On Mon, 11 Nov 2002 12:41:42 -0800 (PST) Alex Small wrote:
Douglas Greene said:http://www.lp.org/lpnews/0211/libsolutions.html Shall we LPers draft a response?The most interesting part of the report is this: "The instant runoff method assures that the least disliked candidate is elected." That's a dubious statement. Even Donald would agree that IRV (just like Condorcet) only uses ordinal information. The IRV procedure doesn't know (and doesn't care) whether your second place candidate is actively disliked, or merely not liked as much as the first place candidate. IRV only knows that the candidate with the fewest first-place votes, however liked he may or may not be, did not attract as much first-place support as the other candidates.
The quoted statement is worse than dubious - it is just plain untrue. IRV's basic problem is that, while it provides for the voter offering ordinal information, IRV selects only part of what is offered in making its decision.
Condorcet accepts the same ordinal information, but uses all of it.
As to "actively disliked", no method should be considered acceptable if it forces voters to list such candidates - unless the method provides for them being labeled as such by the voter.
Maybe a joint response with as many co-signers as possible is in order. I suggest emphasizing the following points: 1) Runoff methods (of which IRV and 2-step are the most common) aren't nearly as effective as PR for promoting multi-party competition. Australia and Louisiana come to mind. Although Louisiana uses 2-step runoff for state-wide races, when there are only 3 candidates with significant support 2-step is roughly equivalent to IRV. I haven't seen many 3rd-party governors or Senators in that state.
I have no comment about packaging legislature seats such that PR might make sense.
Even if that gets done, there are offices such as mayor and governor that are almost certainly best kept as single seat offices.
I do not see 3 candidates as a worthy special case:
2 candidates can happen, but method matters little, if at all, here.
Going past 2, method BETTER be able to handle 3, 4, etc.
NY had about 9 candidates for governor in 98.
FL had 13 candidates for President in 00.
NY had about 9 candidates for governor in 02 - here THREE getting major vote counts.
Does not matter whether a 3rd-party candidate has a chance of winning - what matters is whether the voter can express desire for such AND express choice among the major party candidates.
2) Approval Voting and some PR methods are much cheaper to implement, a significant concern for a party of fiscal conservatives.
If the votes are counted by computer, which they certainly should be, Approval is not that much cheaper than Condorcet. IRV costs a bit more and there could be even more exotic and expensive methods.
3) Although AV doesn't have much field-testing in democracies, it's easy to implement and academic work suggests significant advantages.
Equipment that cannot do better than AV is overdue for replacement anyway.
Any method that allows for selecting more than one candidate (DESIRABLE) should be rejected if, as with AV, it fails to let the voter indicate preference among them (voter can have GOOD cause for selecting a candidate as MUCH DESIRED, and a second who is BARELY TOLERABLE though better than those not selected).
4) Just as the party has left open a variety of PR options, the party
should leave open a variety of single-winner options. Let there be a debate in the marketplace of ideas.
We seem agreed that there are a variety of PR options of more or less equal value.
Seems like enough is known about single-winner options to reject the worst, such as plurality.
--Alex
[EMAIL PROTECTED] http://www.clarityconnect.com/webpages3/davek
Dave Ketchum 108 Halstead Ave, Owego, NY 13827-1708 607-687-5026
Do to no one what you would not want done to you.
If you want peace, work for justice.
----
For more information about this list (subscribe, unsubscribe, FAQ, etc), please see http://www.eskimo.com/~robla/em
