Chris--
I mentioned two separate reasons why Plurality meets Non-Drastic Defense.
Let me quote what I got from a website:
Non-Drastic Defense:
"Each voter must be allowed to vote as many alternatives as s/he wishes tied for top, and if more than half the voters vote some alternative y (tied for) top, then no alternative voted below y by more than half of the voters may be chosen."
[end of NDD definition at the website]
I should have checked your posting before starting this posting, but it seems to me that, in your posting, you worded NDD differently from that, in such a way that Plurality passes because NDD's premise makes a requirement that can't be met in a Plurality example, meaning that there can be no Plurality failure example, because there can be no Plurality example that the criterion applies to, and because your NDD didn't have the requirement for allowing equal top ranking.
It was in regards to that that you said that a pork chop passes NDD for the same reason.
I agree that something's wrong when a method passes for that reason. My criteria are universally and uniformly applicable. Well, all of them but SARC, which I rarely if ever use, and will drop if I can't make it universally applicable without changing it too much.
You stated a few criteria of your own, and they all stipulated, in their premise, something that could only be possible if the ballot has at least 3 rank positions. That's another example of a criterion which Plurality passes for the same reason that a pork chop passes. No one can vote 3 rank positions in Plurality, and so there's no Plurality example that your criterion applies to. Therefore there's no plurality failure example for your criterion
.
Now, if you had added a requirement that the method must allow at least 3 voting levels, then Pluralilty would fail, by fiat, in the way that it fails the actual NDD quoted above. But your criterion didn't have that requirement.
So Plurality passes it for the same reason as pork chop does. I don't remember if there was another reason why Plurality would pass, even if that weren't so.
But, returning to NDD, the version that I quoted above, from the website: Plurality fails that version of NDD, which is more likely to be the official version, since it's at the website. And it fails for a similar reason: Pluralitiy fails because it doesn't allow the voter to vote as many alternatives as s/he wishes tied for top, and NDD requires that a method allow that. It's a "rules criterion", because it says that a method fails if it doesn't have a certain type of rule--in this case it's about balloting rules.
That's something that I avoid, because, as I said, I want my criteria to be universally and uniformly applicable. If some methods can pass simply because their examples can't meet the requirements in the criterion's premise, that isn't very uniform applicabililty. If some methods fail simply because their rules aren't of a certain kind explicitly required by the criterion, then that isn't very uniform applicability either.
Without that rules-requirement, Plurality would pass the criterion, because if more than half of the voters vote Y at top, then no one voted below y by most of the voters can win.
That's why the criterion needs the rules-requirement.
Ok, this posting is a mess, but I hope it clarifies my previous reply.
If there's a point to be found in all this, it's that it would be better if other criteria were universally and uniformly applicable, as the defensive strategy criteria are.
Mike Ossipoff
_________________________________________________________________
On the road to retirement? Check out MSN Life Events for advice on how to get there! http://lifeevents.msn.com/category.aspx?cid=Retirement
---- Election-methods mailing list - see http://electorama.com/em for list info
