convention, each
>member's ballot is replicated to the number of citizens represented by the
>member in the Convention, i.e. the number of votes she received in her
>election to the Convention.

	Definitely.
>
>I suppose that one could argue that IRV is better than this method,
>perhaps even thirty percent better, but at five thousand percent of the
>cost, including the lack of transparency in IRV's non-summable vote
>counting process.

	Well, it's hard to quantify how much better it is like that. IRV has
drawbacks that this system does not, and this system has a quite different
set of drawbacks. The drawback of this system is that, for example,
someone who votes for Nader will not necessarily in every case prefer
Kerry over Bush. The method probably involves making assumptions about
people's later choices, which is bad. The good thing about it is that it
bypasses a lot of the strategic problems which are inevitable in any
singe-balloting election method, because it allows for a period of
collective bargaining where the proxies can work towards a strategic
equilibrium in realtime. 
>
>I think the public would find the televised Election Completion Convention
>to be very informative and interesting, a great educational experience,
>especially if Condorcet and Approval were sometimes used as the completion
>methods.

	I'm not sure whether it would be necessary to have a certain number of
proxy votes (like five thousand or so) in order to participate in the
convention itself. People who received fewer votes than that could be
given the option of pledging their votes to any of the candidates who had
already passed the threshold, or to other candidates in order to push them
over the threshold.
>
>After the public had a few years of exposure to these methods through the
>Completion Convention, perhaps they would want to try them directly.

	Sure, if a Condorcet procedure caught on amongst the electors, it would
be a huge boost to public awareness about Condorcet.
>
>Candidate Proxy might be the best bridge to better methods.

	Well, it does have a certain elegant simplicity to it. In itself it
doesn't necessarily guarantee a transition to other methods, but it would
be a serious step forward towards majoritarianism in itself, and would
certainly shake up the election system while leaving room for further
change.

James