At 05:17 PM 9/14/2005, Simmons, Forest wrote:
(1) Pick 400 registered voters at random. Give them time off work,
and have them study the candidates carefully and decide a tentative
winner by some reputable method (like DMC, Shulze, Approval, Asset
Voting, etc.) It will definitely be worth the time for these
voters, since there will be an appreciable chance that their votes
will be pivotal.
(2) Analyze the ballots from step 1 statistically. If the results
are so decisive that there is less than a one percent chance that a
different winner would emerge from the same method involving the
entire electorate, then submit the tentative winner to the
electorate for ratification by a Yes/No vote. [and more]
The idea of using a committee, essentially, to select a tentative
winner is an excellent one, and parallels my own ideas in a way. If
the committee is statistically unbiased, it should work. Tricky to
ensure that, however.
It would be a good idea to have *any* election result that is below a
clear and general consensus (supermajority definition not stated
here) be submitted to a ratification vote. This effectively forces
voters to approve or disapprove of the result. Thus, in the end, it
is Approval which has been used, in a way, with a very significant
tweak that deals with straight-ticket, truncated votes.
However, it might be advisable to allow more than one candidate on
the runoff ballot. Top two (by undefined standards; what about
Condorcet and Approval winners?) might be appropriate. More than
that, we would start to get strategic voting in the runoff. Another
option in the runoff might be () Start Over.
If a majority of the voters want to redo the whole election process,
it should be their privilege. Presumably the existing office-holders
would remain pending new results, or there could be statutory caretakers.
However, in my own concepts, instead of some abstract (or possibly
corrupt) agency choosing voters at random, *the voters* choose
representatives on the committee, through Delegable Proxy. The
committee functions deliberatively, instead of aggregatively. The
committee could be quite a bit smaller. And the result would *still*
be subjected to a full vote by the general electorate. (Committee
members would have variable voting power according to the votes they
received. Alternatively, Asset Voting would be used to create a PR
Committee that is almost fully representative, while being a peer
committee, though this introduces constraints that produce the "almost."
In a DP committee, one member might represent a very large percentage
of the electorate. If the committee size is fixed, this allows
members to be elected to the committee who represent much smaller
blocks of voters. This is why DP is more fully representative than
Asset. However, there could be advantages to peer committees, in
terms of the encouragement of collegiality.
In the end, the voters should choose their own election methods. And,
again, DP is a way to do it that would be likely to produce an
intelligent result, enjoying the broadest possible consensus.
----
Election-methods mailing list - see http://electorama.com/em for list info
