[EM] Re: Approval is not one person one vote

[MIKE OSSIPOFF]

Of course we've thoroughly discussed one-person-one-vote here before. I used 
to abbreviate it 1p1v, but opov is easier, and so I'll call say "opov" for 
"one-person-one-vote".

Anyone can take the name of a pre-existing standard, and write a new meaning 
for it, and then use that new standard with an old name to judge methods by.

We all agree that opov means that all the voters have the same ways of 
voting available to them, and that, for any particular way of voting,  their 
votes are counted in the same way.

Approval obviously meets that standard.

Some write a new definition of opov that says that everyone gets one vote 
only. That's a rules criterion, as oppposed to a results criterion.

It's well-agreed here that a rules criterion means nothing unless it can be 
justified in terms of a results criterion, one which can ultimatly be 
justified in terms of a fundamental standard, a standard whose importance is 
accepted without being justified in terms of anything else.

Last time this issue came up, I asked the opov advocates to justify opov in 
terms of a widely accepted results criterion or a fundamental standard.

If opov can't be justified in that way, then opov an only be justified as a 
fundamental standard--if you accept it as a fundamental standard.

A rule criterion as a fundamental standard is a rather ridiculous notion.

As I said, I asked to opov advocates to justify opov in terms of accepted 
results criteria, or a fundamental standard. They never did. So I posted the 
last posting in that discussion, entitled "1p1v abandoned?".

Since opov rightfully just means that every voter should have the same ways 
of voting available to hir, counted in the same way for a particular way of 
voting, Approval meets opov. Someone also pointed out that, in Approval, any 
voter has the power to cancel out the effect of any other voter.

It seems to me that Tom Ruen and Ruillon were opov advocates then. To them, 
then, I ask: You didn't answer my question then, about how you justify opov. 
Can you justify it now, or is it just that you wait for a while, and then 
start all over, repeating your refuted claims?

As someone ;pointed out this time (and last time too), Plurality doesn't 
meet the opov-ists' opov:

When you vote, what you're actually votiing is _differences_  in support. In 
Plurality the thing about your Plurality ballot that helps Kerry over Bush 
is the fact that you're voting Kerry over Bush. If you'd given a vote to 
both, that wouldn't help Kerry agains Bush. It's that you're giving a vote 
to Kerry and not to Bush. And that's still just as true when you're only 
allowed to give a vote to one candidate.

So, as was already pointed out, in Plurality you're effectively giving Kerry 
a positive vote, and everyone else a negative vote. There are two levels at 
which you can vote someone: Marked and unmarked. You're voting everyone but 
Kerry at the unmarked level. Plurailty fails the opovists' opov.

Or are the opovists saying that you should only be able to vote one 
candidate over others? Or that, at any point in the count, even if your 
"favor" can transfer around, you must never have more than one candidate in 
your shifting favored set? A silly, unjustifiable limitation of voting 
freedom.

If voting ;power is defined so that it can vary among voters, Plurality has 
more variation in voting power than Approval does.

On the occasions when we discussed this before, we, for the purposes of 
discussion, defined voting power as amount by which the voter can improve 
hir expectation by hir ballot, or the best expectation that s/he can get for 
the benefit brought by hir ballot. I'll call that "ballot expectation".

We discussed that at great length, deriving a formula for ballot expectation 
with some different rating configurations that a voter could have. If I 
could just summarize it here, we agreed that voters with different rating 
configurations could have different voting power in Approval and also in 
Plurality. We also agreed that that voting power varies much more between 
voters in Plurality than in Approval.

Voting power is more equal in Approval than in Plurality. So voting power 
equalization doesn't justify the opovists' opov. When I say opov, I'm 
talking about the opov promoted by those who say that each person should 
have one vote, rather than merely the same ways of voting, counted in the 
same way.

I posted to the Approval mailing list a demonstration of the factor by which 
Plurality can make the voters' voting power vary more than Approval can. 
That posting would be in the first few weeks of the Approval mailing list's 
archives.

Mike Ossipoff

_________________________________________________________________
Get fast, reliable Internet access with MSN 9 Dial-up – now 3 months FREE! 
http://join.msn.click-url.com/go/onm00200361ave/direct/01/