[Election-Methods] Bullet Voting in the wider media

[Abd ul-Rahman Lomax]

At 11:53 AM 10/7/2007, Brian Olson wrote:
>In case anyone's interested in what the general public are hearing
>about voting strategy.
>
>http://www.boston.com/news/local/articles/2007/10/07/ballot_query_to_bullet_or_not_to_bullet

I fixed the link, hopefully, so it should work unless it gets split again.

The article seemed pretty good. Bullet Voting is a choice that voters 
can legitimately make. It is not "insincere", practically by definition.

In Approval, voters effectively consider their internal range of 
preferences, and we assume that voters will rather naturally choose 
their favorite frontrunner, if they have the information, and vote 
for that one, plus any candidates they prefer to that frontrunner. 
This latter is likely to occur, in a two-party system, for only a 
small percentage of people, essentially third-party supporters. Thus 
we can expect most votes in elections where party affiliation is an 
issue to have only a small percentage of additional votes.

Bucklin Voting is alleged in documents provided by FairVote, a 
singularly biased source, to have been dropped because too few voters 
were using the additional ranks, which, in Bucklin, become 
Approval-like because if there is no majority in the first round, 
with first preference votes, there is no candidate-dropping, as with 
IRV, but additional votes are added in. But, in fact, few use of 
those votes is what would be expected under some circumstances, but, 
and this is important, those small number of voters are responsible, 
under Plurality, for the spoiler effect, and Bucklin solved that 
problem, as would basic unranked Approval.

FairVote claims that the small percentage of second rank votes that 
they found in Bucklin elections were because "voting for someone in 
second rank can hurt your first choice," but this is only true if 
both choices are real candidates, i.e., could possibly win, and this 
almost never is the case in a two-party system, as we had then and 
now (but there may have been some places where it was true back when 
Bucklin wa being used, and we have no data yet on this. Eventually, 
it will be gathered, I'm sure.) Bucklin was dropped, I am fairly sure 
from an examination of what evidence I have been able to gather so 
far, not because it wasn't working and was merely an exercise in 
futile complication, but because it was working.

Brown v. Smallwood was based on a case where the plurality winner in 
the first round was defeated by additional votes coming in from the 
second and third rounds, which turned the tide to Smallwood. Brown 
was a voter who didn't like this, and he sued. And his position was 
upheld by the Minnesota Supreme Court in a case which FairVote claims 
turns on one-man, one-vote, though that argument, was, in fact not 
the coreof the Court's reasoning, though the Court's reasoning was to 
bizarrely distorted that it's hard to say what they were actually 
thinking. It reminds me of other result-driven decisions that we have 
seen. In any case, it was very clear that the Court was rejecting the 
very idea of having alternative votes and would almost certainly have 
rejected IRV just as well as it did Bucklin. Somebody did not want to 
allow third parties a toehold, or, more likely, some political 
position required that a major party enjoy the vote-splitting of the 
opposition to gain victories in Plurality elections, which 
alternative voting eliminated. Thus the Republicans in Ann Arbor shot 
down IRV when it was implemented there, because it worked, and they 
lost the mayor's office to a Democrat, the first Black mayor in the U.S.

Bullet Voting? The right of the voter, or it should be. It does *not* 
mean "I have no preference among the remaining candidates." It could 
mean, "I detest them all and could not stand giving any of them any 
support, so I am equal-ranking them bottom."

Robert's Rules, in its alleged "recommendation" of "IRV," as has been 
claimed on the Wikipedia article on IRV, for some time, actually is 
"describing," not exactly recommending, an IRV-like method with one 
critical difference. It is not explicitly stated, but, actually, if 
one simply reads the description without having in mind that one 
already knows what this method is, the majority needed for a victory 
is based on *all* ballots cast, not just those that are not 
exhausted. From other material on Robert's Rules and the meaning of 
majority of votes cast, it is clear that no ballots are to be discarded.

And this makes the method nondeterministic, it can fail, and, in real 
public elections with a majority victory requirement, it would fail, 
if applied. There may be, indeed, some basis somewhere to challenge 
an IRV result if ballot exhaustion led to a winner who did not get an 
explicit vote from a true majority, being more than half of all 
ballots cast. Which includes exhausted ballots containing an 
otherwise valid vote.

I'm under some pretty strong criticism for insisting that the 
Robert's Rules mention be scrupulously correct, which is not 
difficult to do. But this is seriously offensive to the Fairvote 
editor who might once again start an edit war over it. It would help 
if anyone knew of a parliamentarian who could consider the specific 
question. I seem to recall already having seen the opinion, but don't 
remember where it was.

If exhausted ballots are still counted and used to determine 
majority, the meaning of them is clear: they are votes against the 
remaining candidates in favor of a losing candidate. Thus discarding 
them is clearly unjust. They should continue to be counted, and this 
makes IRV, actually, a somewhat better method. Is that better method 
"recommended" by Robert's Rules? Well, not exactly. They point out, 
actually, it's serious problems and do state what they recommend, 
which is repeated balloting until one candidate has a majority. A 
real majority. They only suggest "preference voting" as an option 
where runoffs, or, better, entirely new elections, are considered 
impractical. And they don't specify which form of preference voting 
they are recommending, they just give a common example.

All this is an example of how FairVote has managed to craft 
propaganda, true-seeming statements that only fall apart when 
examined closely. There are many other examples....

For the Instant Runoff Voting article, see
http://en.wikipedia.org/wiki/Instant_runoff_voting

http://en.wikipedia.org/wiki/Talk:Instant-runoff_voting for the 
discussion about this, there are three sections referring to Roberts' Rules.

And, by the way, I am *not* soliciting meat puppets. I hope that more 
editors who support Instant Runoff Voting will join and participate, 
as well as more who oppose it, plus, ideally, some election experts 
and parliamentarians who are neutral. My goal in working with the 
article is that it be truly NPOV, not a carefully crafted propaganda 
piece either for or against Instant Runoff Voting.