Why There Needs to be a New Vote in Florida -- My earlier post elicited comments from one reader (on DailyKos), disagreeing with me, saying basically that one has to respect the vote total and work for electoral reform in the future. Here's my response :

I'd been having this argument with another DKer in context of the Virginia race -- our discussion started when Allen was still ahead in the count. The question was what to do if evidence pointed to the Allen campaign -- maybe Allen himself -- in efforts to supress the vote -- perhaps criminal efforts (the calls threatening voters with arrest, for example).

I argued the other side -- that all we have is the vote -- we can't engage in conjecture as to how the election might have been different without those dirty tricks. No way to know how people might have voted, had they come out to vote -- or maybe, witness Maryland, how they might have voted if they weren't duped by phony endorsement flyers. Presumably hardly anyone fell for those flyers, but hypothetically they could have.

In those cases, I made the point that we would be stuck with the actual vote. Because there is no reason to believe those votes were cast as recorded -- deliberately and knowingly (even if ignorantly) -- and no way to know how some hypothetical voter might have voted -- the vote needs to be respected. At least, the vote should be certified. If the wrongdoing were truly egregious, the elected person could be driven from office -- by expulsion or impeachment. But, that would be different than invalidating the vote.

The situation in Florida calls for a different response. The vote totals are not accurate. It's ridiculous to suggest they are. It would be a terrible precedent to knowingly certify an inaccurate total. In this respect, it's in some ways worse than 2000. The question in 2000 was could you look at ballots and find other votes that might change the result? Here there is no way to do that -- no way to come up with an accurate result. Given the closeness of the race, and the probability (basically 50-50, because the race is so close) that the wrong person is leading, the only logical response is to invalidate the result.

To do otherwise would be to say it's OK to rig an election. I'm not saying THIS election was rigged. I doubt the undercount is due to deliberate malicious code. But, if you certify this vote, you must certify them all.

Will the voter base be different in a second vote? Of course. If nothing else, it will be an expression of people interested in the one race -- not people who came out to vote for governor or senator, and then voted on down the line. That isn't the worst thing -- but that wouldn't be the point of a new election. The idea would be to produce a result that one could be confident reflected the will of all those who turned out to cast their ballots, on that day. Can you ask for more out of an election?