Zeev, I read an article the other day comparing recounting these ballots with quantum physics. The problem is that there is no "true count" because each time you measure them the "true count" changes as more chads fall out. Add in the question of dimpled or pregnant chads with the subjectivity that they bring and the it becomes ever more impossible to determine a "true count". My personal opinion is that with each recount the counts become less accurate, not more accurate.
If you count a dimpled chad, then what do you do with a ballot that has a clear punch but another chad is dimpled? Does that then become a double punch, and an invalid ballot? When you start trying to divine the intent of the voter instead of simply reading what the voter actually did you are asking for trouble.
Suppose you did a statewide recount and came up with a victory by Gore of 400 votes. Should you accept that count? Maybe it would be wrong and the original count right? At that point Bush would have won 3 out 4, so maybe he should be considered to be still ahead. So I think you would need to do another recount. Supposing you did 7 recounts and Bush won 4. Would that, like the World Series, be enough to make him the winner, or would he need 5 of 9? When do you stop?
And if each recount is done with different rules regarding dimpled chads and so forth, how do you agree which rules should control? The whole thing quickly becomes a joke when you use a ballot system that has this high of an error rate and isn't susceptible to repeatable counts. Clearly we need something better in the future, but what can we do about the past. Not much, it would seem to me.
Carl |
